perm filename W81.IN[LET,JMC]1 blob
sn#577221 filedate 1981-04-01 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00678 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00067 00002 ∂01-Jan-81 1254 Mark Crispin <Admin.MRC at SU-SCORE> Re: troubles with dtn
C00069 00003 ∂01-Jan-81 1630 TOB
C00070 00004 ∂01-Jan-81 1708 CLT
C00071 00005 ∂01-Jan-81 1906 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V2 #179
C00084 00006 ∂01-Jan-81 1927 POURNE at MIT-MC (Jerry E. Pournelle) Helping the Boss
C00101 00007 ∂01-Jan-81 1927 Jerry E. Pournelle <POURNE at MIT-MC> Helping the Boss
C00114 00008 ∂01-Jan-81 1927 POURNE@MIT-MC secure systems
C00131 00009 ∂01-Jan-81 1927 Jerry E. Pournelle <POURNE at MIT-MC> Helping the Boss
C00133 00010 ∂01-Jan-81 1932 POURNE at MIT-MC (Jerry E. Pournelle) Let's try to be fair
C00140 00011 ∂01-Jan-81 1936 Henry at MIT-AI (Henry Lieberman) Your Lisp Book
C00142 00012 ∂02-Jan-81 0925 BYY maclisp manual
C00153 00013 ∂02-Jan-81 1032 Baskett at PARC-MAXC Monitors
C00156 00014 ∂02-Jan-81 1224 HENRY at MIT-AI (Henry Lieberman) Your book
C00158 00015 ∂03-Jan-81 0001 POURNE at MIT-MC (Jerry E. Pournelle) meetings and advice
C00166 00016 ∂03-Jan-81 1718 RPG MacLisp
C00167 00017 ∂04-Jan-81 0900 JMC*
C00168 00018 ∂04-Jan-81 1536 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) title of Colloq.
C00170 00019 ∂05-Jan-81 0757 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Faculty Meeting Reminder
C00171 00020 ∂05-Jan-81 0900 JMC*
C00172 00021 ∂05-Jan-81 0906 CLT
C00173 00022 ∂05-Jan-81 0931 FFL
C00174 00023 ∂05-Jan-81 0955 FFL
C00175 00024 ∂05-Jan-81 0958 Buchanan@SUMEX-AIM
C00177 00025 ∂05-Jan-81 1200 JMC*
C00178 00026 ∂05-Jan-81 1311 bledsoe at UTEXAS-11 atp Prize Comm Mtg
C00179 00027 ∂05-Jan-81 1412 FFL
C00180 00028 ∂05-Jan-81 1641 Konolige at SRI-KL (Kurt Konolige) Re: TA possibility
C00182 00029 ∂05-Jan-81 1705 Konolige at SRI-KL (Kurt Konolige)
C00183 00030 ∂05-Jan-81 1830 Kanerva@SUMEX-AIM Now that mail moves again . . .
C00185 00031 ∂05-Jan-81 2048 BYY NYT
C00187 00032 ∂06-Jan-81 0800 Denny Brown <CSD.DBROWN at SU-SCORE>
C00188 00033 ∂06-Jan-81 0900 JMC*
C00189 00034 ∂06-Jan-81 0900 JMC*
C00190 00035 ∂06-Jan-81 0900 JMC*
C00191 00036 ∂06-Jan-81 0920 Buchanan@SUMEX-AIM (Response to message)
C00192 00037 ∂06-Jan-81 1106 FFL
C00193 00038 ∂06-Jan-81 1128 TOB
C00195 00039
C00199 00040
C00200 00041
C00201 00042 ∂06-Jan-81 1155 TOB appointment forms
C00202 00043 ∂06-Jan-81 1159 CSD.SCHREIBER at SU-SCORE Committee meeting
C00204 00044 ∂06-Jan-81 1942 JMC*
C00205 00045 ∂06-Jan-81 2000 JMC*
C00206 00046 ∂07-Jan-81 0847 FFL
C00207 00047 ∂07-Jan-81 0900 JMC*
C00208 00048 ∂07-Jan-81 0900 JMC*
C00209 00049 ∂07-Jan-81 0951 FFL
C00210 00050 ∂07-Jan-81 0959 Nilsson at SRI-KL CBCL
C00211 00051 ∂07-Jan-81 1241 Cmiller@SUMEX-AIM MACLISP MANUALS NEEDED
C00212 00052 ∂07-Jan-81 2022 Marimont at SRI-KL EE PhD Qualifying Exam
C00215 00053 ∂08-Jan-81 0005 Boyer at SRI-F2 (Bob Boyer) Theorem-Proving Prize Meeting
C00216 00054 ∂08-Jan-81 0812 Boyer at SRI-F2 (Bob Boyer) Henry's
C00224 00055 ∂08-Jan-81 0857 FFL
C00226 00056 ∂08-Jan-81 0910 FFL
C00228 00057 ∂08-Jan-81 0915 FFL
C00229 00058 ∂08-Jan-81 1250 Denny Brown <CSD.DBROWN at SU-SCORE> Instructor need for CS105 - NOW!!
C00231 00059 ∂08-Jan-81 2349 POURNE at MIT-MC (Jerry E. Pournelle)
C00232 00060 ∂08-Jan-81 2343 KLC comp committee
C00233 00061 ∂09-Jan-81 0911 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Colloq.
C00251 00062 ∂09-Jan-81 1456 ALR at MIT-AI (Andrew L. Ressler) Selector Notation
C00254 00063 ∂09-Jan-81 1607 FFL
C00255 00064 ∂09-Jan-81 1633 FWH PV+A Seminar
C00262 00065 ∂10-Jan-81 0110 NEUMANN at SRI-KL [Neumann: VERkshop II]
C00268 00066 ∂10-Jan-81 0148 NEUMANN at SRI-KL
C00279 00067 ∂10-Jan-81 1541 Jim McGrath <JPM at SU-AI> Local utilities
C00284 00068 ∂10-Jan-81 1917 Boyer at SRI-F2 (Bob Boyer)
C00301 00069 ∂11-Jan-81 1252 Richard Jay Solomon <Solomon at MIT-Multics> Your mail of 11 Jan 1981 1138-PST
C00303 00070 ∂11-Jan-81 1410 Richard Jay Solomon <Solomon at MIT-Multics> Re: earthquake
C00311 00071 ∂11-Jan-81 1902 Sgt. Sally <CSL.LAB.SALLY> Memo to the Faculty about the Forum Annual Meeting
C00320 00072 ∂12-Jan-81 0900 JMC*
C00321 00073 ∂12-Jan-81 1310 JPM Guest account for Chuck Hedrick
C00324 00074 ∂12-Jan-81 1443 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Honoraria for Colloquia speakers
C00325 00075 ∂12-Jan-81 1519 CSD.ULLMAN at SU-SCORE Binford papers
C00326 00076 ∂12-Jan-81 1556 JPM Chuck Hedrick
C00327 00077 ∂12-Jan-81 2021 JMC*
C00328 00078 ∂12-Jan-81 2021 JMC*
C00344 00079 ∂13-Jan-81 0831 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Re: speaker
C00345 00080 ∂13-Jan-81 1000 JMC*
C00346 00081 ∂13-Jan-81 1231 CSD.LENAT at SU-SCORE COMPUTER FORUM REQUEST
C00348 00082 ∂13-Jan-81 1236 CSD.LENAT at SU-SCORE Computer Forum
C00350 00083 ∂13-Jan-81 1247 CG
C00351 00084 ∂13-Jan-81 1715 FWH PV+A Seminar
C00355 00085 ∂14-Jan-81 0706 Boyer at SRI-F2 (Bob Boyer) Course
C00357 00086 See letter from them in my OUT box.
C00359 00087 ∂14-Jan-81 2000 JMC*
C00360 00088 ∂14-Jan-81 2100 JMC*
C00361 00089 ∂14-Jan-81 2212 CLT
C00362 00090 ∂14-Jan-81 2227 TOB
C00363 00091 ∂14-Jan-81 2244 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
C00365 00092 ∂14-Jan-81 2249 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
C00368 00093 ∂15-Jan-81 0140 REM at MIT-MC (Robert Elton Maas)
C00369 00094 ∂15-Jan-81 0344 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #10
C00378 00095 ∂15-Jan-81 0401 OTA SPACE Digest
C00383 00096 ∂15-Jan-81 0700 TOB
C00384 00097 ∂15-Jan-81 0836 USCHOLD at RUTGERS How to get more water
C00387 00098 ∂15-Jan-81 0847 JGA at MIT-MC (John G. Aspinall) How to get more water?
C00389 00099 ∂15-Jan-81 0900 CSD.JEANIE at SU-SCORE Agenda for Faculty Meeting
C00390 00100 ∂15-Jan-81 0900 JMC*
C00391 00101 ∂15-Jan-81 0904 CSD.ULLMAN at SU-SCORE Re: visitors
C00392 00102 ∂15-Jan-81 0914 Richard Treitel <CSL.VER.RJT at SU-SCORE> Re: my opinion
C00393 00103 ∂15-Jan-81 1147 NEUMANN at SRI-KL VERkshop II progress
C00396 00104 ∂15-Jan-81 1243 DBL Session at the COmputer Forum
C00397 00105 ∂15-Jan-81 1307 SL Computer Forum Speaker/Title
C00402 00106 ∂15-Jan-81 2038 Denny Brown <CSD.DBROWN at SU-SCORE> Gray Tuesday/Course Homework
C00404 00107 ∂15-Jan-81 2043 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
C00406 00108 ∂16-Jan-81 0251 POURNE@MIT-MC conference
C00407 00109 ∂16-Jan-81 0401 OTA SPACE Digest
C00410 00110 ∂16-Jan-81 0427 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #11
C00433 00111 ∂16-Jan-81 0929 FFL
C00434 00112 ∂16-Jan-81 1023 PAM
C00439 00113 ∂16-Jan-81 1120 Konolige at SRI-KL (Kurt Konolige)
C00442 00114 ∂16-Jan-81 1305 RPG Circumscription
C00443 00115 ∂16-Jan-81 1516 Susan Hill <CSD.HILL at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
C00445 00116 ∂16-Jan-81 1545 Susan Hill <CSD.HILL at SU-SCORE>
C00446 00117 ∂16-Jan-81 1600 JMC*
C00447 00118 ∂16-Jan-81 2318 JMC*
C00448 00119 ∂16-Jan-81 2349 Feigenbaum@SUMEX-AIM Re: friedl and disk
C00452 00120 ∂18-Jan-81 2002 TOB
C00454 00121 ∂18-Jan-81 2027 TOB cv
C00457 00122 ∂18-Jan-81 2351 Marimont at SRI-KL EE quals
C00459 00123 ∂19-Jan-81 0401 OTA SPACE Digest
C00473 00124 ∂19-Jan-81 0517 REM at MIT-MC (Robert Elton Maas)
C00474 00125 ∂19-Jan-81 0847 CSD.JEANIE at SU-SCORE ARPA Meeting Fri. 2/13/81
C00481 00126 ∂19-Jan-81 0900 JMC*
C00482 00127 ∂19-Jan-81 0900 JMC*
C00483 00128 ∂19-Jan-81 0900 KRAUSS at MIT-MC (Jeffrey Krauss) Public Key methods
C00485 00129 ∂19-Jan-81 0933 Shortliffe@SUMEX-AIM Medical AI Organization
C00490 00130 ∂19-Jan-81 1010 Konolige at SRI-KL (Kurt Konolige) problems for CS226
C00492 00131 ∂19-Jan-81 1343 CSD.JEANIE at SU-SCORE Visitor from Florida
C00493 00132 ∂19-Jan-81 2039 TOB
C00498 00133 ∂20-Jan-81 0027 LWE ns
C00501 00134 ∂20-Jan-81 0103 Schauble.Multics at MIT-Multics Clipping Service - Solar power satellite-power conversion
C00504 00135 ∂20-Jan-81 0125 POURNE@MIT-MC
C00507 00136 ∂20-Jan-81 0128 POURNE@MIT-MC
C00521 00137 ∂20-Jan-81 0940 CSD.JEANIE at SU-SCORE Jan. 27 Meeting
C00522 00138 ∂20-Jan-81 1056 CSD.JEANIE at SU-SCORE More on Jan. 27 Meeting
C00523 00139 ∂21-Jan-81 0158 POURNE at MIT-MC (Jerry E. Pournelle)
C00524 00140 ∂21-Jan-81 0158 POURNE at MIT-MC (Jerry E. Pournelle)
C00525 00141 ∂21-Jan-81 0900 JMC*
C00542 00142 ∂21-Jan-81 1109 KLC comp
C00543 00143 ∂21-Jan-81 1206 Waldinger at SRI-KL he
C00544 00144 ∂21-Jan-81 2118 SQU This is just a gentle reminder...
C00546 00145 ∂21-Jan-81 2306 LWE inquiry/ns
C00547 00146 ∂22-Jan-81 0038 POURNE at MIT-MC (Jerry E. Pournelle)
C00548 00147 ∂22-Jan-81 0302 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #17
C00571 00148 ∂22-Jan-81 1126 CSL.SSO.OWICKI at SU-SCORE Colloquium talk
C00572 00149 ∂22-Jan-81 1243 TOB NSF site visit
C00574 00150 ∂22-Jan-81 1732 DHM
C00575 00151 ∂22-Jan-81 2145 TOB
C00576 00152 ∂23-Jan-81 0002 LWE ns file
C00577 00153 ∂23-Jan-81 0249 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #18
C00599 00154 ∂23-Jan-81 0421 OTA SPACE Digest
C00603 00155 ∂23-Jan-81 0641 JRA computer science summer institute
C00604 00156 ∂23-Jan-81 1218 CSD.NOWICKI at SU-SCORE News from Usenix
C00609 00157 ∂23-Jan-81 1319 RPG Letter
C00612 00158 ∂23-Jan-81 1400 JMC*
C00613 00159 ∂23-Jan-81 1418 RPG
C00615 00160 ∂23-Jan-81 1627 Nilsson at SRI-KL CBCL
C00618 00161 ∂23-Jan-81 1700 TW
C00619 00162 ∂23-Jan-81 2122 SGR at MIT-MC (Stephen G. Rowley) Bureaucrats & natural gas in Massachusetts.
C00621 00163 ∂23-Jan-81 2345 TOB
C00630 00164 ∂25-Jan-81 0036 POURNE@MIT-MC
C00631 00165 ∂25-Jan-81 0039 POURNE@MIT-MC lots of people don't read sf lovers
C00633 00166 ∂25-Jan-81 0648 Darden@SUMEX-AIM hunger
C00638 00167 ∂25-Jan-81 1030 Feigenbaum@SUMEX-AIM more on ARPA consolidation
C00657 00168 ∂25-Jan-81 1351 DCL
C00662 00169 ∂25-Jan-81 1959 JK ekl
C00663 00170 ∂25-Jan-81 2133 Allen.Newell at CMU-10A Re: The knowledge level
C00668 00171 ∂26-Jan-81 0012 LLW Commuting to Pournelle's Meeting
C00670 00172 ∂26-Jan-81 0028 LLW Friday Dawn Rendevous
C00672 00173 ∂26-Jan-81 0713 JRA summer dates
C00681 00174 ∂26-Jan-81 0828 cracraft@sri-unix SPS article and info sources
C00683 00175 ∂26-Jan-81 0833 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Consolidation
C00686 00176 ∂26-Jan-81 0902 Wiederhold@SUMEX-AIM Re: ARPA Consolidation
C00687 00177 ∂26-Jan-81 0944 Buchanan@SUMEX-AIM Re: more on ARPA consolidation
C00689 00178 ∂26-Jan-81 1145 JMC*
C00690 00179 ∂26-Jan-81 1145 JMC*
C00691 00180 ∂26-Jan-81 1145 JMC*
C00692 00181 ∂26-Jan-81 1145 JMC*
C00693 00182 ∂26-Jan-81 1148 Baskett at PARC-MAXC Re: more on ARPA consolidation
C00696 00183 ∂26-Jan-81 1220 TOB Bloom
C00697 00184 ∂26-Jan-81 1254 CSD.ULLMAN at SU-SCORE arpa consolidation
C00699 00185 ∂26-Jan-81 1310 Feigenbaum@SUMEX-AIM Re: ARPA Consolidation
C00701 00186 ∂26-Jan-81 1316 FWH talk in verification seminar
C00703 00187 ∂26-Jan-81 1427 CSD.JEANIE at SU-SCORE ARPA meeting
C00705 00188 ∂26-Jan-81 1409 Feigenbaum@SUMEX-AIM (Response to message)
C00706 00189 ∂26-Jan-81 1459 CG trip to Sweden
C00707 00190 ∂26-Jan-81 1740 Baskett at PARC-MAXC Re: arpa consolidation
C00709 00191 ∂26-Jan-81 2020 LLW Early AM Commute Times
C00710 00192 ∂26-Jan-81 2132 Brian K. Reid <CSL.BKR at SU-SCORE> Arpa
C00716 00193 ∂27-Jan-81 0853 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Prof. Paz from Israel
C00717 00194 ∂27-Jan-81 0904 CSD.SCHREIBER at SU-SCORE Comprehensive Exam grading session.
C00719 00195 ∂27-Jan-81 0944 DEW via SRI-KL answers to exercises
C00720 00196 ∂27-Jan-81 1207 CSD.JEANIE at SU-SCORE Thursday meeting
C00721 00197 ∂27-Jan-81 1258 Feigenbaum@SUMEX-AIM Brian's message
C00723 00198 ∂27-Jan-81 1424 Edward Feigenbaum <CSD.FEIGENBAUM at SU-SCORE> visit of Gordon Bell on Wed.Jan 28
C00725 00199 ∂27-Jan-81 1824 GREEP at RAND-AI Login on SAIL for CS226 work
C00726 00200 ∂27-Jan-81 2039 Mark Crispin <Admin.MRC at SU-SCORE>
C00729 00201 ∂27-Jan-81 2136 Mark Crispin <Admin.MRC at SU-SCORE> DTN bugs and unprofessionalism
C00733 00202 ∂28-Jan-81 0139 POURNE at MIT-MC (Jerry E. Pournelle) rain on Mongo...
C00735 00203 ∂28-Jan-81 1124 Buchanan@SUMEX-AIM Datamedia
C00738 00204 ∂28-Jan-81 1127 Denny Brown <CSD.DBROWN at SU-SCORE> Stanford football recruiting
C00740 00205 ∂28-Jan-81 1324 Waldinger at SRI-KL temporal logic
C00743 00206 ∂28-Jan-81 1739 Konolige at SRI-KL exercise #2
C00744 00207 ∂29-Jan-81 0900 JMC*
C00748 00208 ∂29-Jan-81 1108 CSD.JEANIE at SU-SCORE
C00749 00209 ∂29-Jan-81 1146 HITCHCOCK at CCA (Chip Hitchcock) Re: natural gas shortage
C00751 00210 ∂29-Jan-81 1523 FWH PV+A Seminar
C00753 00211 ∂29-Jan-81 1810 ES at MIT-MC (Gene Salamin) ''Corporate Greed'' or ''Socialism''?
C00755 00212 ∂29-Jan-81 1937 LLW Tomorrow's Commute
C00756 00213 ∂29-Jan-81 2026 ES at MIT-MC (Gene Salamin) Abundance from the ocean.
C00758 00214 ∂29-Jan-81 2339 MMD maxtex
C00759 00215 ∂29-Jan-81 2340 POURNE at MIT-MC (Jerry E. Pournelle) Abundance from the ocean.
C00761 00216 ∂30-Jan-81 0004 POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
C00762 00217 ∂30-Jan-81 0143 LWE My expanding NS files...
C00764 00218 ∂30-Jan-81 0401 OTA SPACE Digest
C00765 00219 ∂30-Jan-81 0427 JRA finances
C00766 00220 ∂30-Jan-81 0454 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #24
C00777 00221 ∂30-Jan-81 1131 OAF at MIT-MC (Oded Anoaf Feingold) Bostongas and other local quibbles, like, uh, conservation
C00781 00222 ∂30-Jan-81 1252 Denny Brown <CSD.DBROWN at SU-SCORE> Undergrad advising
C00782 00223 ∂30-Jan-81 1341 CSD.NOWICKI at SU-SCORE Ether Tip software
C00786 00224 ∂30-Jan-81 1451 Vaughan Pratt <CSD.PRATT at SU-SCORE> Meeting Monday
C00787 00225 ∂30-Jan-81 1531 VRP via Ethernet NVT @sun
C00788 00226 ∂30-Jan-81 1548 Chiron of Thessaly <FEINBERG at CMU-20C> Pro Nuke Protest
C00789 00227 ∂30-Jan-81 1614 GFS ETHER-TIP
C00790 00228 ∂30-Jan-81 1739 BYY teach sol
C00791 00229 ∂30-Jan-81 2308 Hans Moravec <HPM at SU-AI> Petr Beckmann
C00792 00230 ∂30-Jan-81 2316 MINSKY@MIT-AI
C00794 00231 ∂31-Jan-81 0401 OTA SPACE Digest
C00795 00232 ∂31-Jan-81 0425 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #25
C00808 00233 ∂31-Jan-81 0653 DREIFU at WHARTON-10 (Henry Dreifus) Water situation.
C00809 00234 ∂31-Jan-81 1021 DAM@MIT-AI
C00810 00235 ∂31-Jan-81 1106 BYY on TTY1 1106
C00811 00236 ∂31-Jan-81 1149 MVL at MIT-AI (Michael V. Lease) Boston Gas (advertising for more customers)
C00813 00237 ∂31-Jan-81 1340 Purger
C00814 00238 ∂31-Jan-81 2229 MINSKY@MIT-AI
C00818 00239 ∂01-Feb-81 2146 JK ekl
C00836 00240 ∂01-Feb-81 2351 AVB
C00844 00241 ∂02-Feb-81 1028 CSD.SCHREIBER at SU-SCORE All over.
C00845 00242 ∂02-Feb-81 1114 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Gray Tuesday
C00846 00243 ∂02-Feb-81 1657 FFL
C00855 00244 ∂02-Feb-81 1743 CLT
C00858 00245 ∂02-Feb-81 2134 CSD.LENAT at SU-SCORE Forum
C00862 00246 ∂
C00870 00247 ∂03-Feb-81 0947 FFL
C00881 00248 ∂03-Feb-81 0950 FFL
C00882 00249 ∂03-Feb-81 1039 KGK via SRI-F2 class
C00885 00250 ∂03-Feb-81 1211 CSD.JEANIE at SU-SCORE ARPA Consolidation
C00891 00251 ∂03-Feb-81 1302 DBL forum time: Thu 9:15 -10:30
C00892 00252 ∂03-Feb-81 1339 FFL
C00898 00253 ∂03-Feb-81 1611 FFL
C00913 00254 ∂03-Feb-81 2311 TOB dinner
C00914 00255 ∂04-Feb-81 0000 JMC*
C00919 00256 ∂04-Feb-81 1114 FFL
C00932 00257 ∂04-Feb-81 1606 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Ray Reiter
C00936 00258 ∂04-Feb-81 1652 ARK tex problems
C00937 00259 ∂04-Feb-81 2018 Konolige at SRI-KL (Kurt Konolige) seminar room
C00949 00260 ∂04-Feb-81 2154 LGC Opinion Request
C00954 00261 ∂05-Feb-81 0108 POURNE at MIT-MC (Jerry E. Pournelle) more stuff
C00959 00262 ∂05-Feb-81 0402 OTA SPACE Digest
C00994 00263 ∂05-Feb-81 1300 JMC*
C00995 00264 ∂05-Feb-81 1406 FFL
C00996 00265 ∂05-Feb-81 1524 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
C00997 00266 ∂05-Feb-81 1601 TOB computing
C00998 00267 ∂05-Feb-81 1719 RWF
C00999 00268 ∂05-Feb-81 2248 RWW hi
C01000 00269 ∂06-Feb-81 0401 OTA SPACE Digest
C01012 00270 ∂06-Feb-81 0408 JRA cs institute
C01013 00271 ∂06-Feb-81 0504 Neumann at SRI-KL VERkshop II
C01014 00272 ∂06-Feb-81 0913 CSD.SCHREIBER at SU-SCORE Change of grade
C01016 00273 ∂06-Feb-81 0939 JRA it's a deal
C01017 00274 ∂06-Feb-81 1005 JRA
C01019 00275 ∂06-Feb-81 1157 FFL
C01020 00276 ∂06-Feb-81 1207 FFL
C01021 00277 ∂06-Feb-81 1529 FWH PV+A Seminar
C01025 00278 ∂07-Feb-81 0357 ROD Orals.
C01039 00279 ∂07-Feb-81 1032 CSD.LENAT at SU-SCORE Re: Orals.
C01040 00280 ∂07-Feb-81 1753 ARK LETTER.TEX[TEX,ARK]
C01053 00281 ∂08-Feb-81 2332 LLW Advanced Space Transportation
C01076 00282 ∂09-Feb-81 0902 JMC*
C01077 00283 ∂09-Feb-81 0916 Shortliffe@SUMEX-AIM Medical group within AAAI
C01080 00284 ∂09-Feb-81 1332 AVB
C01081 00285 ∂09-Feb-81 1347 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Lunch with Drs. Kobayashi and Cadiou
C01082 00286 ∂09-Feb-81 1418 FFL
C01083 00287 ∂09-Feb-81 1419 FFL
C01084 00288 ∂09-Feb-81 1516 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
C01085 00289 ∂09-Feb-81 1556 RWW
C01086 00290 ∂09-Feb-81 1617 ROD Orals
C01087 00291 ∂09-Feb-81 1626 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
C01091 00292 ∂10-Feb-81 0223 POURNE at MIT-DMS (Jerry E. Pournelle) that tower
C01093 00293 ∂10-Feb-81 0239 POURNE at MIT-DMS (Jerry E. Pournelle) llw's paper and some curses
C01094 00294 ∂10-Feb-81 0245 Jerry E. Pournelle <POURNE at MIT-MC>
C01102 00295 ∂10-Feb-81 0336 LLW Realpolitic Triumphant!
C01109 00296 ∂10-Feb-81 0855 MERRITT at USC-ISIB Re: campaign
C01111 00297 ∂10-Feb-81 0850 Denny Brown <CSD.DBROWN at SU-SCORE> Gray Tuesday reminder
C01112 00298 ∂10-Feb-81 0900 JMC*
C01113 00299 ∂10-Feb-81 1008 Darden@SUMEX-AIM DINNER ON MONDAY, FEBRUARY 16
C01117 00300 ∂10-Feb-81 1141 Denny Brown <CSD.DBROWN at SU-SCORE> PhD Requirements
C01118 00301 ∂10-Feb-81 1144 FFL
C01131 00302 ∂10-Feb-81 1616 John McCarthy <JMC at SU-AI>
C01137 00303 ∂10-Feb-81 1648 Robert Elton Maas <REM at MIT-MC> Re: campaign [collection of our remarks to President]
C01139 00304 ∂10-Feb-81 1658 ME E macros/variables
C01140 00305 ∂10-Feb-81 2055 Robert Elton Maas <REM at MIT-MC>
C01148 00306 ∂10-Feb-81 2222 AYRES at USC-ISIF Teloperators by Mike Hyson
C01181 00307 My opinion is that tele-operation on the scale envisaged in the
C01183 00308 ∂11-Feb-81 0100 POURNE at MIT-MC (Jerry E. Pournelle) NASA payoff
C01190 00309 ∂11-Feb-81 0127 Jerry E. Pournelle <POURNE at MIT-MC>
C01191 00310 ∂11-Feb-81 0309 REM via SU-TIP
C01192 00311 ∂11-Feb-81 0730 Darden@SUMEX-AIM PLANS
C01195 00312 ∂11-Feb-81 1112 FFL
C01202 00313 ∂11-Feb-81 1305 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Commencement
C01204 00314 ∂11-Feb-81 1323 david at UTEXAS-11 Political messages
C01205 00315 ∂11-Feb-81 1339 MERRITT at USC-ISIB Re: Political messages
C01206 00316 ∂11-Feb-81 1415 FFL
C01207 00317 ∂11-Feb-81 1416 FFL
C01208 00318 ∂11-Feb-81 1600 Janofsky.Tipi at RADC-Multics Re: Political messages
C01210 00319 ∂11-Feb-81 1621 LEWIS at SRI-KL (Bil Lewis) Re: Political messages
C01212 00320 ∂11-Feb-81 1635 KLH@MIT-AI Use of ARPAnet, and letter addresses
C01215 00321 ∂11-Feb-81 1536 DHM
C01224 00322 ∂11-Feb-81 2000 JMC*
C01226 00323 ∂11-Feb-81 2011 MINSKY at MIT-AI (Marvin Minsky)
C01228 00324 ∂11-Feb-81 2207 CSD.ULLMAN at SU-SCORE rumble at chez Lieberman
C01231 00325 ∂11-Feb-81 2356 RMS at MIT-AI (Richard M. Stallman)
C01232 00326 ∂12-Feb-81 0026 Jerry E. Pournelle <POURNE at MIT-MC>
C01236 00327 ∂12-Feb-81 0044 POURNE at MIT-MC (Jerry E. Pournelle)
C01239 00328 ∂12-Feb-81 0115 Jerry E. Pournelle <POURNE at MIT-MC> Re: NASA payoff
C01242 00329 ∂12-Feb-81 0230 Jerry E. Pournelle <POURNE at MIT-MC>
C01245 00330 ∂12-Feb-81 1312 FFL
C01246 00331 ∂12-Feb-81 1518 FFL
C01247 00332 ∂12-Feb-81 1658 FFL
C01248 00333 ∂12-Feb-81 2103 Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE> meeting re: comprehensive
C01254 00334 ∂12-Feb-81 2106 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Policy on Report Distribution to Foreign Countries
C01257 00335 ∂12-Feb-81 1453 RPG letter
C01258 00336 ∂12-Feb-81 2145 Purger
C01260 00337 ∂12-Feb-81 2130 BYY AN ABSTRACT TO ACL THAT MIGHT AMUSE YOU
C01261 00338 ∂13-Feb-81 0530 Darden@SUMEX-AIM PLANS
C01268 00339 ∂13-Feb-81 0731 DISRAEL at BBND Paper request
C01269 00340 ∂13-Feb-81 1056 FFL
C01270 00341 ∂13-Feb-81 1149 100 : frances larson
C01271 00342 ∂13-Feb-81 1259 CSD.JEANIE at SU-SCORE Faculty Meeting
C01272 00343 ∂13-Feb-81 1741 Waldinger at SRI-KL huh?
C01274 00344 ∂14-Feb-81 0250 POURNE at MIT-MC (Jerry E. Pournelle) Comments on space strategy
C01275 00345 ∂14-Feb-81 0856 BYY ACL abstract
C01276 00346 ∂14-Feb-81 1304 RWW
C01277 00347 ∂14-Feb-81 2007 JMC
C01288 00348 ∂15-Feb-81 0545 JMC
C01290 00349 ∂15-Feb-81 2351 Jerry E. Pournelle <POURNE at MIT-MC> Possible bad news
C01294 00350 This may interest you, Jerry or your student. - John
C01295 00351 ∂16-Feb-81 1324 NEUMANN at SRI-KL VERkshop II
C01307 00352 ∂17-Feb-81 1023 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Barbara Huberman
C01309 00353 ∂17-Feb-81 1116 CSD.JEANIE at SU-SCORE Faculty Meeting
C01310 00354 ∂17-Feb-81 1140 TOB task statement for NSF
C01311 00355 ∂17-Feb-81 1627 CSD.ULLMAN at SU-SCORE Lieberman meeting
C01312 00356 ∂17-Feb-81 2008 VRP Wed meeting
C01313 00357 ∂18-Feb-81 0904 FFL
C01314 00358 ∂18-Feb-81 0920 REM via SU-TIP Crunch on microprocessor
C01316 00359 ∂18-Feb-81 1311 FFL
C01320 00360 ∂18-Feb-81 2145 TOB
C01321 00361 ∂18-Feb-81 2311 TOB GM
C01322 00362 ∂19-Feb-81 0003 Brian K. Reid <CSL.BKR at SU-SCORE>
C01323 00363 ∂19-Feb-81 0416 REM via SU-TIP Update on new crunch program
C01325 00364 ∂19-Feb-81 0926 FFL
C01326 00365 ∂19-Feb-81 1230 JMC*
C01327 00366 ∂19-Feb-81 1300 JMC*
C01328 00367 ∂19-Feb-81 1328 REM via SU-TIP LOTS town meeting, expanding LOTS.
C01330 00368 ∂19-Feb-81 1735 Raphael at SRI-KL Address Change
C01331 00369 ∂19-Feb-81 1756 TOB adjunct
C01333 00370 ∂20-Feb-81 2112 REM at MIT-MC (Robert Elton Maas)
C01337 00371 ∂20-Feb-81 2117 CSL.VER.HLO at SU-SCORE Programming Exam
C01343 00372 ∂21-Feb-81 0027 POURNE at MIT-MC (Jerry E. Pournelle)
C01344 00373 ∂21-Feb-81 0033 POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
C01345 00374 ∂21-Feb-81 1014 REM via SU-TIP
C01361 00375 ∂21-Feb-81 1446 CLT
C01364 00376 ∂21-Feb-81 1840 BYY Help
C01365 00377 ∂22-Feb-81 0628 REM via SU-TIP Best possible compression using adaptive model
C01387 00378 ∂22-Feb-81 1223 JK fred abramson
C01388 00379 ∂22-Feb-81 2334 TVR DTN troubles
C01389 00380 ∂23-Feb-81 0348 Mark Crispin <Admin.MRC at SU-SCORE> Re: DTN troubles
C01390 00381 ∂23-Feb-81 0839 FFL
C01391 00382 ∂23-Feb-81 1004 FFL
C01392 00383 ∂23-Feb-81 1051 FWH PV+A Seminar
C01394 00384 ∂23-Feb-81 1107 FFL
C01395 00385 ∂23-Feb-81 1106 FFL
C01396 00386 ∂23-Feb-81 1104 FFL
C01398 00387 ∂23-Feb-81 1143 JMC*
C01399 00388 ∂23-Feb-81 1143 JMC*
C01400 00389 ∂23-Feb-81 1314 CSD.JEANIE at SU-SCORE Meeting 2/24
C01401 00390 ∂23-Feb-81 1408 RWW misc
C01402 00391 ∂23-Feb-81 1535 PP talk by Peter Pepper at SRI
C01405 00392 ∂24-Feb-81 0103 LLW Caught, White-Hatted!
C01408 00393 ∂24-Feb-81 0640 Darden@SUMEX-AIM bread and butter
C01410 00394 ∂24-Feb-81 0658 BYY Moving
C01411 00395 ∂24-Feb-81 1300 JMC*
C01412 00396 ∂24-Feb-81 1505 Hans Moravec at CMU-10A (R110HM60) Nuclear pumped X ray zapsat
C01416 00397 ∂24-Feb-81 1725 Susan Hill <CSD.HILL at SU-SCORE> [ Solomon.Datanet at MIT-Multics: Re: Computer Facilities Charges]
C01419 00398 ∂24-Feb-81 2318 FWH PV+A Seminar
C01421 00399 ∂24-Feb-81 2358 Jerry E. Pournelle <POURNE at MIT-MC> Nuclear pumped X ray zapsat
C01425 00400 ∂25-Feb-81 1343 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Betty Scott
C01429 00401 ∂25-Feb-81 1426 Susan Hill <CSD.HILL at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
C01431 00402 ∂25-Feb-81 1447 TW
C01435 00403 ∂25-Feb-81 1452 FFL
C01436 00404 ∂25-Feb-81 1636 ROY on TTY3 1636
C01437 00405 ∂25-Feb-81 1802 RWW circumscription
C01451 00406 ∂26-Feb-81 0814 Feigenbaum@SUMEX-AIM (Response to message)
C01452 00407 ∂26-Feb-81 0919 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) CS226
C01455 00408 ∂26-Feb-81 1051 FFL
C01456 00409 ∂26-Feb-81 1553 Purger
C01457 00410 ∂26-Feb-81 1608 TW
C01465 00411 ∂27-Feb-81 0021 Jerry E. Pournelle <POURNE at MIT-MC> Caught, White-Hatted!
C01466 00412 ∂27-Feb-81 0034 LLW White-Hatted Money Rustlers
C01468 00413 ∂27-Feb-81 1105 FFL
C01474 00414 ∂27-Feb-81 1212 RPG Appointment
C01477 00415 ∂27-Feb-81 1225 RPG Proposal
C01480 00416 ∂27-Feb-81 1540 CSD.AUBERY at SU-SCORE cs226
C01482 00417 ∂27-Feb-81 1540 CSD.AUBERY at SU-SCORE cs226
C01483 00418 ∂27-Feb-81 1548 TOB
C01485 00419 ∂27-Feb-81 1726 CSD.GOLUB at SU-SCORE
C01486 00420 ∂27-Feb-81 1735 Keith A. Lantz <CSL.LANTZ at SU-SCORE>
C01491 00421 ∂28-Feb-81 2217 TOB
C01510 00422 ∂01-Mar-81 0004 TOB
C01517 00423 ∂01-Mar-81 0033 POURNE@MIT-MC Saving Civilization on the installment plan
C01521 00424 ∂01-Mar-81 0155 TOB computing overload
C01534 00425 ∂01-Mar-81 0515 Admin.Kanef at SU-SCORE (Bob Kanefsky) advice taker in SPHERE
C01544 00426 ∂01-Mar-81 1413 TOB computing
C01548 00427 ∂01-Mar-81 1423 TOB
C01549 00428 ∂01-Mar-81 1428 TOB mail files, etc
C01551 00429 ∂01-Mar-81 1755 RWW CIRCUM
C01553 00430 ∂01-Mar-81 1831 Keith A. Lantz <CSL.LANTZ at SU-SCORE> owicki
C01554 00431 ∂01-Mar-81 1831 MINSKY@MIT-AI Hydrogen
C01560 00432 ∂01-Mar-81 2356 LWE NS TAPE
C01561 00433 ∂02-Mar-81 0040 POURNE at MIT-MC (Jerry E. Pournelle) CHarge!!
C01564 00434 ∂02-Mar-81 0043 POURNE at MIT-MC (Jerry E. Pournelle) Re: Helping the Boss
C01570 00435 ∂02-Mar-81 0101 POURNE at MIT-MC (Jerry E. Pournelle) helping the boss
C01571 00436 ∂02-Mar-81 0104 LLW The Referenced LLW Plaint. . .
C01580 00437 ∂02-Mar-81 0101 LLW Plaint-ively Speaking. . .
C01581 00438 ∂02-Mar-81 0128 POURNE at MIT-MC (Jerry E. Pournelle) Towards A System For The EOP
C01584 00439 ∂02-Mar-81 0134 JPM Re: Late
C01589 00440 ∂02-Mar-81 0150 JPM Re: The Boss
C01593 00441 ∂02-Mar-81 0736 BYY LUNCH THURSDAY
C01594 00442 ∂02-Mar-81 0910 FFL
C01597 00443 ∂02-Mar-81 0928 FFL
C01601 00444 ∂02-Mar-81 1531 FFL
C01602 00445 ∂02-Mar-81 1557 CSD.ULLMAN at SU-SCORE reiter
C01603 00446 ∂02-Mar-81 1642 RPG
C01604 00447 ∂02-Mar-81 1650 RPG
C01607 00448 ∂02-Mar-81 1652 KGK via SRI-KL third exercise
C01608 00449 ∂02-Mar-81 1650 RPG
C01612 00450 ∂02-Mar-81 2116 TOB 2080
C01616 00451 ∂02-Mar-81 2153 BYY LUNCH WED.
C01623 00452 ∂03-Mar-81 1319 MINSKY at MIT-ML (Marvin Minsky) Thanks for pre-text
C01628 00453 ∂03-Mar-81 1332 FFL
C01629 00454 ∂03-Mar-81 1348 FFL
C01630 00455 ∂03-Mar-81 1514 RPG Lisp Timing
C01635 00456 ∂03-Mar-81 1616 RPG Methodology considerations:
C01640 00457 ∂03-Mar-81 1524 RPG Lisp Timing Mailing List
C01642 00458 ∂03-Mar-81 2118 CSD.ULLMAN at SU-SCORE
C01644 00459 ∂03-Mar-81 2256 DEK ershov
C01656 00460 ∂04-Mar-81 0935 FFL
C01657 00461 ∂04-Mar-81 1116 FFL
C01658 00462 ∂04-Mar-81 1117 FFL
C01659 00463 ∂04-Mar-81 1121 CSD.JEANIE at SU-SCORE Faculty Meeting
C01660 00464 ∂04-Mar-81 1358 FFL
C01661 00465 ∂04-Mar-81 1534 CLT Calendar items
C01662 00466 ∂04-Mar-81 1622 FFL
C01663 00467 ∂04-Mar-81 1639 FFL
C01664 00468 ∂04-Mar-81 2216 TOB
C01667 00469 ∂04-Mar-81 2347 BIS london,request
C01669 00470 ∂04-Mar-81 2348 BIS london,reasons
C01671 00471 ∂05-Mar-81 0037 POURNE at MIT-MC (Jerry E. Pournelle)
C01673 00472 ∂05-Mar-81 0107 JMC
C01674 00473 ∂05-Mar-81 0557 RWG
C01675 00474 ∂05-Mar-81 0924 FFL
C01676 00475 ∂05-Mar-81 1043 CSD.ULLMAN at SU-SCORE Reiter
C01678 00476 ∂05-Mar-81 1100 JMC*
C01679 00477 ∂05-Mar-81 1143 CSD.ADAMS at SU-SCORE (lia) Comprehensive exam committee
C01681 00478 ∂05-Mar-81 1301 JMC*
C01682 00479 ∂05-Mar-81 1403 BIS london?
C01683 00480 ∂05-Mar-81 1503 CSD.ADAMS at SU-SCORE (lia)
C01684 00481 ∂05-Mar-81 2200 Buchanan@SUMEX-AIM J.King's application to move from Mathematics to CSD
C01686 00482 ∂05-Mar-81 2304 Wiederhold@SUMEX-AIM Some notes on research and distributed data
C01702 00483 ∂06-Mar-81 0036 POURNE at MIT-MC (Jerry E. Pournelle) a sort of strange thought...
C01704 00484 ∂06-Mar-81 0756 Mark Crispin <Admin.MRC at SU-SCORE> new TOPS-20 TELNET supports SUnet!
C01706 00485 ∂06-Mar-81 0801 FFL Trip to France
C01707 00486 ∂06-Mar-81 0905 Raphael at SRI-KL Party!
C01709 00487 ∂06-Mar-81 0937 CSD.ULLMAN at SU-SCORE David Cheriton
C01710 00488 ∂06-Mar-81 0951 FFL
C01711 00489 ∂06-Mar-81 1000 JMC*
C01712 00490 ∂06-Mar-81 1053 AVB Intel iAPX 432
C01713 00491 ∂06-Mar-81 1128 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Courses and Degrees 1981/82
C01714 00492 ∂06-Mar-81 1652 FFL mail jmc,ffl
C01718 00493 ∂07-Mar-81 0017 Jerry E. Pournelle <POURNE at MIT-MC> So you think our letters did some good...
C01721 00494 ∂07-Mar-81 0058 POURNE at MIT-MC (Jerry E. Pournelle) more policy development
C01727 00495 ∂07-Mar-81 0130 POURNE at MIT-MC (Jerry E. Pournelle) this sounds silly, but think about it...
C01731 00496 ∂07-Mar-81 0148 Jerry E. Pournelle <POURNE at MIT-MC>
C01733 00497 ∂07-Mar-81 0148 Jerry E. Pournelle <POURNE at MIT-MC>
C01734 00498 ∂07-Mar-81 0153 Jerry E. Pournelle <POURNE at MIT-MC>
C01736 00499 ∂07-Mar-81 0206 Jerry E. Pournelle <POURNE at MIT-MC> science advice
C01743 00500 ∂07-Mar-81 1000 JMC*
C01744 00501 ∂07-Mar-81 1442 QCC Kurt
C01745 00502 ∂07-Mar-81 2122 NEUMANN at SRI-KL [dkb at nbs-unix: Motels near NBS]
C01747 00503 ∂08-Mar-81 0753 PDL at MIT-DMS (P. David Lebling) [Re: more policy development]
C01748 00504 ∂08-Mar-81 0900 JMC*
C01749 00505 ∂08-Mar-81 1442 CLT
C01750 00506 ∂08-Mar-81 1745 DEK via SU-TIP datamedia
C01752 00507 ∂08-Mar-81 1909 LLW Swift (If Not Prompt) Deliveries
C01757 00508 ∂08-Mar-81 1920 LLW Frank Press As Science Adviser
C01761 00509 ∂08-Mar-81 1932 LLW Really Heavy Lifting
C01765 00510 ∂08-Mar-81 1924 LLW Discussion
C01766 00511 ∂08-Mar-81 1923 LLW Tactful Teleoperators, Et Al
C01769 00512 ∂09-Mar-81 0106 Jerry E. Pournelle <POURNE at MIT-MC> Really Heavy Lifting
C01770 00513 ∂09-Mar-81 0112 Jerry E. Pournelle <POURNE at MIT-MC> I'm going to say this one more time--
C01773 00514 ∂09-Mar-81 0624 ROD Thesis draft
C01774 00515 ∂09-Mar-81 1000 JMC*
C01775 00516 ∂09-Mar-81 1027 FFL
C01776 00517 ∂09-Mar-81 1027 FFL
C01777 00518 ∂09-Mar-81 1100 JMC*
C01778 00519 ∂09-Mar-81 1117 LGC Appointment?
C01779 00520 ∂09-Mar-81 1206 Konolige at SRI-KL exercise #3
C01781 00521 ∂09-Mar-81 1220 LGC Appointment Time
C01782 00522 ∂09-Mar-81 1327 RWW
C01784 00523 ∂09-Mar-81 1431 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) MTC Qual, Sat. May 23
C01785 00524 ∂09-Mar-81 1530 LGC Papers
C01786 00525 ∂09-Mar-81 1559 CSD.ULLMAN at SU-SCORE
C01787 00526 ∂09-Mar-81 1644 CSD.GRIFFITHS at SU-SCORE Re: GRADE
C01788 00527 ∂09-Mar-81 1809 Vaughan Pratt <CSD.PRATT at SU-SCORE> supply and demand
C01790 00528 ∂09-Mar-81 2045 MINSKY at MIT-ML (Marvin Minsky)
C01796 00529 ∂10-Mar-81 0634 Darden@SUMEX-AIM Re: Washington trip
C01798 00530 ∂10-Mar-81 0848 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
C01800 00531 ∂10-Mar-81 1259 SJF Reference on Lispki's work on nulls
C01801 00532 ∂10-Mar-81 1300 JMC*
C01802 00533 ∂10-Mar-81 1300 JMC*
C01803 00534 ∂10-Mar-81 1305 CSD.ADAMS at SU-SCORE (lia) comprehensive exam
C01805 00535 ∂10-Mar-81 1343 KAPLAN at SRI-KL new DARPA proposal
C01806 00536 ∂11-Mar-81 0035 SAM
C01807 00537 ∂11-Mar-81 0202 SJF Lipski
C01808 00538 ∂11-Mar-81 0530 Darden@SUMEX-AIM (Response to message)
C01809 00539 ∂11-Mar-81 1059 ZM
C01810 00540 ∂11-Mar-81 1155 Wiederhold@SUMEX-AIM Visit to washington and discussion about KBMS
C01812 00541 ∂11-Mar-81 1303 KAPLAN at SRI-KL KBMS portion of new DARPA proposal
C01814 00542 ∂11-Mar-81 1555 RWW fol
C01816 00543 ∂11-Mar-81 2140 LLW Teleoperators For Dirty Businesses
C01818 00544 ∂11-Mar-81 2226 LLW Peddling Blue Sky
C01820 00545 ∂11-Mar-81 2220 LLW Naming Spades
C01824 00546 ∂11-Mar-81 2149 LLW Teleoperating With Public Funds
C01838 00547 ∂12-Mar-81 0114 Jerry E. Pournelle <POURNE at MIT-MC> Teleoperators For Dirty Businesses
C01851 00548 ∂12-Mar-81 0126 POURNE at MIT-MC (Jerry E. Pournelle)
C01853 00549 ∂12-Mar-81 0838 FFL
C01856 00550 ∂12-Mar-81 0910 FFL
C01857 00551 ∂12-Mar-81 0936 FFL
C01858 00552 ∂12-Mar-81 0941 FFL
C01859 00553 ∂12-Mar-81 0945 Konolige at SRI-KL seminar today
C01860 00554 ∂12-Mar-81 1012 Konolige at SRI-KL (Kurt Konolige) Re: seminar notice
C01863 00555 ∂12-Mar-81 1346 LLW The Simple Joys of Coal Mining
C01864 00556 ∂12-Mar-81 1352 FFL
C01868 00557 ∂12-Mar-81 1359 FFL
C01869 00558 ∂12-Mar-81 1354 QCC Paper for Epistemelogigal Problems in AI course
C01870 00559 ∂12-Mar-81 1408 LLW Fitting The Punishment To The Crime
C01874 00560 ∂12-Mar-81 1634 Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE> directed reading
C01876 00561 ∂12-Mar-81 2042 TOB japan trip
C01893 00562 ∂12-Mar-81 2159 JMC
C01894 00563 ∂12-Mar-81 2225 TOB
C01895 00564 ∂12-Mar-81 2242 CLT
C01896 00565 ∂12-Mar-81 2244 CLT q/x
C01897 00566 ∂12-Mar-81 2256 LGC ARPA Meeting
C01899 00567 ∂13-Mar-81 0506 Darden@SUMEX-AIM (Response to message)
C01902 00568 ∂13-Mar-81 0900 JMC*
C01903 00569 ∂13-Mar-81 0902 MERRITT at USC-ISIB Re: duplicated messages
C01905 00570 ∂13-Mar-81 1023 FFL
C01906 00571 ∂13-Mar-81 1155 FFL
C01907 00572 ∂13-Mar-81 1157 FFL
C01908 00573 ∂13-Mar-81 1317 FFL
C01909 00574 ∂13-Mar-81 1506 FFL
C01910 00575 ∂14-Mar-81 0744 JRA msg
C01911 00576 ∂14-Mar-81 0934 QLP how to enter a saved Fol proof in a file ?
C01912 00577 ∂14-Mar-81 1623 JPM Concerned Scientists Meeting
C01914 00578 ∂14-Mar-81 1652 Feigenbaum@SUMEX-AIM Re: Andrei Ershov
C01915 00579 ∂14-Mar-81 1946 CLT
C01917 00580 ∂15-Mar-81 1139 RPG Talk
C01918 00581 ∂15-Mar-81 1158 RPG
C01919 00582 ∂15-Mar-81 1204 RPG
C01920 00583 ∂15-Mar-81 1526 TOB
C01922 00584 ∂15-Mar-81 1548 TOB
C01924 00585 ∂16-Mar-81 1052 LGC Pre-meeting Info
C01926 00586 ∂16-Mar-81 1226 RPG Solicitation
C01928 00587 ∂16-Mar-81 1307 RPG
C01929 00588 ∂16-Mar-81 1404 TOB arpa proposal
C01930 00589 ∂16-Mar-81 1429 HEDRICK at RUTGERS Re: Solicitation
C01935 00590 ∂16-Mar-81 1516 FWH PV+A Seminar
C01937 00591 ∂16-Mar-81 1521 GFS Ethernet article
C01938 00592 ∂16-Mar-81 1829 CSD.DOYLE at SU-SCORE proposal revised
C01940 00593 ∂17-Mar-81 0346 DCL ARPA
C01941 00594 ∂17-Mar-81 0651 BYY
C01942 00595 ∂17-Mar-81 1013 NEUMANN at SRI-KL VERkshop II
C01948 00596 ∂17-Mar-81 1035 Stan at SRI-AI Common Business Communication Language
C01950 00597 ∂17-Mar-81 1204 RPG Proposal
C01958 00598 ∂17-Mar-81 1155 Masinter at PARC-MAXC Re: GC
C01962 00599 ∂17-Mar-81 1218 RPG Bureaucracy
C01963 00600 ∂17-Mar-81 1456 Nilsson at SRI-AI Firdman
C01964 00601 ∂17-Mar-81 1529 CSD.ULLMAN at SU-SCORE Re: IBM grant for proving
C01965 00602 ∂17-Mar-81 1628 FFL
C01966 00603 ∂17-Mar-81 1648 FFL Oral
C01967 00604 ∂17-Mar-81 1921 Bernard S. Greenberg <Greenberg at MIT-Multics> Re: Solicitation
C01977 00605 ∂17-Mar-81 2200 JMC*
C01978 00606 ∂19-Mar-81 0816 FFL
C01979 00607 ∂19-Mar-81 1818 Denny Brown <CSD.DBROWN at SU-SCORE> use of CSD-CF for classwork
C01981 00608 ∂19-Mar-81 1824 Denny Brown <CSD.DBROWN at SU-SCORE> approved courses for using CSD-CF
C01984 00609 ∂19-Mar-81 2203 Wiederhold@SUMEX-AIM visits
C01985 00610 ∂20-Mar-81 0954 CSD.JEANIE at SU-SCORE Faculty Meeting
C01986 00611 ∂20-Mar-81 1555 QEH exercise
C01987 00612 ∂20-Mar-81 1729 MMD Documentation for my TEX macros is finally complete, and comes in two flavours:
C01989 00613 ∂20-Mar-81 2328 GREEP at RAND-AI CS226 term paper
C01990 00614 ∂21-Mar-81 0030 Robert S. Boyer <BOYER at SRI-CSL> Verkshop
C01992 00615 ∂21-Mar-81 1144 KGK on TTY162 1144 via SRI-AI exercise 3
C01994 00616 ∂21-Mar-81 1155 Konolige at SRI-AI Answers to exercise 3
C01996 00617 ∂21-Mar-81 1539 MINSKY at MIT-AI (Marvin Minsky)
C01998 00618 ∂23-Mar-81 0000 JMC*
C01999 00619 ∂23-Mar-81 0000 JMC*
C02000 00620 ∂23-Mar-81 0942 FFL Visit of Bob Grafton of ONR
C02001 00621 ∂23-Mar-81 1048 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Grades are Due
C02003 00622 ∂23-Mar-81 1424 FFL
C02004 00623 ∂23-Mar-81 1535 FFL
C02005 00624 ∂23-Mar-81 1657 SCH greetings
C02006 00625 ∂23-Mar-81 2000 JMC*
C02007 00626 ∂23-Mar-81 2018 RPG
C02011 00627 ∂23-Mar-81 2055 ML cs226
C02012 00628 ∂23-Mar-81 2140 JK
C02015 00629 ∂24-Mar-81 0729 Darden@SUMEX-AIM regards
C02016 00630 ∂24-Mar-81 0833 KAPLAN at SRI-AI ARPA proposal
C02017 00631 ∂24-Mar-81 0916 NEUMANN at SRI-KL VERkshop agenda, first cut
C02028 00632 ∂24-Mar-81 1002 FRG Fran`s sick
C02029 00633 ∂24-Mar-81 1040 Stan at SRI-AI distributed business machines
C02030 00634 ∂24-Mar-81 1120 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) grades
C02031 00635 ∂24-Mar-81 1135 CSD.ULLMAN at SU-SCORE
C02032 00636 ∂24-Mar-81 1220 Walker at SRI-AI net address change for SRI AI Center
C02034 00637 ∂24-Mar-81 1442 RDA phd oral
C02035 00638 ∂24-Mar-81 1553 RPG Results
C02036 00639 ∂24-Mar-81 1604 CSL.JLH.MOGUL at SU-SCORE (Jeffrey Mogul) Comments on the Intel iAPX 432
C02042 00640 ∂24-Mar-81 1719 csl.jlh at SU-SCORE (John Hennessy) Re: Comments on the Intel iAPX 432
C02043 00641 ∂24-Mar-81 1748 LGC
C02044 00642 ∂24-Mar-81 2329 TOB brief discussion
C02045 00643 ∂25-Mar-81 0622 JRA cs institute
C02046 00644 ∂25-Mar-81 1333 FFL
C02048 00645 ∂26-Mar-81 1138 Feinler at SRI-KL Re: EOP
C02051 00646 ∂26-Mar-81 1507 FFL
C02052 00647 ∂26-Mar-81 1520 FFL
C02053 00648 ∂26-Mar-81 1928 Kanerva@SUMEX-AIM DEC-20 High Hurdles
C02055 00649 ∂27-Mar-81 0638 Darden@SUMEX-AIM reference on 3-valued logic
C02058 00650 ∂27-Mar-81 0816 LGC Discussion Today
C02059 00651 ∂27-Mar-81 0902 FFL
C02060 00652 ∂27-Mar-81 1104 Hans Moravec at CMU-10A (R110HM60) Pellet streams
C02062 00653 ∂27-Mar-81 1130 FFL
C02063 00654 ∂27-Mar-81 1244 CSD.ULLMAN at SU-SCORE Faculty meeting
C02064 00655 ∂28-Mar-81 0837 JRA wics
C02065 00656 ∂28-Mar-81 1026 REM at MIT-MC (Robert Elton Maas)
C02068 00657 ∂28-Mar-81 1333 Darden@SUMEX-AIM (Response to message)
C02070 00658 ∂28-Mar-81 1353 Darden@SUMEX-AIM Request for copy of McCarthy paper
C02072 00659 ∂28-Mar-81 2036 RPG Advice
C02076 00660 ∂29-Mar-81 1409 RPG
C02077 00661 ∂29-Mar-81 1455 KDO via SU-SCORE What REM said
C02079 00662 ∂29-Mar-81 1554 RPG
C02081 00663 ∂30-Mar-81 0054 POURNE@MIT-MC HELPING THE BOSS
C02082 00664 ∂30-Mar-81 0156 Jerry E. Pournelle <POURNE at MIT-MC> Re: EOP
C02083 00665 ∂30-Mar-81 1216 NEUMANN at SRI-KL VERkshop II: Formal Application for Participation Needed!
C02085 00666 ∂30-Mar-81 1229 ELLEN@MIT-MC HELPING THE BOSS
C02086 00667 ∂30-Mar-81 1241 RPG
C02089 00668 ∂30-Mar-81 1657 DEK self-proposed visitors
C02090 00669 ∂30-Mar-81 1743 YM new course, abstract data types
C02095 00670 ∂31-Mar-81 1022 FFL
C02096 00671 ∂31-Mar-81 1108 Buchanan@SUMEX-AIM AAAI Council
C02098 00672 ∂31-Mar-81 1326 FFL
C02100 00673 ∂31-Mar-81 1509 ENGELMORE at USC-ISI Re: LISP conference
C02112 00674 ∂31-Mar-81 1553 Stan at SRI-AI Meeting to discuss business communication language
C02113 00675 ∂31-Mar-81 1608 Stan at SRI-AI Meeting
C02114 00676 ∂31-Mar-81 2343 CSD.GENESERETH at SU-SCORE lisp project reports
C02115 00677 ∂01-Apr-81 0800 JMC*
C02116 00678 ∂01-Apr-81 1145 DCL kowalski
C02117 ENDMK
C⊗;
∂01-Jan-81 1254 Mark Crispin <Admin.MRC at SU-SCORE> Re: troubles with dtn
Date: 1 Jan 1981 1253-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Postal-Address: 12155 Edgecliff Place; Los Altos Hills, CA 94022
Stanford-Phone: (415) 497-1407
Subject: Re: troubles with dtn
To: CSD.MCCARTHY at SU-SCORE
In-Reply-To: Your message of 31-Dec-80 0107-PST
I'm trying to get Ralph to hire Tovar, so he won't be able to use
that excuse any more!! Oh well, I'll see if I can't look at it when
I get back from vacation.
-------
∂01-Jan-81 1630 TOB
substantial contribution and effectively supervised
Ruzena Bajcsy
Gerald Agin
Ram Nevatia
Tom Garvey
Kicha Ganapathy
completion by mid 1981
R.D.Arnold
Rod Brooks
Harlyn Baker
effectively supervised
Russell Taylor (with Feldman)
Bob Bolles (with Feldman)
Don Gennery
Hans Moravec
Ron Goldman; mid 1981
∂01-Jan-81 1708 CLT
Come and pick me up when you are ready for supper.
No thanks, I've seen enough movies for a while.
I haven't detected any lossage.
∂01-Jan-81 1906 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V2 #179
----------------------------------------------------------------------
Date: 23 Dec 1980 1218-PST
From: Craig W. Reynolds from III via Rand <REYNOLDS at RAND-AI>
Subject: "secure" systems
John McCarthy thinks that secure systems are possible (while
questioning their need), and correctly ties it to physical
security. Frankly my gut feelings are with Hofstadter, that
it is in fact impossible to have an absolutly secure anything.
The thought is a little scary, if a system is tight enough
to keep out even its designers and operators, who controls
an out-of-control rogue system. That was the basis for the
plot of "The Forbin Project", a secure system in charge of
the national weapons systems, and wasn't doing what the people
in charge wanted. If it is possible for an expert ("wheel",
"God",...) to break into the system, then the most direct way
for a bad guy to get through security is to kidnap the system
hacker and torture him/her until the information is extracted.
However most systems are defenseless against phycisal attack
(rebooting the system with your version of the OS, stealing
disks/tapes, blowing up the hardware, depending on what your
interests in breaking the system are) so it really becomes a
"military" problem: is there any such thing as an attack-proof
place? In a world of nuclear bombs, the answer seems to be no.
Mind you, I do not endorse any of these methods, but when
we talk of SECURE systems (and their penetration), we are
no longer talking about polite society.
- Craig
∂01-Jan-81 1927 POURNE at MIT-MC (Jerry E. Pournelle) Helping the Boss
Date: 27 DEC 1980 0345-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Helping the Boss
To: LAUREN at MIT-MC, ELLEN at MIT-MC, MCLURE at MIT-MC
To: LYNCH at MIT-MC, YORK.MULTICS at MIT-MC, JRDAVIS.MULTICS at MIT-MC
To: AV at MIT-MC, POURNELLE at MIT-MC, OAF at MIT-MC, CENT at MIT-MC
To: STEF at MIT-MC, JMC at MIT-MC, GEOFF at MIT-MC, JNC at MIT-MC
To: PDL at MIT-MC, CBF at MIT-MC, DLW at MIT-MC, RWK at MIT-MC
To: MINSKY at MIT-MC, TAW at MIT-AI
Sorry to subject you to an interminable list of people; you're
the ones who sent informative replies.
I regret that I'm going to give up this attempt to
provide information. It's obvious that I am the WRONG person to
attempt it. I had naively supposd that I was getting a bit
familiar with things, but I find that I'm woefully ignorant of
first principles. I don't even seem to know the difference
between a machine and an operating system! In fact to this
minute I'm confused. I thought a TOPS-20 was a piece of
hardware, but it seems that's not right. Sigh.
I conclude from all this that (1) there is, somewhere,
at least one machine, a Multics (that IS a machine, not an OS,
isn't it?) purchased by the US Gov't, possibly as a result of
the "Report of the Advisory Committee on Information Network
Structure and Functions for the Executive Office of the
President", a copy of whch I have.
That document (it seems to have no reference number, and
I am damned if I'll write that title out again!) is largely a
wish list: that is, it makes (no doubt excellent)
recommendations, but they are all of the "It is recommended that
the EOP INformation Network include a network core service for
queries and on-line help" variety. There are no specifics of
either hardwre or software or operating systems mentioned anywhere.
The document doesn't tell us much about security; and
the discussions I have received in response to my initial
inquiry have alerted me to the fact that security is tougher
than I would have thought, and possibly more dangerous than I
would have supposed. However, I can see how a basic system
could be devised that would be as secure as, say, the telephone
lines between White House and EOP, and no more prone to leaks
and blurting than an office with Xerox machines (no less, either...)
It would have to be a system with very closely
controlled access; meaning that from the outside no user shold
be able to insert any program whatever without physical access
to a cleared facility room and physical equipment. This would,
I thnk, stop the "Trojan Horse" program (the program that plays
ZORK with you and simultaneously accesses all your files and
sends them out to an unauthorized recipient); and also the
"operating system takeover" type of program/invasion. Wouldnt it?
I think there is no security system that would allow
classified access from outside -- say from home -- but there
sholdn't be, anyway; and when one logs in from outside, one
ought not be able to access very much created from inside.
True, it would be convenient as hell to do things another way
and to have ways to tell the system that yo're really all right,
but that's too easily broken.
Thus, it seems to me, you would have a system what
relies largely on preventing physical access by unauthorized
persons (in much the same way that you prevent unauthorized
people from rummaging through paper files); would also have some
levels of access control and logins and passwords and the like;
and which has some very careful controls on access to "foreign"
nets, IE, to nets other than the EOPNET itself. This doesn't
seem to me to be impossible.
JMC points out that you can "blurt" with paper too: if
we're gong to monitor transmissions outside the net (have them
all read by a security officer before they can go out is one
extreme proposed) then perhaps one ought to do that to mail too?
But paper mail doesn't go out the instat you press a button;
thus inadvertent blabbing is harder. It IS MUCH EASIER to
review what yo've written on PAPER than it is ON SCREEN; or that
at least has been my experience. It's easier to create text on
screen, and to CHANGE it; but to review it I find paper easier;
and thus I am less likely to send out something I didn't want to
say in a paper letter than I am in, say, an R-mail Reply...
==============
Second: although I can, I thnk, what with "the document"
plus your various replies, write a meta-language description of
an "ideal" EOPNET, and even scale that to something affordable
(under a couple megabucks), I cannot, from what I have seen,
actuly recommend anything specific. There is NO concensus, and
worst of all, it seems that there is nothng really running in
the sense of turnkey systems available in a few weeks.
If that's wrong, please tell me; but it's the impression
I get. You can't just drop down money and get a net system that
does things.
That's surprising; I had thought you could.
Third: The procurement cycle seems needlessly complex; I don't
see any way out of it, since there is no concensus among
experts; and my suspicion is that bureaucratic rules may
ultimately defeat the purchase of a machine which could
substantially aid the bureaucracy.
But: the incoming EOP people regard EOP quite
differently from the previous; they don't trust civil servant
types within it; or at least several important ones don't.
Still: I see no specific recommendation I could make; no
"take $xxx and hand them to company YYY, making certain to
employ Ms. ZZZZ as a system consultant; you will want to
purchase terminals from FFFF and they will cost $dd each. When
it is done, you will have a minimum net, with electronic mail,
secure files you can get at only from inside the complex, and
"paperless office" capabilities; there will also be a hard-copy
device which will automatically make hard copy of any files you
want that done to. We can teach officers to use the system in
about five one-hour tutorial sessions (or fewer). Secretaries
will also have access to use the system for drafting memos and
other secretarial work, but their access will NOT be the same as
that of officers, and the system is of no real value unless the
officers themselves use it.
Next: (at 300 baud I can't go back and see what point
number I am on): I am going to send this. Continued next memo.
Sigh: I wish I had a way to do that without having to retype the
whole big mailing list...but I dasn't not send this lest I lose
all of it, and I don't know how to save files in rmail; nor how
to use emacs to create a file that I can send
∂01-Jan-81 1927 Jerry E. Pournelle <POURNE at MIT-MC> Helping the Boss
Date: 28 December 1980 03:03-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Helping the Boss
To: POURNE at MIT-MC
cc: LAUREN at MIT-MC, ELLEN at MIT-MC, MCLURE at MIT-MC,
LYNCH at MIT-MC, YORK.MULTICS at MIT-MC, JRDAVIS.MULTICS at MIT-MC,
AV at MIT-MC, POURNELLE at MIT-MC, OAF at MIT-MC, CENT at MIT-MC,
STEF at MIT-MC, JMC at MIT-MC, GEOFF at MIT-MC, JNC at MIT-MC,
PDL at MIT-MC, CBF at MIT-MC, DLW at MIT-MC, RWK at MIT-MC,
MINSKY at MIT-MC, TAW at MIT-AI
Great heavens! 1R worked to create the list! Amazing.
It's well I saved that and sent it, since the entire ITS net
crashed about 35 seconds after I sent it off.
To continue: (1) if someone out there would like to send me a
SHORT note telling me what is machinery and what is software
(micro people are woefully ignorant of what exists in the big
mainframe world) and who makes it and what it costs and--a short
basic primer on how not to sound like an idiot--it would be
useful.
(2) I can't ee that I can recommend anything at the moment; but
I do thnk a short report on possibilities and potentials and
such would be well received, and worth my doing. With all due
respect, it would HAVE to be at least as informative to laymen
as "the document"; and given the situation, would thus be more
useful. Unless there is another and more readable version of
"the document"?
(3) My thanks to you all. I will start a new file called
MC:POURNE;BOSS NEW to contain this second round of correspondence.
JEP
∂01-Jan-81 1927 POURNE@MIT-MC secure systems
From: POURNE@MIT-MC
Date: 12/28/80 02:41:40
Subject: secure systems
POURNE@MIT-MC 12/28/80 02:41:40 Re: secure systems
To: JMC at MIT-MC, POURNELLE at MIT-MC, MINSKY at MIT-ML
If we had a formal proposal for development of such syustems for
use in EOP and White House I could get it to the right people,
but I don't know how to develop a request for proposal out of
that. I do NOT work for the EOP, I merely advise some committees.
On that score: at end of January a number of us
(astronauts and engineers) will meet here in LA to write the
working papers for the national space policy committee. The
committee is already informally known as the Von Neumann
Committee; our papers, consisting of a space plan, will go to it
(and so will undoubtedly other papers). No guarantee they'll
adopt the plan, but it's a good shot. Conference is unofficial
(I have a small grant for support and publication and
secretarial servieces but no money) and will meet at Larry
Niven's house.
If interested let me know; it's getting crowded and LOTS
of people want to come who will not be invitied, but
I could use that other graduate student--not DANNY, but
the other one who was at Pajaro dunes, and I could proabbly get
his way paid here if he's interested, but I got no name for him..
Jerry P
I will come to your end of January space meeting if invited. Perhaps a
previous message to that effect got lost. Also, in case you didn't know,
our computer was down from just before Christmas to Dec 31, so I'm just
catching up on my mail.
In my opinion something can be done to get an EOP system with reasonable
security and capability, but it first it is necessary to know what has
already been done and the state-of-the-art on secure and convenient
time-sharing systems. I even believe that people can work from home with
greater security than taking documents home and having a safe. It
requires an editing terminal that keeps encrypted files and transmits only
encrypted files to the central system. I think that the new developments
in cryptography, i.e. public key, are not needed for this, and I suspect
that there are experts that know almost all relevant facts. The one
possibility for "revolution" is that the experts may have talked
themselves into a more paranoid attitude to computer-based systems than to
manual systems, so that switching from present systems to a computer-based
system with known imperfections may be better than staying with present
systems. Moreover, I'll bet there is an enormous amount of EOP work which
requires no more security than is provided by ordinary operating systems.
∂01-Jan-81 1927 Jerry E. Pournelle <POURNE at MIT-MC> Helping the Boss
Date: 28 December 1980 03:16-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Helping the Boss
To: POURNE at MIT-MC
cc: LAUREN at MIT-MC, ELLEN at MIT-MC, MCLURE at MIT-MC,
LYNCH at MIT-MC, YORK.MULTICS at MIT-MC, JRDAVIS.MULTICS at MIT-MC,
AV at MIT-MC, POURNELLE at MIT-MC, OAF at MIT-MC, CENT at MIT-MC,
STEF at MIT-MC, JMC at MIT-MC, GEOFF at MIT-MC, JNC at MIT-MC,
PDL at MIT-MC, CBF at MIT-MC, DLW at MIT-MC, RWK at MIT-MC,
MINSKY at MIT-MC, TAW at MIT-AI
Bottom line: I will try to edit a report out of the information
I get from all you helpful people. It will go in in the form of
a recommended system, or a description of a recommended system.
THere is no particular urgency; I will already have reported
that I was mistaken, you cannot simply "drop in" a system that
does the things I was describing to my friends, so this is a
more liesurely project.
There is also the chance that the civil service types
will have implemented something; but our discussion can, I thnk,
do no harm, and might eventually do some good. But it does take
on a LOT less urgency.
∂01-Jan-81 1932 POURNE at MIT-MC (Jerry E. Pournelle) Let's try to be fair
Date: 29 DEC 1980 0332-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Let's try to be fair
To: RWK at MIT-MC, JNC at MIT-MC, CSTACY at MIT-MC, MINSKY at MIT-MC
To: JMC at MIT-MC, POURNELLE at MIT-MC, HQM at MIT-MC, BYTES at MIT-MC
This LISP nonsense is getting to me. I just took down
my poor article, and goddammit I still don't see that I deserve
flaying alive.
(1) Recall that BYTE is not a hacker's magazine; it is
subtitled the small systems journal, it has no advertisements or
articles for large computer users, and concentrates almost
exclusively on micros.
(2) Many people read BYTE to learn what they ought to be
doing in the micro world.
(3) I remain unrepentant, then: "If you like playing
with powerful languages, mu-LISP-79 is recommended--but don't
blame me if you don't use it very often after the first wave of
enthusiasm." That is what I said, and I fail to see how that is
"personally offensive" to, one of you, a hell of a lot of people
at MIT.
(4) I am going to do a second piece on LISP in another
column. I still don't know of a commercially available text
that I would recommend, and I don't thnk telling people to write
MIT or Stanford for documents is a winning way to make friends.
If you've suggestions on introductory materials for people out
there in micro-land to read so that they can then play
meaningfully with the LISP they can get out of Microsoft, please
tell me what it is, how obtained, etc., and is it suitable for
the kind of people who read my columns?
(5) As to the insistence that LISP programs are
self-documenting, probably they are--if you understand the
language. Me, I open LET'S TALK LISP by Siklossy at random and
I find:
(EVLIS (LAMBDA (NONEVALARGS ALIST)
(COND (NULL NONEVALARGS) NIL)
(T CONS (EVAL1 (CAR NONEVALARGS) ALIST)
(EVLIS (CDR NONEVALARGS) ALIST))) )))
Which, I presume is self-documenting and clear as
English to all of you; but to me it is pure jargon. It is all
very well to have an argot with cars and cdrs and lambdas and so
forth; but why be surprised when it has the same effect as ANY
argot does, that is, it sorts out those on the insider and in
the know from the peasants; but it also makes the peasants less
likely to join the inside.
(6) I know that none of you are responsible for other
people's textbooks, but I assure you that such paragraphs as:
"In many LISP systems, when a function is DEFINEd, the
LAMBDA expression defining the function becomes the property of
the indicator EXPR (for EXPRession) on the property list of the
atom naming the function. for example, in EVQ-LISP, the effect
of the program
DEFINE ((
(DONOTHING (LAMBDA (SEX) SEX))
))
is equivalent to the program
PUTPROP (DONOTHNG EXPR (LAMBDA (SEX) SEX) )
Both programs define the function DONOTHING. The values of the
programs are different, though."
And so forth--
I assure you that finding that sort of thing in the book
to teach the language does not, at least does not for me,
produce inspiration and a burning desire to learn more. Perhaps
I am just perverse?
(Like many textbooks, this one seems to put the most nteresting
cases as questions, which it doesn't bother to answer, so that I
at lest found it hard even to use it like a cookbook. Ah well.)
If, then, it makes any of you who find my views on LISP
"personally offensive" feel any better to thnk of me as simply
too old and too stupid to comprehend, then by all means do so.
meanwhile: if you know of a better (or indeed alternative) way
to learn LISP than to be around people who are using it; please
tell me; and I will meantime retract the term "obscure" and use,
instead, soemthing like "specialized, and difficult to learn
without tutorial instruction"; or is that also offensive?
I want to be fair, you see; I have alrady lost much
money over this silly question, and I wish to heaven I had never
seen the mu-LISP program sent me for evaluation.
JEP
∂01-Jan-81 1936 Henry at MIT-AI (Henry Lieberman) Your Lisp Book
Date: 30 DEC 1980 0016-EST
From: Henry at MIT-AI (Henry Lieberman)
Subject: Your Lisp Book
To: JMC at SU-AI, CLT at SU-AI
Carl Hewitt has told me that you are working on an introductory book on
Lisp. I am doing research on programming environments for Lisp which I
hope will make Lisp programming easier to learn for beginning users, and
for AI applications. Surveying introductory textbooks will help me see
how programming concepts are initially presented to users and provide
examples of simple kinds of programs new users write. Would it be
possible for me to obtain a draft copy of the book? [I can FTP and Dover
it over the net if sending hard copy is a hassle.] Thank you.
Henry Lieberman, MIT AI Lab
To FFL: Please send him one.
I have asked my secretary to send you a copy.
∂02-Jan-81 0925 BYY maclisp manual
To: JMC
CC: JMC
I'll be in at the office today. If you're there, perhaps
you cpould put it back in your mail box. Thanks. Jon
∂02-Jan-81 1032 Baskett at PARC-MAXC Monitors
Date: 2 Jan 1981 10:31 PST
From: Baskett at PARC-MAXC
Subject: Monitors
To: JMC@Sail
John: You might look at this monitor. Maybe you would want the 20 inch
version? -Forest
---------------------------
Mail-from: Arpanet host SU-AI rcvd at 1-JAN-81 1655-PST
Date: 01 Jan 1981 1344-PST
From: Andy Bechtolsheim <AVB at SU-AI>
Subject: philips video monitor
To: "@SUN.DIS[P,DOC]" at SU-AI, LCW at SU-AI, REG at SU-AI,
TM at SU-AI, TOB at SU-AI
We have a north american philips monitor on loan, currently connected
to the SUN graphics prototype in MJH 433. The monitor has the following
characteristics: non-interlaced display, 70 frames per second,
606 vertical lines, 1024 horizontal points, white P4 phosphor.
Due to video bandwidth limitations of the prototype, a displayed point
is actually two pixels wide; the visible area is therefore 512 by 606
and characters thus appear wider than normal.
The monitors we plan to order for the first batch of SUN workstations are
electrically and mechanically similar. However, their tube is mounted
in portrait mode and their visible display area is 800 by 1024.
Philips also offers a model with a 20 inch wide screen that might be
interesting for some applications.
We have to return the evaluation unit on January 7, so please have
a look before then and let me know your comments right away.
------------------------------------------------------------
∂02-Jan-81 1224 HENRY at MIT-AI (Henry Lieberman) Your book
Date: 2 JAN 1981 1525-EST
From: HENRY at MIT-AI (Henry Lieberman)
Subject: Your book
To: JMC at SU-AI
Thank you for sending me a copy.
∂03-Jan-81 0001 POURNE at MIT-MC (Jerry E. Pournelle) meetings and advice
Date: 3 JAN 1981 0301-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: meetings and advice
To: POURNELLE at MIT-MC, JMC at SU-AI
Meeting will be full weekend working session to write a
feasible plan for man in space, exploioting technology to best
feasible levels. If yo're interested in coming, let's discuss
that after I return from AAAS (I leave at dawn).
EOP computer project shelved; I can 't get a concensus
for a quick and dirty system that can be installed, and there
exist (I think) mechanisms for the slow grind it out civil
service method.
JEP
∂03-Jan-81 1718 RPG MacLisp
To: REG, ME
CC: JMC
I'm going to try to standardize our MacLisp progammers more to the
MIT style of doing things (ie, using their macro packages where
once ours were used). This is for two reasons: 1, we continue to
diverge in even more ways and it is difficult to maintain the
two sets of macros as time goes on; 2, MIT's macros imitate NIL,
which we will surely run on the VAXs later, and maybe on the 2080
or even 2060 eventually. This way we can start being compatible
with that now. If I take off for hunoz where in March, it's better to
do it now than after I leave.
-rpg-
∂04-Jan-81 0900 JMC*
tasks in cs226
∂04-Jan-81 1536 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) title of Colloq.
Date: 4 Jan 1981 1534-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: title of Colloq.
To: JMC at SU-AI
John, you are scheduled to give a colloquium on Tuesday, Jan. 27.
I need a title.
Carolyn
-------
∂05-Jan-81 0757 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Faculty Meeting Reminder
Date: 5 Jan 1981 0757-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Faculty Meeting Reminder
To: CSD-Faculty:
cc: TOB at SU-AI, SL at SU-AI, REG at SU-AI, CSD.BSCOTT at SU-SCORE,
CSD.DBROWN at SU-SCORE, WOL at SU-AI, JEB at SU-AI
Faculty Meeting, Tuesday, Jan. 6, 2:30 p.m., Jacks 252.
-------
∂05-Jan-81 0900 JMC*
Lindstrom
∂05-Jan-81 0906 CLT
Did you remember to call the architect?
Also, would you make a copy the last installment of Boyer proof for me, please?
∂05-Jan-81 0931 FFL
To: JMC, FFL
Bob Wagoner, Physics, 7-4561, has set April 5 as tentative date for your
Sunday seminar. Is that OK with you? Pls. let him know.
∂05-Jan-81 0955 FFL
To: JMC, FFL
Bruce Buchanan asks that you call him today as it is important to speak
with you about D. Lenat reappointment. 7-0935.
∂05-Jan-81 0958 Buchanan@SUMEX-AIM
Date: 5 Jan 1981 0951-PST
From: Buchanan@SUMEX-AIM
To: jmc@SU-AI, rwf@SU-AI, tw@SU-AI
I must talk with you today about Doug lenat's reappointment. Please call
me (7-0935) when you come in.
The responses to my letter (from Nilsson, Bledsoe, & Hayes-Roth) are
mixed: very positive on assessment of brilliance and potential, but
guarded & cautious on productivity.
Ed Feigenbaum was in favor of a one-year reappointment to send a strong
message to Doug that he has to start contributing. I am in favor of
that, even though it seems a little harsh.
Please call me as soon as possible.
thanks,
Bruce
-------
∂05-Jan-81 1200 JMC*
Buchanan terminal
∂05-Jan-81 1311 bledsoe at UTEXAS-11 atp Prize Comm Mtg
Mail-from: ARPANET site UTEXAS-11 rcvd at 5-Jan-81 1306-PST
Date: 5 Jan 1981 at 1457-CST
From: bledsoe at UTEXAS-11
Subject: atp Prize Comm Mtg
To: mccarthy at score
cc: bledsoe at utexas
Thanks for your letter. Your idea seems like a good one and I
hope we can discuss it at the meeting Thursday. If you arrive before
12:30 (even by 1:00) there will still be time I believe. I'm delighted
that you plan to attend. Best regards, Woody
-------
It now looks like I'll be on time.
∂05-Jan-81 1412 FFL
To: JMC, FFL
Mr. Oakner says that it would be fine for you to see him on Friday,
January 9, at ll a.m. His telephone is 857-9400.
∂05-Jan-81 1641 Konolige at SRI-KL (Kurt Konolige) Re: TA possibility
Date: 5 Jan 1981 1633-PST
From: Konolige at SRI-KL (Kurt Konolige)
Subject: Re: TA possibility
To: JMC at SU-AI
In-Reply-To: Your message of 4-Jan-81 1542-PST
It sounds good, as I was hoping to fulfill the TA requirement
without too much work; the fewer students, the better. I'll be over at
Stanford sometime this week, and I'll try to see you then to talk. --kk
-------
The class is Tuesday and Thursday at 11 in 380C starting tomorrow.
There will be no class on Thursday this week. You should try to
attend at least half of the classes at least in the early part of
the quarter.
∂05-Jan-81 1705 Konolige at SRI-KL (Kurt Konolige)
Date: 5 Jan 1981 1703-PST
From: Konolige at SRI-KL (Kurt Konolige)
To: JMC at SU-AI
In-Reply-To: Your message of 5-Jan-81 1643-PST
Ok, I'll be there at 11 tomorrow. --kk
-------
∂05-Jan-81 1830 Kanerva@SUMEX-AIM Now that mail moves again . . .
Date: 5 Jan 1981 1830-PST
From: Kanerva@SUMEX-AIM
Subject: Now that mail moves again . . .
To: JMC@SAIL, REG@SAIL
Date: 23 Dec 1980 1910-PST
From: Kanerva@SUMEX-AIM
To: JMC@SAIL, REG@SAIL
Subject: DEC-20 for the Text Network Program: Progress Report
Ed Shaw's meeting with the vice-presidents yesterday produced
the permission to order the DEC-20 with June delivery, with an option
to cancel until two months before delivery but also to ask for an
"early" delivery. This gives Stanford the 25% discout that DEC had
offered, plus time to think about how well the service will sell here
at Stanford. Jon Sandelin will be busy marketing the service and has
the IMSSS system available for demonstrating it and for getting ready
for the arrival of the 20. - Pentti
-------
-------
∂05-Jan-81 2048 BYY NYT
Thanks for the news flash. I was aware of the fight, but not the latest
skirmish. By the way, Ruth Marcus just turned down an offer from us
to stay with the pluraists at Yale, in spite of her running battles wtih
them.
How do you get the news?
Also, how do you update the msg file so that it automatically
calls the first page you havet seen? Jon
∂06-Jan-81 0800 Denny Brown <CSD.DBROWN at SU-SCORE>
Date: 6 Jan 1981 0756-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
To: JMC at SU-AI, DPB at SU-AI
In-Reply-To: Your message of 6-Jan-81 0050-PST
OK on Konolige. -Denny
-------
∂06-Jan-81 0900 JMC*
call susie about sales slip
∂06-Jan-81 0900 JMC*
Call Miro about Watt.
∂06-Jan-81 0900 JMC*
stereo and buchanan terminal
∂06-Jan-81 0920 Buchanan@SUMEX-AIM (Response to message)
Date: 6 Jan 1981 0916-PST
From: Buchanan@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI
In response to your message sent 05 Jan 1981 1944-PST
I am having a modem installed in a replacement for the Datamedia you
loaned me but it is not ready yet. You may certainly reclaim the
DM any time you need it. If your need is not pressing, though, I
will be able to use it for some short but indefinite time until
my terminal is hooked up.
It has been very convenient having it. Thanks.
bgb
-------
∂06-Jan-81 1106 FFL
To: JMC, FFL
Neil Lindstrom returned your phone call. His number is 9 321 0900
∂06-Jan-81 1128 TOB
Dr. Binford will continue to lead a research group in Artificial Intelligence
in robotics and graphics.
His appointment serves special needs in computer-aided design and in
computer science in manufacturing, topics of strong interest now and in the
future. These areas are not otherwise covered in the department.
He will participate in teaching the course in robotics with enrollment about 100
and in one-to-one research supervision.
He is on the ad-hoc committee for
the Medical Information Sciences Program and will participate in its development.
He is involved with the MS program in Computer Science/Artificial Intelligence.
∂06-Jan-81 1155 TOB appointment forms
If you wish any of that rewritten, please let me
know how and I will do so quickly.
Tom
∂06-Jan-81 1159 CSD.SCHREIBER at SU-SCORE Committee meeting
Date: 6 Jan 1981 1159-PST
From: CSD.SCHREIBER at SU-SCORE
Subject: Committee meeting
To: csd.ossher at SU-SCORE, jeb at SU-AI, csd.gischer at SU-SCORE,
csd.clarkson at SU-SCORE, jmc at SU-AI, csd.yao at SU-SCORE,
csd.rwf at SU-SCORE, csl.lantz at SU-SCORE, csd.schreiber at SU-SCORE,
als at SU-AI, csl.sso.owicki at SU-SCORE
We met Tuesday, 6 Jan. Decided to allow Jim Bennett up to 9 hours
to take the exam. Jim Boyce and Jay Gisher will prepare a draft
of our policy on the programming project. Jay and Ken Clarkson
volunteered to help with questions in MTC and AI. We agreed to
draft our proposed questions, make 10 copies, and give them to Carolyn
Tajnai, who will then give each of us a complete set to review
before our next meeting.
NEXT MEETING: Monday, 12 JANUARY at 10:30 AM in 301 MJH.
EXAM date: Saturday 31 JANUARY.
Rob
---
-------
∂06-Jan-81 1942 JMC*
elliott
∂06-Jan-81 2000 JMC*
call susie about sales slip
∂07-Jan-81 0847 FFL
To: JMC, FFL
Carolyn Tajnai asked me to remind you that she needs to know if you or
another person is going to speak at the Colloquium on January 27. She neds
to know today or tomorrow.
∂07-Jan-81 0900 JMC*
stereo
∂07-Jan-81 0900 JMC*
cate
∂07-Jan-81 0951 FFL
To: JMC, FFL
which has priority - the class notes' paper by McCarthy and Hayes or
the permission letters?
I suppose we should get the class notes out serially to give us
more time. I would like "Programs with Common Sense" to distribute
next Tuesday, and 50 copies will be required, since the enrollment
was about 40. It is ready to go now. After that the permission
letters have priority.
∂07-Jan-81 0959 Nilsson at SRI-KL CBCL
Date: 7 Jan 1981 0957-PST
From: Nilsson at SRI-KL
Subject: CBCL
To: jmc at SU-AI
cc: Nilsson
I rcvd your write-up about the common business communication language.
Some folks over here are reading it. Maybe we'll come up with some
sort of plan to collaborate with you in writing a proposal for a
project on which you could be a consultant.
I have some sample galley proofs of some of your AI&Phil material
back from my typesetter. Will send them to you by ID mail and will
then call to get your reactions. -Nils
-------
∂07-Jan-81 1241 Cmiller@SUMEX-AIM MACLISP MANUALS NEEDED
Date: 7 Jan 1981 1221-PST
From: Cmiller@SUMEX-AIM
Subject: MACLISP MANUALS NEEDED
To: STANFORD USERS DISTRIBUTION:
cc: CSD.BERLIN@SU-SCORE
Due to late delivery from MIT there is a shortage of MacLisp manuals
for CS222. If anyone has a manual they can loan them for one or two
months they would be extremely grateful. Make sure your name is in the
manual and get in touch with either Doug Lenat (CSD.LENAT@SU-SCORE) or
Danny Berlin (CSD.BERLIN@SU-SCORE).
-------
∂07-Jan-81 2022 Marimont at SRI-KL EE PhD Qualifying Exam
Date: 7 Jan 1981 2021-PST
From: Marimont at SRI-KL
Subject: EE PhD Qualifying Exam
To: jmc at SU-AI
cc: Marimont
Professor McCarthy,
The EE PhD qualifying exam is only a week away, and I wanted to check
to see whether you had had the chance to discuss your attendence of the
EE faculty meeting at which the exam rsults are determined with Tom
Binford or Larry Manning (the EE professor who is in charge of the exam).
As I mentioned in my message to you of a few weeks back, your willing-
ness to speak on my behalf at the meeting could very possibly be the
deciding factor in my case. Anyway, to make a long story short, I will
be forever in your debt if you talk things over with Tom and Pro-
fessor Manning and decide to attend the meeting.
By the way, Professor Manning told me on Tuesday that the meeting is
tentatively shceduled for Friday 23 January, and that it will probably
take place at three or four p.m. in McCullough 240. Louise Peterson,
a secretary in the EE department office, is in charge of the scheduling,
and I can check back with her to get the final word if you like.
My progress on other fronts includes wangling a research assistantship
out of Tom for this quarter. I'll be working in the AI Lab, presumably
on something related to low-level vision, but I have no details as yet.
My coursework this quarter tentatively consists of statistical signal
processing, partial differential equations, and vector calculus.
Again, I'm very grateful to you for your willingness to consider helping
me in this matter. Naturally I would appreciate hearing from you as soon
as you reach a decision.
Thanks for your help.
Dave.
-------
∂08-Jan-81 0005 Boyer at SRI-F2 (Bob Boyer) Theorem-Proving Prize Meeting
Date: 8 Jan 1981 0002-PST
From: Boyer at SRI-F2 (Bob Boyer)
Subject: Theorem-Proving Prize Meeting
To: jmc at SU-AI
I spoke with Woody tonight. He said you have a proposal. From
what he said, it sounds good. I'm sorry to hear you might be
late. Woddy said he thought the meeting would run from 12 to
2 at the latest. Hope to see you there.
-------
I shall be on time. Woody apparently hasn't yet got the habit of
reading his ARPAnet mail. Do you know where Henry' is?
∂08-Jan-81 0812 Boyer at SRI-F2 (Bob Boyer) Henry's
Date: 8 Jan 1981 0809-PST
From: Boyer at SRI-F2 (Bob Boyer)
Subject: Henry's
To: JMC at SU-AI
Henry's is at the top of the downtown SF Hilton on the corner
or Mason and O'Farrell.
-------
∂08-Jan-81 0857 FFL
To: JMC, FFL
∂07-Jan-81 2317 JMC reprints for cs226
To: FFL at SU-AI
CC: konolige at SRI-KL
Kurt Konolige aka KONOLIGE%SRI is the TA for the course.
He will distribute the reprints and collect.
How long is the delay to get something printed, with and without declaring
an emergency. There is one other paper I would like by Tuesday if it
isn't too hard, namely the article by Fodor in the Scientific American
I left on your desk. I will have some material on knowledge this weekend.
Will it be reasonable to distribute it on Thursday if it is ready to be
printed Monday morning? I won't be in except briefly today (Jan 8).
You will have the PROGRAMS WITH COMMON SENSE on Tuesday. I will try to get
the Fodor article by then. If no serious interruptions, it is possible.
Again the material I receive on Monday probably can be done (if it is straight
manuscript) if I am able to concentrate on it. I hope I have plenty of time
on the McCarthy-Hayes paper. It is difficult (and lengthy) for me in Tex
because of the math. I will just do my best.
∂08-Jan-81 0910 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
The speaker for the Knowledge and Action Seminar on Thursday, January l5,
at 4:15 p.m., in Room 301, Jacks Hall, will be Richard Waldinger of
SRI International.
Bombs, Divorce and the Weather:
The Problem of Action in a Changing World
ABSTRACT
A formalism for planning should be able to deal with the following phenomena:
* Actions may change the names of objects.
* Plans may have loops and branches.
* Once you have performed an action in a
given state, that state no longer exists.
We will concentrate on NONmonotonic reasoning in real-world, blocks-world,
and program-synthesis domains.
∂08-Jan-81 0915 FFL
To: JMC, FFL
Don Chewning of Shugart Associates called. He would like to speak with
you about robotics. He asks that you please call him at 56 733 0100,
ext. l529.
∂08-Jan-81 1250 Denny Brown <CSD.DBROWN at SU-SCORE> Instructor need for CS105 - NOW!!
Date: 8 Jan 1981 1249-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Instructor need for CS105 - NOW!!
To: bboard at SU-SCORE, bboard at SU-AI,
CSD-Faculty: ;
cc: reg at SU-AI
Due to a last-minute schedule conflict, I am without an instructor for
one 10 o'clock section of CS105, the introductory programming course.
The class is being covered temporarily while I try to find an instructor.
Please send suggestions, or even volunteers, to me asap. 7-2274 or
CSD.DBROWN@SCORE. Any visitors, RAs, students, ... out there who
might be interested? -Denny
-------
∂08-Jan-81 2349 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 9 JAN 1981 0250-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Well, I'm back, having learned very little; but the grant
sponsor is happy, and one does get a little somethng from AAAS
meetings. More on upcoming conference tomorrow or next day
including US Snail type notice.
JEP
∂08-Jan-81 2343 KLC comp committee
Hi...I'm supposed to be working on AI questions, and am having a hard time finding
real problems not of the identify terms/work through A*/compare and contrast
type of thing. Could I talk to you tommorrow perhaps? At the meeting Tuesday
we thought it would be a good idea to have some problems distributed to
other committee members before next ∃onday's meeting, but I'm not sure I
have anything good.... Ken Clarkson
I'll be in tomorrow afternoon, but I don't have anything yet either.
∂09-Jan-81 0911 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Colloq.
Date: 9 Jan 1981 0911-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Colloq.
To: jmc at SU-AI
cc: fml at SU-AI
John, time is growing short for colloq. planning for Jan. 27.
Please let me know your plans or those of your substitute.
Thanks,
Carolyn
-------
∂09-Jan-81 1456 ALR at MIT-AI (Andrew L. Ressler) Selector Notation
Date: 9 JAN 1981 1638-EST
From: ALR at MIT-AI (Andrew L. Ressler)
Subject: Selector Notation
To: JMC at SU-AI
Dear Professor McCarthy,
Hi, My name is Andrew Ressler and I am a graduate student working for
Ed Fredkin. I am finishing up a master thesis on the design of a
Conservative Logic Computer. As part of my thesis I refer to the
logic representation for Conservative Logic that we have been calling
Selector Notation. This is where the expression (A -> B, C) means the
value of B if A is one, otherwise the value of C.
Anyway, Ed Fredkin says that you were the original creator of this
notation or something nearly like it. I would like to include a
reference in my thesis to you specifying this notation. Has this ever
been discussed in any papers by you or do you have any suggestions
about what I could use as a reference?
Thank you,
Andrew Ressler
McCarthy, John (1963): "A Basis for Mathematical Theory of Computation",
in Braffort, P. and D. Hirschberg (eds.) %2Computer
Programming and Formal Systems%1, North-Holland Publishing Co. (1963).
treats the mathematical properties of conditional expressions.
The first printed proposal to use conditional expressions in programming
languages is probably in a note in CACM that was printed as a letter to
the editor in 1959 or 1960.
The purely logical ternary connective, i.e. where B and C are required
to be 0 or 1 is much earlier. There is a reference in "Mathematical Logic"
by Alonzo Church, towards the end of one of the chapters on propositional
calculus.
∂09-Jan-81 1607 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
Richard Waldinger will be speaking on
NONnon-monotonic Reasoning rather than
Non-monotonic Reasoning, as previously
announced, at the Knowledge and Action
Seminar on January l5.
∂09-Jan-81 1633 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, January 13
SPEAKER: Vaughan Pratt, MIT and Stanford University
Title: tba
∂10-Jan-81 0110 NEUMANN at SRI-KL [Neumann: VERkshop II]
Date: 10 Jan 1981 0107-PST
From: NEUMANN at SRI-KL
Subject: [Neumann: VERkshop II]
To: JMC at SU-AI
John, I sent the first round of queries out to last year's list of
attendees, but want also to ask whether you are interested. The results
from last year were mostly contained in the SOFTWARE ENGINEERING NOTES
July 1980 issue (ACM SIGSOFT). I don't think you have been particularly
concerned with the security community (which is the origin of Walker's
interest in verification), but you certainly have had a long-standing
interest in verification. HAPPY NEW YEAR in any event! Best regards, Peter
-=-=-=-=-=-=-=-=-=-=-=
Yes, I'm interested, but I'll have to look at the Proceedings to see
whether I would want to make a presentation.
---------------
Date: 7 Jan 1981 1826-PST
From: Neumann
Subject: VERkshop II
To: VERKshop:
Steve Walker appears to have some good news for the verification and
security communities, and is very eager to convene a VERkshop II. He
has asked me to round up the usual suspects. He would like us to meet
in Washington in mid-April (a location that would be of great help to
the government people and less help to the west-coasters among us --
unless you might be returning from April in Paris at the INRIA
Distributed Systems meeting). If there is a great clamor against DC,
the west coast would be OK. (I suspect a DC meeting would cut into
the nongovernmental attendance, while a west-coast meeting would lop
off many government people. Meeting in Texas or Chicago or Omaha
would probably lop off most of both.) How about a one- to two-day
meeting of the verification-oriented people followed by a one- to
two-day meeting of the security-oriented people, maybe three or four
days in all? This note is a questionnaire to solicit your answers to
the following questions.
1. ARE YOU INTERESTED IN ATTENDING THE MEETING IN DC?
2. WOULD YOU ATTEND THE MEETING IF IT WERE IN THE LA OR SF AREA?
3. WHICH OF THE FOLLOWING ARE POSSIBLE FOR YOU?
(3A) 13-15 April?
(3B) 15-17 April?
(3C) 20-22 April?
(3D) 22-24 April?
A fast reply would be appreciated. The hoped-for response is
something like "YES,YES,ACB", where the acceptable dates are listed
in priority order.
Please add any suggestions, outspoken opinions, desires, etc. For
your info, the mailing list is appended. Let me know if you think
there are any glaring omissions. I hope we can avoid last year's
unfortunate conflict with the protocol community.
Happy New Year!
Peter
-=-=-=-=-=-=-=-=-=-=-=
THIS MAILING WENT TO THE FOLLOWING:
VERKshop:
Adrion@NBS-10,
Ames@ISI,
Cerf@ISI,
Cohen@BBN-TENEXB,
MCorasick@BBN-TENEX,
Cullyer,
Darr@RADC-TOPS20,
DEdwards@ISI,
Druffel@isi,
Faust@RADC-Multics,
Fisher@isi,
Denicoff@ISI,
Heafner@NBS-10,
NSF-CS@ISI,
Keeton@ISI,
Lubbes@isi,
Marmor@ISIA,
Millen@RADC-MULTICS,
NRL@ISI,
BPrice@ISI,
Shotting,
JTaylor@ISI,
Tinto@BBN,
Walker@ISI,
Wilson@RADC-Multics,
Balzer@ISIE,
Bledsoe@Utexas,
Bochmann,
Boebert@MIT-Multics,
Bonyun@UCLA-S,
Boyer,
Cohen@UTexas,
Crocker@isie,
Elspas,
Feiertag,
Flon@ISIC,
Gaines@RAND-UNIX,
Gerhart@ISIC,
Goguen,
Good@UTexas,
CCG@SAIL,
DHare,
GHaynes@bbn-TENEXA,
Dick@UCLA-S,
Lamport,
Levitt,
London@ISIC,
Luckham@SU-AI,
Manna@SAIL,
Melliar-Smith,
Moore,
Moriconi,
Musser@isic,
Neumann,
Oppen@SU-AI,
Owicki@SU-SCORE,
Polak@SU-AI,
Popek@UCLA-SECURITY,
Pratt@MIT-AI,
LRobinson,
Schwartz,
SDIM,
SDIN,
Sunshine@ISIE,
DThompson@ISIC,
vonHenke@SU-AI,
Bruce@UCLA-S,
Weissman@isia
-------
---------------
-------
∂10-Jan-81 0148 NEUMANN at SRI-KL
Date: 10 Jan 1981 0147-PST
From: NEUMANN at SRI-KL
To: JMC at SU-AI
cc: NEUMANN at SRI-KL
In-Reply-To: Your message of 10-Jan-81 0116-PST
Don't worry about that so much. Last year was the first in a series.
Polak and Luckham were there. This one may be somewhat different. Once
we know who might be interested, we will try to set the guidelines....
Thanks for your prompt response! P.
-------
∂10-Jan-81 1541 Jim McGrath <JPM at SU-AI> Local utilities
Date: 10 Jan 1981 1532-PST
From: Jim McGrath <JPM at SU-AI>
Subject: Local utilities
To: energy at MIT-MC
I am not aware of why local utilities are granted a monopoly over
thier areas (historically that is). But there are some grounds
for justification:
Technological - with power transmission losses, your local plants
normally supply the bulk of local power. However, superconducting
lines will eliminate this (as SPS will to a certain etent), so in
the future utilities could be organized along national, or continental
lines.
Political control - power (and utilities in general) are a god given
right (we all know this, right?). Thus they are political issues
when something goes wrong (delay in power transmission, raise in
rates). So local politicians want a hand in things. This will not
change, but the feds might muscle the locals out of things (as they
have done in France). But with our history of federalism and the
courts as a means of last resort, do not hold your breath.
Economic - monopolies have some funny economic characteristics.
When well controled (either externally or internally), they can be
more efficient than a pure competitive market.
Organizational - it is easier to keep track of one company than
many. Thus locals like to deal with monopolies (local politicians
that is).
I think we could introduce several power distribution companies,
which are each continental in operation and looked after only by
the feds. They would buy and sell power in an optimum fashion,
and will be the people who set end user rates. Since they will
all use the existing power grid (which can be held by another
company (monopoly) or the fed government), they will be able
to compete for customers.
Power generation would be in the hands of a lot of smaller companies
(each operating only a few stations). They would be regulated
by the feds (ie all nuclear power stations under NRC), and the locals.
They will compete with each other to sell power. I imagine they will
all have slight advantages selling power to be used in the local region
(superconductors will have some loss in application), but generally the
true price of power will be passed on.
Externalities (pollution) - the locals apply a tax to the plant for
air, water polltution. This will tend to drive plants out to
non-populated areas, since damage to human life would be taxed
higher than damage to the general environment. Perhaps Interior
would rec a fed tax to Congress for some of these areas.
Any comments of this plan? Think it could (and should) be done?
Jim
∂10-Jan-81 1917 Boyer at SRI-F2 (Bob Boyer)
Date: 10 Jan 1981 1914-PST
From: Boyer at SRI-F2 (Bob Boyer)
To: JMC at SU-AI
J and I have decided to stay at SRI instead of going to Texas. The
decision was excruciating, but was based upon a preference for doing
research fulltime.
Peter Neumann has promised to invite you to a workshop on program
verification. It will be the second such workshop. The first was
held at SRI last April and I thought it would be awful, but it turned
out to be fairly interesting. I hope that both you and Woody
can make it to the next one, so that theorem-proving is better
represented.
-------
Well, now that you and J have decided to stay at SRI in order to do
research full time, let me mention that for some time I have been
intending to suggest that the Computer Science Department ask one
or both of you to give a course some time. Delay only due to my
limited bureaucratic initiative. Might you be interested?
Anyway, it's good for us here that you are staying.
Peter invited me, and I have accepted.
∂11-Jan-81 1252 Richard Jay Solomon <Solomon at MIT-Multics> Your mail of 11 Jan 1981 1138-PST
Date: 11 January 1981 1545-est
From: Richard Jay Solomon <Solomon at MIT-Multics>
Subject: Your mail of 11 Jan 1981 1138-PST
To: JMC at SU-AI
I forget when I last told you, so forgive me if I repeat anything.
Ithiel had his spleen removed before Christmas and it had a malignancy.
He seems to have lymphoma, but just how far it has spread, or how
serious it is I don't know. He has recovered from his operation and
was in the office last week for the first time on Wednesday; in fact he
gave a 45-minute lecture at the CATV conference I have been organizing!
He is on the phone several hours per day and on the computer as often as
ever. You can call him at 617 547-5717. Other than general lethargy
(from his point of view) from chemotherapy, he seems to be coping fine.
The FEMA project is moving along, and we are scheduling a seminar on
crisis communications for Feb 26. I will be at an OECD meeting in Paris
the week after this; if you have any ideas as to useful people to see in
this area send them on.
richard
∂11-Jan-81 1410 Richard Jay Solomon <Solomon at MIT-Multics> Re: earthquake
Date: 11 January 1981 1630-est
From: Richard Jay Solomon <Solomon at MIT-Multics>
Subject: Re: earthquake
To: JMC at SU-AI
Cc: Solomon.Datanet Pool.Datanet Neuman.ComRes
I have been reading my mail backwards, so I didn't see your earthquake
message before the last reply.
I just typed a long response to your message, and what happened is a
case in point of technological disasters in the highly leveraged
infrastructure towards which we are rapidly moving -- meaning dependence
on computers and word processing and telecommunications.
First, my son tripped over the maze of wires I have going into and out
of terminal, printer, modem, radio & etc. I was fortunate in being able
to reattach to my process after redialling (sometimes I cannot), but
then I hit the wrong key and this marvelous screen-oriented mail system
that Multics has, buried my file in some obscure corner of the disk and
I can't find it. If it was sent on to you, fine. But I think it was
erased. Needless to say I did not make a paper copy.
The second portent of danger was your case-in-point: the head crash.
Fortunately the portable terminal I brought with me on my Xmas vacation
didn't work (somebody at the rental place had wired the modem so it
would only connect with an EIA interface and not via the acoustic
coupler, and of course, TI maintenance didn't work, either, over the
holidays). Therefore I didn't lose any files on SAIL.
The maze of wires is related to the other crisis. I am working in my
living room because my office is minus 10, and I have turned off the oil
heat. Most of the house is cozily, and very economically warmed by a
wood stove; my farmer neighbors get a real kick out of seeing me work
with high-tech equipment next to a wood pile. Next year I have to get a
second wood stove for the rear of the house. All of which makes me
wonder if we would save much in the way of total energy if people work
at home and turn up their thermostats, not to mention make more trips to
the store since they are in the area more often.
Now for earthquakes: we must separate contingency planning between short
and long term, and between that necessary for saving lives and property
for a short-term disaster and providing for some semblance of
continuity, maintenance of productivity, public order (and as FEMA calls
it, "continuity of government" COG), after a major disaster. The last
time I was in Southern Italy I did not notice much in the way of
infrastructure; but apparently what little they had is pretty much wiped
out. For the U.S. today, some complex interconnections could be disprupted
and not only telecommunications. But the disaster being planned for may
be 5, 10 years ahead, when things will be even more complex. There is
also the problem of cross-effects between necessary short-term
preparations and long-term emergency infrastructure. Preparing for one
could be counterproductive for the longer-haul of reconstruction of
simple maintenance of life afterwards. I am thinking of water gone bad
in air-raid shelters, and other examples of lack of follow-through.
Do we know enough to advise agencies now? A book, or videotape or a set
of skill training courses would be a good product from our study. What
we should do now is to locate and capture what we do know. I'm sure you
have a host of computer disasters which could go awry. I am
very concerned about Ma Bell and the dependence of the public on
telephone information during a disaster. Though our entire tiny program
is fully electronic as far as word processing goes (or could be) since
everyone has a terminal at home and even our associates and program
sponsors are getting connected, without the network, I will have
problems out here in the woods. I may back myself by getting standalone
capability, but that still doesn't solve the communication end. I don't
see the slightest evidence that AT&T can handle anything but short term
disasters, nor are they at all prepared for a very likely overload of
the network due to word processors being used in the near future, (other
than to get rid of flat rates). Admittedly, they do handle short-term
outages very well, indeed.
I will be in Paris Jan 19 for an OECD meeting. If you have any
suggestions of people to see in this regard, send them on. Maybe I
should go to Italy, though judging from the papers, what would I learn
except how not to prepare for an emergency?
CC: replies to Pool and Neuman.ComRes%Multics. (Note the different
project names.)
richard
∂11-Jan-81 1902 Sgt. Sally <CSL.LAB.SALLY> Memo to the Faculty about the Forum Annual Meeting
Date: 11 Jan 1981 1145-PST
From: Sgt. Sally <CSL.LAB.SALLY>
Subject: Memo to the Faculty about the Forum Annual Meeting
To: CSD.Dbrown, CSL.JLH
cc: csl.sally at SU-SCORE
Remailed-date: 11 Jan 1981 1901-PST
Remailed-from: Denny Brown <CSD.DBROWN at SU-SCORE>
Remailed-to: CSD-Faculty: ;
I have prepared a memo that needs to be sent out immediately
to the CSD/CSL faculty concerning the Annual Meeting. I know
John is going to be mad about this, but I don't know how to do
it. Could each of you send out the memo to your respective people?
Thanks very much.
To top it all off, the xerox machine in ERL is down this morning
and I can't even send out the stuff in hardcopy form until tomorrow.
Here it is:
STANFORD COMPUTER FORUM
DATE: 11 January 1981
To: Computer Systems/Computer Science Faculty
From: Doug Lenat
Subj: l3th Annual Meeting of the Stanford COmputer Forum
As Conference Chairman for the l3th Annual Meeting of the Stanford
COmputer Forum, it is my pleasure to invite your participation in this
very important event. The dates for this year's meeting have been de-
signated as Thursday, 5 February and Friday, 6 February. Please also
set aside the evening of Thursday, 5 February, for our banquet.
Please note that the site for our Technical Sessions has been moved
to CERAS - room 112. This has an amphitheater configuration and seems
the proper size for our expected attendance. Once again, we will feature
two "Highlight Sessions," as well as five standard sessions. The Forum
Committee wishes to request that first-year faculty members speak for
themselves and that other faculty members designate a finishing student
who has not spoken before to speak for them. Faculty members are dis-
couraged from releasing their time slot to another member. Instead,we
wish to request that they release their claim to that amount of time
this year, as our program will have an abundance of speakers as it is.
Please get in touch with your Session Chairman immediately to give
him the details of your selected topic and speaker. Again, 20 minutes
have been allotted for each talk which also includes the brief discussion
at the end. Highlight sessions have a slight expansiveness, depending
upon the number of speakers selected.
Session Chairmen: Thank you for agreeing to assume responsibility
for this task. Please submit your portion of the program to Sally Burns
(CSL.Sally @ SCORE) by Thursday, 15 January. We are running behind our
usual tardiness this year, so this deadline is a critical one. Would
you also please keep in mind the necessity (sometimes) for moving your
speakers along during the Session. We have had a tendency to slip behind
in the past years. I would appreciate your cooperation in this matter.
Please ask your speakers to submit clear xerox copies of the view-
graphs for their talks to Sally - ERL 234 - by 23 January so that they
amy be compiled into a booklet for distribution at the Meeting. Limit:
10 per speaker. The title of the talk and the name of the speaker
should appear on the first viewgraph.
"Dry runs" are strongly encouraged to enhance the clarity of the talks
and to tighten some of the points. Sheila Manchester at 7-1440 would be
glad to assist you in scheduling ERL 237 for these practice sessions.
The Technical Sessions have been organized as indicated below.
Your enthusiastic participation is enlisted to see that this l3th
Annual Meeting maintains the high standards set by our past twelve
Forum programs.
Session I (Thursday, 5 Feb: 9:l5 - l0:35 a.m.)
J. Ullman, Chr.; M. Flynn, S. Owicki, R. Floyd, D. Luckham
Session II (Thursday, 5 Feb: ll:00 - 12.20 p.m.)
R. Schreiber, Chr; G. Dantzig, J. Oliger, J.Herriot, G. Golub
Session III (Thursda, 5 Feb, 1:30 - 3:00 p.m.)*
B. Reid, Chr; F. Tobagi, D. Knuth, F. Baskett, researchers
Session IV (Thursday, 5 Feb: 3:20 - 5:00 p.m.)
Z. Manna, Chr; T. Winograd, S. Liebes, J. McCarthy, T. Binford
Session V (Friday, 6 Feb: 9:00 - 10:20 a.m.)
B. Buchanan, Chr; E. Feigenbaum, H. Brown, D. Lenat
Session VI (Friday, 6 Feb: 10:50b- 12:15 p.m.)
K. Lantz, Chr; R. Tarjan, J. Hennessy, A. Yao, G. Wiederhold
Session VII (Friday, 6 Feb: 1:30 - 3:00 p.m.*
E. McCluskey, Chr; J. Wakerly, A. Peterson, W.vanCleemput, researchers
* denotes highlight session
-------
∂12-Jan-81 0900 JMC*
reiter and meeting, lindstrom
∂12-Jan-81 1310 JPM Guest account for Chuck Hedrick
John,
Chuck at Rutgers would like to be able to access SAIL if at all
possible. He is responsible for the new PASCAL compiler on LOTS/
SCORE, and in general does good work. I do not anticipate heavy
usage by him, so the account would be more in the way of a courtesy.
Could you fix up an account for him, or forward this to one who could?
Ralph would not have any objections to this as far as I know (although
I have not broached this topic with him yet).
Thanks.
Jim
Ralph has the authority to act on this, not me, so I suggest you take
it up with him.
∂12-Jan-81 1443 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Honoraria for Colloquia speakers
Date: 12 Jan 1981 1440-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Honoraria for Colloquia speakers
To: CSD-Faculty:
For Winter Quarter there are openings for Tuesday, Feb. 3 and
Tuesday, March 3. I would appreciate suggestions for possible
speakers. We are now able to pay honoraria for appropriate speakers.
Thanks,
Carolyn
-------
∂12-Jan-81 1519 CSD.ULLMAN at SU-SCORE Binford papers
Date: 12 Jan 1981 1517-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Binford papers
To: jmc at SU-AI
I hate to hassle you, but I am concerned that we not let the Binford case
slide much longer. How is it coming??
-------
∂12-Jan-81 1556 JPM Chuck Hedrick
Thanks, have forwarded the request for an account for him to Ralph.
Jim
∂12-Jan-81 2021 JMC*
reiter
∂12-Jan-81 2021 JMC*
Connolly about Bryson on jan 26
∂13-Jan-81 0831 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Re: speaker
Date: 13 Jan 1981 0825-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Re: speaker
To: JMC at SU-AI
In-Reply-To: Your message of 12-Jan-81 2218-PST
John,
March 3 is the next open date. Shall I put him in that slot?
I will need a title.
Will you be able to speak on Jan. 27? Time is short.
Carolyn
-------
I'll speak on Jan. 27. Title: Ordinary logic is better than temporal logic.
Reiter's title for March 3 is
Towards a logical reconstruction of database theory.
∂13-Jan-81 1000 JMC*
mazda
∂13-Jan-81 1231 CSD.LENAT at SU-SCORE COMPUTER FORUM REQUEST
Date: 13 Jan 1981 1228-PST
From: CSD.LENAT at SU-SCORE
Subject: COMPUTER FORUM REQUEST
To: JMC at SU-AI
cc: CSD.DBROWN at SU-SCORE, CSL.LAB.SALLY at SU-SCORE
John,
Zohar will be out of town during this year's forum meeting, and
we need to assemble the schedule very soon. I would greatly
appreciate it if you would chair the session you're in
(represetnatives of you, Liebes, Binford, Winograd, and Manna).
In return, I will be happy to arrange the actual list of
speakers, and you need only serve as chairman during the
meeting itself. Furthermore, I wanted Doyle to talk during
the Forum, but we had to exclude him due to his category of
appointment. Doing this favor for us allows me to make an
exception of your session, and include him, too, as a speaker.
Since Doyle's talk will be one of the high points of the
Forum, everyone will benefit from this arrangement.
Thanks in advance -- let me know ASAP.
Doug
(Thu Feb 5, 3:20-5:00pm)
-------
∂13-Jan-81 1236 CSD.LENAT at SU-SCORE Computer Forum
Date: 13 Jan 1981 1234-PST
From: CSD.LENAT at SU-SCORE
Subject: Computer Forum
To: zm at SU-AI, tob at SU-AI, sl at SU-AI, tw at SU-AI, jmc at SU-AI, jd at SU-AI
cc: dbl at SU-AI
Please send the name of the student (or yourself) who will
be speaking in your "slot" at this year's Forum meeting to ME (DBL@SAIL).
The official Forum policy this year is to encourage people not to
trade slots with other faculty members, so if (i) you have no
near-terminal student who hasn't spoken at the Forum ever, and
(ii) you have personally spoken at the Forum within the last
few years, then it is OK to simply leave your slot vacant this
year. With 6 people on the schedule for one session, we HOPE
that one or two of you will not be represented this year. THis
should even out in the long run, and will give YOUR students
more time to speak when you DO have someone to present. Thanks.
I will let you know who your session chariman will be shortly.
-- Doug
-------
I guess I've spoken fairly recently, and I don't have a finishing
student who hasn't spoken.
∂13-Jan-81 1247 CG
Richard asked that I remmind you that the SEUS demo is at 3:30 today
∂13-Jan-81 1715 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, January 20
SPEAKER: Amir Pnueli, The Weizmann Institute, Rehovot, Israel
TITLE: Impartiality, Justice and Fairness:
The Ethics of Concurrent Termination
ABSTRACT:
The method of proving termination of programs by well-founded
sets is extended to apply to concurrent programs. The interesting case
is programs that terminate only for fair executions. Different flavors
of fairness are defined which lead to different proof methods. The
methods are shown to be complete.
∂14-Jan-81 0706 Boyer at SRI-F2 (Bob Boyer) Course
Date: 14 Jan 1981 0707-PST
From: Boyer at SRI-F2 (Bob Boyer)
Subject: Course
To: JMC at SU-AI
I'm really pleased at your suggestion about a course at
Stanford, but right now neither J nor I has the time or
inclination to work up a course.
However, I definitely look forward to giving a lecture in
your LISP course next Fall. It's possible that by then J
and I will have converted our theorem-prover from Interlisp
to Maclisp, so some of the students could even try it out.
I'm glad you're coming to the Verkshop. Woody has also
indicated he will come.
-------
See letter from them in my OUT box.
They have evidently changed the name of their reports distributor
on their computer. Please write them a letter requesting that
the report announcements be sent by ARPAnet mail.
∂14-Jan-81 1646 MAILER failed mail returned
The following message was undeliverable to recipient(s)
doles at RUTGERS:
∂14-Jan-81 1644 JMC mailing list
To: doles at RUTGERS
This is to request the transfer of my name from your paper mailing
list for reports to your ARPAnet mailing list.
John McCarthy
Computer Science Department
Stanford, CA 94305
the ARPAnet address is JMC@SU-AI
The following message was received from the remote host:
No such mailbox at this site.
∂14-Jan-81 2000 JMC*
keyboa
∂14-Jan-81 2100 JMC*
nail
∂14-Jan-81 2212 CLT
i am at jax to work for a while
∂14-Jan-81 2227 TOB
Dear John
I urge you to attend the meeting in EE concerning quals, Jan 23. I think
Dave Marimont has mentioned this to you. In case his score is
in the middle, your support would help greatly. I do plan to support
him and think well of him. Please let me know what you plan to do.
Best regards
Tom
Can't I find out whether myy presence is likely to be decisive first?
Has he taken the exams yet?
∂14-Jan-81 2244 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
Date: 14 Jan 1981 2237-PST
From: Keith A. Lantz <CSL.LANTZ at SU-SCORE>
Subject: Re: Comp Ctte Meetings
To: CSD.SCHREIBER at SU-SCORE, csd.ossher at SU-SCORE, jeb at SU-AI,
csd.gischer at SU-SCORE, csd.clarkson at SU-SCORE, jmc at SU-AI,
csd.yao at SU-SCORE, csd.rwf at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
cc: csd.tajnai at SU-SCORE, csd.dbrown at SU-SCORE, CSL.LANTZ at SU-SCORE
In-Reply-To: Your message of 18-Dec-80 1235-PST
A sample SCRIBE format file for comprehensive questions (and solutions)
is <CSL.LANTZ>COMPMODEL.MSS. The resulting PRESS files are
[SAIL]
COMPM.PRE[1,KAL] and [SAIL]
COMPMA.PRE[1,KAL], with and without
solutions respectively. Read the comments embedded in the manuscript
file and you should do alright. Fire away if you have any problems.
Keith
-------
∂14-Jan-81 2249 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
Date: 14 Jan 1981 2248-PST
From: Keith A. Lantz <CSL.LANTZ at SU-SCORE>
Subject: Re: Comp Ctte Meetings
To: CSD.SCHREIBER at SU-SCORE, csd.ossher at SU-SCORE, jeb at SU-AI,
csd.gischer at SU-SCORE, csd.clarkson at SU-SCORE, jmc at SU-AI,
csd.yao at SU-SCORE, csd.rwf at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
cc: csd.tajnai at SU-SCORE, csd.dbrown at SU-SCORE, CSL.LANTZ at SU-SCORE
In-Reply-To: Your message of 18-Dec-80 1235-PST
Subject: [Keith A. Lantz <CSL.LANTZ>: PRESS/DOVER]
Note the following message. It impacts on YOU when you print your
questions/solutions. If the file is created on SCORE it will almost
certainly be filed under your login account at SCORE. Then again, you
shouldn't print anything unless you are going to be there to pick it
up!
Keith
---------------
Date: 14 Jan 1981 2243-PST
From: Keith A. Lantz <CSL.LANTZ>
Subject: PRESS/DOVER
To: reg at SU-AI, csd.nowicki
cc: csl.lantz
There is still at least one major annoying bug in these programs --
the fact that they take both the file name and the user name OUT of
the PRESS file. This is incredibly difficult to track down when I print
Internet Experimental Notes created at ISI -- they are usually filed under
(Jon) Postel, for example. Similarly with any PRESS files FTPed from
SCORE. Please fix this before I lose something sensitive -- e.g. the
comprehensive exam problems!!!!
Keith
-------
---------------
-------
∂15-Jan-81 0140 REM at MIT-MC (Robert Elton Maas)
Date: 15 JAN 1981 0441-EST
From: REM at MIT-MC (Robert Elton Maas)
To: JMC at SU-AI
We're in the process of changing from a territorial/nationalist mentality
to a world-view, from USA is better than USSR toward all peaceloving
scientists are better than all military warmongers. I'm not sure what
the best transition method is. (Gee, if only we had a "transition team"
for that mounumental task!)
∂15-Jan-81 0344 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #10
Date: 15 JAN 1981 0641-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #10
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Thursday, 15 Jan 1981 Volume 3 : Issue 10
Today's Topics:
Security - Control of Cryptography Research,
Home Info Services - Source Experience & CompuServ,
Humor - You're Under Arrest
----------------------------------------------------------------------
Date: 14 January 1981 14:47 est
From: Richard Lamson at MIT-Multics
Subject: Cryptography and National Security
Reply-To: Lamson at MIT-Multics
I can't believe that the red-herring ghost of *National
Security* has been brought out on the cryptographic research
issue. Frankly, judging from what the probable new leader of
the "free" world's diplomatic corps, Alexander Haig, had to
say about "National Security" this week during confirmation
hearings, I would prefer to forgo it in favor of some genuine
attempt to cooperate with "our" enemies.
-- Richard Lamson (Lamson@MIT-Multics)
------------------------------
Date: 14 Jan 1981 1023-PST
From: MOORE at USC-ISIB
Subject: Home Info Services - Source Experience
Subject: Not Ready for Business Applications
I agree with the comment that the SOURCE is not appropriate
(yet) for business usage. My wife & I were investigating the
possibility of setting up a bookkeeping system on the SOURCE,
using the packages they advertised as working. What followed
was similar to the reported experience of others:
Documentation was nearly impossible to get, there was no one
there who was a competent consultant on the package; it did
seem to work on the initial test data, but no satisfactory
way was found to bring up a version for private use, there
were clearly necessary customizations but no advice on how
to achieve them. I am sure there is some way to run their
system with private data, without a complete copy of the
programs in our own directory -- but I could find no way,
with the time/effort I was willing to spend.
I maintain my account & occasionally peek in to see what's new.
I do think that a la{oo'f)1β;∂#'?;>K∪∃βNs≠?Kn S'?rβWS'fKSd4VKMβπrβ'∪↔
β←#?≡)βS'n)β'Mε≠3?O*↓55βε+K#ππ→βS#*αN>V∀~∃β'~β'Q8hRπ3OzaβS#/Iβ←?.c⊃β∪zβS#↔o≠↔3[/→↓#πv!βS#.KIβW≡+KM%ε β≠π6{Iβπv 4+∂F ;∨∃π##↔'∩β '3fK;≥βε{3'∂Jq↓αO'∪π'∨G!β∂?vs↔∂Qπ#'7∃ε ∂∂?.sS';8h+∂K. S↔Mεs↔∨π&K[∃βNs∂↔;&K[∃β6{IβSF+5βSzβ ∃βn{K∃β.3≠'∂N+;Qβ∞s⊂4+6{Iβ;/9βWO/∪MβSzβ∪=β&C∃βO␈∪Qβ?2↓ C3∂K';≥∩βS#π"β←'3bβ3↔π"βS<4V3';∪Ns≥βSF)β7?∨!βπS'∪π∂SO3∃βπ∨β↔∂S~β?→β&C∃βOO≠S↔5r↓αC↔⊗CπCLhSS#↔JβO#?.c⊃βπg≠=βK.≠?;ON#↔Iβ&C↔'Iα∪';'&KπS'}qβ≠↔*⊃↓55∧K→βSF+'H4W≠↔K[N≠∃β'~β[π3. 3∃bβ7?K*βC↔?εc∃β←Nc1β∪O≠∂?[/⊃βS#O→βπ;"β ↔∂}k∀4+⊗+∨W3∂⊃β∂W∨#?7↔↔→β'→π##↔K*β'O9?!β¬β⊗K≥1βv{97K.≠?[↔⊗ 3∃ε≠?OPhSWAβ7∪?;Qph(4*VK5α7}{K∀4Ph)55ji555ji555ji555ji555ji555ji44(hR∪πS+Q↓EQ∧Sπ9↓ IaE↓E;↔∪;/≠∪πeJ↓AeA~jBNPhR≠K?kQα"&∀:⊗2Qε Qα2da66~(h*OW⊗S↔∂QRα∂?7π*O↔K2↓=α7N≠K>;/ 4(4TK9βK/≠C?;≡)βS=π##∃α≡{7CV≡+KYβ∂+↔KeαB
R]εC?]β∂∪∃βSF+eβOε+33'v84+SF+'Iβv 7∃βv{]⎇%ph(4*Jβ3??↑+⊃βπ"α7'∂⊗z;↔Qε ?W"↓aβ7␈→9βπ>yβ≠K}iβS#*βC↔K∨β↔∂SO3∃β?0h+¬β6{K7↔∩β↔7Cf{g↔∃r↓α%β&yβ;?"β/;?:βπ ?/!βS#.KIβ7∞K1βOO≠S↔5ε{H4+&C↔'Iε3π∂'fKSeβ6{Iβ7N≠K=β≡{≠S←∂∪∃9↓¬##↔eε K∃β↔+;;'v9α∪↔~iECM`h+π3bα.%β␈⊃α.1∧~BWMr↓αg?*β#π[*βπ∂∂/≠MβSzβS#∃εkπ∂#Ns∃βπv!β∧4W≠W O/!β?→π##↔'∩βO?≠';πK∃r↓αS#*βOW ≡+Qβg␈)β∨↔"β↔c∂g+∪↔Mε ;e4hSS#'v9βS#∂!β'MπβK?C⊗K↔Sπ↔Iβπ;"βK↔G.KK↔Mπ##πQπ##∃β∂+S#?↔→β ∀hS∨'[.qβ¬β⊗{gπ3'I↓#πv!β#↔v≠∃β↔F+∂WSN{9βSNk∃β7∂∪-βWαI84(hRO';≡)βS#/IβπCε+πIβ&yβK↔∨#K'∂"βπ∂∂/≠MβSzβ;?9oβK'7*βS'7*`4+SF+K∃βO→β3''#3∃βNsS↔K6+K↔;≡)β≠K}iβS#.KIβ∂}k7↔K≡Kπ1β≡c'↔;'_4)#∞cS#?.;!βSF+K∃βn eβ *βO?7*I84(hR≠?Iπ##?O*β←#=ε≠πK∃bβS#∃ε{C↔K∂#';≥π≠gOS.iβ'Mεs?Qα$zBM5 ↓84*≡{7CV≡+KYβ&K[↔K>+⊃βπ⊗{WQβ⊗+3↔π≡)↓U9β↓β?→¬">BMk A9↓¬##'LhS∨↔;/∪π33Jβ7↔πw→βS#∂!β≠↔∂#WK↔~β'9β&C∃β↔∂∪3'↔∩α∩⊗
εk?;'&{KL4V K∃βπ∪↔O↔w!1β /!βS#*β3πS/⊃β≠↔∂#WK↔~βπK↔r;Q1β∂!β3↔∂≠Qβ'ph+S#*βOπ7*β←πeph(4*&C∃βC⊗{∪W∂"β7π;∞;↔Iβ/≠↔⊃β&yβ ∃∧k'/∃¬;πK⊃αCπ;⊃πβK? ∞∪3d4W≠S'3bβ'M%r↓αS#*βC#?v)β'9∧≠?3Wn∪WMβO→↓!Y !%↓Q+95aYβ4)#.s3↔O~βO↔;Nc'SeεKMβ∨/#S';:β7∃%r↓αS#/Iβ#π6)β?≠6K∂↔MεK84+n{OQβn +?Iε≠'S'/→↓#πv!βO?n)β;?"βO=βn +?Iε≠'S'/→%84Ph*S#.KIβS.≠!βO/βC?K"β≠?Iε≠?77/∪∂'πbβ∂?7ε ;'↔~β'Mβ∞∪?WQπ##πPhSOπ7*βπMβ∞sg?;*β'9β&C∃βSNk↔O#∂∪';≥ε∪WO'v+OM9αα%β∪}q∨Qβ↑s?\4W;#πQπ##↔eπ;'31ε#=β←O#!α7N≠K>;/!βWO/∪M9↓¬##↔'∩βOSπ61β#π_h+O?n)β[↔↔Iβ∨?}!βC↔␈β3∃β∞s⊃βSF+eβO.+5βSzβSKeεCπK⊃π#=β/.+@4+&C↔K∃π∪↔π3Jβ∂WO&{7↔K~β#πCπI9↓α∞;π'9bα%β∪}q∨Qβ↑s?]β>CπP4W;'31εCπCC.qβ←'&Aα7'∨∪>;↔"βC↔?εc∃84Ph)55ji555ji555ji555ji555ji555ji44(hR∪πS+Q↓EM∧Sπ9↓ IaE↓ AM6¬~P4*7∪?5i∧
∩BN~βπQα-~
6&≤H4*O.∪+↔∂#Qα?;gIβ'9∧ 7↔KN≠¬1β␈⊃α7?⊗)α≠π⊗c↔MβNqβ?W∩αS'7(h(4*7∪?5β&C∃α←∂≠#';?#?9αε{OQα↔+O';/≠MβO.≠S'?rα7?;& eα+∞q↓EIb↓Eea
`4+?⊗K∨';∞c3eβ7∪?5α∨#W∪↔w!α3π?K↔I8hP4))∧ α∂?f{Kπ∪zβ←?7∞qβOS∞∪ ↔⊃εC↔IβG+O πv!β←'&Aβ¬βn+πQβ6{K-9αα#∀4R↓β ↔g#↔⊃βF+Iβ←O#!β¬π#'K∃εKK?9r↓α∪'≡{K∪↔⊗ceβ∂}s∪W∂#y↓α;zq↓αSF(4)↓εc?∂πbβ∪'O'∪'∂Qε SS?⊗s↔eβ≡CπK∨.!βS#*β∂?Wεc∃β←O#!β∪.+3';:p4)↓¬##πQ?→β¬β6+3?;Jβ'9α≡{3?K∞#=84Ph))α
α∪↔M∧k?';/→β+W&;∃βC⊗+CπKNs≥βSzβO↔;&+;∂∃ε β∂?w3'∂S.!β W⊗;3πHhQ↓β←∂→β3?≡[↔⊃β␈+Qβ?2β#'Mε≠?WK'∪??5r↓απMπ##∃βW+∪∨∃ε ;⊃β↔+'3∪Ns≤4)αβ+π;O#?Iβ7+7 3.!β←'&AβS#*β∪??∩aβS#*β WK>cπIβ∨#↔CC.!β'9ε ;⊃0hQ↓βW≡K;≥β
βCπC/⊃β∂3O↓1β;∞K1β≠Nc∃βπv!βC3.k ↔I?→β←K.s∂!1ε{C↔;. 4)↓εKQβ←O##'9π≠↔∂?v#M84R↓↓↓↓∃##';Zβ;?SFK;≥β}1β'Qb⊃βS#*β WK>cπIβ⊗+C3'.!β←#.qβS#*β+W∪>(4)↓α↓↓β?63↔K↔"βS#πv[M9↓α∩'Q∨~β¬β7∂#S↔Iε{→βC⊗{≠↔O≡K?;πbβ∂?W↔#↔Oer⊂4)↓¬##∃βW+∪∨∃π##↔9π≠↔;S.s∂↔⊃εC'5β&yβS#*β7πcNkW5↓ ↓βg↔∂∪Mβ'rβ+π'bp4)↓α↓↓↓
N1α%βv+↔⊃βFK5βπ> '91∩β#'MεC?;?∩β'Mβ⊗+C?K&+⊃βSzβ#π[(h)↓↓α↓↓βK.kπK/.!1↓
Jβ/;?:β←#↔⊗)β#∃>c1β *qλ4(hQ555ji555ji555ji555ji555ji555jh4(4T+;⊃β}1α"Vl
96:-"Mα∪N;↔OPhQ)))RQ)))RQ)))RQ)))RQ)))Ph(4(0=E*j+π9kAE↓↓β!AD&⎇"¬↓M~Bε∞*α∪'∨/≠Q↓hRS=i¬~Bε∞,α6&Qlj 4(hRNBε≤)α∪'>+OP4Ri555ji555ji555ji555ji555ji555ji555ji555ji555ji555ji555ji555ji555ji555hh(4*& S∃iβ Qα+∞q↓EeC ↓EQ+96BN h*≠K}iiαS.!απ;&+KO?r↓r>R
βπQα≥)6ε%ph*OW⊗S↔∂QRα;?]π##πQπK?Uβn+;S'}qβS#*βO#W'#3∃1π≠↔[↔⊗ 1βK.cπS↔"βGW↔∨#'?;~β∂?7*β∧q to mind.
To: space-enthusiasts at MIT-MC
1) Has NASA seen the light WRT taking the shuttle tanks all the way to
LEO instead of throwing them away 5% (or whatever) short of orbit?
2) Has anyone heard who slated for head of NASA in the Reagan administration?
3) Will the first flight have any means of checking the tile shield for
integrity once in orbit but before reentry?
4) If the answer to the previous question is yes, consider the following
horrible situation: Shuttle launches without too much trouble but
after arriving in orbit they discover that a whole patch of tiles
was ripped away and there is virtually no chance of the shuttle surviving
reentry. As I understand it they are not bringing along a tile repair
kit on the first (several?) mission. What do they do? They have at least
3 days to stew about it, then what? Does anyone know what the plans for
this sort of a scenario are?
5) Related to the above question, are they bringing along on the first
mission a space suit? It seems that this would be invaluable for in orbit
repair. The tile problem is just the most publicized of the possible
external problems.
6) Will the astronauts be able to enter the payload bay on the first mission?
Is it pressurized?
I would appreciate answers to these questions if anyone has them.
Thanks,
Ted Anderson
------------------------------
Date: 15 JAN 1981 0353-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: [OTTO at WHARTON-10 (George Otto): Thought for the Day!]
To: REM at MIT-MC, SPACE at MIT-MC
Dyson spheres in 100 years? Phil Morrison told Carl Sagan and
Bruce Murray at the "Saturn and the Mind of Man" sumposium that
"the protons will decay first" before we build Dyson spheres.
I suspect that 100 years is a bit optimistic, but closer
to truth than Morrison.
Unless, of course, the Black Box people have their way;
after all, Sagan and Murray used the "Saturn Symposum" as a
forum to badmouth the Apollo program...
------------------------------
Date: 15 JAN 1981 0357-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: The Planetary Society (and other space advocacy groups)
To: SPACE at MIT-MC, Hamilton.ES at PARC-MAXC
List of space organizations was well done, but
unfortunately left out:
American Astronautical Society (AAS)
which stands somewhere between L-5 and AIAA in
"far-outedness". Although I am a member, In confess I don't
recall whether AAS insists on all grades of members having
professional qualifications (there are at least three grades).
JEP
------------------------------
End of SPACE Digest
*******************
∂15-Jan-81 0700 TOB
John
I will find out.
Tom
∂15-Jan-81 0016 JMC
Can't I find out whether myy presence is likely to be decisive first?
Has he taken the exams yet?
∂15-Jan-81 0836 USCHOLD at RUTGERS How to get more water
Date: 15 Jan 1981 1125-EST
From: USCHOLD at RUTGERS
Subject: How to get more water
To: energy at MIT-AI
cc: uschold at RUTGERS
The current serious drought here in the east got me
thinking... Everybody knows that to make electricity, one boils
water. Why not, for coastal areas at least, boil sea water instead
and add the resultant distilled water to the water supply? Can anyone
think of why this may not be feasible? One advantage is the lower
boiling point of salt water. The efficiency of conversion of coal,
oil, or whatever to electricity should increase. The only major
problem I can think of is how to deal with the salt. It would take
some redesigning, of the plants, but it shouldn't be /that\ difficult.
Besides we can use the salt for roads in winter and for health nuts
who only buy "sea salt" at their local health food store.
The water may not be very tasty, but we could certainly flush
our toilets with it! Perhaps the amount of water produced this way
would be insignificant, but it seems worth looking into. Can anyone out
there more knowledgeable than I shed some light on the feasibility of
this idea?
Mike
-------
∂15-Jan-81 0847 JGA at MIT-MC (John G. Aspinall) How to get more water?
Date: 15 JAN 1981 1140-EST
From: JGA at MIT-MC (John G. Aspinall)
Subject: How to get more water?
To: USCHOLD at RUTGERS
CC: ENERGY at MIT-MC
Date: 15 Jan 1981 1125-EST
From: USCHOLD at RUTGERS
The current serious drought here in the east got me
thinking... Everybody knows that to make electricity, one boils
water.
The water that is boiled in electricity-producing power plants is
the working fluid in a ** CLOSED CYCLE ** ! It must be kept quite
pure (cleaner than our tap water) or else crap will plug up heat
exchangers, turbines, etc. And it is certinly not boiled off to
atmosphere under normal operating conditions. Thermodynamically,
that would be disaster.
John.
∂15-Jan-81 0900 CSD.JEANIE at SU-SCORE Agenda for Faculty Meeting
Date: 15 Jan 1981 0855-PST
From: CSD.JEANIE at SU-SCORE
Subject: Agenda for Faculty Meeting
To: CSD-Faculty: ;
Agenda for Faculty Meeting Tues. Jan. 20 2:30 pm
New Business:
1. MS Enrollments
2. Comprehensive Exam Contents
3. Industrial Contracts Policy
4. Western Institute for Computer Science
-------
∂15-Jan-81 0900 JMC*
Call Fredkin
∂15-Jan-81 0904 CSD.ULLMAN at SU-SCORE Re: visitors
Date: 15 Jan 1981 0904-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Re: visitors
To: JMC at SU-AI
In-Reply-To: Your message of 14-Jan-81 1702-PST
Naturally I'll approve anyone you or any other member of the faculty sponsors
(although if space is involved, you have to provide it from your "territory").
My personal feeling is that people who invite themselves to learn what we
are doing should only be taken if there is some hope we shall learn something
in turn.
-------
∂15-Jan-81 0914 Richard Treitel <CSL.VER.RJT at SU-SCORE> Re: my opinion
Date: 15 Jan 1981 0914-PST
From: Richard Treitel <CSL.VER.RJT at SU-SCORE>
Subject: Re: my opinion
To: JMC at SU-AI
cc: CSL.VER.RJT at SU-SCORE
In-Reply-To: Your message of 14-Jan-81 1739-PST
I was largely trying to needle ARK, who believes that SAIL/WAITS is the
ultimate pinnacle of human achievement and that nothing can possibly be
better unless it includes all the features of the old system. Of
course SAIL has many good features, which I would be glad to try and keep.
-------
∂15-Jan-81 1147 NEUMANN at SRI-KL VERkshop II progress
Date: 15 Jan 1981 1130-PST
From: NEUMANN at SRI-KL
Subject: VERkshop II progress
To: VERKshop: ;
The most likely dates for VERkshop II appear to be 21-23 April,
Tues-Thurs, probably in the DC area. Steve is currently negotiating
with NBS to see if we can hold it there.
Firmer plans, specific goals of the VERkshop, topics to be discussed,
desired contributions, etc., will follow shortly to aid in your
planning. Please continue to send in your suggestions. You might
begin to think about what position statements you might wish to write.
I would again like to collect them on-line, and distribute them in
advance. They could be self-standing or incremental to last year's.
If you wish, they should be suitable for inclusion in the July 1981
issue of the SIGSOFT Sofware Engineering Notes (SEN). (You might also
care to submit something even if you cannot attend!)
This scheduling will leave you West Coasters Monday and/or Friday to
travel if you prefer, or otherwise to visit in the East. Almost all
Westerners indicated willingness (or in some cases even desire) to go
East. Thus it seemed reasonable to accommodate the government people
who were urging us to hold it in the East this time. Excellent
government attendance is expected, including Vint Cerf -- who has
taken over verification and security work at ARPA.
Many thanks to those of you who responded to the initial query. The
response was very strong. More soon. Peter
-------
∂15-Jan-81 1243 DBL Session at the COmputer Forum
To: TOB, SL, JD, JMC
OK, the way the session looks,
TOB, SL, and JD wll talk (or be represented by a student of theirs);
I would like JMC to chair. This will be the opening session,
and you will have one hour for these three talks (we had to
rearrange things a bit).
Tom, Sid, John D.: I need your speaker name and talk title TODAY.
John M.: please let me know TODAY if you are willing to chair; I
would like you to introduce the other folk and put their
work in context for the audience.
Thanks; sorry about the last-minute rush!
Doug
∂15-Jan-81 1307 SL Computer Forum Speaker/Title
To: DBL at SU-AI
CC: JMC at SU-AI, TOB at SU-AI, JD at SU-AI
Speaker: Sidney Liebes, Jr.
Title: "Stereo Image Analysis: Object Space Invariants and Cultural Objects"
∂15-Jan-81 1243 DBL Session at the COmputer Forum
To: TOB, SL, JD, JMC
OK, the way the session looks,
TOB, SL, and JD wll talk (or be represented by a student of theirs);
I would like JMC to chair. This will be the opening session,
and you will have one hour for these three talks (we had to
rearrange things a bit).
Tom, Sid, John D.: I need your speaker name and talk title TODAY.
John M.: please let me know TODAY if you are willing to chair; I
would like you to introduce the other folk and put their
work in context for the audience.
Thanks; sorry about the last-minute rush!
Doug
I would be glad to chair if I am here. There is a possibility that
I will have to make a trip to France at that time. I had hoped to
know by now, but maybe I can make a phone call and find out.
∂15-Jan-81 2038 Denny Brown <CSD.DBROWN at SU-SCORE> Gray Tuesday/Course Homework
Date: 15 Jan 1981 1702-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Gray Tuesday/Course Homework
To: CSD-Faculty: ;
cc: csd.tajnai at SU-SCORE, CSD.BSCOTT at SU-SCORE, reg at SU-AI, jeb at SU-AI,
sgn at SU-AI
We plan to schedule the Gray Tuesday Meeting for Feb. 10. If that
date poses problems, contact me quickly and we'll re-plan.
Also, I'd like to remind/notify everyone that at the faculty meeting
last week it was decided that we would refuse to accept homework done
on CSD facilities (SCORE, SAIL, ALTOS, SUMEX) for CS classes. The
minutes of the meeting should be available soon, but students should
be notified of the new policy immediately.
-Denny
-------
∂15-Jan-81 2043 Keith A. Lantz <CSL.LANTZ at SU-SCORE> Re: Comp Ctte Meetings
Date: 15 Jan 1981 1735-PST
From: Keith A. Lantz <CSL.LANTZ at SU-SCORE>
Subject: Re: Comp Ctte Meetings
To: CSD.TAJNAI at SU-SCORE, CSD.SCHREIBER at SU-SCORE, csd.ossher at SU-SCORE,
jeb at SU-AI, csd.gischer at SU-SCORE, csd.clarkson at SU-SCORE,
jmc at SU-AI, csd.yao at SU-SCORE, csd.rwf at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
cc: csd.dbrown at SU-SCORE, CSL.LANTZ at SU-SCORE
In-Reply-To: Your message of 15-Jan-81 1326-PST
SEveryone should submit at least one copy of their questions to you by
tomorrow morning, so they can pick up complete sets by afternoon.
I, at least, will have things SCRIBED and taken off the Dover. Actually,
if everyone makes a SCRIBE file of their questions, I will volunteer
to merge them... unless someone else wants the pleasure.
Keith
-------
∂16-Jan-81 0251 POURNE@MIT-MC conference
From: POURNE@MIT-MC
Date: 01/16/81 04:17:16
Subject: conference
POURNE@MIT-MC 01/16/81 04:17:16 Re: conference
To: JMC at MIT-MC, POURNELLE at MIT-MC
Conference growing, but in hand. Looks good. We may not write
THE space plan of the US, but we've at least an input at a place
where inputs count.
If you have clearances, I urge you to get yours in to
Col. Kane, since the phasing with military space plan looks
critical. There will almost undoubtedly be a military presence
in space, but our job is to save a civilian space program...
Love and plenty kisses, JEP
∂16-Jan-81 0401 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 15 Jan 1981 1356-PST
From: Alan R. Katz <KATZ at USC-ISIF>
Subject: answers toOTA
To: space at MIT-MC
cc: katz at USC-ISIF
1. NASA does not now have any plans to put the tanks in orbit, but maybe
we can convince them otherwise.
2. I heard that Hans Mark (previous head of Ames research center and Sec. of
the Air Force) was to be head of NASA (rumor only)
3. They will not be checking the tiles, which is why they are not bringing
a repair kit. If some come off, we will find out about it when then try
to land. The rational is that EVA to check the tiles introduces more risk
than its worth.
Alan
-------
------------------------------
Date: 15 January 1981 22:10-EST
From: Daniel L. Weinreb <dlw at MIT-AI>
Subject: SPACE Digest
To: SPACE at MIT-MC
About the space shuttle, the story I get is that they are definitely
taking space suits some time, as there are planned EVAs. With slightly
less certainty on the part of my source, the general plan is to always
have space suits for the crew, but that when there are a lot of
passengers there will not be room, so for each passenger (presumably
with spares) there will be a cute little device which is basically a
balloon that you get inside (you have to scrunch). I don't know about
the other questions.
------------------------------
End of SPACE Digest
*******************
∂16-Jan-81 0427 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #11
Date: 16 JAN 1981 0708-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #11
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Friday, 16 Jan 1981 Volume 3 : Issue 11
Today's Topics:
Security - Control of Cryptography Research
----------------------------------------------------------------------
Date: 15 Jan 1981 0757-PST
From: HARMON at USC-ISI
Subject: National Security & Cooperation
In reply to Lamson's comments on national security:
Cooperation with our enemies really only works right if they
wish to cooperate with us. However, it is a nice idea if the
situation is right.
Scott Harmon
------------------------------
Date: 15 Jan 1981 1936-PST
From: Mike Leavitt <LEAVITT at USC-ISI>
Subject: CRYPTO RESEARCH
Does anyone know whether the government has set up a
system of prior restraint as NSA is now suggesting, in any
other academic field? Nuclear physics? Things related to
chemical/biological warfare? If there is no history of
prior restraint in peacetime, where the government is not
sponsoring the research, I don't believe NSA has a leg to
stand on. If there is such a history of peacetime prior
restraint on publication, I suspect NSA can get away with
it. This is totally independent of whether or not one
approves of such governmental actions.
Mike
------------------------------
Date: 15 Jan 1981 (Thursday) 0747-EDT
From: DREIFU at WHARTON-10 (Henry Dreifus)
Subject: Crypto & facts of Life.
There is the 'other side of the looking glass perspective'
which HNT really has not explored in too much detail. I
realize that this will raise lots of eye-brows and blood
pressures so I have changed the scheduling of our mail
server accordingly.
Basically, if we 'consider' the NSA / CIA / ??? / ??? side
of the coin it looks somewhat different: All this fantastic
Cryptology will help preserve our integrity worldwide.
Unfortunately we live in a world that is not ideal, and
secrets are mandatory. In order to maintain the integrity
of these secrets, we would need unpublicized algorithms.
In addition, if say MIT or PENN came out with some really
beautiful solution to all other Encrypters, it would be
somewhat less-than-desirable to give it to some of our
'less-favored-nations'.
If the N.S.A. and the like are sponsoring the research, and
soliciting help from Universities, then some of the results
should not be published. There are probably many professors
on this list who consult in TopSecret areas for some reason
or another, and their perspective on this issue might be a
useful insight.
The issue is 'tell or not to tell'. The decision ultimately
comes from the government, not the researcher. For example,
if the research was on a Neutron-Corbimite-Aspirin Bomb, which
could immobilize 1E6 people, being developed and modeled by a
University, would the investigators want to publish their
results in International Publications?
When the results of some research endangers (a) a person's
life, (b) the immediate security of the country or (c) is
so sensitive; it best be kept confidential, then I think
the Principal Investigator has an obligation to resist
publicising.
Realizing this places lots of pressure on the Principal
Investigator, there must be some guideline as to what
gets published how and where. Adding to the stack of
knowledge has a price, the cost is not yet decided.
Hank
------------------------------
Date: 15 Jan 1981 (Thursday) 1957-EDT
From: PATTI at WHARTON-10 (Tony Patti)
Subject: Human-Net'ers side of discussion
Subject: about cryptology restrictions
Ever since Eckert and Mauchly created the ENIAC here at U. Penn
for the Army, the military has been understandably interested
in the uses of computers. But we, as Human-Net'ers, are less
concerned about the military uses, and more concerned about
the uses of computers by other Human Beings. The latter uses
certainly require discussions of the privacy issues involved
with the use of computer networks, and hence the subject of
cryptology (which is a very intellectually interesting subject).
Unfortunately, I see that the military is trying to constrain
this discussion. It appears to me that they are trying to
restrict not only
(1) the international shipment of cryptographic devices
(which are already restriced as a Category XIII
article under the International Traffic in Arms
Regulations (ITAR)), but also
(2) the subjects of cryptology and even mathematics
(because it might have cryptologic applications)
among civilians.
I guess each person will make up his/her own mind concerning
this, but it is my belief that these two things are quite
different/separate, and that the latter should not also be
restricted by the military.
Tony Patti
P.S. I want to correct two bits of information which appear
in Hank (Dreifus@Wharton)'s message:
1. The military already has vaults full of unpublished
information about cryptology -- and I don't think
that they need to add (voluntarily or otherwise)
stuff that civilians write.
2. Shipments of articles on the Munitions List (which
includes cryptographic devices) are already prohibited
(under section 126.01 of ITAR) to such countries as
Albania, Bulgaria, the USSR, etc. (and in general
to any area where such shipment would not be in
furtherance of world peace and the security/foreign
policy of the U.S.).
------------------------------
Date: 15 Jan 1981 (Thursday) 2246-EDT
From: DREIFU at WHARTON-10 (Henry Dreifus)
Subject: Re: Cryptology.
[ At First glance it would seem that we (T. Patti) and I have
taken opposite views on the subject of Cryptology, however
the point that is being debated is the degree of 'control'
the government should exercise on this form of 'research' ].
Continuing, the United States government could in theory take
one of the following stands:
First: Not touch Cryptology at all, and not use encryption as
a means of confiding information. Along with this, I
would assume a world of absolutely no secrets, and each
and every citizen has equal and un-restricted access to
any piece of information he desires [ Under laws laid
down in the Constitution of the United States ].
Second: Maintain a leasch of sorts, keeping certain information
flowing only in certain circles. The state of the world
today is such that we must protect the integrity of our
nation, and in doing so, certain information must *not*
be revealed, since it could pose a threat to our welfare.
Third: Impose drastic regulations, such that anyone doing any
form of Crypto-graphic research is very carefully
monitored, and constantly kept under wraps - his work
never seeing the public world.
The third argument asks by itself an interesting question: If we
had "N" Tons of documents of unpublished Crypto-stuff, how much
is enough? Indeed this raises the question - which governments
constantly do, I am told, ''One can never say enough is enough,
so why not work on it some more?''. Indeed this is the case
in almost anything; let it be technical, social, offensive,
defensive, artistic, innovative, or a myriad of other neat
things.
The reason we have cryptology is to try and preserve what we
have as a nation, or at least make it real tough for those
(and there will always be enemies) to try and crack our
backbone.
On the other hand, if one's idea's really change the course of
thinking, for example some algorithm which will revolutionize
the data encoding scheme, and thus aid man in furthering his
society as a whole, it should not be supressed.
Governments at times become 'over cautious' in some instances.
This is due to a number of things, as so very simple as a piece
of dust catching the lower portion of one's eye, irritating like
crazy, and reading a report on releasing some thing or another.
A leaky roof has been known to start wars -- or at least I
suspect?
Backing up... The formal channels set up in our governmental
system are really a double edged sword: One cannot get too
close to revealing too much crypto stuff, or it is useless
to use by our own, yet to deprive man-kind from the knowledge
and insight of some new discovery is equally as saddening,
yet the decision process is quite paradoxical.
The issue is far from resolution, and even though there are no
stops to prevent some person in this nation to come up with some
crypto scheme that is better than anything else from printing it.
Contrast this to the research and development efforts in some
other nations on this planet. Philosophically it is ludicrous to
believe that people would want to hide things from others to the
extent of cryptology, however if one's survival is dependant on
it, one would tend to use it as best as possible. Other nations
supress *all* results of cryptological research. If I may cite
a specific example, the ACM [Association for Computing Machinery]
reviews over 300 publications in piecing together the Computing
Reviews. Looking at this publication carefully for quite a few
years, I have noted that virtually none of the Mathematical or
Computer-computational articles dealt with encryption. This
might be a 'non-exportable' item from Russia and other such
nations. The thought of too tight a control shivers down very
poorly. Exercising Option 3, as probably Russia etc. use, would
stifle some very meaningful work.
The fundamental ideas behind Cryptology have really not changed
since they were used (back in the 1600's I believe? perhaps
even earlier) to conceal information. Though it is true that
(a) Computers can jumble things so very very very well with
so very many neat algorithms the security ultimately
depends on the people who control the information.
Security through obsecurity only holds true up to a
certain point.
(b) The larger the radius of people who know certain facts,
the more liable the chance of this information 'leaking'.
In fact, with all of our super fantastic security, leaks
still occur, and it is rather disappointing, as it
ultimately depends on the people! People will always
be people -- that is to say VERY UNRELIABLE, as opposed
to the sequencing of computers. But people - and some
of them -- unreliable people control those very same
'reliable' computers.
The next issue is morals. Not much has been said of morality
recently. People always seem to feel that the highest prin-
ciples are always instantiated whenever they deal with topics
like Cryptology or other fun stuff. Let me say that people
on a whole are NOT honest, NOT law abiding [Yes Ozzie Myers
was my Congressman - and YES I wrote him many a letter he
never answered - and now it matters not!!] and usually it is
personal goals (greed, sex, power, you name it) that comes
first, and if and only IF it is convenient for the country
will that interest be observed. Part of the responsibility
of having all this super technology is proper maturity. Many
a person lacks this maturity, especially when an intelligent
decision based on trying to achieve the best goals for this
nation are trying to be reached.
Last semester I attended a Psycology class, which much of the
information I chose to filter. One fact that I did receive
and did check further was that 20% or so (depending on who
you listen to) of the world's population has some mental
disorders. They can range from habitual actions, depression,
paranoia or worse. It is impossible to keep that portion
of the population separate from the 'normals'. In fact who
really knows who the normals are??? The bottom line of this
is that people - for their own reasons chose to view the
world as they choose. What might seem like a correct and
useful technological breakthough may be viewed as easily
through some other reasoning process to be extremely
dangerous and suppressed ...
Life is fun isn't it? The infinite variations of opinion only
serve to swing a Gigantic Pendulum. As I have observed many
issues, things move at about the same rate from the extremes.
At one point this country invites millions of poor and homeless
people to migrate and find a home, and then put them to hard
work - at no substantial pay - to build some thing or another....
And so it goes, we 'open' our hearts to the Cuban Refugees, only
to have a good percentage of them hijack our planes and endanger
our lives - thus swinging the pendulum the other way to an
"anti-cuban" stance. Their at one-time-gracious sponsors have
all but disappeared, and what is left is very very sad indeed.
Using this swing analogy, the 'publication/supression' of
cryptographic material will probably swing from side to side
as does most everything in our world. It all sounds so easily
describable, however, there is an associated danger in accepting
this deduction: It is true that we can from each side PUSH the
Pendulum to the other side ''My'' side as each will say, but when
the arc begins to go unchecked, then the danger arises. It is
said that desperate people do Desparate things. No mention of
wrongness or correctness - just desparate. Such actions can
cause people to be subject to physical danger; this is wrong.
How to control the swing? Good question. Just checking
to see how many people were expecting the answer. I have
no answer, only some thoughts: Maturity, wisdom and common
sense are not used enough by the vast majority of the world
most of the time. Morals and ethics are only there to be
used when they are beneficial... We will progress no where
but in reverse unless the general attitude of the world
returns to the basic premises. Things at times become so
very convoluted that another ripple in one's path changes
and distorts the decision process.
There is much more that should be brought out into the open,
I would like to see other views and mind-reasoning appear in
Human-Nets.
------------------------------
End of HUMAN-NETS Digest
************************
∂16-Jan-81 0929 FFL
To: JMC, FFL
Elliott Bloom's secretary called to remind you that you have a lunch
engagement with Prof. Bloom at Faculty Club at 12:30 today.\
et phi
∂16-Jan-81 1023 PAM
To: duffey at MIT-AI, JMC at SU-AI, lewis at SRI-KL,
wilkins at SRI-KL
Crypto controversy simplified...
At risk of exposing my ignorance, I claim that the whole issue of control
of cryptographic research has been made too complicated in some of the
discussions in this forum. I propose that it is actually much simpler,
and will field comments intended to remove this conforting ignorance if
that is indeed the case.
First, it seems that the introduction of Public Key encryption methods has
made obsolete all the previous methods which required a secret key. If
this is true, then we can assume that whenever security is more important
than cost, Public Key methods will be used, by governments, businesses,
and individuals.
A number of consequences follow directly from this conclusion:
All secrets about less secure encryption methods are obsolete and might as
well be released to public view. Keeping them secret is analogous to
supressing the details of dynamite manufacture in a nuclear age.
Anyone who knows MacLisp and algebra can construct a Public Key encryption
method, so it is too late to try to suppress this technology. Advances
can make it cheaper to use, but it is already practical to use.
All the published methods have in common the feature that a means of
cracking the encryption involves solving an NP complete problem. Any
mathematician who does so can be famous for centuries by publishing his
results. Anyone who chooses to keep a solution secret could make a great
deal of money, but would be probably be imprisoned or killed to keep the
secret. Under these circumstances, I think any amount of restrictions on
publication would be ineffective.
This line of reasoning suggests that the only legitimate or useful purpose
for restricting distribution of encryption research is to stop the wider
use of these new, very effective techniques. As such, it appears that
this attempt could only be aimed at allowing governments to retain access
to data that is encrypted more weakly, while using the stronger methods
for their own data. Again, I cannot see how this could succeed given the
amount of the cat that is already out of the bag.
To summarize: All older methods are obsolete, and thus need no protection.
The public Key methods are in the public domain, and cannot be suppressed.
These methods resist cracking for such fundamental reasons that a solution
to them would be as imporant to math as Godel's work. The discoverer
could either be famous or dead, and would in all likelihood choose fame.
I see no legitimate grounds for restrictions on the research.
Paul Martin.
∂16-Jan-81 1120 Konolige at SRI-KL (Kurt Konolige)
Date: 16 Jan 1981 1113-PST
From: Konolige at SRI-KL (Kurt Konolige)
To: JMC at SU-AI
cc: konolige at SRI-KL
In-Reply-To: Your message of 16-Jan-81 0244-PST
Here is my tentative answer, with the proviso that I need to think
a lot more about the relevant model theory.
I will use the predicates PR (provable) and CS (consistent), which are the
relevant ones for knowledge in my formalization, and for which I have better
intuitions. Also let Mx mean "if x is a man then x is mortal". Then "John
doesn't know that all men are mortal" is expressed as ~PR('AxMx') in John's
theory of the world. I've used 'AxMx' to name the formula AxMx in the object
language, with "A" being the universal quantifier (I'll use "E" for
existentials). Because of the duality between PR and CS, we get the
following series of equivalent formulas:
~PR('AxMx')
~(~CS('~AxMx'))
CS('Ex~Mx')
At this point we do the skolemization trick, converting from 'Ex~Mx'
to '~Mg' for some skolem constant g. I'm not real happy about doing
it inside the CS predicate (although for the PR predicate it's got to
be ok), and this is where I need to think harder about the model
theory. At any rate, the skolem constant g is just another name, so
we can continue the equivalent formulas:
CS('~Mg')
Eu CS('~Mu') where u ranges over object language terms,
Eu ~PR('Mu') by duality again
But this makes sense to me: one can't prove the universal formula
'AxMx' just in case there is some instance of it that can't be
proven.
By using the dual predicate CS, it is possible to convert any
universally quantified object language expression into an
existentially quantified one, and pull the quantifier out after
skolemizing. So there doesn't really seem to be any need for inner
variables.
--kk
-------
∂16-Jan-81 1305 RPG Circumscription
In your talks you mention the missionaries and cannibals problem.
Do you discuss this in any recent papers? Has anyone else? Is there
a clear discussion of what you see as the issues anywhere?
-rpg-
∂16-Jan-81 1516 Susan Hill <CSD.HILL at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
Date: 16 Jan 1981 1452-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
Subject: [John McCarthy <JMC at SU-AI>:]
To: jmc at SU-AI
Is this message complete? Do you want me to open an account for him? I think
this message is a little too cryptic for me.
Susan
---------------
Mail-from: ARPANET site SU-AI rcvd at 15-Jan-81 2126-PST
Date: 15 Jan 1981 2126-PST
From: John McCarthy <JMC at SU-AI>
To: csd.hill at SU-SCORE
Michael Gordon, Computer Science Dept. U. of Edinburgh, Scotland
---------------
-------
Aren't you the Susan who asked for his address in a system message?
∂16-Jan-81 1545 Susan Hill <CSD.HILL at SU-SCORE>
Date: 16 Jan 1981 1543-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
To: JMC at SU-AI
In-Reply-To: Your message of 16-Jan-81 1518-PST
Sorry, no. Maybe it was Susan Seabrook at the reception desk. Can't think of
any other Susans.
-------
∂16-Jan-81 1600 JMC*
Binford papers
∂16-Jan-81 2318 JMC*
lisp question
∂16-Jan-81 2349 Feigenbaum@SUMEX-AIM Re: friedl and disk
Date: 16 Jan 1981 2142-PST
From: Feigenbaum@SUMEX-AIM
Subject: Re: friedl and disk
To: JMC@SU-AI
In response to your message sent 16 Jan 1981 1722-PST
Hi, John!! Having just come back from Europe, I feel relatively
uncoupled from Stanford business, and would like to remain so for
a while longer. Would appreciate it if you would pursue that with
Friedl.
Hope everything has been going well for you.
Ed
-------
∂18-Jan-81 2002 TOB
There are others relevant to the longer term program, but I did not
supervise: J.M.Tenenbaum, Quam, Scheinman, Paul
∂18-Jan-81 1905 JMC
The part about " Graduates of the program are
leaders throughout academic and industrial research." needs the substantiation
of specific names and positions. One sentence for each of no more than
5 such graduates would suffice.
∂18-Jan-81 2027 TOB cv
John
See CV.6[PUB,TOB]. I will respond quickly to comments.
Tom
∂18-Jan-81 1822 JMC cv needs more
"Include
A. Identifying data
date and place of birth, nationality
B. Complete academic record
Colleges and universities attended,
degrees and dates,
scholarships and academic honors
fellowship and residency training
other study or research appointments
C. Complete employment record
LIst all academic and non-academic positions held. Account for all time
since graduation from secondary school.
D. Public service.
E. Post degree honors and awards
Include major invited papers and addressses, memberships in professional
associations and learned societies, etc.
cv.6[pub,tob] looks adequate. I need a paper copy.
∂18-Jan-81 2351 Marimont at SRI-KL EE quals
Date: 18 Jan 1981 2350-PST
From: Marimont at SRI-KL
Subject: EE quals
To: jmc at SU-AI
cc: marimont
Professor McCarthy,
I had my qualifying exam on Thursday, and it went somewhat better than
I had expected. Unfortunately, since my performance is evaluated
solely relative to that of other students, I don't have much of a feel
for my chances of passing. I will check with Larry Manning, however, to
see if he would be willing to give you any advance information as to
whether your attendence at the meeting would be decisive in my case.
Thanks for your help, and I'll get back to you as soon as I have
some word from Manning.
Dave
-------
∂19-Jan-81 0401 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 18 Jan 1981 1133-PST
From: Ted Anderson <OTA at SU-AI>
To: space at MIT-MC
Subject: 3 stage version of Interim Upper Stage scrubbed
n028 1033 18 Jan 81
BC-ROCKET
By JOHN NOBLE WILFORD
c. 1981 N.Y. Times News Service
NEW YORK - The National Aeronautics and Space Administration has
notified Congress that because of excess costs and technical problems
it plans to discontinue development of a new rocket system for
launching planetary spacecraft from aboard the space shuttle,
replacing it with a modified version of the proven Centaur rocket.
Dr. Robert A. Frosch, the space agency administrator, said that the
decision would mean another delay, from 1984 to 1985, in the
launching schedule for the Galileo mission to orbit Jupiter and
deploy an instrumented probe into the Jovian atmosphere. Under the
new schedule, the spacecraft would not reach Jupiter until late 1987.
Dr. Frosch said that the schedule change would add extra costs, as
yet undetermined, to the Galileo project. Officials at the Jet
Propulsion Laboratory in Pasadena, Calif., which manages the project,
said the increase could be as much as $75 million above Galileo's
current cost estimates of $650 million.
This would be in addition to the costs of converting the Centaur
rockets for operation from the shuttle.
Centaurs have been used since the mid-1960s as an upper stage in
rockets launched from the Earth, particularly those carrying large
communications satellites into a high orbit of the Earth and sending
Viking spacecraft to Mars and Voyagers to the outer planets. But
launching Centaurs from the cargo bay of the shuttle will to be a
technological challenge because the rocket's liquid hydrogen
propellant must be maintained at super-cold temperatures.
The Centaurs are produced by the General Dynamics Corporation under
the direction of NASA's Lewis Research Center in Cleveland.
Because the re-usable shuttle is limited to flight in low Earth
orbit, spacecraft to be launched from its cargo bay and sent to
higher altitudes or to other planets must receive an added boost from
an attached rocket. The shuttle itself is running about three years
behind schedule; its first orbital test flight is now expected to get
under way in late March or April.
The rocket system being discontinued is a three-stage solid-fueled
booster called the Interim Upper Stage, which was under development
for the space agency by the Boeing Co. The action does not affect the
two-stage version of the same system, which is being built for the
Air Force to use in launching its communications and surveillance
satellites from the shuttle.
But Air Force officials, concerned about delays in the two-stage
version as well, have indicated that they may be forced to extend the
production of Titan 3 rockets as backups in case either the space
shuttle or the Interim Upper Stage vehicles encounter further delays.
The NASA decision and its effect on Galileo and possibly other
projects was seen as confirming the worst fears of those who had
criticized the agency for its failure to allow for backup rocket
systems to be kept in production while awaiting the shuttle.
Dr. Bruce C. Murray, director of the Jet Propulsion Laboratory, said
that this was ''a sharp break from the practice of the l960s and
1970s.'' The reason for ''our present dilemma,'' Dr. Murray added, is
that the Titan-Centaur rockets were not kept in production until it
was certain that the space shuttle and its accompanying interim
rocket systems were ready for operation.
nyt-01-18-81 1330est
***************
------------------------------
Date: 18 Jan 1981 2355-PST
From: Jim McGrath <JPM at SU-AI>
Subject: Voyager pictures
To: sf-lovers at MIT-AI
CC: space at MIT-MC
The following is from MSK at SAIL. I thought it would be of interest
to you all. - jpm
New from the Astronomical Society of the Pacific
PHOTOGRAPHS FROM THE VOYAGER/SATURN ENCOUNTER
In keeping with its goal to provide its members and the public
with authoritative information concerning new developments in
astronomy, the A.S.P. is pleased to announce the availability of sets
of prints and slides of Saturn, its rings, and its satellites,
selected from the images returned by the Voyager spacecraft. These
photographs have been chosen for both their visual impact and
scientific importance. Detailed captions and information about the
mission accompanies each set.
Prints are 8 X 10 PHOTOGRAPHIC prints (four in color, one in
black-and-white in each set) carefully produced from originals
provided by NASA's Jet Propulsion Laboratories. Slides are standard
35mm format, carefully reproduced to bring out the subtle details
captured by Voyager's cameras.
PRINTS
Saturn System Collage -- full color dramatic 8 X 10 montage of
the Saturn system, assembled from the best Voyager images of the
planet and 5 of its satellites. $2.65 each.
Print Set I -- Saturn and Its Rings. $10.60 per set.
1. Color. A full view of Saturn and its ring system from 18 million km
as Voyager approached.
2. Color. Sweeping close-up of the ring system against the limb of the
planet.
3. Color. Saturn casts a dramatic shadow on its rings in a view from
the departing spacecraft.
4. Color. False color enhancement of the ring system as seen from
underneath.
5. B&W. Detailed close-up of the ring system, showing more than 90
individual ringlets.
Print Set II -- Saturn and Its Satellites. $10.60 per set.
6. Color. Saturn, its rings, the ring shadows, Tethys, and Dione as seen
from 13 million km.
7. Color. Dione, crossing the clouds of Saturn, shows its two different
hemispheres.
8. Color. Close-up of Dione showing impact craters and fault line.
9. Color. A false-color close-up of the various layers in the atmosphere
of Titan.
10. B&W. Close view of Mimas, showing its huge impact crater.
SLIDE SET
Slide Set -- Saturn, Rings, and Satellites. $13.78 per set.
1.-10. The ten images from above.
11. Color. False-color image of Saturn and its rings, designed to bring
out details in the bands of the planet's upper atmosphere.
12. Color. Close-up of Saturn's cloud deck.
13. Color. False-color enhancement of Saturn's northern hemisphere,
showing belts and weather systems.
14. Color. Color-enhanced view of the rings from beneath.
15. B&W. Close-up of the "braided" F ring.
16. B&W. A sequence of photos showing the spokes in the B ring.
17. B&W. Mosaic of surface of Rhea.
18. B&W. Saturn-facing side of Tethys with its large valley.
19. B&W(?). Iapetus with its two very different faces.
20. B&W(?). S 11, one of the newly-discovered co-orbiting satellites, and
the shadow cast on it by a previously unknown ring.
Please note that, except for the collage of the Saturn system,
neither prints nor slides are available singly -- they MUST BE
PURCHASED IN SETS. Prices above include 6% tax, but do not include a
$1.50 Handling and Postage charge on each order.
--------------------------------------------------------------------
I intend to send in an order on Jan 30. If you want to make up a
pool order (to split the handling and postage charge) mail to
CSD.MSK@SU-SCORE. If enough people want individual photos, it might
also be possible to split a set. I assume that slides number 19 and
20 are B&W, since the flyer didn't specify.
Note: I'll need the money from you before the order is sent, so
that I won't be broke for the 4-6 week delivery time. The handling
and postage charge will be split proportionately after all orders are
in (i.e. the bigger your order, the greater the percentage of the
shipping charge that you'll pay). Checks may be sent to me through ID
mail (to Michael Kenniston, Computer Science Dept, Margaret Jacks Hall
420), or you can put them in the "K" mailbox on the second floor of
Jacks, or drop them off at Jacks 420 in person. If you prefer to send
in your own order for any reason, stop by my office and you can Xerox
my order form. (Call first to make sure somebody's there: 497-2513.)
------------------------------
End of SPACE Digest
*******************
∂19-Jan-81 0517 REM at MIT-MC (Robert Elton Maas)
Date: 19 JAN 1981 0748-EST
From: REM at MIT-MC (Robert Elton Maas)
To: JMC at SU-AI
It's not such a nice idea to send the ASCII ESC character to non-SU-AI
places like HUMAN-NETS when you mean not-equal. It may print as not-equal
on your Datadisk, but that's not a standard ASCII code. Here it
prints as $ and I guessed you actually sent $ instead, but some places
it may cause trouble with the editing programs attempting to read mail.
∂19-Jan-81 0847 CSD.JEANIE at SU-SCORE ARPA Meeting Fri. 2/13/81
Date: 19 Jan 1981 0844-PST
From: CSD.JEANIE at SU-SCORE
Subject: ARPA Meeting Fri. 2/13/81
To: csd.ullman at SU-SCORE, csd.mccarthy at SU-SCORE, buchanan at SUMEX-AIM,
csl.fb at SU-SCORE, csl.lantz at SU-SCORE, tob at SU-AI, zm at SU-AI,
dcl at SU-AI, acy at SU-AI, wiederhold at SUMEX-AIM,
csd.feigenbaum at SU-SCORE
cc: csd.bscott at SU-SCORE
.The meeting will be Friday 2/13 at 9:00 in Room 252 MJH.
To: Present or Aspiring ARPA Principal/Co-Principal Investigators
From: Ed Feigenbaum
Subject: ARPA's desire to consolidate our monies
At ARPA's request there will be a meeting on Friday, Feb. 13
(let's assume 9am) to discuss the desire of IPTO to consolidate
the Stanford CSD funding (plus Luckham's funding if possible)
into a single contract. Attending for ARPA will be Bob Kahn,
Bob Engelmore, and Duane Adams. Attending for Stanford should
be all of us who are responsible for ARPA projects, plus
those who currently have unfunded proposals in the mill.
This is a very significant development (for better or for worse;
we have to decide which) and all should attend who can possibly
do so. We should also have a vigorous net message correspondence,
perhaps supplemented by a meeting, before the ARPA people arrive.
Background
Back in 1965, ARPA established funding patterns at MIT and CMU
that were different from that established at Stanford. MIT and
CMU have large-contract "lump sum" funding under the control
of a team of Principal Investigators. MIT, because of
historical reasons, has two contracts (LCS and AI); CMU has but
one. In addition, all the funding for ISI is under one contract.
Stanford has had separate contracts for every project that has
come on line (except for a few that have been consolidated
under the SU-AI contract).
This has had a number of effects, becoming more strongly felt
each year as the number of individual contracts has
proliferated:
a. it has created a blizzard of Stanford-related paperwork at
ARPA. This is becoming resented at ARPA. the amount of time
spent in handling us is an order of magnitude more than the
time spent handling MIT or CMU.
b. it has created considerable confusion on our end, sorting
out the various monies that arrive at random times in random
lumps from the various contracting services of the DOD. Betty
has to spend a lot of time sorting this out.
c. ARPA is reluctant to even think about the "small" requests
that come from various of our faculty members because each
!small request generates as much paperwork for them as a big contract.
They would like us to manage as a group, rather than as a
collection of individuals. This has become feasible, perhaps
even desirable, with the reintegration of the department in one
place and the departmental management of the SAIL computer facility.
Some considerations are (and I discussed these with Engelmore):
1. The philosophy of "each tub on its own bottom" keeps the
internal coordination and internal conflict levels low at Stanford
(to which the ARPA response is: yes, at the expense of increasing
the hassle level at ARPA).
2. It is easier for a contract that "sticks out" at the level of
several million dollars per year to get cut in adverse times
(thought by ARPA not to be a great risk, but should be explored
further).
3. Large funding levels might make it harder for us to get "long"
contracts (two or three years). This is denied by ARPA, citing
as evidence CMU and ISI (three year contracts, each year many
millions).
Conclusion: I urge each of you to think hard about what this
means for you; and I urge a vigorous discussion. I personally
am feeling mildly positive about the idea on the basis of
the flexibility it will give us, the possibility of large
equipment acquisitions accompanying a large contract, and
the long-term health of the department (allowing young
faculty members and those with small needs the possibility of
participating in ARPA funding).
P.S. I suggest MJH 252 for the place of the meeting on Feb. 13,
if it is free.
-------
-------
∂19-Jan-81 0900 JMC*
estimate to Lindstrom
∂19-Jan-81 0900 JMC*
call nils 859-2311 about Firdman book.
∂19-Jan-81 0900 KRAUSS at MIT-MC (Jeffrey Krauss) Public Key methods
Date: 19 JAN 1981 1201-EST
From: KRAUSS at MIT-MC (Jeffrey Krauss)
Subject: Public Key methods
To: jmc at SU-AI
In your message to Human-nets dated January 18, you referred to
the "major cryptographic weakness" of the public key method of
Rivest. Could you provide me with a reference to an article
where that weakness is discussed, and the modification is described?
Thank you.
---Jeff Krauss---
While I believe their present method doesn't have the weakness in question,
neither they nor anyone else has pointed out in print the weakness of the
original memo which also applies to the version described by Martin Gardner
in Scientific American.
∂19-Jan-81 0933 Shortliffe@SUMEX-AIM Medical AI Organization
Date: 19 Jan 1981 0915-PST
From: Shortliffe@SUMEX-AIM
Subject: Medical AI Organization
To: AAAI-Executive-Committee:
cc: lindberg@SUMEX-AIM
I am writing this brief note on behalf of Don Lindberg and myself.
Don is Chairman of the AIM (AI in Medicine) Advisory Committee associated
with the SUMEX-AIM resource at Stanford and a prominent member of the US
medical computing community. He and I, among others, are interested in
promoting recognition and use of AI techniques in medical computing research
activities. "Mainstream" medical computing work had largely ignored AI until
recently, and AI work is still largely viewed with suspicion by the larger
community.
Recently, at the time of a national medical computing meeting held in
Washington in November, several researchers in the field met to form a U.S.
Committee on Computers in Medical Care. The new committee was viewed as
necessary to coordinate the disparate medical computing activities and
organizations around the country and to provide them with a more cohesive
organization and political force. Representatives of essentially all the
medical computing organizations in the country were there (e.g., SIGBIO, the
IEEE medical computing group, TC-4 of IFIPS, the Society for Computer
Medicine, the Society for Advanced Medical Systems, the MUMPS Users Group --
you get the idea). Don Lindberg represented the AIM community on an ad hoc
basis, but it was clear to all that in doing so he was representing a
community of researchers and a set of methodologies but not an organization
as such (SUMEX itself is, of course, a government-sponsored research
resource, and not a formal organization with a membership that can be
identified or that votes for officers). He and I discussed this problem
later and felt it was appropriate to start thinking about forming a Medical
AI Society of some kind, one that could have a formal membership list and
a set of officers who could represent the community's interests at meetings
such as the Washington session in November. Rather than setting up an
entirely new organization, we also thought it made sense to propose that this
organization be a subset of the membership of the AAAI. We discussed the
idea briefly with Bruce Buchanan, who mentioned that he was unaware of any
efforts to establish formal subgroups within the AAAI but that we seemed to
have a good reason for wanting to do so. He therefore suggested that we
propose the idea to the AAAI Executive Committee to get your opinions and
suggestions on how to proceed. Since such a "special interest group" within
the AAAI would presumably set some precedents for other groups that might
wish to form in the future, we assume you would like to consider establishing
formal guidelines or rules.
Your consideration of this idea is much appreciated. Please let me
or Don Lindberg know if we can answer any further questions that might
arise.
Regards,
Ted Shortliffe
(Co-PI and Medical Liaison, Sumex Computer Project, Stanford Univ.)
-------
∂19-Jan-81 1010 Konolige at SRI-KL (Kurt Konolige) problems for CS226
Date: 19 Jan 1981 0959-PST
From: Konolige at SRI-KL (Kurt Konolige)
Subject: problems for CS226
To: jmc at SU-AI
John, Kowalski's book "Logic for Problem Solving" has some good
examples of the use of logic in solving blocks-world problems; chapter 6
in particular has plan-formation as a deductive process, and illustrates
the frame problem in a simple example. Although it's all in clause form,
it's easily adaptable to the FOL formalism.
--kk
-------
∂19-Jan-81 1343 CSD.JEANIE at SU-SCORE Visitor from Florida
Date: 19 Jan 1981 1341-PST
From: CSD.JEANIE at SU-SCORE
Subject: Visitor from Florida
To: csd.feigenbaum at SU-SCORE, csd.lenat at SU-SCORE,
csd.genesereth at SU-SCORE, tob at SU-AI, csd.mccarthy at SU-SCORE,
buchanan at SUMEX-AIM
If any of you are interested in having a visitor from Florida who would like
to discuss the cognitive capabilities of dolphins, please let me know by
noon 1/20. He is particularly interested in acoustic analysis and real time
data management.
jeanie
-------
∂19-Jan-81 2039 TOB
John
I could not send this last night before I went to bed; system was down.
Tom
Binford, T.O.;
%2"Computer Integrated Assembly Systems";%1
Proc NSF Grantees Conf on Production Research and Technology,
Cornell Univ, September 1979.
The paper reviews progress in automation research in areas of advanced
programming systems for robots and inspection. Research is described
in force sensorimotor control: theoretical analysis enabling calculation
of motor torques required to implement an arbitrary force or coompliance,
specification of high level force commands, and software for realtime
control. Extensions are described to the AL programming system,
especially the AL interpreter for interactive programming. AL is
an advanced and well-developed programming system. Several assembly
experiments in AL are reported.
A new program of research is reported in future generation
programming systems for robots, including geometric modeling,
problem solving, and simulation, integrated with computer vision.
The ACRONYM system provided the basis for GRASP, a system for automatic
choice of grasp points to carry out tasks. GRASP is implemented as a
set of rules specifying strategies for maximum stability, accessible
faces, and secure grasp positions. ACRONYM shows great generality
for inspection and for the difficult task of picking parts placed
randomly in bins. A preliminary test of ACRONYM in photointerpretation
is described.
∂20-Jan-81 0027 LWE ns
My AP digest files are accumulating and are approaching
120 pages of disk space, so I shall send another courier tape shortly
in order not to contribute unduly to the disk space crunch on your system
(this way the files can be dumped onto tape and deleted any time they
begin to look excessive--assuming Ralph Gorin will be kind enough to
help out with the file transfer/tape write procedure again).
---
Also, I would like to add a second output request to the ns program,
temporarily: starting with Reagan's inauguration day, for a period of
one month, I would like to accumulate the full text of all digest stories
(and only digest stories) that involve Reagan; this will permit
a validity check on the "slippage" between the digest summaries and the
matching full stories (NSF reviewers insisted we do this as a pilot study
before they are willing to "buy" our use of the summaries as the only trace
of national media content for the full-scale project). Since the resulting
output file will grow rapidly (I am aiming for a total sample of about
100-120 stories maximum), there is all the more reason I get a tape to
receive the file dump. However, I shall await your response before I proceed
with this--I may be causing more trouble than I think--and before I
surprise Ralph Gorin with this (I forgot to send him a CC of this message,
as I noticed after I got started typing it--will contact him after I hear
from you first). Please do let me know if you foresee a problem, ok?
∂20-Jan-81 0103 Schauble.Multics at MIT-Multics Clipping Service - Solar power satellite-power conversion
Date: 20 January 1981 03:24 est
From: Schauble.Multics at MIT-Multics
Subject: Clipping Service - Solar power satellite-power conversion
To: energy at MIT-AI, space at MIT-AI
The November 20 issue of Electronic Data News (EDN) contains an article
on a new version of klystron tube the performs a direct conversion from
light to microwaves. The article is too long and too technical to
transcribe, so I will summarize.
The "photoklystron" is being developed by Dr John Freeman and his
research team at Rice University in Houston Texas. The device is
derived from a reflex klystron by substituting a photocathode for the
normal thermonic cathode. This removes the need for a cathode heater,
leaving only bias power (a few microwatts) required. Their present
experimental model works in the range of 5 to 240 MHz with a conversion
efficiency of about 1%. They forsee operation at higher frequencies and
with conversion efficiencies up to about 6%. For comparison,
high-quality solar cells driving high-efficiency microwave tubes can
achieve efficiencies of 9 to 12%. However, the photoklystron's
simplicity, small size, high reliability, and potentially very low cost
may outweigh the lower efficiency
The device was specifically designed for use in the Solar Power
Satellite project sponsored by NASA and the Dept of Energy. The NASA
Lewis Research Center has funded a large part of the design, with ITT
fabricating test units.
∂20-Jan-81 0125 POURNE@MIT-MC
From: POURNE@MIT-MC
Date: 01/20/81 04:25:41
POURNE@MIT-MC 01/20/81 04:25:41
To: JMC at MIT-MC
FYI re llw
Date: 20 JAN 1981 0411-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
I got a real problem with this thing, since it's getting out of
hand in size. We DO want people who can get some work done; I
have Poul Anderson coming down to help write this up, and Harry
Stine coming in from Arizona for the same purpose. But
everywhere I turn are more people coming out of the woodwork! A
couple of big aerospace outfits called today and damned near
demanded "seats" as if somehow it is now established that we
really are doing the REAL THING space plan of the US.
SO: I added you because JMC said you were about the
best person he could thnk of to come. If your associate comes,
he MUST understand that as this gets larger, I'm gong to have to
get more dictatorial in assigning people to sub-committees and
sending them off to smaller rooms with others; I'll then try to
circulate around and see that everyone is getting the work done,
and shuffle resource people around to loan them where needed,
and put some of the "big wheel" types who are showing up not to
work but to say they've been there off with some of the "worker"
types who may not be needed just then to keep them amused--
IE: Far be it from me to say no to someone productive,
but I want it clearly understood that I need a PRODUCT out of
this conference, and I'll get as arbitrary and capricious as I
have to be to get the documents I need prepared!
∂20-Jan-81 0128 POURNE@MIT-MC
From: POURNE@MIT-MC
Date: 01/20/81 04:28:45
POURNE@MIT-MC 01/20/81 04:28:45
To: JMC at MIT-MC
I don't know how to concatenate messages, so you get this to
which the other was a reply also FYI.
I presume the packet of information on the conference has come?
Date: 19 Jan 1981 2137-PST
From: Lowell Wood <LLW at SU-AI>
Dear Jerry:
I suspect that you're keenly interested in keeping the attendance at your
Space Program Planning Conference down to a productive size, but would you
be willing to have my Systems Studies Section Leader, Dr. Rod Hyde, come
along with me?
Rod (who got all three of his degrees in MIT's top-ranked Aero&Astro
Department during the past decade) is my Program's Systems Studies Section
Leader and my co-author on everything serious I've ever written about
space (for excellent reason--he's not only educated to the gills in the
relevant subjects, but he's a True Blue space fanatic; he also has the
incidental advantage of being brilliant). I sincerely believe that he
could contribute a great deal to your conference, both in the form of
ideas and as a quantitatively oriented resource person--besides knowing
most everything that's ever been done re space work, he's faster with his
calculator than most guys are with a decent-sized computer (and when he's
riding a CRAY-1 computer, he's absolutely peerless; however, he'll come
equipped only with his HP-67). I certainly hope that you can spare room
for him.
In any event, I'll show up at the appointed hour of commencement and stay
for the conclusion, holding forth into the nights with the most nocturnal
of your conferees; my secretary will forward a Secret-level clearance to
the TRW contact point, and I'll be staying at the ancestral home (as will
Rod, if he comes) over in Simi Valley on Friday and Saturday nights. My
Snail label locally is
Lowell Wood
639 Hazel Street
Livermore, CA 94550
re the packet of conference materials which you kindly offered to send.
I'll try to get some suggestions of my own documented and sent off to you
in advance.
Thanks,
Lowell
∂20-Jan-81 0940 CSD.JEANIE at SU-SCORE Jan. 27 Meeting
Date: 20 Jan 1981 0935-PST
From: CSD.JEANIE at SU-SCORE
Subject: Jan. 27 Meeting
To: tenured-faculty: ;
A meeting has been scheduled for Jan. 27 at 2:30 pm to discuss the tenured
case of Andy Yao. It will be held in MJH 252. The papers on this case are
in my office (214 MJH) if you wish to look them over before the meeting.
--jeanie
-------
∂20-Jan-81 1056 CSD.JEANIE at SU-SCORE More on Jan. 27 Meeting
Date: 20 Jan 1981 1051-PST
From: CSD.JEANIE at SU-SCORE
Subject: More on Jan. 27 Meeting
To: tenured-faculty: ;
In addition to the tenured case of Andy Yao, the faculty will also be voting
on the case for Doug Lenat. His papers will be in my office, also.
--jeanie
-------
∂21-Jan-81 0158 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 21 JAN 1981 0459-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Date: 20 Jan 1981 0200-PST
From: John McCarthy <JMC at SU-AI>
I think Rod Hyde is as good as Lowell says.
He was invited, with stipulation that I intend to be a tyrant if
need be to get the job done.
JEP
∂21-Jan-81 0158 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 21 JAN 1981 0458-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
No, that was it.
Please bring anything you thnk we-or you-will need to
prepare the documents etc.
We intend to be both general and specific in this.
FYI Lowell Wood is coming.
Have NOT heard from Minsky. Doubt he wants to come, but
wonder why he isn't answering maiol?
When Minsky is undecided, he often can't bring himself to answer mail.
∂21-Jan-81 0900 JMC*
voinovich, lindstrom, nilsson
∂21-Jan-81 1109 KLC comp
Hi...did you just send me a message? I saw something flick by on the
screen while reading the AP...when can we go over the comp questions?
No. I just repeated a news bulletin from NS. I should have something
by late this afternoon.
∂21-Jan-81 1206 Waldinger at SRI-KL he
Date: 21 Jan 1981 1205-PST
From: Waldinger at SRI-KL
Subject: he
To: jmc at SAIL
cc: ffl at SAIL
will He be able to get an invitation letter.
richard
-------
∂21-Jan-81 2118 SQU This is just a gentle reminder...
...that your income tax is due in april and that MY RECOMMENDATION from you
is needed soon. Very soon. Ah wouldn't want ta hafta (at this point I roll
a toothpick from the left side of my mouth to the right) tell my boys ta
come an persuade ya.
Quite seriously, John, I'd like to get all of my application in on time
and I realize its not easy to write something up overnight -and I'd rather not
cut things down to the wire.
Apropos of nothing -do you happen to have three old golf balls, perhaps?
(for juggling) I put a notice on BBOARD but no response as yet.
jk
∂21-Jan-81 2306 LWE inquiry/ns
just thought i'd check in to see if you had read my message yet.
I read it, but I've been too busy to do anything. I'll get to it.
∂22-Jan-81 0038 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 22 JAN 1981 0339-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Date: 21 Jan 1981 0924-PST
From: John McCarthy <JMC at SU-AI>
When Minsky is undecided, he often can't bring himself to answer mail.
AHA
∂22-Jan-81 0302 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #17
Date: 22 JAN 1981 0551-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #17
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Thursday, 22 Jan 1981 Volume 3 : Issue 17
Today's Topics:
Computers and the Deaf - CA PUC Recommendations,
Terminal Design - Microwave Hazards
----------------------------------------------------------------------
Date: 21 Jan 1981 1802-PST
From: FYLSTRA at SRI-KL (David Jon Fylstra)
Subject: California PUC makes landmark decision!
The California Public Utilities Commission today adopted a land-
mark decision in the matter of State Bill 597 that will be of
significant national benefit to deaf Americans. The Commission
overruled the PUC staff recommendation of Baudot/Weitbrecht-only
devices and adopted instead a proposal to require dual ASCII/Bell
103 and Baudot/Weitbrecht devices in the state distribution
program.
[ Please note that these notes are based on a telephone
conversation with an attorney who attended the
Commission meeting today. We will receive the official
written decision in 2 or 3 days. A press release was
issued today and should appear in tomorrow's newspapers.
When I receive more detailed information I will submit
a more accurate account. ]
Administrative Law Judge Daly submitted his recommendation to
the Commission as a synthesis of the public hearings held in
this matter. (As you may recall, the California State bill
requires the telephone companies to distribute a teletype-
writer device to qualified deaf subscribers. The program was
to be funded through a monthly surcharge to be levied on each
of the state's 12 million rate-payers.) In his recommendation,
Judge Daly essentially adopted the PUC staff proposal, which
required Baudot/Weitbrecht-only teletypewriters as the stan-
dard equipment. Daly suggested that future technology might
eventually lead to lower costs for ASCII equipment that might
cause dual ASCII/Baudot equipment to fall within the permitted
5% range above the minimum qualifying equipment.
Commissioner Gravel objected to this proposal, calling the
Baudot code a "dinosaur." He wrote an alternative decision
which was adopted by the Commission. This decision says
that the equipment for the statewide program shall be
compatible with both ASCII/Bell 103 and Baudot/Weitbrecht
standards during the near future, so that a transition may
be facilitated that will lead ultimately to an ASCII/Bell
103-only system. He suggested that it was the Commission's
responsibility to promulgate standards that are current
with today's technology, and that the staff proposal would
not sufficiently stir competition. The Commission president
commented that the PUC was "pleased and honored" to initiate
this program to integrate the deaf into a more equal level
of telecommunication access.
The decision is based on an estimated 90,000 devices to
be distributed over the coming three years. The initial
surcharge will be $0.15 per month per subscriber-line, and
the funds will accrue into a trust fund to be administered
by a committee with representatives from Pacific Telephone,
General Telephone, and the independent companies. The
program will commence in 120 days, which is judged to be
sufficient time to permit the manufacturers to modify their
equipment to meet these standards. The minimum device will
include a keyboard, a visual soft-copy display or paper
printer, an acoustic coupler, and AC and battery operation
and shall be less than 12 lbs in weight. A ring signalling
device will also be included.
In my view, this decision was extremely far-sighted and will
lead ultimately to the national adoption of ASCII/Bell 103
as a standard for telephone communications among the deaf.
Such a decision was crucial because the California state
bill would have certainly tripled and possibly quintupled
the size of the deaf teletypewriter network in the United
States.
------------------------------
Date: 21 Jan 1981 0849-PST
From: Zaumen at SRI-KL
Subject: Physics of radiation
Recently there have been several contributions to this
digest about the effects of radiation from either VDTs or
a SPS. The following is a summary of the physical issues
involved.
A. High Frequency Radiation (X-rays, Gamma-rays, etc.)
Even a little is bad for you. Each photon ionizes something
or other with all sorts of biological side effects. These
effects are unpredictable on a photon by photon basis and
range from the creation of unusual chemicals to changes in
DNA.
Low-level radiation is safer only because the damage is
done slowly enough that your body can recover (usually).
Also, we live in an environment with radiation present
from natural sources, and presumably we are not noticeably
hurt by radiation levels small compared to the background
level.
B. Low Frequency radiation (microwaves, etc.)
Here, there is not enough energy for a single photon to ionize
an atom or even effect chemical reactions on a molecular level
(this last possibility depends on the molecules involved.) If
a large number of microwave photons are simultaneously absorbed
by an atom, ionization is possible; however, the probability of
this is very small -- the order of (the photon density times
(1/137) ) to a large power.
The effects of low-frequency radiation include thermal heating
and direct electrical effects on the nervous systems. There
have been suggestions that certain parts of the body (eyes,
for instance) are particularly sensitive to microwave radiation
because of poor heat dissipation. There is also a question of
whether or not microwave radiation can electrically affect the
nervous system. Both possibilities are open questions, as far
as I know.
------------------------------
KRAUSS@MIT-MC 01/21/81 08:57:52
Re: terminal safety; low level non-ionizing radiation
Contrary to Jerry Pournelle's claim that low level microwave
radiation has zero effects on biological systems, there is now
apparently some evidence that low level radiation affects the
permeability of the blood-brain barrier. See, for example, the
abstracts of papers delivered at Session 7 of the Bioelectro-
magnetics Society, in the Proceedings of the National Radio
Science Meeting of June 18-22, 1979 held in Seattle Washington.
Available from:
USNC/URSI
National Academy of Sciences
2101 Constitution Ave NW
Washington, DC 20418
---Jeff Krauss---
------------------------------
Date: 21 January 1981 1745-EST (Wednesday)
From: R.Kamesh <Kamesh.Ramakrishna at CMU-10A>
Subject: Effects of microwaves
With respect to Pournelle's claim that
"To the best of my knowledge, all data on low-level microwave
radiation indicates ~zero effects on biological systems.
However, there are no studies in which low levels have been
employed for long (years) times."
Is ~zero supposed to be "non-zero" or is it "approximately zero"?
I assume that it is the latter (if it means non-zero, my
apologies to Mr. Pournelle). Simply speaking, this is not the
case. Let me make the following observations about the second
sentence:
(1) What does "no studies ... for long times" mean? In the last
one year? the last five? or the last twenty? There was a
two-part article in the New Yorker two years ago that gave a
detailed history of particular researchers in this country
whose efforts at establishing a lower limit were quite
crudely suppressed by manufacturers (threatened by possible
changes in microwave heater leakage standards) and the US
Department of Defense (threatened by lawsuits from people
exposed to excessive radiation, as well as by people in
neighborhoods close to military stations emitting high-levels
of microwaves). The work referenced in that 2 year old
article had been done over the previous 15 years and was
still in progress. Undoubtedly, there has been work done
since, even if I am not familiar with it.
(2) What does "low-power" mean? Enough to gently warm you as you
plough through it on a cool day? US radiation leakage
standards? or is the aside in the note about quantum-level
power supposed to indicate the level of power? The US
standard for domestic microwave equipment leakage is 5000
microwatts; the standard for the military is 10000
microwatts. The corresponding standards in the Soviet Union
are 5 and 20 microwatts. Thats 3 orders of magnitude (if you
count in decimal, 10 if you do your rithmetix in binary).
The Soviet standards were established in the early sixties
after some researchers there had reported effects on humans
at the 20-100 microwatt levels. In the last 30 years there
have been incidents in the US armed services of microwave
exposure in the 1000+ range. These have generally not been
allowed in the courts as the exposure is within the standard,
and there have been no "proven" effects of microwaves on
humans -- just more of them lying statistics.
(3) What does "~zero effect on biological systems" mean? What is
~zero? If nobody dies immediately (foam trickling out of
mouth, eyes glaring, hands twitching, etc.) does that
qualify? If nobody dies in a month? What if the exposed
person is left with a feeling of unease all his/her life?
Where do you draw a line, and who is qualified to play god in
such situations? The 20-100 microwatt studies made in the
Soviet Union claimed the following two effects:
o increased incidence of heart disease
o increased incidence of cataracts.
At lower powers, the incidence probabilities dropped.
A report I saw recently (I think in Science, I can get the
reference easily) claimed that these incidence probabilities did
not suddenly shoot up from base level at some level of power, but
appeared to be smoothly increasing functions of total exposure.
This argues that "non-zero effects" appear on "humans" at all
"powers", however "low". There are base probabilities of getting
heart disease and getting cataracts that are due to unknown
ambient effects on humans; any unusual exposure to microwaves
increases the probability of suffering from these diseases.
Microwave exposure appears to be cumulative in its effects on
humans - in particular, the studies with respect to heart disease
seem to show this effect. There have been other reports that
cumulative effects appear in all muscles that function
cyclically; the mechanism for this effect is not known (my
favorite explanation suggests a hysteresis-like phenomena in
these muscles, maybe microwaves keep some catalyzed chemical
cycle in the muscle cell from returning to its initial state,
thus hastening cellular death).
It does not make sense to protect people from every conceivable
harm. I do not advocate such a thing. I do not even wish to
prevent people from harming themselves. But I have the right to
keep other people from harming me, and, by god!, that is
precisely what microwaves can do. Furthermore, I will very often
not even realize that I have been harmed. In many cases, courts
will not accept statistical relations between some causes and
their effects. There are some good reasons not to let our
statistics run our courts, but this does not seem to be one of
them.
What I find difficult to comprehend (though easy to understand)
is the belief (seemingly held by the majority of American
corporations, and the US Department of Defense) that any effort
at quantifying this harm should be suppressed. I can comprehend,
and even sympathize with efforts to suppress information of
military value -- so that even though I find the attempt to
control cryptography research outrageous, and some of the
proposed measures draconian and threatening, I can understand why
somebody thinks it is important. It may genuinely be a matter of
life-and-death and I believe that that makes a difference. I
have problems when somebody equates a matter of profit-and-loss
to a matter of life-and-death. I am especially disturbed when
decisive control over life-and-death issues is in the hands of
entities primarily concerned with profit-and-loss.
In this matter, the Department of Defense is largely an
institution that expresses the opinions of its sub-contractors
who stand to lose the most (what they gain in contracts will
possibly go out in the form of mucho payments for damages). I am
quite unwilling, and do not expect anyone else to be willing, to
give up a single minute of my life to make somebody's
balance-sheet look better. (I might do it for emotional reasons;
for friends; and I can understand doing it for personal pleasure,
and so on).
So the dialogue quoted by Pournelle in the last part of his
message makes eminent sense to me -- if low-power microwaves,
even sub-quantum-level microwaves, MIGHT have an effect, I would
rather not be exposed to them.
What do I mean here by MIGHT? Well, I would like to see a theory
that proposes effects, or I would like to see an analogy with an
existing theory. In the absence of either, I would want data to
be gathered about possible effects and want to be able to
exercise control if the data indicates positive correlations
between microwaves and human disease. In any case, I want to be
able to decide the levels of risk I will be exposed to and not
let somebody else make this decision. I think there is much
justification for paranoia about the current level of microwave
exposure and the lack of mechanisms for exerting control over
this level.
Kamesh
------------------------------
End of HUMAN-NETS Digest
************************
∂22-Jan-81 1126 CSL.SSO.OWICKI at SU-SCORE Colloquium talk
Date: 22 Jan 1981 1125-PST
From: CSL.SSO.OWICKI at SU-SCORE
Subject: Colloquium talk
To: mccarthy at SU-AI
John,
Your title has me fascinated. Unfortunately, I'll be in POPL at that time.
Do you have anything written on the subject?
Susan
-------
∂22-Jan-81 1243 TOB NSF site visit
Schedule for site visit, Friday, Jan 30, 1981
draft, (Jan 22, 81)
Dr. William Spurgeon, NSF, Program Manager for Production Research and Technology
Division of Applied Research, NSF
Dr. Richard Schoen, NSF, Division of Applied Research
9:00 John McCarthy
9:30 Debra, Wm. Reynolds
Institute for Advanced Manufacturing
10:00 Roth
Collaboration with Mechanical Engineering, 1965-1980
10:30 Binford
Technical History and Overview, 1965-1980
Context of Robotics Program and collaborations
Relation to other efforts
Relation to Productivity
11:30 Demos
12:30 Lunch
1:30 '78 Proposal Goals: How we stack up
AL: Goldman
Mujtaba
Vistnes
Force: Salisbury
Simulator: Soroka
Vision: TOB
3:00 Goals from now on
∂22-Jan-81 1732 DHM
Professor McCarthy,
Contrary to what I thought he had told me earlier, Professor Manning now tells me
that he does not think anyone will object to Tom Binford's attending the EE quals
meeting on my behalf. Since he knows me better and is willing to support me as a
doctoral candidate, I think Tom's attendence at the meeting will get me by if any-
one's will. If I'd known earlier what Manning is now telling me, I never would
have had to make such a pest of myself to you, so, sorry about that, and thanks
for considering going in the first place.
Dave.
OK, good luck.
∂22-Jan-81 2145 TOB
John
Thanks.
Tom
∂22-Jan-81 2111 JMC
The papers have gone to the dean.
∂23-Jan-81 0002 LWE ns file
To: JMC
CC: REG
I have tentatively entered an additional search-and-store request
to the ns system for the full text of certain stories, intended for a
limited time period (c. 4 weeks max), in addition to the accumulation
of ap digests. If this poses problems for disk file size please let
me know--I certainly don't want to impose unduly on limited resources.
If it's ok I'll send another courier tape to relieve disk space (if
I may prevail on Ralph Gorin's help once again with the disk-to-tape dump
at the appropriate time...). Thanks very much for all your help...
∂23-Jan-81 0249 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #18
Date: 23 JAN 1981 0539-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #18
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Friday, 23 Jan 1981 Volume 3 : Issue 18
Today's Topics:
Television - Standards Conversion,
Computers and the Deaf - CA D.E.A.F. Surcharge,
Terminal Design - Microwave Hazards, Home Info Services - MicroNet
----------------------------------------------------------------------
Date: 20 January 1981 03:24 est
From: Schauble.Multics at MIT-Multics
Subject: Video standards
There are three standards in common use worldwide for television
signal transmission: NTSC (American), PAL (German), and SECAM
(French). Of these, NTSC provides by far the lowest picture
quality.
Clearly it would be desirable to convert the US to a better
standard. Equally clearly, conversion presents tremendous
problems. Do you think that a conversion is possible? How
would you go about it?
Some data: There are over 100 Million television sets in
operation. The average useful life of a set is 7 to 9
years, with newer sets expected to have a longer life.
------------------------------
Date: 22 January 1981 1206-EST
From: Hank Walker at CMU-10A
Subject: deaf phones
Has anything been resolved about the surcharge to pay for deaf
TTYs? I know that a surcharge will be used, but is this the
final answer? Are court cases still pending?
This sets a precedent. When the phone company starts offering
fancy graphics terminals, will they have to build text to voice
things for the blind?
Competitive, and basically unregulated industries don't appear
to have to pay for special equipment for the handicapped and
charge other users. For example, cars for paraplegics. Now
that the phone company is becoming more and more unregulated,
will the PUC still have the power to make such decisions?
It's clear to me that they wouldn't if the phone company
wasn't a utility. This issue seems like one that should be
solved by the electorate.
Note that I haven't decided yet whether the surcharge idea
is good or bad.
------------------------------
KRAUSS@MIT-MC 01/22/81 08:57:46 Re: Low Level Microwaves
The following abstract is reproduced almost verbatim from
the Proceedings of the 1979 meeting of the Bioelectromag-
netics Society:
ULTRASTRUCTURAL NEUROPATHOLOGY IN AREAS OF INCREASED
BLOOD-BRAIN PERMEABILITY AFTER MICROWAVE IRRADIATION
by Ernest N. Albert, Dept. of Anatomy, George Washington
University Medical Center, Washington, DC 20037
Adult Wistar rats and Chinese hamsters were exposed to CW
microwave irradiation of 2450 and 2800 MHz at 10 mW/cm**2
for 2 hours. Immediately after the exposure animals were
anesthetized and an electron dense protein, horseradish
peroxidase, was injected via the femoral vein for identi-
fication of areas of brains whose blood-brain barrier had
become leaky.
The following ultrastructural changes associated with the
blood-brain barrier alterations were observed.
1) Endothelial cell flooding with horseradish
peroxidase indicating increased permeability
of the plasma membranes,
2) swelling of astrocytic end feet,
3) myelin figures in dendrites indicating
possible degeneration,
4) membranous proliferation in dendrites,
5) occurance of platelet microthrombi in a
few capillaries.
Some of the observed changes could present serious health
hazards while others may be of lesser concern. Details of
these observations will be discussed.
------------------------------
I don't claim to understand the details of this report,
but apparently researchers have identified a physiological
mechanism, other than cell heating, whereby microwave
radiation interacts with biological systems.
---Jeff Krauss---
------------------------------
Date: 22 Jan 1981 (Thursday) 1350-EDT
From: WESTFW at WHARTON-10 (William Westfield)
Subject: Russian Microwave standards....
My E & M professor once told us that the Russian standards
for microwave emission were rather ridiculous in that they
were on the same order of magnitude as the microwave power
emitted by a human body (black body radiation, remember?).
Thus if you attend a party, the microwave energy densities
exceed the Russian standards. Maybe this is why Russians
have fewer parties than we Americans ?
( On the other hand, he never gave us any numbers, and I
never bothered to figure it out, so maybe this isn't true. )
Bill W
------------------------------
Date: 18 Jan 1981 (Sunday) 1107-EDT
From: DREIFU at WHARTON-10 (Henry Dreifus)
Subject: From the Wire: MicroNet
The Weekly VIDEO column
By John Teets
(c) 1981 Chicago Sun-Times (Field News Service)
Everybody loves a gadget, and I admit an instant addiction
that has just developed. It grew out of last week's column about
the strange videotapes of "indies," independent video producers,
but in a rather indirect way.
Some weeks ago, I was talking with the indie who put together
most of the TV spots for Chicago Mayor Jane Byrne and former Cook
County State's Attorney Bernard Carey in recent years. He asked
me if I remembered one of Carey's ads about crime rates that
featured bar graphs, in color. Of course I did; the feature was
pretty interesting. What hardware did he use to produce it? He
laughed: The whole thing had been done on a Bally Arcade, a home
video game that the indie had programmed to paint bars across
the screen.
So much for sophisticated machines - but the seed had been
planted.
And last week, while I was thinking about Arcades and Atari
video games, one thing led to another. Before I knew what was
happening, I was typing on a keyboard hooked up to a computer
in Columbus, Ohio, displaying its wonders on a regular tele-
vision screen - a loaned mini-computer terminal, spinning the
stuff dreams are made of, at least for a gadget-lover.
I've mentioned the MicroNET national computer system before;
it has since become one facet of the CompuServe Information
Service, a company offering computer hookups anywhere in the
country for any lucky soul with a home terminal. Besides news
and features from the Columbus Dispatch and the Associated
Press, it offers menu plans, NFL injury statistics, advice on
wood stoves, electronic mail (either from user A to user B, or
from user A for systemwide broadcast) and a few hundred other
functions. One plunge into MicroNET and the system becomes a
regular computer, ready to run standard programs or help you
in writing your own.
All you need is a mini-computer with adequate memory-like an
Apple, a TRS-80, even an Atari with the proper extras. Prices
start at $400 or so, but be warned: They can top $2,000 for the
biggest home units. Add a hookup so the computer can tap into
phone lines and you're ready to apply for a pre-arranged password
with a computer network. (CompuServe is one of two emerging
leaders in the field; the other is called The Source, based in
McLean, Va., and offering a comparable but different data base.)
Fire up the computer, dial the phone, type in the password
and, voila, news, features and computer capabilities.
And then there are the games - and more games and more
games. Hour on hour I sat in front of the keyboard, alternately
struggling through a Tolkien-esque world of magic words and
caves in a game called Adventure, or shooting down Klingon
ships in Star Trek.
Actually, my Enterprise cruiser suffered more damage than any
alien vessel, and usually the computer tried to humiliate me as
well: "Foolish mortal, you have attempted to venture out of the
galaxy into sub-space. On the third time you try this, the living
god, Korp, will destroy you...." He (or it or whatever) finally
did - in several successive games.
Most ended with a rather sad report: "You have been taken
to Klingon headquarters. If you had a starbase, you would be
repatriated and given a new starship to command. Since you
have no starbase, you will be mercilessly tortured to death
by the archfiend Cecil Dybowski, scourge of the universe...."
Adventure was a little kinder. When my time limit was
up, the screen suddenly told me: "A large cloud of green
smoke appears in front of you. It clears away to reveal a
tall wizard, clothed in gray. He fixes you with a steely
glare and declares, 'This adventure has lasted too long.'
With that, he makes a single pass over you with his hands,
and everything around you fades away into gray nothingness....
You scored 0 out of a possible 350, using 50 turns. You are
obviously a rank amateur...."
True, though, after switching to more conventional
games, I won $500 in computer money at blackjack, lost it
all on craps and won it back on blackjack. Unfortunately,
none of the computer money applies to the hookup fee-$5 an
hour after 6 P.M. local time, and a rate almost three times
that for business hours service, when commercial clients put
the heaviest demand on the system. It's all automatically
billed to a bank card, so you needn't note the magnitude of
your obsession till the end of the month. (You can call up
an accounting of your bill at any point, however.)
Still, the real fun began with a switch into the main
computers - on which users have 126,000 spaces of active
memory, for those who know about such things. It's quite a
thrill for a total novice like me to be told by a computer,
in essence, to strap yourself in: "The next page will take
you into MicroNET," the screen says with a tinge of fore-
boding - and before you know what has happened, bingo. The
commands aren't quite as plain-English as in the CompuServe
mode, but they're easily understandable after a few minutes'
practice, even for a technical ninny like me.
The first computer "job" was serious stuff, of course.
I ran my biorhythm chart for the first three months of 1981.
From what I learned, I shouldn't even be typing this, judging
from the severity of the first part of the year. "JAN 19: A
critical day, try to avoid overexertion.... JAN 31: A critical
day, decisions may be poor, avoid physical excess.... FEB 4:
Emotional upset possible, take care...." (I wondered if it
was singling me out for abuse, much as the "rank amateur"
and "foolish mortal" responses had been inserted into the
CompuServe games - but no. I ran a friend's program and
learned that, for him, the month was a computerized string
of pearls: "Your mind is exceptionally sharp today.... A
great day, go get 'em, tiger.... Things should go quite
smoothly today...." I should have it so good.)
Other programs, of course, will do everything from
text editing and word processing to explanation of program
languages, statistical tabulations, small business appli-
cations and exchanges of programs between users (if both
parties consent). Many mini-computers will perform similar
stuff, given the right programming - and that's the big
selling point of MicroNET: You don't have to design the
system yourself; just plug into standard, tested software
at a reasonable hookup charge.
Yes, indeed, a wondersome gadget. There are only two
things blocking total bliss: the lack of infinite time,
and the lack of infinite money.
------------------------------
End of HUMAN-NETS Digest
************************
The California P.U.C. decision to require the telephone companies
to provide teletypes to the deaf sets or follows a bad precedent.
If the State considers that the deaf should be provided with free
teletypes, then the Legislature should pass a bill appropriating
money to pay for it. To the first order it amounts to the same
assuming that everyone in California uses the telephone. The non-deaf
pay for the teletypes for the deaf. However, there is an important
second order difference. Suppose an issue arises about the
adequacy of the teletypes. For example, suppose the rest of
the country adopts, as seems likely, a different standard of
communication among the deaf than California's. If the legislature
is paying, it can decide whether and when to upgrade according
to whatever criteria for spending money on social benefits are
then current. Under the P.U.C. decision, the costs are open-ended,
and a lawsuit requiring that "adequate" teletypes be provided
leaves the issue up to a judge.
The P.U.C. decision is an example of a practice of the 70s that has fallen into
disrepute - namely, instead of appropriating money to provide a
benefit, the government passes a law requiring someone else to
provide it. The examples getting the most publicity involved the
Federal Government requiring the states to provide a service.
Unfortunately, there isn't an absolute criterion separating the
obligation to provide a service to the whole of a body of users
in exchange for the monopoly given by a franchise. Thus it
has been illegal since the first telephone franchises were given
in the 1880s for a telephone company to provide inaudible phones
in part of the region it has undertaken to serve in order to
discourage business in an expensive area.
∂23-Jan-81 0421 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 23 Jan 1981 0054-PST
From: Ted Anderson <OTA at SU-AI>
To: space at MIT-MC
a004 2146 22 Jan 81
AM-Rocket Accident, 1st Ld - Writethru, a620,240
Eds: Subs 1st graf to change rammed to hit, subs 2nd graf to CORRECT
that whole rocket cost $17 million and subs 6th graf to CORRECT that
door didn't fall.
Rocket Knocked Into Tower During Assembly Accident
CAPE CANAVERAL, Fla. (AP) - A launch tower door hit a Delta rocket
being prepared for launch Thursday, knocking the rocket from its
upright position, officials said. No one was hurt.
National Aeronautics and Space Administration spokesmen said the
damage was ''considerable,'' but had not been assessed. Delta rockets
cost $17 million in 1977.
The rocket, owned by McDonnell Douglas Corp., was being prepared for
a March 12 launching of a weather satellite. The accident at 10:25
a.m. EST had no connection with the space shuttle, the first reusable
U.S. spacecraft, which is scheduled for its maiden launch March 17
from another pad here.
''Officials don't know how much damage has been done, and nobody
knows whether the launch will have to be delayed,'' said NASA
spokesman Hugh Harris.
The rocket's first stage and inter-stage had been set up on the
launch pad earlier this week, and the first of nine solid fuel
boosters used for the launch were placed in a sling inside the mobile
launch pad tower, officials said.
As the tower moved into position to install the booster, a large
door struck the Delta, officials said.
The moving tower pushed the Delta several feet from its upright
position and pulled free two of the three bolts that hold the rocket
to the launch pad, they said. The Delta tilted back into the tower and
was left leaning into it, they said.
ap-ny-01-23 0046EST
***************
------------------------------
End of SPACE Digest
*******************
∂23-Jan-81 0641 JRA computer science summer institute
john,
i'd asked you many months ago about possibility of your guest lecturing at
summer CS institute course on LISP i'm giving. the course is in early
july and therefore won't conflict with ijcai. is your answer still "yes"?
if so, i need a biography and a picture for the poster very soon. btw,
the course will be at santa clara univ. this year, and topic is up to you.
i'd suggest an afternoon session in mid-late week; by then they know
what's going on (hopefully).
What are the dates of the CS institute?
∂23-Jan-81 1218 CSD.NOWICKI at SU-SCORE News from Usenix
Date: 23 Jan 1981 1212-PST
From: CSD.NOWICKI at SU-SCORE
Subject: News from Usenix
To: @sun at SU-AI
There were some interesting seesions at the Unix conference Thursday
on networking. There were three implementations of TCP discussed.
MITRE has one running in a front end processor (Z8000 based) with
their own host-frontend protocol (sound familiar?). This one sounds
like a loser. The BBN implementatin is not much better. They claim
to be installing some test sites in March, but privately one the
programmers admitted it won't be out till the end of this year.
Of course they require all sorts of Kernel hacks, though it will
be part of the Berekely system soon. The most impressive implementation
was that of 3COM. Half of 3COM was there, Bruce Borden and Greg Shaw.
They had a terminal hooked up to a system running TCP and everything
was up and running: user/server telnet, ftp, and mail. There are
two major problems, however. They let me look at a bit of their code,
and it was a typical Unix Hack. No comments except for the copyright
notice, obviously done in a week. The other problem is the ridiculous
price: $7,000 for the first source liscense and $4,000 for each aditional
CPU. Unless we can work out some special deal, that is probably too
expensive to justify. I did notice that the only system that 3COM has
inhouse is an Onyx Z8000 workstation (there were several on display)
and perhaps they could exchange some computing power....just a wild idea.
-------
∂23-Jan-81 1319 RPG Letter
I am applying for a research position at the Co-ordinated Science Lab at Illinois
to work with Dave Waltz. Winograd wrote me a marginal letter, which I think
was an accurate reflection of our mutual displeasure. Dave is looking for a
number of strong letters from senior people to discredit Winograd's letter,
both now and in the future.
To give you a brief outline of my interaction with Winograd,
we have constantly disagreed about my research and also about
what constitutes a valid line of attack in AI. To a large extent
I view a lot of this failure with the breakdown of communications
while he was a Xerox pursuing whatever he was pursuing. You and
I discussed this during his tenure battle.
I would be appreciative of any action you could take to help me
get out of Winograd's clutches.
The letter should go to:
Dr. R.T. Chien
Co-ordinated Science Lab
University of Illinois
Urbana, Ill, 61801
If you don't think that this is appropriate, please let me know.
-rpg-
On the basis of my present knowledge, I can say that you do good
systems work and are helpful. I suspect that to do good I would
have to be in a position to praise your thesis work, and to do that
I would have to see a copy, which I haven't. I wouldn't be able
to look at it seriously for two weeks. Would that be soon enough,
and can you give me a copy?
∂23-Jan-81 1400 JMC*
file equipment
∂23-Jan-81 1418 RPG
∂23-Jan-81 1406 JMC
On the basis of my present knowledge, I can say that you do good
systems work and are helpful. I suspect that to do good I would
have to be in a position to praise your thesis work, and to do that
I would have to see a copy, which I haven't. I wouldn't be able
to look at it seriously for two weeks. Would that be soon enough,
and can you give me a copy?
Yes that is soon enough. I will be able to give you a copy soon
also. It may be sufficient to comment only on those things you
know about now. I am preparing a shortened version of the more
important parts of my thesis (excluding almost all the NL stuff),
which should be easier to read and more informative about the general
ideas behind it. I might prefer giving you that since the thesis has
a large amount of laborious detail, which I'm sure you would prefer to
not think about.
-rpg-
∂23-Jan-81 1627 Nilsson at SRI-KL CBCL
Date: 23 Jan 1981 1627-PST
From: Nilsson at SRI-KL
Subject: CBCL
To: JMC at SU-AI
cc: Nilsson
Several of us had a meeting to discuss your CBCL language idea today.
We concluded that we would like to see a project to develop such
a language. We did not yet identify an SRI individual who would
see to it that a proposal got written and that the project gets off
the ground.
I am going to Washington in a couple of weeks. It was decided that
I should inquire of various sponsors who I'll be seeing whether or
not they would be interested in funding such a project. May I leave
a copy of your memo with people like Marvin Denicoff?
In the meantime, we might be able to get started by aiming some of
our distributed ai work in the direction of machines talking to
each other about business problems. We should keep in touch about
that possibility.
Also, I might talk to a friend of mine in the business school about
your idea. I presume that eventually we'll need to become familiar
with the sorts of things that business oriented machines will want
to say to each other.
So, the idea is now somewhere between the back burner and the
middle burner.
On another subject: have you had a chance to talk to Ed about the
possibility of closer SRI-Stanford collaboration on natural language
research (where we really are interested in how to represent things
that are said in natural language)? To summarize, we would be very
excited if Stanford were to add an appropriate nat. lang. person to
its faculty (and could help finance the project by providing half-time
support, maybe). We could also help support students in that area.
--Nils
-------
I have no objections to your giving people copies of the CBCL memo.
I haven't seen Ed, and I suspect he will evade involvement until
he returns to being department head.
∂23-Jan-81 1700 TW
To: BKR at SU-AI, KAL at SU-AI, DBL at SU-AI,
genesereth at SUMEX-AIM, JMC at SU-AI, RWW at SU-AI
Does anyone have an answer for Brian McCune?
----------
∂20-Jan-81 1452 BPM Distributed AI
Terry,
Who is doing distributed AI at Stanford? Keith Lantz or Brian Reid?
Any RAs or students? Thanks.
Brian
∂23-Jan-81 2122 SGR at MIT-MC (Stephen G. Rowley) Bureaucrats & natural gas in Massachusetts.
Date: 24 JAN 1981 0020-EST
From: SGR at MIT-MC (Stephen G. Rowley)
Subject: Bureaucrats & natural gas in Massachusetts.
To: ENERGY at MIT-MC
As some of you may know, Massachusetts has had quite a severe
shortage of natural gas the past few weeks. The Governor declared
an energy emergency, required industrial & commercial thermostats
turned down to 55F, etc.
However, a bit of madness appears to be prevailing in the state .
Today, the state Dept. of Transportation issued its guidelines for
the transport of hazardous substances. Shipments must go through more
miles of red tape than road, must not pass near large population
centers, etc.
The kicker is that they declared liquefied natural gas to be a
hazardous substance right along with the REAL dangerous ones.
So we are faced with an extreme shortage at the same time the state
decides to place restrictions on how the stuff can be shipped... Sigh.
-$teve
∂23-Jan-81 2345 TOB
To: EAF at SU-AI, JMC at SU-AI, FB at SU-AI, ZM at SU-AI, DCL at SU-AI
Joint ARPA funding
I favor consolidating Stanford's ARPA funding for reasons stated in
Feigenbaum's message. I believe that we should aim for a much larger
computer support, on parity with CMU and MIT. I would like arrangements
to simplify participation on our parts and favor something like
the following:
1. Funds of an individual investigator are to be separately identified
in the budget, are to be under the discretion and control of that investigator,
are to be maintained in separate accounts, and any overspending by one
investigator is not to affect the budget of any other.
In the past, it has been impossible for us
to make accurate projections of our part of a joint ARPA
account because expenditures depended on decisions of others.
2. We want representation in decisions affecting centralized or overhead
costs, e.g. managerial and accounting personnel and policies.
Overhead and centralized costs are to be directed by a group with broad
representation. This includes choice of centralized research equipment
especially computer equipment and its support in a cost center.
Surprise decisions regarding
CSD-CF are an example of taxation without representation.
Decisions have been largely reasonable and CSD-CF accomodating, but we have
found computing costs more than double for a grant after submitting
the proposal. I reason that by some participation in decision making,
we will know more about cost ctors, and the sooner we know of cost factors,
the sooner we can make reasonable predictions.
We would like an unspecified class of decisions on CSD-CF to be largely
responsive to users, not to administrative convenience.
3. Research equipment, especially computers, should be reserved for
research purposes, not burdened with general teaching. With the price
of computing, it is false economy if researchers are idle so that computers
are not idle. That is, research equipment should be used only for course
work directly related to the research supported (perhaps including course
work of students supported by the research contract).
I do favor adequate computing facilities for student course work, but I
believe that it should be clearly distinguished. Also, I believe that
there is adequate financial support for academic computing for computer
science, that organizing energy is a major limitation.
Tom
∂25-Jan-81 0036 POURNE@MIT-MC
From: POURNE@MIT-MC
Date: 01/25/81 03:35:36
POURNE@MIT-MC 01/25/81 03:35:36
To: JMC at MIT-MC
Possony is coming down Friday AM, if you're coming then you
might want to come together.
We're getting more and more showing up. This is likely to
REALLY be THE space planning conference of the year.
∂25-Jan-81 0039 POURNE@MIT-MC lots of people don't read sf lovers
From: POURNE@MIT-MC
Date: 01/25/81 03:39:28
Subject: lots of people don't read sf lovers
POURNE@MIT-MC 01/25/81 03:39:28 Re: lots of people don't read sf lovers
To: JMC at MIT-MC, REM at MIT-MC, MINSKY at MIT-MC
As why should they?
POURNE@MIT-MC 01/24/81 04:41:38 Re: Cheers?
Awaken me early in the morning, mother, for I'm to be queen of
the may..
Actually, Niven and Pournelle are pleased (and relieved) to
announce that after ten years they have FINISHED the novel OATH
OF FEALTY, which wil go to the publisher this week and will
probably be in print about Fall of 1981.
OATH stars a cast of splendid characters, telepathic
love-making, gratuitous zaps at Luddites, needles stuck into
Proxmire, and the best city you've ever seen.
Don't fail to miss it if you can.
JEP
I look forward to reading your new book in the Fall. When electronic
publication comes, the waits will be much shorter.
∂25-Jan-81 0648 Darden@SUMEX-AIM hunger
Date: 25 Jan 1981 0644-PST
From: Darden@SUMEX-AIM
Subject: hunger
To: JMC@SAIL
Hi, John. Is there any chance you will be getting hungry this spring?
Do let me know if and when you have any plans from coming to Washington.
As I think back on my visit t Stanford, the conversations we had
certainly are among the most vivid and interesting of my memories.
I am now teaching philosophy of science (what is science?) nd
history of modern biology (why would anyone ever think that
species change over time?) and feel all too fr removed from the
AI material (can we use analogies to be creative?) that I most
want to think about right now. Keep me posted on common sense.
Regards, Lindley.
-------
I also remember our conversations with pleasure, and I will surely
be hungry the next time I come to Washington.
∂25-Jan-81 1030 Feigenbaum@SUMEX-AIM more on ARPA consolidation
Date: 25 Jan 1981 1021-PST
From: Feigenbaum@SUMEX-AIM
Subject: more on ARPA consolidation
To: jmc@SAIL, fb@SAIL, zm@SAIL, dcl@SAIL, buchanan@SUMEX-AIM,
To: csd.ullman@SCORE
cc: feigenbaum@SUMEX-AIM, CSD.JEANIE@SU-SCORE
In response to Binford's message (which will be forwarded to all addressees
on my previous message, along with this further discussion; please
keep everyone in the loop on this interchange):
1. One of the characteristics of a consolidation is that overspending by
one PI inevitably affects the others. For who is the backup to an
errant soul who overspends? The department (No, it has nothing to do with
ARPA research projects)? ARPA (No, because they pass lump-sum funds to us)?
Hence the administrator of the consolidated project must make sure that
no one PI overspends (long before the overspending occurs). This is more
easily said than done, but it is do-able. This is one of the dangers
of consolidation.
There may be times at which we all agree that person X, needing the funds,
should go ahead and overspend, using person Y's money (since Y is
underspending for some reason). These will be explicit decisions and
constitute a source of flexibility in consolidation.
2. On the question of control of policies (overhead, central equipment,
personnel), I envision the control of the consolidated monies by a
"board" of PIs (anyone who gets ARPA money). In the practical day-to-day
working of this scheme, there would not be a "chief PI" who is
organizing or dictating. However, a PI with a 50K grant that includes
one programmer and one student should not expect to have as much say in
the "central " decisions as a PI administering a $500K chunk of money. The
latter has a much great "interest" at stake in the decisions.
3. It is both scientifically wrong and illegal to run student classwork
on research machines. I'll have to find out what happened in the past few
months to make this a salient issue. Only student RESEARCH is allowed on
our research machines!
Will some of you others please contribute to this discussion? Please?
Ed
-------
∂25-Jan-81 1351 DCL
To: "@ARPA.LST[1,DCL]" at SU-AI
Questions concerning central administration of Stanford CS ARPA contracts.
1. Where exactly does ARPA save (or expect to save) on paper work?
It would seem that each project manager has just as much to do in analysis
of his interest in a joint proposal as in a separate proposal in his specific
area. In fact, a joint proposal might be more work.
The saving may only be evident after ARPA has made the funding decisions,
say at the appropriation, audit, and award phase?
An accurate answer to this question may help us to design a sensible format
for a "joint" proposal. There might be several possibilities, ranging from the
old AI Lab. format with a joint budget, to a set of separate proposals and budgets
bound together in one volume. The format clearly affects the number of
duties to be performed by a "board". It seems the less the board has to do, the
better.
2. What exactly are the formats of the joint ARPA proposals at MIT and CMU?
Do they have "boards", and if so, what are their duties and functions?
In short, how do they do it and what problems have they had?
3. What exactly would a "board" do?
I think this question a bit premature. Even so, we seem to have started discussing
the list of duties. The items discussed so far seem to fall into
at least three classes:
decisions subject to weighted vote (e.g., equipment), decisions requiring
unanimity (e.g., can A spend B's money), and decisions possibly beyond the scope
of the board but possibly affected by department policies (e.g., can C hire D).
-David
∂25-Jan-81 1959 JK ekl
Seems ready. I will re-write the manual this week and prove Ramseys
theorem.
The new features added contain running of programs with deletion
of intermediate lines and the global unparseflag which
determines whether formulas are printed in normal or lisp form.
∂25-Jan-81 2133 Allen.Newell at CMU-10A Re: The knowledge level
Date: 26 January 1981 0035-EST (Monday)
From: Allen.Newell at CMU-10A
To: John McCarthy <JMC at SU-AI>
Subject: Re: The knowledge level
In-Reply-To: John McCarthy's message of 23 Jan 81 03:11-EST
Message-Id: <26Jan81 003508 AN02@CMU-10A>
John: (1) I will put another copy in the mail. I will probably
scribble some more warnings on it -- that is it preliminary.
I really may make some changes in it given all the feedback
(diverse, you can be assured) I've gotten on it.
(2) Thanks for feedback a few days ago. I will attend to it
carefully. I have the paper on Ascribing mental qualities
to machines, though interestingly I didn7t go back over it
when checking out some of the stuff. I do not seem to have
Michies #9, so would appreciate a copy of the paper on
First order theories of individual concepts and propositions.
(3) In fact, I took the Pat and Mary example not from your
IJCAI77 paper on Epistomological problems ..., but from
the little not entitled "Predicate Calculus" on page 987,
which was part of Danny B's session. There it states:
"3. When told that Mary has the same telephone at Mike,
that Pat knows Mike's telephone number, and that Pat dialed
Mike's telephone number, the program should assent to the
statemennt that Pat dialed Mary's telphone number, and
should express ignorance about whether Pat knows Mary's
telephone number."
My "Not know if Pat knows .." seems an ok translation of
"should express ignorance about ..." This example doesn't
appear in the big talk, cause I thought it did and went
there and look. Pat and Mike and safes. I was just after
the simplest possible exaample, since I didn' make use
of the details.
AN
∂26-Jan-81 0012 LLW Commuting to Pournelle's Meeting
To: JMC
CC: LLW, RAH
∂25-Jan-81 2339 JMC Pournelle meeting travel
When are you going to L.A.? It would be interesting to talk en route.
My present plan is to go from San Jose to Burbank Friday morning.
[John: Rod and I are planning to drive down Friday AM and return via the
same mode Sunday afternoon/evening, partially in order to avoid the dual
airport hassles on each trip leg and also to have time to talk. We would
be delighted to have you join us, if you'ld care to do so; since the drive
down IS 5 is about 5 hours, we are planning to leave the Lab about 0700
Friday. Lowell]
Tentatively, I will join you. I assume there is a place I can leave my
car at the Lab. I might chicken out, but I think I won't. Thanks.
∂26-Jan-81 0028 LLW Friday Dawn Rendevous
To: JMC
CC: LLW, RAH
∂26-Jan-81 0023 JMC
Tentatively, I will join you. I assume there is a place I can leave my
car at the Lab. I might chicken out, but I think I won't. Thanks.
[John: We'll plan to meet you at the Lab's South Main Gate (the one that
enters from East Avenue, about half way down the Lab's southern edge from
the East Avenue-Vasco Road intersection) at 0700, unless we hear from you
by 2200 hours Thursday to the contrary. There's an always-accessible
parking lot right at hand for your car. I hope you decide to make it!
Lowell]
One last question. How much time should I allow at that hour from Stanford?
I drive fairly fast.
∂26-Jan-81 0713 JRA summer dates
dates are july 6-10, with a guest lecture by you around 9-10 if your schedule
allows.
history question: did the name "OVERLORD" as the monitor for 704 lisp, come
from clarke's "childhood's end"?
OVERLORD is Steve Russell's name; I don't know were he got it. Are the
dates right for that?
I have no conflicts with the dates. Did you mention a financial
arrangement?
∂26-Jan-81 0828 cracraft@sri-unix SPS article and info sources
Date: 26 Jan 81 3:39:17-PDT
From: cracraft@sri-unix
To: energy@mc
Subject: SPS article and info sources
There's an interesting article in February's issue of OMNI (SPACE section)
about SPS. A couple of addresses for further technical information are given:
"Satellite Power Systems Concept Development and Evaluation
Program, Preliminary Assessment" DOE/ER-0041, Sept 1979
Price $4, from
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
"Some Questions and Answers about the Satellite Power System"
also $4 from NTIS
∂26-Jan-81 0833 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Consolidation
Date: 26 Jan 1981 0831-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: ARPA Consolidation
To: CSd.Feigenbaum at SU-SCORE, CSD.Ullman at SU-SCORE, JMC at SU-AI,
CSD.Buchanan at SU-SCORE, CSL.FB at SU-SCORE, CSL.Lantz at SU-SCORE,
TOB at SU-AI, ZM at SU-AI, DCL at SU-AI, ACY at SU-AI, Wiederhold at SUMEX-AIM
cc: CSD.BScott at SU-SCORE
o
With regard to overspending, underspending, etc., it is not my understanding
that all principal investigators would be operating out of one account. This
would be absolutely impossible--from my point of view--to administer.
There is a much easier way: the various research efforts and dollars would
be set up in separate accounts--suballocations from the prime contract.
By doing this each P.I. would know at all times just how much money he
has, and I can control overspending because I watch the balances.
I don't think it necessary to go into all the administrative details in this
message, but separating the money is not a difficult task IF ARPA ALLOCATES
SPECIFIC AMOUNTS TO SPECIFIC RESEARCH EFFORTS in the various modifications
it issues.
By the way, don't you think a meeting to discuss the pros and cons would save
everybody's time?
Betty
-------
∂26-Jan-81 0902 Wiederhold@SUMEX-AIM Re: ARPA Consolidation
Date: 26 Jan 1981 0900-PST
From: Wiederhold@SUMEX-AIM
Subject: Re: ARPA Consolidation
To: CSD.BSCOTT@SU-SCORE, CSd.Feigenbaum@SU-SCORE,
To: CSD.Ullman@SU-SCORE, JMC@SU-AI
cc: sagalowicz@SRI-KL, kaplan@SRI-KL
In response to the message sent 26 Jan 1981 0831-PST from CSD.BSCOTT at SU-SCORE
I think a joint meeting would be helpful. I have been pleased with the
support through the intermediate agency we have ( NAVELEX) and I7d hate to
loose that due to consolidation. Gio ( KBMS).
-------
∂26-Jan-81 0944 Buchanan@SUMEX-AIM Re: more on ARPA consolidation
Date: 26 Jan 1981 0942-PST
From: Buchanan@SUMEX-AIM
Subject: Re: more on ARPA consolidation
To: Feigenbaum@SUMEX-AIM, jmc@SAIL, fb@SAIL, zm@SAIL, dcl@SAIL,
To: csd.ullman@SCORE
cc: CSD.JEANIE@SU-SCORE
In response to the message sent 25 Jan 1981 1021-PST from Feigenbaum@SUMEX-AIM
A single grant from ARPA can still be split into separate University
accts. Each PI would still have responsibility for balancing the
specific acct, but we gain a little flexibility in moving money across
accts without involving ARPA.
I have no strong objection. I believe a central board of PI's would
only need to decide on (a) initial proposal amounts for each sub-project,
(b) negotiated actual amts to allocate to each acct if proposed and
awarded amts differ, (c) transfers of money between pairs of accts.
Bruce
-------
∂26-Jan-81 1145 JMC*
Friedl
∂26-Jan-81 1145 JMC*
Elliott to be sure we aren't meeting.
∂26-Jan-81 1145 JMC*
possony
∂26-Jan-81 1145 JMC*
dr appointment
∂26-Jan-81 1148 Baskett at PARC-MAXC Re: more on ARPA consolidation
Date: 26 Jan 1981 11:13 PST
From: Baskett at PARC-MAXC
Subject: Re: more on ARPA consolidation
In-reply-to: Buchanan's message of 26 Jan 1981 0942-PST
To: Buchanan@SUMEX-AIM
cc: Feigenbaum@SUMEX-AIM, jmc@SAIL, fb@SAIL, zm@SAIL, dcl@SAIL,
csd.ullman@SCORE, CSD.JEANIE@SU-SCORE
Having had some experience with coordinated Arpa contracts, I feel that
you should not underestimate the additional effort that will be needed at
proposal creation and editing time. The fact is, as I see it, that a
coordinated proposal has to LOOK coordinated if it is to effectively
compete with the other proposals Arpa receives. That means that, at the
very least, it needs to be done in one single typesetting system with a
consistent style across the different sections. (A consistent style is
more than just the same fonts, too.) I haven't yet found secretaries who
can do what needs to be done but maybe they exist. My estimate, based on
two experiences with this so far, it that 2 to 3 man months went into this
additional effort on the VLSI contracts (these were NOT secretary months).
The coordinated look will be more difficult for Stanford, I think, than
for MIT or CMU where they have been more coordinated for some time and we
have been fairly oblivious to coordination for years. Now, I do think
that those high level man months can be easily recaptured in savings of
time in running the resulting programs but don't forget this necessary up
front overhead.
Forest
∂26-Jan-81 1220 TOB Bloom
John
I am not sure that we got to the point of your visit
a couple of days ago. Did you mean for me to try to
interest Elliott Bloom in path calculation and dynamics.
I will sound him out about it. He seems interested in
character recognition. I loaned him the proceedings
of Int Conf on Pattern Recog from Dec. He called to
say he will return them tomorrow. We arranged that he
will present a survey of the literature for our vision
seminar in March. Meantime, he has a report and won't
have much available time.
Comments?
Tom
∂26-Jan-81 1254 CSD.ULLMAN at SU-SCORE arpa consolidation
Date: 26 Jan 1981 1249-PST
From: CSD.ULLMAN at SU-SCORE
Subject: arpa consolidation
To: baskett at PARC-MAXC
cc: csd.feigenbaum at SU-SCORE, jmc at SU-AI, zm at SU-AI, dcl at SU-AI,
csd.jeanie at SU-SCORE, csd.buchanan at SU-SCORE
The notion that arpa wants us to combine unrelated resarch projects into one
proposal seems reasonable. What is unreasonable is that they should then
expect us to pretend that the mix is one monolithic project. Can they
seriously expect this?
-------
∂26-Jan-81 1310 Feigenbaum@SUMEX-AIM Re: ARPA Consolidation
Date: 26 Jan 1981 1309-PST
From: Feigenbaum@SUMEX-AIM
Subject: Re: ARPA Consolidation
To: CSD.BSCOTT@SU-SCORE, CSd.Feigenbaum@SU-SCORE,
To: CSD.Ullman@SU-SCORE, JMC@SU-AI
cc: csd.kuhn@SCORE
In response to the message sent 26 Jan 1981 0831-PST from CSD.BSCOTT at SU-SCORE
Jeff Ullman and I wish to call a meeting to discuss the issue
of ARPA funding consolidation, on Thursday, January 29.
The proposed time is noon--brown-bag lunch-- and the place will
be the Chairman's conference room in MJH. RSVP.
Ed Feigenbaum
-------
Unfortunately, I have a conflicting meeting at that time
of the Mathematical Sciences committee. I would like to be present
at the ARPA funding consolidation meeting.
∂26-Jan-81 1316 FWH talk in verification seminar
Last quarter you expressed interest in giving a talk on Elephant in the
program verification seminar. Because of scheduling problems I did not
propose a date in fall quarter. I now wonder whether you are still
interested in giving a talk; I'd be glad to arrange for a seminar this
quarter.
Also, David Luckham would like to learn about non-monotonic logic
and is interested in a talk on this topic, too. (Or would Jon Doyle be
someone to ask for such a talk?)
- Friedrich
I would be glad to give an Elephant talk towards the end of the quarter.
You would get quite different non-monotonic reasoning talks from me and
from Doyle.
∂26-Jan-81 1427 CSD.JEANIE at SU-SCORE ARPA meeting
Date: 26 Jan 1981 1409-PST
From: CSD.JEANIE at SU-SCORE
Subject: ARPA meeting
To: csd.feigenbaum at SU-SCORE, csd.ullman at SU-SCORE, jmc at SU-AI,
csd.buchanan at SU-SCORE, csl.fb at SU-SCORE, csl.lantz at SU-SCORE,
tob at SU-AI, zm at SU-AI, dcl at SU-AI, acy at SU-AI,
wiederhold at SUMEX-AIM, csd.kuhn at SU-SCORE
cc: csd.bscott at SU-SCORE
.John McCarthy cannot make it at 12:00 on Thurs 2/29.....how is 11:00 for
everyone? Please let me know if you cannot make it at this time.
--jeanie
-------
I teach at 11.
Also I won't be here on Friday.
∂26-Jan-81 1409 Feigenbaum@SUMEX-AIM (Response to message)
Date: 26 Jan 1981 1358-PST
From: Feigenbaum@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI
In response to your message sent 26 Jan 1981 1331-PST
John, let me see what I can do...Ed
-------
∂26-Jan-81 1459 CG trip to Sweden
I'll be leaving for Sweden this weekend (on the 31st of January).
I'll spend two weeks visiting the Mathematics department at
the University of Stockholm (from the 1st through the 13th) and
two weeks visiting the Computer Science Department at the University
of Gothenberg. I return to Stanford on the 28th of February.
∂26-Jan-81 1740 Baskett at PARC-MAXC Re: arpa consolidation
Date: 26 Jan 1981 16:31 PST
From: Baskett at PARC-MAXC
Subject: Re: arpa consolidation
In-reply-to: CSD.ULLMAN's message of 26 Jan 1981 1249-PST
To: CSD.ULLMAN at SU-SCORE
cc: baskett, csd.feigenbaum at SU-SCORE, jmc at SU-AI, zm at SU-AI, dcl at
SU-AI, csd.jeanie at SU-SCORE, csd.buchanan at SU-SCORE
"The notion that arpa wants us to combine unrelated resarch projects into one
proposal seems reasonable. What is unreasonable is that they should then
expect us to pretend that the mix is one monolithic project. Can they
seriously expect this?"
They won't seriously expect this until they are looking at the resulting proposal
and comparing it to the others and thinking what THEIR management will
think. Then you can be certain that the appearance of monolithicness will
matter.
∂26-Jan-81 2020 LLW Early AM Commute Times
To: JMC
CC: LLW
∂26-Jan-81 0030 JMC
One last question. How much time should I allow at that hour from Stanford?
I drive fairly fast.
[John: I get from the Lab Gate to Edward's home (a couple of blocks from
yours) in 54-57 minutes, driving legally most of the way. Lowell]
∂26-Jan-81 2132 Brian K. Reid <CSL.BKR at SU-SCORE> Arpa
Date: 26 Jan 1981 2127-PST
From: Brian K. Reid <CSL.BKR at SU-SCORE>
Subject: Arpa
To: jmc at SU-AI, fb at SU-AI, zm at SU-AI, dcl at SU-AI, buchanan at SUMEX-AIM,
csd.ullman at SU-SCORE, feigengaum at SUMEX-AIM, csd.jeanie at SU-SCORE
cc: Hennessy at SU-SCORE, Lantz at SU-SCORE
I have seen the exchange of messages from current PI's about Arpa
consolidation, and decided that it would not be out of order for me to
comment since I am a future PI. I also have the newcomer's perspective,
as Alexis de Tocqueville had while observing America in 1831.
The issue is not a common project, but a common purpose. We are all
doing research in Computer Science, and our effectiveness at that
purpose cannot possibly be increased by the divisiveness and
factionalism that characterize our department now.
We should not be concerned about grant management, but about
establishing a research community. A research community is a very
special thing, more than the sum of its parts. Princeton Physics in the
1940's; Thomas Edison's laboratory in Menlo Park in the 1920's. Caltech
and JPL in the middle 1960's during the heyday of the space program.
And for that matter, Xerox PARC in the mid 1970's.
CMU when I left it four months ago was a research community, alive and
vital and just humming with that je ne sais quoi that exudes from people
who know they are in the right place at the right time, and who knew
that the experience of working together as a community made everyone's
mind sharper.
Stanford when I arrived here four months ago was not so much a community
as it was a bunch of buildings, occupied by scientists too busy to talk
to each other, to whom unity meant sharing a receptionist and a Dover.
In learning the emptiness of Stanford, the closed doors and the
invisible faculty, I was overcome not by nostalgia for the way of life I
left behind, but by sadness in the realization that Stanford was so
vacant and so ineffectual by comparison. I probably should have seen it
as a harbinger when no faculty at all showed up for my job interview
talk; I certainly took it as confirming evidence when I saw only one
faculty member in my colloquium talk last week. It's not that I think
I'm terribly worth listening to, but at CMU it is fairly common for
every faculty member to be present at a job interview talk: they
are interested, because this applicant might someday be their colleague,
and they want a vote.
I don't really have much faith that things will ever change. There is
too much of a tradition of isolation and individual enterprise at
Stanford, and this ``every tub on its own bottom'' approach works well
in the more traditional scientific disciplines. Stanford is drowning in
Masters' students, and it is not reasonable to eliminate them, even
though they are by definition not contributing members of the community.
And finally, nobody ever likes to change the way they live--indeed, it
was the shock at being forced to change the way I did research by coming
here that made me think about these issues at all.
It seems to be possible to do good theoretical work without a full
research community, but it is not possible to do good systems work in
isolation. I don't think that Stanford will ever have a significant
research effort in the systems area without the formation of a research
community.
Brian
-------
∂27-Jan-81 0853 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Prof. Paz from Israel
Date: 27 Jan 1981 0842-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Prof. Paz from Israel
To: CSD-Faculty:
Prof. A. Paz from Israel presently visiting at Berkeley will speak
today at the AA Lunch Bunch, 12:30, Jacks 301. "A Generalization of
the Extended Euclid's Algorithm."
He will be available after 2 p.m. Would like to talk to anyone.
Please contact Andy Yao on 71979 if you are interested.
-------
∂27-Jan-81 0904 CSD.SCHREIBER at SU-SCORE Comprehensive Exam grading session.
Date: 27 Jan 1981 0901-PST
From: CSD.SCHREIBER at SU-SCORE
Subject: Comprehensive Exam grading session.
To: csd.ossher at SU-SCORE, jeb at SU-AI, csd.gischer at SU-SCORE,
csd.clarkson at SU-SCORE, jmc at SU-AI, acy at SU-AI, rwf at SU-AI,
csl.lantz at SU-SCORE, csd.schreiber at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
We will meet at 12 noon on Sunday Feb. 1 in room 252 MJH to grade the
exam. I expect 8 graders, which means each of us will grade about
3/2 sections). Lunch will be provided. If you can't come, please let
me know now.
Rob
---
-------
∂27-Jan-81 0944 DEW via SRI-KL answers to exercises
John, I finished the answer to the first exercise. It's the file
grab2.prf[1,dew]. It contains a few comments, the revised axiom system,
and a proof from FOL. The axioms and command file are also set up as
the separate files grab.ax[1,dew] and grab.cm[1,dew]. --kk
∂27-Jan-81 1207 CSD.JEANIE at SU-SCORE Thursday meeting
Date: 27 Jan 1981 1147-PST
From: CSD.JEANIE at SU-SCORE
Subject: Thursday meeting
To: csd.feigenbaum at SU-SCORE, csd.ullman at SU-SCORE, jmc at SU-AI,
csd.buchanan at SU-SCORE, csl.fb at SU-SCORE, csl.lantz at SU-SCORE,
tob at SU-AI, zm at SU-AI, dcl at SU-AI, acy at SU-AI,
wiederhold at SUMEX-AIM, csd.kuhn at SU-SCORE
cc: csd.bscott at SU-SCORE
NThe meeting has been scheduled back to the original time...12:00.
-------
∂27-Jan-81 1258 Feigenbaum@SUMEX-AIM Brian's message
Date: 27 Jan 1981 1249-PST
From: Feigenbaum@SUMEX-AIM
Subject: Brian's message
To: csl.bkr@SCORE, jmc@SAIL, fb@SAIL, zm@SAIL, dcl@SAIL,
To: buchanan@SUMEX-AIM, csd.ullman@SCORE, csl.lantz@SCORE
Thanks to Brian for his views, with "fresh eyes" of our situation.
However, let us not becloud the actual issue that
stands before us, to be discussed on Feb. 13. The issue for ARPA
is NOT (repeat definitely NOT) common purpose, but common paperwork.
ARPA wants to reduce the number of MROAs it sends upstairs
and the number of contracting actions that have to be taken each year
for stanford. If Brian's opinions are worthy ones, they
should be discussed in some other forum (like a faculty meeting).
In my view, we have more shared activity and purpose than Brian
thinks, but much less than CMU. The research eenvironment
here is DIVIDED(by choice) not DIVISIVE. Possibly we forego
some possible synergistic effect, but perhaps we get so much
work done because we stay out of
each others' hair.
Ed
-------
∂27-Jan-81 1424 Edward Feigenbaum <CSD.FEIGENBAUM at SU-SCORE> visit of Gordon Bell on Wed.Jan 28
Date: 27 Jan 1981 1421-PST
From: Edward Feigenbaum <CSD.FEIGENBAUM at SU-SCORE>
Subject: visit of Gordon Bell on Wed.Jan 28
To: jmc at SU-AI, dek at SU-AI, csl.fb at SU-SCORE, csl.lantz at SU-SCORE,
csd.buchanan at SU-SCORE
I found about by phone a couple of hours ago that Gordon Bell will
stop by Wednesday afternoon, between 2 and 4. If any of you are interested
in talking with Gordon about common interests, you'll find him
in or near the Chairman's conference room on the second floor.
Forest/Keith: perhaps the SUN terminal group should show Gordon new
developments.
Don: perhaps you or Luis wants to tell Gordon about new developments re TEX
Bruce: you and I should arrange to demo some HPP things for him.
Ed
-------
∂27-Jan-81 1824 GREEP at RAND-AI Login on SAIL for CS226 work
Date: 27 Jan 1981 1804-PST
From: GREEP at RAND-AI
Subject: Login on SAIL for CS226 work
To: JMC at SU-AI
I'm sending this to you instead of the TA because I don't know his name or
mailbox. I told him that I had a guest login [GRP,SJT] on sail but it
was scheduled to disappear. He said he would arrange for it not to vanish
but it seems to have done just that.
Steven Tepper
-------
The TA is Kurt Konolige, and he is konolige%sri-kl. However, I'll
resuscitate GRP,SJT for now, but you should send him a message
anyway.
∂27-Jan-81 2039 Mark Crispin <Admin.MRC at SU-SCORE>
Date: 27 Jan 1981 2028-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Postal-Address: 12155 Edgecliff Place; Los Altos Hills, CA 94022
Stanford-Phone: (415) 497-1407
To: JMC at SU-AI
In-Reply-To: Your message of 27-Jan-81 0146-PST
I haven't put any work into DTN, if that's what you mean.
-------
It seems unprofessional to me to put up programs and then not
fix gross bugs.
∂27-Jan-81 2136 Mark Crispin <Admin.MRC at SU-SCORE> DTN bugs and unprofessionalism
Date: 27 Jan 1981 2134-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Subject: DTN bugs and unprofessionalism
To: JMC at SU-AI, REG at SU-AI
1. The DTN bug you refer to only occurs when you are simulating
a Datamedia while you are using a Datamedia, something that only
somebody on an IMLAC would want to do.
2. I have no way of reproducing the problem, except by logging
into SAIL from a grIMLAC.
3. I am completely uninterested in doing any further system
programming on SAIL unless forced to, because:
. I have many other tasks at hand to do on SCORE, and
SAIL has its own system programmer.
. I'm convinced I'd be wasting my time doing so since
WAITS is a dead operating system. Every time I am
forced to program on WAITS I am repeatedly repulsed
by the "features" it has which prevent me from doing
a task in a reasonable manner.
. Having been told repeatedly that Frost (or before,
Harvey) is in charge of all SAIL system software
including programs I've written AND having control of
some of those programs taken away from me, I am quite
happy to leave Frost with the responsibility for
maintaining all the old WAITS programs I wrote.
4. Unsupported software, offered on an "as is" basis, exists on
every system. I doubt that you are advocating banning the
installation of unsupported software just because it may have
bugs that might not get fixed. Most of WAITS is unsupported!
5. I think it is grossly unfair to clutter up Ralph's mailbox in
order to try to pressure me to work on this.
-- Mark --
-------
1. The bug strikes when one is logged into SAIL on a Datamedia and
goes from there to SCORE, so it can be investigated without an Imlac.
2. The arguments you have offered for not looking at bugs in programs
you wrote reinforce my opinion regarding professionalism or lack thereof.
3. As you know I have given up control of SAIL, but I still take an
interest, and Ralph is the person to whom otherwise unresolved
complaints about CSD computer facilities must be directed.
∂28-Jan-81 0139 POURNE at MIT-MC (Jerry E. Pournelle) rain on Mongo...
Date: 28 JAN 1981 0440-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: rain on Mongo...
To: POURNELLE at MIT-MC, JMC at SU-AI
Guilty, except that the officers WERE willing to buy
drinks for anyone on the planet who'd make it easy for them by
coming to that bar at the right time....true, the number who
were likely to take advantage of the offer was considerably
smaller than the population of the planet.
∂28-Jan-81 1124 Buchanan@SUMEX-AIM Datamedia
Date: 28 Jan 1981 1122-PST
From: Buchanan@SUMEX-AIM
Subject: Datamedia
To: jmc@SAIL
I have a new terminal for my office so I am returning the one you loaned
me to Marty Frost. Thanks very much.
bgb
-------
∂28-Jan-81 1127 Denny Brown <CSD.DBROWN at SU-SCORE> Stanford football recruiting
Date: 28 Jan 1981 1124-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Stanford football recruiting
To: CSD-Faculty: ;
The football program is entertaining a set of potential football
players this weekend. At least one of them is interested in computing.
They would like one of us to attend a luncheon at Bill DeMent's
house (on campus) from about 12:30-2:30 on Saturday. Each faculty member
says a little about his department and a little about his own interests.
There is a little time for one-to-one questions with the appropriate
recruit. I usually do this, and would do so again, but I have a prior
commitment for Saturday. If anybody is interested, contact me and I'll
provide the appropriate introductions. -Denny
-------
∂28-Jan-81 1324 Waldinger at SRI-KL temporal logic
Date: 28 Jan 1981 1305-PST
From: Waldinger at SRI-KL
Subject: temporal logic
To: jmc at SAIL
about your talk. it seems that by using your approach you might be able
to get rid of the "fairness" restrictions that temporal logic requires.
they need to impose fairness because they are essentially modelling
concurrency as a nondeterminism, so they have to be sure that one process
doesnt hog all the time. if you use a global time parameter, say time
is a real number. then all you have to do is require that there be some
minimal time to execute any single insturction. then there is no way one
process can get too far ahead of the others.
He has asked whether he will get a letter. He says he needs in in
two weeks to get his money from the chinese.
best regards
richard
-------
∂28-Jan-81 1739 Konolige at SRI-KL exercise #2
Date: 28 Jan 1981 1740-PST
From: Konolige at SRI-KL
Subject: exercise #2
To: jmc at SAIL
John, I'm not sure my SAIL-sent message got through, so this is
a repeat. exer2.txt[1,kgk] has my writeup of the next exercise. If you
approve, it needs a due date, and then I can make copies for the next
class. --kk
-------
∂29-Jan-81 0900 JMC*
call cate
∂29-Jan-81 1108 CSD.JEANIE at SU-SCORE
Date: 29 Jan 1981 1103-PST
From: CSD.JEANIE at SU-SCORE
To: tenured-faculty: ;
At the tenured faculty meeting Tuesday we voted to reccommend the promotion
of Andy Yao to associate professor (with tenure) and to reappoint Doug
Lenat for one year.
If you were not at the meeting, please let me know of your approval or
disapproval of such actions, or of and reservations you care to express.
Because of deadlines from the dean's office, I must proceed fairly quickly,
and ask that you respond by Feb. 6.
-------
I approve both actions by the tenured faculty.
∂29-Jan-81 1146 HITCHCOCK at CCA (Chip Hitchcock) Re: natural gas shortage
Date: 29 JAN 1981 1440-EST
From: HITCHCOCK at CCA (Chip Hitchcock)
Subject: Re: natural gas shortage
To: ES at MIT-MC, ENERGY at MIT-MC
cc: HITCHCOCK at CCA
In response to the message sent 27 JAN 1981 2102-EST from ES@MIT-MC
Granted that it was stupid of Congress not to pass the deregulation bill,
the fact remains that the would-be producers are pigs. The current price
is well above the price that existed before the 1973 embargo (I think it's
higher even if you allow for inflation); they're just assuming that they'll
get their way (and arabesque profits) eventually and aren't willing to
commit until they get the highest possible return. Normally this would be
sound business practice; considering the business they're in, its
inexcusable.
(Note also that Boston Gas went around looking for more customers last
spring and summer even after they knew of the likelihood of a shortage;
someone pointed this out a few days ago but in view of your msg it should
be emphasized.)
-------
∂29-Jan-81 1523 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, February 3
SPEAKER: Peter Pepper, Technical University, Munich, Germany,
and Stanford University
TITLE: Algebraic Specification of Programming Languages
ABSTRACT:
Programming languages are considered to be (hierarchies of) abstract
data types. The major difficulty in this approach is the specification
of least fixed points within the framework of first order conditional
equations. This problem is solved by introducing particular
homomorphisms. The corresponding congruence relations allow us to
consider the semantic models of a language as quotient structures of
the term algebra given by the context-free syntax. In this way the
relationship between denotational and operational models can be
analyzed.
∂29-Jan-81 1810 ES at MIT-MC (Gene Salamin) ''Corporate Greed'' or ''Socialism''?
Date: 29 JAN 1981 2104-EST
From: ES at MIT-MC (Gene Salamin)
Subject: "Corporate Greed" or "Socialism"?
To: ENERGY at MIT-MC
I wish to express my disagreement with HITCHCOCK's comment about
greedy oil corporations. A corporation puts the interests of its
stockholders first; this is the service for which the stockholders
invested their money. Those who feel differently should pool their
assets and organize "People's Petroleum, Inc." In any case, the
corporations were accused of curtailing the production of oil in
order to receive higher prices in the future. This seems like a
reasonable thing to do. Should they sell their oil only to have
the value of their cash decay through inflation? Perhaps the real
culprit is not "Corporate Greed", but rather "Socialist Economic Policy".
If our money were backed, and on demand convertible into, gold or silver
or some such easily divisible, easily stored, non-perishable commodity,
then runaway inflation would be replaced by more mild inter-commodity
price fluctuations.
∂29-Jan-81 1937 LLW Tomorrow's Commute
To: JMC
CC: LLW, RAH
∂29-Jan-81 1924 JMC
See you tomorrow morning.
[Indeed--we're looking forward to travelling down with you!]
∂29-Jan-81 2026 ES at MIT-MC (Gene Salamin) Abundance from the ocean.
Date: 29 JAN 1981 2323-EST
From: ES at MIT-MC (Gene Salamin)
Subject: Abundance from the ocean.
To: ENERGY at MIT-MC
Good old Petr Beckmann is at it again. According to Feb. 1981
"Access to Energy", not only is there millions of years worth of
uranium and thorium in the oceans (at anticipated 21st century energy
consumption), but should fusion prove feasible, there is enough
lithium and deuterium for the life of the solar system. He doesn't
say how much of that fusion fuel will be needed to leave the solar
system after those billions of years.
∂29-Jan-81 2339 MMD maxtex
I do not know what was wrong with PRESS, but you can now print MAXTEX.MD[UP,DOC]
if you wish.
∂29-Jan-81 2340 POURNE at MIT-MC (Jerry E. Pournelle) Abundance from the ocean.
Date: 30 JAN 1981 0236-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Abundance from the ocean.
To: ES at MIT-MC, ENERGY at MIT-MC
Date: 29 JAN 1981 2323-EST
From: ES at MIT-MC (Gene Salamin)
Good old Petr Beckmann is at it again. According to Feb. 1981
"Access to Energy", not only is there millions of years worth of
uranium and thorium in the oceans (at anticipated 21st century energy
consumption), but should fusion prove feasible, there is enough
lithium and deuterium for the life of the solar system. He doesn't
say how much of that fusion fuel will be needed to leave the solar
system after those billions of years.
(1) IS THERE some reason to doubt the statements Dr. Beckmann
gives in ACCESS? In my experience he is one of the most careful
of men when it comes to "statements of fact" (as opposed to
opinions).
(2) If mankind is to survive the solar system we MUST eventually
have means to leave it. That will, of course, take energy in
large quantities; but I fail to see why this is a subject for humor.
∂30-Jan-81 0004 POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
From: POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
Date: 01/30/81 03:05:15
POURNE@MIT-MC (Sent by COMSAT@MIT-MC) 01/30/81 03:05:15
To: JMC at MIT-MC
[COMSAT: This was a failing QSEND.]
POURNE@MIT-MC 01/30/81 03:05:13
Are you still up? Most of the troops are already here.
∂30-Jan-81 0143 LWE My expanding NS files...
To: REG
CC: JMC
I am putting another courier tape at your disposal as I fear that
my rapidly growing NS files my be tying up scarce disk space on your
system - they are already totaling 180 pages. So please feel free to
dump them to tape any time via DUMPER; just leave a message if and
when you do so, and I shall delete them . Put differently,
this is also an implicit request to help out again with the tape
dump since I can't get into ARPAnet from Chicago (with your help,
it worked fine last time, and was greatly appreciated). The file
names are NOVEMB, DIGEST, APDIGS, and REAGAN (all NS) on [1,LWE].
Thank you !!!
∂30-Jan-81 0401 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 30 January 1981 02:55-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Puzzling item on Shuttle checklist
To: TAW at SU-AI
cc: SPACE-ENTHUSIASTS at MIT-MC
Obviously it is a reminder that all the myriad weapons:
Schmeissers, Tommy guns, nerve gas, leprosy bombs, and the
like--be seen to be aboard. After all what's space flight
without your nukes?
------------------------------
End of SPACE Digest
*******************
∂30-Jan-81 0427 JRA finances
i'll check with bill mckeeman about money. i have to find out what i'm gettng
paid too.
childhood's end was published in 1953, so could be source. do you know how to
reach steve?
∂30-Jan-81 0454 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #24
Date: 30 JAN 1981 0747-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #24
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Friday, 30 Jan 1981 Volume 3 : Issue 24
Today's Topics:
Query - Photocopier Allergy, Security - ATM Cards & Bookstore Systems
& Control of Cryptography Research,
Home Info Systems - Source/MicroNet Claims & Public Exposure
----------------------------------------------------------------------
DGSHAP@MIT-AI 01/29/81 15:06:00 Re: Allergy letter for Human Nets
Lately, I have managed to develop a contact allergy to something
in the output of a dover printer, which is a fancy xerox machine
at the AI lab. Furthermore, this reaction appears to be extending
itself to normal xeroxed material. I touch it, and I itch for an
hour. It has been suggested to me that fuser oil (a substance used
to prepare the paper for the carbon or ink) might be an allergen.
Does anyone know if this is true? And does anyone know where (what
department of which xerox plant) I can get more information about
the chemicals involved in the printing process?
Aside from the fact that I look silly wearing gloves to do my
reading, I am getting worried that this itching is the first
symptom of bigger problems to come.
------------------------------
Date: 28 Jan 1981 (Wednesday) 1140-EDT
From: MORGAN at WHARTON-10 (Howard Morgan)
Subject: Schlage cardkey
We looked into the system, which is quite interesting. There
are tuned capacitive circuits embedded in the card, with some-
thing over 1,000,000 possibilities. They also said they assign
them with a computer system that will not assign the same code
to two customers closer than 100 miles (as if that is super
protection). Of course, one can use an RF generator to "read"
a card, and copy the frequencies which are used, but it is
a non-trivial task. The nicest thing is that the card can
be read while still in your pocket (or pocketbook), and the
detector can be mounted in the wall, so there is no visible
action needed to send the code.
------------------------------
Date: 29 Jan 1981 (Thursday) 0804-EST
From: GERMAN at HARV-10
Subject: electronic security in bookstores
A query for human-nets readers: Does anyone know how the
electronic security systems in bookstores work? I have
seen electronic security in department stores for a long
time, where I think it is based on some special price tag
which can be detected. But I don't see any special tags
on books. These systems are not just for appearances --
I once heard the alarm sound (in this instance it did not
help the store very much, since the alarm was triggered
when someone walked in).
Steve
------------------------------
Date: 24 Jan 1981 8:39:54 EST
From: Bernie Cosell <cosell at BBN-UNIX>
Subject: NSA and nongovernmental cryptography
In the recent discussions on cryptography, NSA and their
interest in inhibiting certain kinds of research has been
mentioned. The director of NSA delivered a speech, some-
where (I don't know where) wherein he discussed NSA's
perspective on two matters that had caused something of
a controversy for the Agency: the shortening of the key
length for the DES, and the attempts to interfere with
research on the Public Key cryptosystems. The following
(rather long) article was printed in CRYPTOLOGIA, Vol 3
No 3, July 1979) copyright 1979 by CRYPTOLOGIA, Albion
College.
/Bernie
[ This article is quite long by digest standards ( >6000 words ).
Therefore, I have had copies of the paper established in files
at the sites listed below. Everyone interested should obtain
the paper from the site which is most convenient for them. If
you are unable to do so, please send mail to HUMAN-NETS-REQUEST
at MIT-AI and I will be happy to make sure that you get a copy.
Please obtain your copies in the near future however, since the
files will be deleted in one week. A copy of the material will
also be available from the HUMAN-NETS archives. Thanks go to
Richard Brodie, Richard Lamson, Doug Philips, and Don Woods for
providing space for the materials on their systems, and to
Bernie Cosell for making the paper available to us.
Site Filename
MIT-AI AI:DUFFEY;HUMNET NSAPOS
CMUA TEMP:NSAPOS.HMN[A210DP0Z]
PARC-MAXC [Maxc2]<Brodie>HUMNET-NSAPOS.TXT
SU-AI NSAPOS.HUM[T,DON]
MIT-Multics >udd>sm>rsl>human-nets>nsa-crypt.text
[Note, you can TYPE or FTP the file from SAIL without an
account.] -- RDD ]
------------------------------
Date: 21 Jan 1981 1017-EST
From: Steven J. Zeve <ZEVE at RUTGERS>
Subject: National Security
While I don't believe that the ends justify the means, I do
believe that National Security is a valid issue. I am amazed
by the naivete of some of the human-nets people, especially
those who don't seem to believe that it is a serious concern.
Would they really have preferred to see Nazi Germany or
Imperial Japan win WWII, or perhaps they would like to
live in a US controlled by the USSR?
steve z.
------------------------------
Date: 28 Jan 1981 10:52:58-PST
From: E.jeffc at Berkeley
Subject: MicroNet PR man
Sorry, the person introduced himself as a former PR man for the
Source, and later revealed that he is now working for MicroNet,
to the amusement of the audience. My memory failed me there for
a moment.
------------------------------
POURNE@MIT-MC 01/30/81 02:59:58
Re: Introducing the public to computers
Isn't it interesting how geniuses like that mind everyone's
business but their own...
------------------------------
REM@MIT-MC 01/30/81 05:14:17 Re: Introducing the public to computers
Reversi (Othello) doesn't have elements of Go. Except
for taking turns putting down a stone (like Tic Tac Toe)
and counting the territory-score at the end to see who
won, Reversi has nothing in common with Go. Rules for
moves, captures etc. are all different. There is virtu-
ally no skill that Go and Reversi have in common, where
practice in one will help the other, except minmax look-
ahead which is common in virtually all two-person games
(in fact, Reversi is more like Chess in that aspect, in
being highly combinatoric and not much strategic). That
article distorts the two games and should be reprimanded.
------------------------------
End of HUMAN-NETS Digest
************************
∂30-Jan-81 1131 OAF at MIT-MC (Oded Anoaf Feingold) Bostongas and other local quibbles, like, uh, conservation
Date: 30 JAN 1981 1423-EST
From: OAF at MIT-MC (Oded Anoaf Feingold)
Subject: Bostongas and other local quibbles, like, uh, conservation
To: ENERGY at MIT-MC
Foo! The t'ing that bothers me about this recent shortage is that
it exposed how underemphasized conservation and "wintering" efforts
have been in this state.
New construction standards in Massachusetts specify 3" of insulaton
in the walls, 6" of insulation (of some value of insulating goodness)
in the roof. Standards for upgrade of existing buildings do not, to
my knowledge, exist. Considering what Boston weather is like ordinarily
times, that's a joke. Example: My friend the small-planet/no-nukes/
hamster-powered-treadmill-in-your-backyard freak does solar greenhouse
construction, and often collaborates with contractors building energy
efficient housing. New residential construction with which he's involved
typically has 8" of insulation in the walls, 12" in the roof, and a
greenhouse/heat pump for profit and veggies. Typical fuel usage for these
houses is in the range of ** 10% ** that of similarly sized buildings built
to spec. (That statement applies also to this cold snap.) That kind of
investment pays back within 3 years! (More numbers available on request.)
There have been complaints that BostonGas used its money to push
for new customers rather than finance energy-efficiency surveys
and improved insulation efforts. No doubt true. I also don't
doubt that pushing for new customers is more profitable, so I
would always expect them to do that. (Perhaps our beloved state
government should take some blame, god forbid.) Certainly there
is an underdeveloped market, at least partly due to unfavorable
tax and lending structures, in getting value for one's energy
dollar around here. That's what I object to.
(Actually, almost all construction in this benighted state would gag a
maggot, but that belongs in SIGWHYTHISSTATESHOULDBENUKEDINTHEINTERESTSOF-
URBANRENEWAL, and won't be reprinted here.)
∂30-Jan-81 1252 Denny Brown <CSD.DBROWN at SU-SCORE> Undergrad advising
Date: 30 Jan 1981 1249-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Undergrad advising
To: CSD-Faculty: ;
We just got our annual call for general advisors. Anybody interested
in advising undergrads should tell me. You can choose undeclared
freshmen, transfer students, or reassigned sophomores. Many more
details available from me if you want them.
-Denny
-------
∂30-Jan-81 1341 CSD.NOWICKI at SU-SCORE Ether Tip software
Date: 30 Jan 1981 1336-PST
From: CSD.NOWICKI at SU-SCORE
Subject: Ether Tip software
To: @sun at SU-AI
Out of no other motivation besides personal satisfaction, I propose the
following design for an ether tip. Note that I implemented a terminal
concentrator at NU in my Junior year, and have been maintaining the telnet
software on vaxes here for the last year, so I think I may have something to
offer. My plan is quite revolutionary since it uses NO operating system.
I know I'll get flak from many of you, but here it goes....
The design is based on interrupts. There are queues for ethernet output, input,
and each terminal's input and output. In the discussion below whenever I say
put something on queue X I also mean start I/O on X if I/O is not already
in progress. I also assume a timer. All devices cause an interrupt when
ready for another transfer.
Main program: take packets off the ethernet input queue.
If the packet was an ADATA for a user, and it matches his bytepointer,
put the characters in his output queue. Put an ACK in the
ethernet output queue, with the number of bytes left in his output
buffer for flow control.
If the packet was an ACK, and we were waiting for such an ACK,
delete those characters from that user's input queue.
If the packet was END or ABORT, terminate the connection.
Ethernet input interrupt:
put the packet on the ethernet input queue, unless it would result
in overflow. Note that the "main program" logic could be done at this
level if you didn't mind throwing away back to back ethernet packets.
Ethernet output interrupt:
If the ethernet output queue is not empty, get the first packet
on that queue and send it. If it was an ADATA, set the byte-pointer
acordingly so that the timer will retransmit if not acked.
Terminal input (keyboard) interrupt:
Put the character into the terminal input queue, and put an
ADATA packet into the ethernet output queue if there is not already
one there.
Terminal output interrupt:
output the next character in the queue if there is one.
Timer interrupt:
For each outstanding ADATA packet that has not been ACKed,
queue up another ADATA packet in the ethernet output queue.
If this has already been done many times, declare the connection
dead.
This scheme would be lightning fast, and I think it could be implemented
in a few months.
-------
∂30-Jan-81 1451 Vaughan Pratt <CSD.PRATT at SU-SCORE> Meeting Monday
Date: 30 Jan 1981 1446-PST
From: Vaughan Pratt <CSD.PRATT at SU-SCORE>
Subject: Meeting Monday
To: @sun at SU-AI
There will be a Sun organizational meeting Monday 4 pm in MJH 402.
Agenda:
Present hardware status
Near-term hardware plans
Present software status
Near-term software plans
My apologies for the short notice; the week filled up from Friday backwards,
and the meeting was almost pushed into the past. I suggest we push our luck
and hold the next meeting on Friday the 13th...
-------
∂30-Jan-81 1531 VRP via Ethernet NVT @sun
To: "@SUN.DIS[P,DOC]" at SU-AI
Just a reminder: to mail to the 40 people on SUN.DIS[P,DOC] (maintained
at Sail):
From Sail: mail @sun
From Score: mail "@sun"@sail
elsewhere: by analogy with Score
To add to the list please send me mail.
This is the only active sun mailing list on any host attached to the ARPAnet.
There is a list called sunhax on Shasta, mainly for people presently
generating Sun software.
To find out more about the Sun environment on Shasta, do "man sun" on
Shasta.≥
∂30-Jan-81 1548 Chiron of Thessaly <FEINBERG at CMU-20C> Pro Nuke Protest
Date: 30 Jan 1981 1833-EST
From: Chiron of Thessaly <FEINBERG at CMU-20C>
Subject: Pro Nuke Protest
To: Energy at MIT-MC
What a fine idea! A pro-nuke protest!! I aprove.
--Chiron
-------
∂30-Jan-81 1614 GFS ETHER-TIP
To: "@SUN.DIS[P,DOC]" at SU-AI
Nowicki's idea sounds good: simple, straightforward, and feasible. It is also an
elegant implementation of Fudd's First Law: It goes in, it must come out.
The fury it will cause should be a sight to behold, since it requires neither
memory-management nor multi-tasking.
The Ethernet packet receiver code could be in the main routine since the
Ethernet interface board buffers back-to-back packets automatically. Nice, huh?
∂30-Jan-81 1739 BYY teach sol
jon is teaching sol how to send a message
∂30-Jan-81 2308 Hans Moravec <HPM at SU-AI> Petr Beckmann
Date: 30 Jan 1981 2300-PST
From: Hans Moravec <HPM at SU-AI>
Subject: Petr Beckmann
To: energy at MIT-MC
Spider Robinson has a fantastic review of "The Health Hazards of NOT going
Nuclear" now being published by Ace, in the March 2, 1981 issue of Analog.
(Oh yes, he follows with Jerry Pournelle's "A Step Farther Out" but I wouldn't
want to embarass anyone on this list with excessive praise. Besides
its an Ace re-release, right?).
∂30-Jan-81 2316 MINSKY@MIT-AI
From: MINSKY@MIT-AI
Date: 01/31/81 02:12:38
MINSKY@MIT-AI 01/31/81 02:12:38
To: energy at MIT-MC
I still don't see why people hassle about BostonGas. What's wrong
with trying to get customers to switch from oil -- it's not as though
they were creating energy eaters? It seems to me gas heat is
preferable for several reasons:
- equipment to heat with it is simpler.
- pipelines now exist, and better than oil trucks, no?
puts substantially les CO/2 into the atmo.
--- if Tom Gold is right, there's a lot more where it came from.
I heard on radio some really technically false claims about oil heat
being "better".
As for all those hazards, I agree that they are serious. But if all
you losers would get to work on the real problem, it might do a lot
more good. Let's get out of these ridiculous organic bodies.
∂31-Jan-81 0401 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 30 Jan 1981 1553-PST
From: Alan R. Katz <KATZ at USC-ISIF>
Subject: Shuttle status phone number
To: space-enthusiasts at MIT-MC
cc: katz at USC-ISIF
For up to date info on the shuttle, dial:
(213)922-INFO
Alan
-------
------------------------------
End of SPACE Digest
*******************
∂31-Jan-81 0425 DUFFEY at MIT-AI (Roger D. Duffey, II) HUMAN-NETS Digest V3 #25
Date: 31 JAN 1981 0720-EST
From: DUFFEY at MIT-AI (Roger D. Duffey, II)
Subject: HUMAN-NETS Digest V3 #25
To: HUMAN-NETS at MIT-AI
HUMAN-NETS AM Digest Saturday, 31 Jan 1981 Volume 3 : Issue 25
Today's Topics: Security - Control of Cryptography Research
& ATM Cards & Bookstore Systems
----------------------------------------------------------------------
Date: 30 Jan 1981 1026-PST
From: MJMARCUS at USC-ISI
Subject: Cryptography and classification
A lot of interesting background material on the issues
in classifying cryptography research are contained in the
34th report of the House Committee on Government Operations
entitled "The Government's Classification of Private Ideas",
(House Report No. 96-1540, Union Calendar No. 908) which
can be purchased from GPO or probably requested from your
congressperson. The report also discusses atomic energy
issues.
Mike Marcus
------------------------------
Date: 30 Jan 1981 1037-PST
From: ROODE at SRI-KL (David Roode)
Subject: national security
Just because it is believed that USA cryptography efforts aided
in winning WW II, it does not follow that in some future conflict
the shoe could not be on the other foot. In particular, I do not
doubt that there is a Russian NSA possibly intercepting this very
message as I type it. For all we know, the recent developments in
this country in cryptography have been in clandestine use abroad
for some time. The NSA position paper reveals the basic weakness
of their argument all the more. They defend their support of
DES and then go on to argue that code cracking techniques be
suppressed lest their own codes be cracked. It is suppression
of technique and not proliferation that threatens the national
security in substance and basic concerns of democracy in spirit.
------------------------------
Date: 30 Jan 1981 10:39 PST
From: Deutsch at PARC-MAXC
Subject: Inman's statement
Given the importance that intelligence agencies attach
to their activities, the numerous occasions in the past on
which even less sensitive government agencies have covered
up or lied, and the documented activities of NSA in the
communications-monitoring area, I don't see why one should
believe that Inman is telling the whole truth, or even the
unaltered partial truth. From his point of view, he has
more to gain (in the interest of national security, of
course, not his own personal interest) by lying than by
telling the truth if he thinks he can get away with it,
which, judging from past experience, he probably can, at
least for quite a long time.
I don't see cryptographic technology as any different from
any other that is of interest to non-government-restricted
researchers. If the security agencies can't PERSUADE
researchers that their work must be kept secret, by demon-
strating a compelling connection with the national interest
and an absence of less drastic remedies (as is required,
by current judicial standards, in all other cases of
restrictions on the First Amendment), I don't think they
should have any right to COMPEL them. Inman's argument
extends to every form of technology: do we compel people
to keep secret the design of better computers, because
revealing the principles would compromise our superiority
in the area of being able to analyze intelligence data
more thoroughly or quickly? No. We just don't export
the computers themselves, in some cases. But it seems
to me an argument could be made here that is almost
identical to the one Inman makes in the article.
------------------------------
Date: 30 JAN 1981 0647-EST
From: REM at MIT-MC (Robert Elton Maas)
Subject: Schlage cardkey
A disadvantage of being able to read the resonances while
they are still in your pocket is that if you have two cardkeys
because you have two different places where you are permitted
access, the system would read both cards in parallel and get
confused, furthermore somebody could install a cardkey reader
on a sidewalk and be able to read the cards of everybody who
walks by, and thus be able to forge cardkeys for all the local
security installations if any of their employees walk by that
location. At least it would seem so to me.
------------------------------
Date: 30 Jan 1981 1055-MST
From: Spencer W. Thomas <THOMAS at UTAH-20>
Subject: Bookstore security
The security systems used in bookstores (and libraries) rely
on a thin strip of some sort of magnetic tape in the binding.
I had a friend who worked at the bindery in a library, and
she had to put these into the older books. I think that most
new (hardback) books are manufactured with the strips already
installed in the binding. I'm not sure what the process of
'sensitizing' and 'desensitizing' is, but it's something
magnetic. I have seen some ring binders which set off the
alarm and have to be 'desensitized' every couple of weeks.
-S
------------------------------
Date: 30 Jan 1981 0944-PST
From: Tom Wadlow <TAW at SU-AI>
Subject: Library book sensors.
Apparently, each book contains a small metal strip or wire placed
in the binding. These wires are easily magnetized/demagnetized.
The wires are (I suspect) initially magnetized, and thus when you
try to carry the book through the sensor archway, it affects an
EM field in a detectable manner. When you check a book out of
the library, they run the binding past what must therefore be a
bulk eraser. This demagnetizes the metal past the point where
it can have a detectable effect on the electromagnetic field and
thus the alarm is not triggered (unless you forgot to give the
librarian that book in your backpack/briefcase).
------------------------------
Date: 30 January 1981 1512-EST
From: Hank Walker at CMU-10A
Subject: bookstore security
They could use microdot tags.
------------------------------
Date: 30 Jan 1981 (Friday) 0913-EDT
From: SHRAGE at WHARTON-10 (Jeffrey Shrager)
Subject: Bookstore Security
We recently had a very long detailed discussion of this topic
in our local BULLetin board system. Several experiments were
carried out on the Penn Bookstore's system -- someone even went
as far as running one of the books thru a device looking for
metal in it. We decided that it was mostly a bluff but that
certain of the more expensive books had wires shoved into
their bindings which set off the detector. The bookstore has
to run ALL purchases over the little "demagnetizing" boxes on
the counters both to complete the bluff and because the clerks
do not know which objects are tagged and which are not. Those
boxes, by the way contain a magnet and are connected to nothing
(at least not by wires). This brings up an interesting problem
-- if the process of "desensitizing" is to simply scramble some
magnetic code in the wire (which is all that a powerful magnet
will be able to do, not demagnetize it) then how are those
systems able to detect a moving specialized magnetic sequence???
It would be pretty tough to decode (or even to encode) parts of
a wire with binary magnetic "bits" and expect it to stay put in
different parts of the wire.
By the way, the desensitizing process gets pretty silly when
they start running toothpaste, greeting cards, and pen refills
over the countertop boxes.
Our particular machines are made by 3M and have two columns with
a little photodetector at the base of one (they count the number
of passers-by).
------------------------------
End of HUMAN-NETS Digest
************************
∂31-Jan-81 0653 DREIFU at WHARTON-10 (Henry Dreifus) Water situation.
Date: 31 Jan 1981 (Saturday) 0944-EDT
From: DREIFU at WHARTON-10 (Henry Dreifus)
Subject: Water situation.
To: energy at MIT-MC
It seems that bad planning, poor management, and dumb decisions contributed
to the current water crisis on the East coast. I'm sure the drought (that
means no rain for an extended period) has been a major catalyst in this
situation, however I read that the water table has been on a steady decline
over the years anyway.
What can be done? Water-catchers and still-suits?
/Hank
∂31-Jan-81 1021 DAM@MIT-AI
From: DAM@MIT-AI
Date: 01/31/81 13:15:38
DAM@MIT-AI 01/31/81 13:15:38
To: ENERGY at MIT-MC
I simply wish to add a note of agreement to Minsky's statement and
say that all this talk about getting off the planet misses the point. "Let's
get out of these rediculous organic bodies", Amen.
∂31-Jan-81 1106 BYY on TTY1 1106
can I mail files on ARPA that contain ⊂,∀,↓etc?
You can mail them, but hardly anyone else can read them on their
terminals or print them. Only M.I.T. AI Lab, I think.
∂31-Jan-81 1149 MVL at MIT-AI (Michael V. Lease) Boston Gas (advertising for more customers)
Date: 31 JAN 1981 1438-EST
From: MVL at MIT-AI (Michael V. Lease)
Subject: Boston Gas (advertising for more customers)
To: energy at MIT-MC
CC: MLease at BBNC
There is nothing wrong with BostonGas trying to get more customers -- except
that (as noted in my previous letter) they weren't all that sure that they had
the ability to serve all those customers. As a result, we were faced with the
recently ended shortage. I personally see nothing wrong with gas heat per se,
it's just that I object to these shoddy business practices.
Mike Lease
∂31-Jan-81 1340 Purger
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging.
If you don't delete some of your files, the purger will.
Your disk quota is: 4080
Your files occupy 4524
∂31-Jan-81 2229 MINSKY@MIT-AI
From: MINSKY@MIT-AI
Date: 02/01/81 01:20:44
MINSKY@MIT-AI 02/01/81 01:20:44
To: ENERGY at MIT-AI
As for water, I read somewhere that Boston and many other cities lose
HALF their water because of system leaks, because the mains are very
old. Repairing them would cost many billions. The article estimated
some ten billion or so for Boston, as I recall.
Assuming that's true, I bet that if we invested a billion in advanced
teleoperator technology, we could repair these systems for much less
than those costs, which are based on digging up a lot of system.
Repairing pipes from the inside is very favorable, since the pressure
is on your side and many major repairs can be done by injecting
hydraulic cements and plastics.
The trouble is that I can't find any way to get that done. The
government doesn't seem to have any place to consider such ideas.
Nor does MIT, so far as I can see. My article in OMNI got lots of
good comment, but ---.
Now that population in the US has more or less stabilized -- or would,
with sane immigration policies -- a variety of problems are
practically settled. The East won't need more water than it has, so
we just have to fix things. The parking problem in the cities will
get only better as the cars get a little smaller. If we had a little
more energy, things could be cleaned up better. If we only had some
technical leadership. Sob.
∂01-Feb-81 2146 JK ekl
The new manual is written; hopefully more readable than the previous
one: EKL.MAN[EKL,JK]. I will try to get hold of Weening tomorrow.
∂01-Feb-81 2351 AVB
To: "@SUN.DIS[P,DOC]" at SU-AI
hardware status report
I am sending this out as background information for the SUN meeting
tomorrow (today), Monday, 4 pm in MJH 402.
See you then - Andreas
-----------------------------------------------------------------
Hardware Status Report:
I am pleased to announce that both the 68000 board and the Ethernet interface
are working. Testing of the Ethernet interface was completed January 15,
in conjunction with one Alto on a local Ethernet. The 68000 board
had surprisingly no design errors at all, save for omitted pullups.
I started debugging of the graphics controller board, and
we yet have to fabricate the frame buffer PC board.
A realistic date for completing testing of the graphics subsystem
is the end of this month (February).
-----------------------------------------------------------------
Inventory:
Currently we have two working Ethernet interfaces and one working 68000 board.
Two more Ethernet interfaces have been wirewrapped and assembled, and
we have parts for three more 68000 boards to be wrapped this week.
Thus very soon we will have four sets of 68000/Ethernet interface boards.
In addition, we have a total inventory of 6 Motorola 68000 Development Boards.
These systems will support the ongoing hardware hardware development
and initial software development.
How many additional SUN systems do we need in the short-term future?
Counting the immediate applications, I came up with the following list:
-----------------------------------------------------------------
Simple SUN systems:
3 gateways (SUMEX, LOTS, CIT)
3 TIPs (TERMAN, ERL, MJH)
1 Sumex (Ethernet Load Scheduler)
2 hardware development (AVB, Binford)
8 software development (Archer, Baskett, Hennesy, Lantz, Nowicki, Reid, Pratt)
? others (operating system course, home usage, etc).
In addition to these display-less SUN systems, we are preparing to
build a batch of 15 single-user SUN Workstations for the following groups:
11 VLSI project
3 TEX project
1 Robotics
-----------------------------------------------------------------
Costs:
The net parts cost for the smallest usable SUN system,
consisting out of a 68000 board and an Ethernet interface, is about $2400.
Additional 128 kbyte RAM is $400. Uart Boards with 8 lines for TIPs are $350.
Replacing the wirewrap boards with PC boards will save about $300 per board.
Parts cost estimates for a small SUN system:
1 68000 board with 128k RAM, wirewrap $1300
1 Ethernet interface Board $800
1 4-slot card cage $200
1 Small power supply $100
------
$2400
Parts cost estimates for one SUN workstation:
1 68000 Processor Board with 256 kByte memory 1700
1 Ethernet Interface Board 800
1 Display Controller Board 450
1 Frame Buffer Board 700
1 high-resolution Video Monitor 850
1 Keyboard 400
1 6-slot Cardcage 300
1 Switching Power supply 300
1 Packaging 500
-----
Total $6000
-----------------------------------------------------------------
Fabrication:
It is important to decide soon how many systems we are going
to build in the first round, in order to make the most economical
fabrication and implementation decisions, such as
in-house versus outside manufacturing and
such as wirewrap versus printed circuit boards.
We are investigating ways to get SUN systems produced commercially for us.
No agreements have been reached yet, but we received two offers to build
SUN workstations for Stanford at a cost of $10,000 per unit.
By similar token, the display-less system could probably be built
commercially for about $4000.
-----------------------------------------------------------------
Other Components:
Mass Storage:
SMD Storage Module Controllers are about $2500.
Wincherster Disk Controllers about $2000.
Checking on SMD/Winchester costs.
What about getting 2 SMD Controllers and 300 Mbyte Disks?
∂02-Feb-81 1028 CSD.SCHREIBER at SU-SCORE All over.
Date: 2 Feb 1981 1027-PST
From: CSD.SCHREIBER at SU-SCORE
Subject: All over.
To: csd.ossher at SU-SCORE, jeb at SU-AI, csd.gischer at SU-SCORE,
csd.clarkson at SU-SCORE, jmc at SU-AI, acy at SU-AI, rwf at SU-AI,
csl.lantz at SU-SCORE, csd.schreiber at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
Thanks for your generous assistance. I'm glad we got it all done yesterday.
Good luck to those of you on next quarter's commitee.
Rob
---
-------
∂02-Feb-81 1114 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Gray Tuesday
Date: 2 Feb 1981 1110-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Gray Tuesday
To: CSD-Faculty:
cc: wol at SU-AI, jeb at SU-AI
Gray Tuesday
Tuesday, February 10
2:30 p.m.
Jacks 252
-------
∂02-Feb-81 1657 FFL
To: JMC, FFL
Whit Diffie called. Wants to see you briefly on Tuesday. If you come
in very shortly, he will be available at 9 969 9l70.
∂02-Feb-81 1743 CLT
have gone to play my flute, back around 7ish
∂02-Feb-81 2134 CSD.LENAT at SU-SCORE Forum
Date: 2 Feb 1981 2131-PST
From: CSD.LENAT at SU-SCORE
Subject: Forum
To: jmc at SU-AI
John,
Rod Brooks won't be able to talk at the forum this year,
as he has just come down with mononucleosis. Tom Binford
will be speaking instead. You should check with Tom about
his title (e.g., at the session itself), and make an announcement
explaining the program change. Thanks.
Doug
-------
When am I to show up at the forum?
∂
Jerry: This excerpt from HUMAN-NETS might be relevant to your
student. I forget whether you see it regularly anyway.
Date: 2 Feb 1981 2238-EST
From: JMCKENDREE at BBNB
I'm a User Consultant on the EIES System, and will look forward
to meeting Hank (DREIFU at WHARTON-10) there. You can message me
as JACK, or as 112 or as JOHN MCKENDREE, Hank.
You ask if the commercial messaging or conferencing systems may
evolve in a similar direction as EIES. There are many mansions
on that dedicated conferencing computer, so, in that sense, how
can they miss? For example, there is the menu of choices
(leading one to messaging, conferencing, notebooks, etc.), a
parallel menu of commands using mnemonics, and several simplified
smaller menus devised by groups for their use and kept in the
system repertoire when the groups disbanded. In fact, using one
particular command, any user can rename any menu choice to suit
his or her taste - so there's no limit in the design of an
interface. For example, INITIAL CHOICE? has been re-phrased by
one user as:
WHAT WOULD YOU ENJOY FROM TODAY'S MENU AT THE MOMENT, SIR?
Translation into French, German, or other languages is an obvious
possibility for this general purpose (and fun) command. The
freedom to structure group communications only begins here,
moreover.
A network of State Legislative research staffs (about 40 states I
believe have regular use of EIES) developed their LEGITECH menus
and separate text files for lightly formatting inquiries and
responses. A response from an expert on line saves time and
effort by the staffs at other State Houses. Inquiries and
responses are filtered by EIES so that only those of interest are
delivered to each participant. When there is pending legislation
(e.g. hazardous waste regulation) at one location, inquiries and
responses all in the same light format can be compiled and
retrieved for rapid analysis. This value-added service on EIES
has been in operation for over two years now. The separate menus
and formats have evolved several times until they are pretty
stable. That is a speed of evolution which commercial systems
are not likely to emulate, but the stable design could very
easily be adopted by other networks, it seems to me....
EIES also serves handicapped. Experimental terminals which favor
particular handicaps (foot operation, one-handed operation, giant
keyboards, etc.) have been employed by the New Jersey Institute
of Technology staff at their Computerized Conferencing and
Communications Center.
Graphics capabilities, interconnected minicomputers, and other
communications oriented hardware and software concepts are in
trials.
Numerous research reports of various group efforts, facilitated
by EIES, and their evaluated experience have been produced.
Subscribers to the network often are invited to browse through
drafts of such reports before they go to final publication. This
early access to research is a side benefit of networking on EIES
in which you might very well want to participate, Hank.
One of the fun things to do is co-author a paper with others you
meet there and whose face you may not see until after final
publication! There is a PAPER FAIR (Public Conference 1017) in
which authors not only grant access to their works but then get
involved in feedback and discussion about issues which their
papers raise. After all there are over 500 individuals from
U.S., Canada, Europe, and even Australia, and they represent
private industry, banking, university R&D, state and federal
government agencies - quite a likely source of feedback to any
author who wants to brainstorm ...
Just a few days ago, Robert Bezilla, an author and speaker, was
wined and dined in Cambridge by an admirer who read his paper on
the SOURCE! I had copied the paper from EIES' PAPER FAIR to the
SOURCE' FORUM. (with Bob's permission, to be sure) Now other
authors have asked me if I can do the same for them...
Why shouldn't EIES be a reflection of the future? Just a matter
of time for the word to get around to designers and their
management. After working out the bugs on the NJIT research
facility, transfer the concept and the easy to use interface to
your own environment!
John McKendree
∂03-Feb-81 0947 FFL
To: JMC, FFL
Sorry, I was interrupted while doing the first message about Mr. Giesbert. Anyway
he says he works for the TIME of Paris and wants to speak with you. I asked
if he were interested in robotics, but he refused to be led down that garden
path and said he wanted to interview you because you were one of the leaders in
AI. I told him to call back in a couple of days and I would ask if you could
arrange it. He wants to come 12th or 13th of February and says he will not
take much of your time. What should I tell him?
∂03-Feb-81 0950 FFL
To: JMC, FFL
Call from New Mexico State wanted the latest version of LISP language.
Where do I get this? Cindy Walters, New Mexico State U., Box 3AT, Las
Cruces, NM 88003
∂03-Feb-81 1039 KGK via SRI-F2 class
To: JMC
CC: KGK
John, I won't be in to class today unless you think it really
necessary. I'll be in Thursday to collect more checks for the Xeroxing,
and anything else that needs to be done.
In reviewing the AIRPO6.AX axioms for the exercise, it strikes
me that one won't be able to derive "holds(at(suitcase,I),Sf)" unless
some modifications to the frame conditions in the drive and walk actions
are made. The problem is that if x and y are at the car, and x is at y,
then after the car is driven one can show that x and y are still at the
car, but not that x is at y. I think that modifying the disjunct "y=car"
to "y=car OR holds(at(y,car),s)" should do the trick.
One final thing-- since I don't have an office at Stanford, I'm
considering spending Thursday afternoons from 1 to 3 at Stanford to
help students with using FOL and other problems that may come up.
--kk
I haven't looked at what modifications of the axioms might be required.
I hope the students understand that modifying the axioms is part of the
problem rather than a defect in the problem. If the modification shows
that the original examples were too ad hoc, this should be stated, and
your comment indicates that they were.
Holding an office hour at MJH is a good idea. I think a seminar room
should be free at that time. Why not send Fran a message asking if she
can reserve 301 or some other? They're having problems with FOL syntax.
∂03-Feb-81 1211 CSD.JEANIE at SU-SCORE ARPA Consolidation
Date: 3 Feb 1981 1002-PST
From: CSD.JEANIE at SU-SCORE
Subject: ARPA Consolidation
To: csd.feigenbaum at SU-SCORE, csd.ullman at SU-SCORE, jmc at SU-AI,
csd.buchanan at SU-SCORE, csl.fb at SU-SCORE, csl.lantz at SU-SCORE,
tob at SU-AI, zm at SU-AI, dcl at SU-AI, acy at SU-AI,
wiederhold at SUMEX-AIM, csl.bkr at SU-SCORE, csl.meindl at SU-SCORE,
icl.isl-morf at SU-SCORE, csl.fat at SU-SCORE
cc: csd.bscott at SU-SCORE, csd.kuhn at SU-SCORE
i
To: various PIs of ARPA-IPTO Projects
From: Ed Feigenbaum
Subject: meeting to discuss consolidation
of ARPA contracts
Bob Kahn, Duane Adams, and Bob Engelmore of ARPA-IPTO have
requested a meeting on Friday, February 13 to discuss their
desire to consolidate the many Stanford ARPA contracts
into one or two contracts. The meeting will take place at
9am in the Chairman's conference room of Margaret Jacks Hall,
and will probably last all morning (at least).
I would urge you to take whatever steps are necessary to allow
your attendance, since these discussions are very important.
Some of us in CSD have already met to discuss the issues involved.
Those in EE who have not should communicate with Baskett,
Luckham, or me to brief themselves on the issues.
-------
∂03-Feb-81 1302 DBL forum time: Thu 9:15 -10:30
John,
Your session is the first one. kennedy speaks at 9,
will probably end betw 9:15 and 9:30. Your session
lasts for one hour, then. See you Thursday, CERAS 112.
Doug
∂03-Feb-81 1339 FFL
To: JMC, FFL
David Black of NEXT magazine in NYC wishes to do an article on khoe home
computers will transform family life. Your name has been mentioned in other
interviews he has done and he would like to speak with you.
212 662 0557. He will be in all day Wednesday.
∂03-Feb-81 1611 FFL
To: JMC, FFL
Paul Bannister called from Almaden (56 997 7214 or 7305). He is writing
something about Christopher Evans' book, MICROMILLENIUM, in which you are
mentioned twice. He wishes to speak to you about it for about three
minutes.
I'll speak to Bannister.
∂03-Feb-81 2311 TOB dinner
Is this Friday, Feb 6, still ok for dinner?
If so, 8 pm. I will send a message about
how to get to our house.
Tom
∂04-Feb-81 0000 JMC*
recommendations
∂04-Feb-81 1114 FFL
To: JMC, FFL
Deborah Gee of KGO-TV called. She wishes you to call her, please, at
55 565 7896.
∂04-Feb-81 1606 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Ray Reiter
Date: 4 Feb 1981 1603-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Ray Reiter
To: JMC at SU-AI
You didn't mention an honorarium for Prof. Reiter. The Dept. will pay
$150 for honorarium and up to $200 in expenses.
We will need his address and social security number.
Carolyn
-------
OK, I'll pay the balance of his expenses from ARPA money. I suggest you
phone him for the SS number. He is at U. British Columbia in Vancouver.
Reiter,Ray 604-228-4142.(office) 604 733 9185 (home)
∂04-Feb-81 1652 ARK tex problems
∂04-Feb-81 0123 JMC letter.tex[tex,ark]
r tex
\input letter.tex[tex,ark]
(letter.tex[tex,ark] 1 2 3
!Input page ended while scanning def of \letteroutput
p.3,l.0
??
I'll look at this when I get back to Stanford. I'm now at a conference
at Berkeley.
Arthur
∂04-Feb-81 2018 Konolige at SRI-KL (Kurt Konolige) seminar room
Date: 4 Feb 1981 1722-PST
From: Konolige at SRI-KL (Kurt Konolige)
Subject: seminar room
To: ffl at SU-AI
cc: jmc at SU-AI
Frances, I'd like to reserve a seminar room (perhaps 301) for
Thursday afternoons, say from 1pm to 3pm, so that I can offer some help
to students in CS226. Could you take care of this? I'd like to start
tomorrow.
-------
∂04-Feb-81 2154 LGC Opinion Request
It's likely that I'll send a paper off tomorrow or Friday to be published in
the Pacific Philosophical Quarterly. The paper is to some extent a byproduct
of my work on causal reasoning and the frame & qual. problems for the Advice
Taker project. I'd like your opinion as to the appropriateness of the
acknowledgment footnote as it concerns ARPA. If you are unable to vouch for
the accuracy of the order and contract numbers, perhaps you could give me a
pointer to an appropriate source (the numbers are the ones I used 2 years ago
in my IJCAI-79 paper).
The title is: CAUSAL EXPLANATION AND THE REALITY OF NATURAL COMPONENT FORCES.
The acknowledgment footnote:
*I have benefitted from stimulating discussions of earlier versions
of this paper with Nancy Cartwright. The paper is based in part on research
supported by the Advanced Research Projects Agency of the Department of
Defense under ARPA Order No. 2494, Contract MDA903-76-C-0206.
Fran should know proper form of acknowledgment, but otherwise Betty Scott
will know.
∂05-Feb-81 0108 POURNE at MIT-MC (Jerry E. Pournelle) more stuff
Date: 5 FEB 1981 0409-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: more stuff
To: JMC at MIT-MC, llw at SU-AI
mc:pourne;space concl has a first cut at some logic for the
second paper for the report. It also has obvious places
suggesting things you two can do.
llw-- can you clean up the Whither NASA paper making it a little
less of a pitch for Mark as NASA administrator and send me a
good REPRODUCIBLE copy for circulation? Needed quickly.
You should have some draft stuff on SPS soon. I suppose I could
read that into the net, but it takes forever.
∂05-Feb-81 0402 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 5 FEB 1981 0336-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: well maybe you do it this way...
To: SPACE-ENTHUSIASTS at MIT-MC
.M10 H11 S1 F0 G60 L74
.C
|THE JOINT AMERICAN ASTRONAUTICAL SOCIETY--L5 SOCIETY
.C
|CONFERENCE ON PLANNING AMERICA'S FUTURE IN SPACE.
.C
|STATEMENT ON PROPOSED FY 82 NASA BUDGET
.c
|SPACE: THE CRUCIAL FRONTIER
1. The |rediscovery of progress| is a reasonable and feasible
national goal for the United States in the 1980's.
Progress is possible. We do not have to accept limits to
growth; but we do need specific strategies for progress. Growth
requires investment and continuous expansion of the resource base.
The United States has a world mission. We influence by
example; we are the showplace of freedom; and in the present era we
must also be the sword and shield of liberty. To fulfill this role we
must do more than survive. We must remain militarily, economically,
and ideologically strong.
We need visible goals: a reason for the nation to exist. If
we have no dreams and goals, we have no nation.
Insuring progress for ourselves and the world is a reasonable
and feasible goal for America. Space activities can be a significant
part of our rediscovery of progress.
2. The vast majority of resources accessible to mankind are
NOT here on Earth. The solar system abounds with minerals and energy.
Other nations are even now claiming those resources and developing
capabilities for using them. If the United States does not compete,
we will have effectively abdicated economic leadership to those who
do.
There is more at stake than that. Space has very great military
potential. Although no one is certain that strategically decisive
weapons can be deployed in space, no reasonable person can be certain
that they can |not| be. Space based beam weapons may develop into
reliable missile defenses. At the very least, the United States
|must| retain the option to compete in space.
Space also has symbolic importance, if for no other reason
than the United States made the "Moon race" critical to our national
prestige. To abandon space after announcing its crucial importance
hands the Soviets an unearned but enormously important ideological
victory. It is obvious from their space activities that the Soviets
realize this. We must, therefore, retain the option to move
effectively and quickly into space.
Retaining that option is not simple. No one can be sure what
capabilities will be needed. Our adversaries have more experience in
the space environment than we do.
Since we cannot know which space capabilities may prove to be
decisive, we cannot design robots or artificial intelligence systems
in advance. The only truly versatile space system is man; and the
only way to insure a capability to do a wide variety of tasks in
space--including construction of the military systems that may be
needed in future--is to make entry to and operations in the space
environment routine.
We must continue both manned and unmanned exploration of
space. Our survival may depend on it.
AAS--L5 Statement on FY 82 Budget Page Two
3. The "Revoluton of Rising Expectations" concides with the
"era of limits" to aggravate international instabilities. Most of the
world will remain poor in the remaining years of this century--and
this in a "global village". The wretched of the Earth are very much
aware that everyone doesn't live their way. World economic growth is
not merely desirable on ethical grounds; it is very much in the U.S.
national interest.
Rapid economic growth is not easy. It requires investment.
It also requires technological growth, and expanded resources. We
cannot abandon technology; indeed, we must rapidly expand our entire
technological and industrial base.
4. All the above factors combine to make space an important
option. To preserve and increase capabilities for military activities
in space we must expand our space activities. If we are to extend our
technological base, we must actively seek renewed interest in the hard
disciplines of science and engineering. The economic growth of the
U.S. and the world will be enhanced by exploitation of the space
environment. Ignoring space abandons the major resource base of the
next century.
5. Retaining space options is time dependent. The lead time
for space activities is long. Decisions made NOW in 1981 have
consequences stretching far into the future. Decisive programs must
be underaken NOW or many capabilities will be lost; and once lost,
they cannot be regained without costly and wasteful crash programs.
Much that we should accomplish before 1988 cannot be done without
immediate changes in our national space policies.
6. The space question is crucial: if we do not preserve space
options, we are betting national survival in order to save a miniscule
fraction of the national budget. This is neither reasonable nor
prudent.
7. It is also possible to make space pay for itself--indeed,
to use space to feed a new period of rapid economic growth. The
opportunities are there. The resources and energy are there. It is
now obvious that some nations will gain great wealth from space. The
only controversy is over the time scale.
8. If humanity survives at all--which we fully expect--then
there is no doubt that civilizations in the centuries to come will
spread across the entire solar system. As Arthur Clarke has said,
except for a fleeting instant in the beginning of history, the word
'ship' will mean space ship.
This generation can take mankind and freedom into the solar
system. Much can be lost by delay; still more can be gained by
beginning now. |The nation and statesmen who give mankind the planets
will be remembered forever.
AAS--L5 Statement on FY 82 Budget Page Three
.C
|PRESERVING SPACE OPTIONS.
The United States needs, but does not have, a comprehensive
strategy for exploiting space. We must have a unified plan which
abandons the artificial division of space into "military" and
"civilian" programs.
Such a plan cannot be devised in a few days. Space plans are
by nature technologically complex, and require considerable study.
However, it is clear that certain capabilities ought to be preserved,
so that strategists will retain a full menu of options.
These options must include the capability to:
Move quickly to a permanent manned presence in space.
Develop economic returns from the space investment.
The FY 1982 NASA Budget prepared by the previous
Administration forecloses significant options which should be
preserved. We therefore recommend that the following items be added
to the FY 82 budget as insurance. Note that we do NOT recommend that
all of these systems be constructed; but we do think it vital that
they remain possible.
While the costs of these systems is not low, it is small
compared to many other elements of the national budget; and the
options retained thereby may be vital to the preservation of the
United States in future times.
FY 82 recommendatons
1. LEO BASE ONE (Space Industrial Park)
FY 82 Funds required: $5 million
TOTAL SYSTEM COST: $4 billion
System Operational Capability: 1988
Preliminary plans already exist for LEO Base under the concept
of the Space Operations Center; a general-purpose permanently manned
space station capable of supporting privately-financed space
industrial activities. It may be thought of as a "space industrial
park".
LEO Base One could be the most important new start of this
decade. It will place the U.S. permanently in space, demonstrating
unequivocally that we have not abandoned the high frontier to the
Soviet Union. Moreover, this operations facility provides
opportunities for the creative energies of private enterprise to be
brought to this crucial area. It has been the historic role of
goverment to build roads to new frontiers and protect the early
settlers. This space facility meets that need.
LEO Base One also provides a splendid opportunity for
international cooperation. Its modular design would allow not merely
experiments, but industrial research and development, in cooperation
with allies and friends.
The total cost of the LEO Base, ready for operations, is
approximately $4 billion in 1981 dollars.
The space station has a significant possibility of bringing a
very high return on investment. We might get filthy rich from it.
LEO Base One could be made operational before 1988 if
intelligent management and procurement procedures are employed.
AAS--L5 Statement on FY 82 Budget Page Four
We have studied the possibilities of having significant
hardware components of LEO Base One in orbit by Fall of 1984. We
conclude that while this is possible, it is a high-risk venture, and
requires an immediate crash program to be given highest national
priority. It does not cost a great deal more in total costs to go for
the 1984-85 target date, but it does require more money invested much
earlier in the program.
2. Halley Comet Flyby (Scientific/ National Prestige)
FY 82 $20 Million
TOTAL COST: $600 Million
The Halley mission is the only competition with the Soviets
that is fixed in advance. We will look good if we try it. We will
show that we have not abandoned space to others.
Halley's Comet will be visible in the solar system in 1986-87.
Unlike the Kahoutek "Christmas Comet", Halley's has been known for
centuries, and has never disappointed us yet. It is likely to be
spectacular.
There is very little that the United States can do in 1986-87
that will be impressive in comparison with the known Soviet space
capabilities and intentions. Therefore, it is reasonable to exploit
the few advantages we have--and the capability for spectacular
pictures from within the gasseous coma of Halley's Comet is nearly
certain. The mission could fail, but that is highly unlikely.
The Halley Comet mission requires an immediate funding of $20
million. If that is not put in the FY 82 budget, the opportunity is
lost for this generation.
3. Space Solar Power Systems
FY 82 $30 Million
TOTAL SYSTEM COST Up to $200 Billion
Systems Operational Capability 1990-2000
Although the most spectacular use of Space Solar Power Systems
is to provide electricity for Earth, they will also be vital for
exploitation of space resources.
There is no question that Space Solar Power Systems will
←work.← Many expensive studies have proved that. The controversy is
over the economics of using them as a means of providing significant
electric power to Earth. (One currently considered design would have
each satellite generate as much electricity as does Grand Coulee Dam.)
There are sufficient uncertainties as to preclude making
Space Solar Power Systems a national goal at this time.
HOWEVER: the option to build Space Solar Power Systems should
be preserved and the economic feasibility of the concept investigated.
A reasonable funding level for this kind of space power system
for FY 82 would be $30 million. Most of that would go to technology
studies; the resulting technology will be useful for other space
programs, including development of long-term capabilities for
exploitation of lunar resources. Therefore, the money spent in the
Space Solar Power Systems program is largely a recoverable investment.
As an aside, most investments in technology have more than
paid for themselves. Knowledge is indivisible, and is useful no
matter under which program it is developed.
AAS--L5 Statement on FY 82 Budget Page Five
The Space Solar Power Systems option adds another means of
national survival. Our present energy policies are unlikely to
provide the energy resources for rapid industrial growth until the
year 2000. The Space Solar Power Systems option is cheap insurance
against failure of more conventional energy supplies--and all the
money for Space Solar Power Systems is spent in the US, developing US
technological capabilities.
4. ASTEROID OR LUNAR POLAR INVESTIGATION
Industrial Exploitation with strong Scientific Value
FY 82 $50 Million
TOTAL SYSTEM COST Under $300 Million
Mission Completion: Before Fall 1984
Note that this is the only new start that could begin and go
to completion before 1985.
APOLLO gave us valuable knowledge about 20% of the lunar
surface. This knowledge can be rapidly and economically extended to
the entire lunar surface by means of an unmanned satellite in a polar
orbit around the Moon. A large number of well-conducted scientific
and engineering studies have defined the mission, which employs
off-the-shelf spacecraft.
The lunar polar mission can discover an even greater range of
useful resources than were found by Apollo. It may confirm the
existence of vast quantities of water ice, which theory predicts must
exist in the eternally cold polar regions.
Instrumentation for the lunar polar mission would be
applicable to exploration of asteroids, other moons, possible
near-Earth space debris, and planets with tenuous atmospheres.
Asteroids are a potential source for a variety of industrial
raw materials. Investigation of asteroids will expand the potential
material resource base for the United States and all of humanity.
This mission is important for eventual exploitation of space
resources, and also commands great support within the scientific
community.
Either or both of these missions: asteroid or lunar
polar--could be accomplished in 1984 (if we begin now). Much of the
equipment--satellite and launch vehicle--required for either mission
is common to both. Thus, provided that we commit now to doing one of
the missions, we have a few months in which to decide which one
actually to accomplish.
.c
|CONCLUSION
The above recommendations preserve significant options at
relatively low cost. This will allow more liesurely study and the
development of a comprehensive national space policy. Failure to
preserve these options dictates a number of limits on our space
program in advance of development of an integrated space policy and
plan.
A comprehensive strategy for using space as a means to pursue
vital national interests is urgently needed. It should not be
crippled in advance through pretended savings.
------------------------------
Date: Feb 5, 1981
To: SPACE@MC
From: OTA@SAIL
Subject: OMB cuts again, this time: NASA
n525 0133 05 Feb 81
BC-NASA-02-05
By William Hines
(c) 1981 Chicago Sun-Times (Field News Service)
WASHINGTON - The Reagan administration is tentatively proposing a
slash of more than 9 per cent in the budget of the National
Aeronautics and Space Administration.
Such a cut would bring the U.S. interplanetary exploration program
to a virtual halt for the rest of this decade.
The cuts were included in a wide-ranging ''hit list'' prepared by
the Office of Management of Budget and obtained this week by the
Chicago Sun-Times. Disclosure of the proposals took NASA's top
management by surprise.
If OMB Director David Stockman prevails, NASA will lose $629 million
of the $6.726 billion budget that President Carter submitted to
Congress shortly before he left office last month.
No element of the space program - including even the hitherto
sacrosanct Space Shuttle manned rocketship - would be shielded from
the OMB axe. The new proposals call for deleting the option to build
a fifth Shuttle craft to augment the four now authorized as the
mainstays of both civilian and military space activities beginning
about 1985.
Also doomed under the OMB formula would be a project called Galileo,
designed to continue scientific investigations of Jupiter and its
ring and moon systems that began with the Pioneer and Voyager
missions of the 1970s.
The cuts also would ''defer'' - for the 1982 fiscal year, at least -
an advanced program for the mapping of cloud-shrouded Venus, whose
surface features were first disclosed in crude detail by a Pioneer
spacecraft last May.
The de-emphasis on interplanetary exploration, if approved, would
virtually eliminate the famed Jet Propulsion Laboratory in Pasadena,
Calif., as a significant center of space research after the middle of
this decade. It also would yield supremacy in this field of space
science to other countries now beginning to take a lively interest in
deep space.
Decisions not to go to Jupiter again and to delay the Venus mission
almost certainly would dash whatever hopes U.S. scientists had for a
close-up look at the famous Halley's Comet, which is due to make a
rare flight through the inner solar system in 1986.
Japanese, European and Soviet space scientists are working on
spacecraft designed to fly by Halley's, which is seen about once
every 75 years. Without a go-ahead this year, American scientists are
doomed to stand on the sidelines as the celebrated comet dashes past.
In addition to a $40 million saving in the 1982 budget by deferring
the Venus Orbiting Imaging Radar (VOIR) project and $108 million by
canceling the Galileo-Jupiter mission, OMB proposes to save $155
million by cutting off the fifth Shuttle and $52 million by deferring
work on an Earth-orbiting spacecraft designed to study some of the
universe's most mysterious objects.
This ''Gamma Ray Observatory'' would zero in on stars and other
remote objects that emit not just light, but also invisible radiation
akin to X-rays. Some of these gamma-ray emitters, scientists believe,
may be dying stars just on the verge of collapsing into themselves
and becoming ''black holes.''
Two projects with more practical applications to everyday life on
Earth also are targeted in the OMB hit list.
One is a set of scientific experiments intended for research on the
upper atmosphere, and the other is an aerodynamic simulator that
would be expected to help in the design of advanced airplanes. The
upper atmosphere research package is in the Carter version of the
1982 budget for $20 million, and the simulator for $16 million.
Deferring of the Venus radar mission had definite political
overtones, as did its inclusion in the Carter budget.
VOIR - an acronym that in French means ''to see'' - had long been
proposed as a logical next step in the exploration of the nearest and
most Earthlike of all our planetary neighbors. Because of its dense,
perpetual cloud cover, Venus' surface can never be observed by
ordinary telescopes or TV spacecraft.
VOIR would carry advanced side-looking radar equipment to map in
fine detail the surface of Venus, which is now known to contain huge
mountains, continents, valleys and other Earthlike features.
Carter kept NASA's VOIR proposal on the back burner until Nov. 1,
the weekend before the election, and then announced it under
circumstances that were widely interpreted as a vote-getting bid to
the aerospace community.
Probably the biggest surprise in the OMB hit list was the
cancellation of Galileo, which has had wide scientific support and
represents another step in a field of space exploration where the
United States has been unchallenged up to now.
The schedule for Galileo was launching in 1985 and arrival at
Jupiter in 1987. It would have consisted of two spacecraft,
simultaneously launched. One would be an orbiter that would circle
Jupiter, studying the planet, its rings and moons over a period of
years. The other would be a probe designed to fly into Jupiter's
dense atmosphere, giving insights into the nature of the giant planet
that could not be obtained in any other way.
The harsh cuts proposed by Stockman were at sharp variance with the
kind words he had for the space program in speech to the National
Press Club here the day after the Reagan administration took over.
While specifying that ''we will be looking for some waste'' in NASA,
the OMB director said he did not ''foresee any major changes or major
reductions in the space budget.''
He added: ''I think (the space effort is) a very constructive and
important investment for the country to make, not only because of its
technological spin-off, but simply because of the boost that (it)
gives to our economy and our aspirations and imaginations as a
society as a whole in general.''
END
nyt-02-05-81 0433est
***************
------------------------------
End of SPACE Digest
*******************
∂05-Feb-81 1300 JMC*
juilland, donnelly franz mazda doctor paolucci
∂05-Feb-81 1406 FFL
To: JMC, FFL
Les Dugan asks that you call him please today. 55 495 5669.
∂05-Feb-81 1524 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Date: 5 Feb 1981 1519-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
To: JMC at SU-AI
In-Reply-To: Your message of 5-Feb-81 1455-PST
Do you have his home address?
-------
∂05-Feb-81 1601 TOB computing
Jim Bell, now of Hewlett-Packard, offered to brief
me on a non-disclosure basis about their plans for
modern computing. Do you want to take part? I would
value your opinion.
Tom
No. I don't have time to take part. If you are impressed, perhaps
we can then arrange something.
∂05-Feb-81 1719 RWF
∂18-Jan-81 0137 JMC passwords for equipment
The file PASSWO[W81,JMC] contains a writeup of the idea, more or less as
we left it. Shall we try to patent it? My idea is to try the Stanford
patent people. Stanford and the Department would then get 2/3 of any
money made, but we would get something if they succeeded in marketing
it. If you want to proceed, please look at the file and make revisions.
The file seems OK to me as is. The idea,in print, looks very promising.
Sure, I don't mind getting rich. -Bob
∂05-Feb-81 2248 RWW hi
i am back lets talk.
richard
∂06-Feb-81 0401 OTA SPACE Digest
To: SPACE@MIT-MC
SPACE Digest
----------------------------------------------------------------------
Date: 5 FEB 1981 0930-EST
From: REM at MIT-MC (Robert Elton Maas)
Subject: OMB cuts again, this time: NASA
To: OTA at SU-AI
CC: SPACE at MIT-MC
Bletch! Well, I guess it's time for all us science-loving people,
and there are many of us, to band together and either make Congress
reinstate these budget items or else pool our funds and purchase
NASA and JCL and keep all scientific findings to ourselves until
those others realize their folly and purchase the findings from us.
I'd be glad to pay NASA&JCL for my share of the pictures of
Ganymede and Io and Dione and Saturn etc. (if enough share costs that
my share isn't more than I can afford) and would be glad to do the same
for future missions such as Galileo and Asteroid-sampler.
Yeah, I know, purchasing a government agency isn't quite like purchasing
a corporation, but something might be arranged somehow.
------------------------------
Date: 6 FEB 1981 0602-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: maximum effort
To: SPACE-ENTHUSIASTS at MIT-MC
They're now testing the trial baloon:threaten to cut our liver
out, to see if we have any constituency. It's an old ploy, and
if we have no constituents, they will in fact hack away.
The Star Trek outfits are doing something.
Soms sf people are (if anyone wants to forward this to SF
lovers, feel free)
The space budget is threatened. Psace is like defense, we say:
to be increaed, not cut. It's INVESTMENT in mnkind's future.
If you believe that send letter to Hon. David Stockman, OMB EOP
Wash DC 20503 and say so. You might also urge that he read the
AAS/L5 paper he has (the one we circulated here).
President Ronald Reagan The White House Wash DC 20500
"Dear Mr. President..."
It's aa standard test. If nobody squeals they will cut out
orbiters and asll the rest of the programs.
On the other hand we could show them space has a
constitency. Or does it?
------------------------------
End of SPACE Digest
*******************
∂06-Feb-81 0408 JRA cs institute
bill mckeeman suggests $300. is that reasonable? if not, what do you suggest?
i'm sure we can work out something. i need a biography today, and a picture
almost as soon, for the poor people are pulling their hair to get this to the
printer. i'll try to get hold of you before i leave stanford this morning
john
Make it $400 and you have a deal. BIO[1,JMC] may be suitable for your
purposes.
∂06-Feb-81 0504 Neumann at SRI-KL VERkshop II
Date: 6 Feb 1981 0504-PST
From: Neumann at SRI-KL
Subject: VERkshop II
To: VERKshop:
Steve has now confirmed NBS 21-23 April.
More details soon. Plan ahead. Peter
-------
∂06-Feb-81 0913 CSD.SCHREIBER at SU-SCORE Change of grade
Date: 6 Feb 1981 0909-PST
From: CSD.SCHREIBER at SU-SCORE
Subject: Change of grade
To: csd.ossher at SU-SCORE, jeb at SU-AI, csd.gischer at SU-SCORE,
csd.clarkson at SU-SCORE, jmc at SU-AI, acy at SU-AI, rwf at SU-AI,
csl.lantz at SU-SCORE, csd.schreiber at SU-SCORE, als at SU-AI,
csl.sso.owicki at SU-SCORE
cc: csd.tajnai at SU-SCORE
After reconsidering his answer to the LISP program question, both
AI graders, Harold Ossher and Ken Clarkson, agreed that Marvin Theimer's
should have gotten more points. That raises him over the PhD pass threshold
and removes his subminimal performance in AI. He still is shy of minimum
competance in MTC. I will, therefore, inform Carolyn that he has a PhD
pass, conditional on satisfactory performance in a course or other
work agreed to by Bob Floyd, unless I here a loud complaint by Monday
AM.
If you haven't prepared a solution set for the questions you made up,
please send me one in the next few days. Rob
-------
∂06-Feb-81 0939 JRA it's a deal
....however bio[1,jmc] is protected. could you transfer a copy
onto bio[1,jra], please?
∂06-Feb-81 1005 JRA
can you copy bio[1,jmc] onto [1,jra]?
∂06-Feb-81 1157 FFL
To: JMC, FFL
Rob says they need a clean copy of solution to your comprehensive question
by sometime Monday.
∂06-Feb-81 1207 FFL
To: JMC, FFL
Something is at fault in Energy.pre and I cannot handle it. Sorry.
∂06-Feb-81 1529 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, February 10
SPEAKER: Olaf Owe, University of Oslo, Norway,
and Stanford University
TITLE: The Response Concept as a Specification Tool
for Concurrent Programs
ABSTRACT:
Some extensions of Ole-Johans Dahls ideas about time sequences as a tool
for describing cuncurrent programs will be presented. For a given object
(say a module, process or monitor) both input and output are regarded as a
sequence of events, ordered by time. The object is viewed as a sequence
transformer, accepting an input and producing an output. The behaviour of
the object is characterized by a response function (or relation). The
response to a certain input q includes future output that results from q.
It will be shown how to define response functions, using temporal logic,
that apply to more or less complex objects. Every response function will be
monotonic in the sense that response increases with time, and satisfy that
the response to q will appear regardless of subsequent input to q.
Response functions can be used to discuss fairness as well as deadlock
properties.
∂07-Feb-81 0357 ROD Orals.
To: TOB at SU-AI, JMC at SU-AI, csd.lenat at SU-SCORE
CC: ROD at SU-AI
I would like to schedule my Ph.D. oral exam for the week of March 9th to
13th. I think the university requires it to be at 2:15. Are you free on
any of the days that week, or not free on any? If that week doesn't fit I'll
try for the previous one. I expect to have a substantial thesis draft by then.
Any afternoon that week is ok, though I have a slight preference that it
not be Tuesday.
∂07-Feb-81 1032 CSD.LENAT at SU-SCORE Re: Orals.
Date: 7 Feb 1981 1030-PST
From: CSD.LENAT at SU-SCORE
Subject: Re: Orals.
To: ROD at SU-AI, TOB at SU-AI, JMC at SU-AI
In-Reply-To: Your message of 7-Feb-81 0357-PST
M,W,F is best for me.
Doug
-------
∂07-Feb-81 1753 ARK LETTER.TEX[TEX,ARK]
...has been corrected.
Arthur
∂08-Feb-81 2332 LLW Advanced Space Transportation
To: POURNE at MIT-MC
CC: LLW at SU-AI, RAH at SU-AI, JMC at SU-AI
THE NATIONAL REQUIREMENT FOR SPACE TRANSPORT TECHNOLOGY DEVELOPMENT
BACKGROUND. The costs--direct and indirect--of transporting men,
materials, and systems from one point to another aggregate to well over
90% of the total costs to date of the space program of the United States.
Cutting these costs drastically would seem to be the first priority of a
rationally managed National space exploration and exploitation program.
Large--in contrast to incremental--gains can be realized only through
non-trivial extensions of the space transport technology base.
However, the National technology base in space transportation has been
totallly abandoned by NASA during the past two decades. Essentially no
really new idea for transportation of men or materials has received
serious support from NASA since its inception. Apollo and Shuttle propulsion
technology is a straightforward extension of that pioneered by Goddard in
the 20s and developed most of the way to its present level by von Braun
and his colleagues in the 30s and 40s. Indeed, the first two major,
technically viable American initiatives in advanced propulsion for space
transport--nuclear rocket engines (Project NERVA) and bomb-propelled
rocket propulsion (Project ORION) died under NASA sponsorship in the 60s,
and nothing has replaced them through the present time.
At best, NASA development of the advanced space transport technology base
has been ineffective; the less charitable might label it pernicious. The
effort clearly needs to be re-chartered in more innovative directions, and
possibly re-staffed with more forward-looking management.
But what should a rejuvenated space transport technology base effort do?
All present space transportation technologies are limited by the rocket
equation--which specifies how much propellant must be expended to
transport a given payload to a specified destination from a given starting
point--and by the use of chemical propellants. It's therefore pertinent
to pose two quite specific questions: Can one avoid the constraint of the
rocket equation when moving mass, and are propellants other than chemical
ones possible? The answer is clearly 'Yes' to both questions.
What then are some examples of advanced but still near-term technologies
for getting mass from the Earth's surface into Earth orbit?
ADVANCED EARTH-TO-EARTH ORBIT TRANSPORT. There are several systems based
on beaming energy to the spacecraft which is then used to power a rocket,
thereby breaking the classic coupling in chemical rockets between the
energy realizable from a propellant and the propellant's mass; one can in
principle put far more energy into a rocket exhaust stream than is
possible through chemical reactions, and thus make much more efficient use
of the ejected mass stream. These systems typically use either microwave
or near-visible radiation beams (usually originating on the Earth's
surface) to heat on-board reaction mass to much higher temperatures (and
thus substantially higher exhaust velocities and specific impulses) than
can be obtained via any possible chemical reaction. (However, it's not
clear that such systems have sufficiently large advantages over advanced
conventional rocket technology, such as single-stage-to-orbit launchers,
to warrant the required developmental effort.)
Another related set of quasi-conventional propulsion technologies are the
nuclear fission-based systems, both pulsed and continuous. The NERVA-type
technologies use very compact nuclear reactors to heat reaction mass
streams (typically hydrogen) to temperatures which correspond to specific
impulses in the 850 seconds regime; these values can be enhanced to levels
over 1000 seconds by oxygen injection into the rear section of the rocket
nozzle to burn the exhausting hydrogen. (By comparison, the most advanced
conventional rocket technology, the hydrogen-oxygen engine, achieves
exhausting-to-vacuum specific impulses of the order of 450 seconds.)
Orion-type rockets propelled by a succession of small nuclear fission
explosions at their rear offer the promise of specific impulses in the
7,000-12,000 second range, and are particularly promising for transporting
very large payloads. However, development of such technology has been
halted--perhaps permanently--by unabashedly political considerations.
(Why such political considerations are germane for space transport far
from any planet, including Earth, in an already thoroughly radioactive
universe has never been rationalized.) NERVA systems represent very
nearly flight-ready technology, while ORION rockets were within five years
of their maiden flights when the program was cancelled by the White House
in the mid-60s.
But is it possible to forsake reaction mass-based systems entirely, and
thus to avoid having to pay for the reaction mass and the corresponding
mass ejection systems of even exotic rocket-type technologies? It is only
by such advances that the costs of space transportation can be really
drastically reduced, i.e., by more than an order-of-magnitude.
The most notable prospect is the mass driver. Such systems permit the use
of Earth- (or lunar-) based electrical energy systems to provide the
energy necessary to climb out of planetary gravity wells into orbit.
Their use potentially reduces the cost of putting mass into Earth orbit
from the present $1000-$10,000/lb to values less than $1/lb, thereby
profoundly altering the economic feasibility of orbiting large quantities
of mass, e.g., for space station construction activities.
The best-known mass driver is the linear motor, which throws its moving
element out the end of its stationary element every motor `cycle.' Such
systems have been conceptually designed for putting mass from the lunar
surface into the Lagrangian points of the Earth-Moon system, and have been
proposed for use in orbiting mass from Earth, a far more formidable
undertaking. These systems have the impressive advantage of being
technically non-challenging--their realizations are straightforward pulsed
power engineering exercises--and they are also essentially guaranteed to
function as specified. All near-term systems of this type impose
substantial geometric and acceleration constraints on the payloads which
they orbit--the payloads need to be slender and tolerant of high
accelerations--and thus are best suited for large-scale transport of raw
(or at least reasonably compact and strong) materials. For at least the
near term, they are completely unsuitable for personnel transport.
Another type of mass driver, less familiar but probably more promising,
would use high performance hydraulic systems to eject continuous `wires'
of virtually any material from the surface of the Earth or Moon into
orbit. Such systems could be exceedingly compact--the volume of a large
bed for a complete system which could launch a million tons per year of
lunar material into L-4 or L-5 from the lunar surface--and are cheap to
build and simple to operate. Operating the lunar surface, such a system
would require only the photovoltaically converted power of sunlight
falling on a few acres being inputted to a commercial hydraulic pump.
Such a system operating in the much deeper gravity well of the Earth would
be a substantial extension of current technology. However, it could form
escape speed wires which would efficiently penetrate the Earth's
atmosphere. To put such wire streams into the same circularized high
Earth orbits, however, angular momentum would have to be `borrowed' from the
orbital motion of the Moon, by sending some mass fraction of the wires (in
the time-average, perhaps) into initial orbits which would pass reasonably
close to the Moon prior to taking up its final orbit around the Earth. (Of
course, delivery of such wires to the surfaces of the Moon or other
planets from the Earth's surface does not require such expedients.) The
use of magnetohydrodynamic motors as wire extruders may substantially
reduce the already modest cost and complexity of such high rate mass
orbiting systems operating on the Earth's surface.
One of the earliest space transport technologies proposed in any detail--
the very large cannon shooting a projectile into orbit featured in Jules
Verne's From The Earth To The Moon--has a modern version which would not
only work (as Verne's would not), but which could put compact,
acceleration-tolerant payloads into orbit massing as much as a million
tons, thirty thousand times that of a Shuttle launch. Driven by nuclear
explosive-created steam out of a shaft sunk to several kilometers depth in
rock, the payload projectile of such a system could carry all the material
needed for the largest space stations considered to date into orbit in one
launch event whose total cost--about that of an Apollo mission--would be
well under that of the materials being orbited. Moreover, such a launch
event would input far less radioactivity into the human environment than
did the Mt. St. Helens volcanic eruption.
Several recent proposals exploit advanced materials technology to create
any of several very different varieties of high-capacity skyhooks--devices
for relatively gently lifting large quantities of materials high into
space, perhaps putting them into orbit in the process. The best-known of
these proposed systems are 40,000 km-long cables dangling from
geostationary orbits down to the Earth's surface which can be used as classic
skyhooks; some of these rotate in orbit, and thus can be used to lift
cargos into space from the Earth's surface, or from one orbit to another.
Unfortunately, all of the most interesting skyhooks require the use of
hypothetical materials substantially stronger than could possibly exist.
A recent skyhook-type proposal obviates the materials problem by using
dynamic stiffening of a tower extending from the Earth's surface to
geostationary orbit. In this scheme, a (semi-) continuous stream of
high-speed metal hoops circulating up the tower from its Earth footing to
its top donates momentum (semi-) continuously all the way up and repeats
the process going down after being direction-reversed at the tower's top.
This stream of hand-sized hoops thus dynamically supports a tower built of
even relatively weak materials, and moreover permits it to transport very
large loads to geostationary distance (and thus into geostationary orbit).
A 300,000 ton tower composed mostly of petrochemicals--synthesized from a
few supertankers' cargos and grown upward entirely from the Earth's
surface--could loft a few million tons/year of any type of cargo into
geostationary orbit, while using only a couple of percent of the present
US electrical generating capacity. The development and use of this
technology would provide Americans with a `Star Bridge' over which a
National move into space could be made in a very rapid and extremely
large-scale fashion--all without ever having to leave the Earth beforehand.
ADVANCED INTERPLANETARY TRANSPORT TECHNOLOGY. Mass drivers and skyhooks
are also potentially useful for flinging mass into interplanetary orbits,
either from the surfaces of the Earth and the Moon or from orbits in
cis-lunar space. Really extended mass drivers--of the order of 1000 km
length--may even be used to launch payloads containing personnel into
interplanetary trajectories.
However, advanced transport systems for general-purpose interplanetary use
seem likely to fall into two classes for the rest of this century:
high-performance ion rockets and fusion torch-type rockets.
Ion rockets have been discussed for nearly 3 decades, and some extremely
modest ones have actually been employed for moving small payload very
slowly between Earth orbits. Most such systems have limited interest in
the advanced space transport context, due to very limited specific thrusts
which imply very long travel times. However, providing the electricity
required for a relatively very high thrust ion rocket with a highly
mass-economized (e.g., very thin, direct band-gap) semiconductor solar
photovoltaic power system offers the prospect of a ion drive with a few to
several tens of watts per gram specific power, corresponding to continuous
specific thrust levels of several dozens of milligees in Earth-Mars-Venus
service, i.e., interplanetary trip times in the inner part of the solar
system of 1-4 months. Such systems are the hydrofoils in an analogy in
which conventional solar sail-propelled vehicles are the recreational
sailboats.
For order-of-magnitude improvements in specific power (and roughly
half-order-of-magnitude reductions in trip times) and solar system
transport which isn't limited to the Earth's proximity to the Sun, fusion
torch drives are the primary alternative. These propulsion plants will
feature a sustained series of fusion microexplosions capable of generating
specific impulse levels as high as 600,000 seconds at specific thrusts of
milligees (and correspondingly higher specific thrusts at lower specific
impulses). Research to demonstrate high energy output fusion
microexplosions is presently being funded for both military applications
and civilian power generation purposes. For propulsion applications,
fusion pellet drivers capable of rejecting waste heat at high temperature
(at least 1000 K) are required; rare gas-halide excimer lasers (e.g., KrF
and XeF systems) are currently the leading candidates for such a role.
CONCLUSIONS. The three space transport areas in which advanced technology
is likely to provide the largest near-term increases in what the US can
accomplish are moving materials from the Earth to Earth orbit,
transferring materials between the Moon or asteroids and useful Earth
orbits, and exploring the inner half of the solar system. The
technologies sketched in the foregoing are indicative of the gains which
can be made by a significant, progress-oriented program of space
transportation technology development by the United States. Any
economically rational US space program for the 80s will emphasize
development and utilization of some fraction of these possibilities,
simply because devotion to six-decade-old space transport technology has
made space exploration and exploitation almost forbiddingly expensive for
even the richest nation on Earth.
∂09-Feb-81 0902 JMC*
call Franz, Fredkin, Fordham
∂09-Feb-81 0916 Shortliffe@SUMEX-AIM Medical group within AAAI
Date: 9 Feb 1981 0911-PST
From: Shortliffe@SUMEX-AIM
Subject: Medical group within AAAI
To: AAAI-Executive-Committee:
cc: LINDBERG@SUMEX-AIM, clancey@SUMEX-AIM, KULIKOWSKI@RUTGERS
Council Members:
I have heard from a few of you regarding my recent message suggesting
the formation of a subgroup on medical AI within the AAAI. Although I'd
appreciate hearing from everyone, there seems to be enough support for the
concept that Don Lindberg and I will go ahead and prepare a formal proposal
for your consideration. We'll try to get this material to you within the
next 4-6 weeks so that there will still be time to do some organizational
work prior to IJCAI/AAAI in Vancouver. That meeting may provide an
appropriate setting for getting our subgroup off the ground.
Thanks for your interest and suggestions.
Regards,
Ted Shortliffe
-------
I can't see any problem with the existence of a medical subgroup.
∂09-Feb-81 1332 AVB
∂09-Feb-81 1212 JMC
Please put RWW on SUN list and send him message with ccst breakdown.
--- okey, done --
∂09-Feb-81 1347 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Lunch with Drs. Kobayashi and Cadiou
Date: 9 Feb 1981 1344-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Lunch with Drs. Kobayashi and Cadiou
To: CSD-Faculty:
Dr. Kobayashi and Jean-Marie Cadiou of IBM would like to invite members
of the faculty to lunch tomorrow. If you are interested, please
send message to me.
Carolyn
-------
Lunch with IBM fine, but my class ends at 12:15.
∂09-Feb-81 1418 FFL
To: JMC, FFL
Your termite man, Franz returned your call. His phone is 328-0590.
∂09-Feb-81 1419 FFL
To: JMC, FFL
The person from N.M. State U., who wanted the latest version of Lisp, called
again. You said you needed to know what kind of computer they had. It is
an Amdahl 470, IBM compatible. What may I now tell her?
Cynthia Walters, PO Box 3AT, NMSt. U., Las Cruces 88003. 505 646 l443.
She should write or phone Dr. Anthony Hearn, RAND Corporation,
1700 S. Main St., Santa Monica, CA
∂09-Feb-81 1516 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Date: 9 Feb 1981 1513-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
To: JMC at SU-AI
In-Reply-To: Your message of 9-Feb-81 1437-PST
I'll find out where and perhaps you can meet them.
-------
∂09-Feb-81 1556 RWW
who do i ask about getting on the sun terminal mailing list?
It would be AVB, but I already asked and you're on it.
∂09-Feb-81 1617 ROD Orals
To: TOB at SU-AI, JMC at SU-AI, csd.lenat at SU-SCORE
CC: ROD at SU-AI
I've scheduled my oral exam for 2:15 pm Friday 13th March.
∂09-Feb-81 1626 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Date: 9 Feb 1981 1622-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
To: JMC at SU-AI
In-Reply-To: Your message of 9-Feb-81 1437-PST
They will wait for you in Jacks 252.
Carolyn
-------
∂10-Feb-81 0223 POURNE at MIT-DMS (Jerry E. Pournelle) that tower
Date: 10 Feb 1981 0523-EST
From: POURNE at MIT-DMS (Jerry E. Pournelle)
To: JMC at SU-AI, LLW at MIT-MC, pournelle at MIT-MC
Subject: that tower
Message-id: <[MIT-DMS].184829>
Easier than that tower I can sell maybe a case of tuberculosis. The tower
goes in the Space Council report over my cremated body and exorcised
ghost.
MC is dead, so I am sending this from memory; I was working on reply to JMC
comment on LLW transport paper when MC died.
We are a citizen's advisory group on space policy. We may become THE
citizen's advisory group on policy. We have at the moment a pretty good
reputation (having ovecome the L-5 is flakes image we used to have). All I
need do is talk about that tower, and maybe they listen to me again sometime
in early 1996.
Papers for internal stimulation are all right; that is what that tower is; but
let us not confuse that with reports to the public and the authorities.
∂10-Feb-81 0239 POURNE at MIT-DMS (Jerry E. Pournelle) llw's paper and some curses
Date: 10 Feb 1981 0539-EST
From: POURNE at MIT-DMS (Jerry E. Pournelle)
To: jmc at SU-AI, llw at MIT-MC, pournelle at MIT-MC
Subject: llw's paper and some curses
Message-id: <[MIT-DMS].184830>
MC keeps crashing when I am halfway through answering jmc's comment
on llw's paper.
We will see that all necessary reference materials are included in
all papers; if we need to explain isp we'll do that. I'm for not
doing it, but maybe. I've used it in some papers before.
WHAT EVER HAPPENED TO DUMBO and is it worth mentioning?
∂10-Feb-81 0245 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 10 February 1981 05:45-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, LLW at SU-AI, RAH at SU-AI
Date: 09 Feb 1981 0127-PST
From: John McCarthy <JMC at SU-AI>
Comments on LLW draft on "Advanced Space Transportation"
Question for Jerry: To what extent is it important to reduce the amount
of technical knowledge and familiarity with space technology required
to understand the proposals?
1. I think a paragraph should be devoted to the importance of specific
impulse by giving its relation to mass ratio. This will make clear its
relation to costs.
2. We should soft-pedal the tower till the paper is published. We can
mention it, but perhaps make a soft-pedalling comment, such as that
the idea is new and hasn't received critical comment yet.
3. We should somehow mention that the political problems of NERVA are
less than those of ORION, since ORION was scuttled for frankly political
reasons and its revival will be opposed by the science establishment,
while NERVA is nominally ok with them.
4. We should mention that fusion torch propulsion is a follow-on to
laser fusion for energy.
5. The value of separating the problem of transporting materials to
orbit by a cheaper system than is used for transporting people should
be made explicit.
6. Question: Can we afford to recommend pursuing the development of
alternative space technologies in parallel possibly with the help
of allies?
If you agree with some of these proposed changes and you want
me to, I can prepare some paragraphs. Some prose follows:
The cost of getting material into orbit by rocket depends
critically on the specific impulse of the rockets. Present specific
impulses (450 seconds for the Shuttle engines which are almost the
best possible for chemically fueled rockets) require that
the mass leaving the Earth be at least 20 times that which reaches
low earth orbit. For various reasons, the mass ratio of the Shuttle
is actually about 100. The NERVA rocket would permit a mass ratio
of xxx, and the ORION a mass ratio of xxx. The numbers for longer
journeys are even more spectacular. The Apollo system takeoff mass
was 1000 times the mass delivered to te moon, whereas the mass ratio
for a NERVA system would be xxx and an ORION system xxx.
The large mass-ratios are the main reason for the high cost
of space transportation. The only way to reduce these costs is
to use high specific impulse rockets or to avoid rockets as much
as possible.
Until we can use resources from the moon or the asteroids,
space travel is a pyramid with its base the step from the Earth
to low earth orbit. Even when these resources become available,
moving people and material to Earth orbit will still be a major
cost. Therefore, it makes sense to consider every possible way
of reducing the costs of transportation to low earth orbit.
While people and delicate equipment must be transported
into Earth orbit, the bulk of what has to be transported is material,
and the single most important material is rocket fuel. What material
is used for rocket fuel depends on the kind of rocket, but all
materials can stand much higher accelerations than people can.
This is fortunate, because many of the possible systems for economical
transportation involve high accelerations. Thus we may envisage
using rockets to transport people and delicate equipment and
other systems for bulk material and less delicate equipment.
Many ideas for inexpensive space transportation are old
and have been laid aside by NASA's concentration on the project
of the moment. However, in spite of the lack of encouragement,
the last few years have seen the proposal of several promising
new ideas.
Many of the ideas mentioned above are in mutual competition,
and it will eventually be necessary to make choices. Also they are
available on different time scales. Nevertheless, it is prudent
to develop several technologies in parallel to be sure of having
the best choice available in the end. Perhaps the best way to
do this is to enlist our allies and divide the work of pursuing
promising possibilities. Today both the Europeans and the Japanese
have the technology to develop major systems.
We need rational goals and a new charter for NASA. We can then
start work on specific technologies--which must be deducible
from our goals.
We require: internal information and laundry lists; and a
concensus on what we say externally. These are two different
things. Externally we do NOT confuse the paymaster with dreams
nor with too many decisins; we give them choices, but choices
which move them in our direction.
∂10-Feb-81 0336 LLW Realpolitic Triumphant!
To: pourne at MIT-MC
CC: LLW at SU-AI, RAH at SU-AI, JMC at SU-AI
∂10-Feb-81 0312 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 10 FEB 1981 0613-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: llw at SU-AI
Date: 10 Feb 1981 0523-EST
From: POURNE at MIT-DMS (Jerry E. Pournelle)
Easier than that tower I can sell maybe a case of tuberculosis. The tower
goes in the Space Council report over my cremated body and exorcised
ghost.
MC is dead, so I am sending this from memory; I was working on reply to JMC
comment on LLW transport paper when MC died.
We are a citizen's advisory group on space policy. We may become THE
citizen's advisory group on policy. We have at the moment a pretty good
reputation (having ovecome the L-5 is flakes image we used to have). All I
need do is talk about that tower, and maybe they listen to me again sometime
in early 1996.
Papers for internal stimulation are all right; that is what that tower is; but
let us not confuse that with reports to the public and the authorities.
[Jerry: All right, all right, I'm sorry my enthusiasm seduced my programmatic
orientation. You'll eventually come to realize that the Star Bridge is just
that--the human race's big, broad, affordable bridge to the stars--and it'll
work more better than any alternative. In the meantime, you write to the
Authorities, and I'll console myself with Emily Dickinson's reminder:
Tell the truth,
But tell it slant;
Success
In circuit lies.
Too bright
For our infirm delight
The truth's
Superb surprise!
Lowell]
∂10-Feb-81 0855 MERRITT at USC-ISIB Re: campaign
Date: 10 Feb 1981 0849-PST
From: MERRITT at USC-ISIB
Subject: Re: campaign
To: POURNE at MIT-MC, ENERGY at MIT-MC
In-Reply-To: Your message of 10-Feb-81 0323-PST
Perhaps we (SPACE mailing list) should archive some of the conversations
and send transcripts (printouts) to somebody in washington. This could
be VERY effective, especially considering the sources.
I am VERY dissappointed at the decision to cut the space budget; it should
be increased!. If some of the members of this list could get together and
collaborate on a 'letter' (angry) directed at the President, perhaps there
could be SOME effect. Anyway, I am open to suggestions. Maybe a digest of
comments from each of the members, say, about one paragraph per comment. I
don't know. Anyway, SOMETHING MUST BE DONE.
<IHM>
-------
∂10-Feb-81 0850 Denny Brown <CSD.DBROWN at SU-SCORE> Gray Tuesday reminder
Date: 10 Feb 1981 0846-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Gray Tuesday reminder
To: CSD-Faculty: ;
cc: csd.tajnai at SU-SCORE, jeb at SU-AI, sgn at SU-AI, tob at SU-AI,
csd.bscott at SU-SCORE
Remember the Gray Tuesday meeting today, Tuesday, Feb. 10.
2:30 in MJ 252. We plan to start promptly, and finish by
4 pm. -Denny
-------
∂10-Feb-81 0900 JMC*
Darden fordham fredkin
∂10-Feb-81 1008 Darden@SUMEX-AIM DINNER ON MONDAY, FEBRUARY 16
Date: 10 Feb 1981 1001-PST
From: Darden@SUMEX-AIM
Subject: DINNER ON MONDAY, FEBRUARY 16
To: JMC@SAIL
HI, JOHN. WHAT A PLEASANT SURPRISE TO GET A DINNER INVITATION FROM
YOU. I ASSUME YOU PLAN TO BE IN WASHINGTON, D.C. ON THAT DAY. HERE IS
MY SITUATION: I HAVE HAD VIRAL PNEUMONIA AND I AM SLOWLY RECOVERING.
I TAUGHT TWO CLASSES FOR THE FIRST TIME IN ABOUT FOUR WEEKS
YESTERDAY. I HOPE THAT BY NEXT WEEK I WILL BE BACK ON SOMETHING
LIKE A NORMAL SCHEDULE IN WHICH I WILL TEACH THREE CLASSES ON MONDAYS.
SO, ASSUMING SO RELAPSE, ETC., I WOU D BE ABLE TO GO OUT FOR DINNER,
BUT I MAY BE A BIT LESS LIVELY THAN USUAL. WILL YOU HAVE A CAR? IT WOULD
BE GUSTITORIALY LESS INTERESTING TO EAT OUT HERE IN THE SUBURBS NEAR
MY HOUSE IN GREENBELT MARYLAND, BUT I WOULD BE LESS TIRING THAN MY DRIVING
INTO THE DISTRICT. IF YOU WON'T HAVE A CAR BUT ARE STAYING NEAR A METRO
STATION, THAT OPENS UP OTHER POSSIBILITIES. WHILE REREADING THIS I NOTICE A NUMBER OF TYPOGRAPHICAL ERRORS, BUT
I HOPE YOU CAN MAKE SENSE OF IT ANYWAY. WHAT YOU HAVE HERE IS A CONDITIONAL
ACCEPTANCE AND A REQUEST FOR MORE INFORMATION FROM A SOMEWHAT SICK FRIEND WHO WOULD LIKE TO SEE YOU IF POSSIBLE.
PERTINENT INFORMATION: HOME ADDRESS: 6 P RIDGE ROAD, GREENBELT, MD20770;
HOME PHONE: 301-474-0037; OFFICE PHONE: 301-454-2850 OR 301-454-3978
I HAVE DECIDED THAT I HAVE TO LEARN LISP AND WOULD LIKE YOUR ADVICE
AS TO THE BEST WAY TO DO IT. IS THERE A TUTORIAL PROGRAM ON LINE SOMEWHERE
THAT I CAN ACCESS TO LEARN LISP?
I LOOK FORWARD TO HEARING FROM YOU.
REGARDS, LINDLEY
-------
I'll have a car. I don't know of a tutorial program for LISP. I'll send
you a copy my book with Carolyn on LISP. Winston's book is also usable.
I'll phone you later today or this evening. I don't know yet, but Tuesday
night might also be possible if more convenient. Take it easy and don't
have a relapse.
∂10-Feb-81 1141 Denny Brown <CSD.DBROWN at SU-SCORE> PhD Requirements
Date: 10 Feb 1981 1132-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: PhD Requirements
To: bboard at SU-SCORE, bboard at SU-AI, csd.bbd at SUMEX-AIM,
PHD-Distribution-list: ;
cc: CSD-Faculty: ;, csd.tajnai at SU-SCORE
The file {SCORE}<CSD.FILES>PHDREQUIREMENTS.DOC contains the current
memo on requirements for the PhD. See Carolyn Tajnai for hard-copy.
-Denny
-------
∂10-Feb-81 1144 FFL
To: JMC, FFL
Lester Dugan asks that you call him. 9 595 5669.
∂10-Feb-81 1616 John McCarthy <JMC at SU-AI>
Date: 10 Feb 1981 1550-PST
From: John McCarthy <JMC at SU-AI>
To: energy at MIT-MC
Capital costs for energy projects
1. Capital costs have been a very large part of all means of
generating energy. Dams have been paid for in the U.S. often by
allocating parts of their costs to irrigation, flood control and
recreation as well as energy production. Nuclear energy is mainly
capital cost and will remain so even with much higher uranium
prices. Oil, gas and (I believe also) coal are mainly fuel cost, but
the capital costs are not negligible. Capital costs of transmission
and distribution are also important.
2. Low power density energy technologies require large structures,
and these dominate the capital cost if the power density is low enough.
Thus tidal power systems have always lost out on evaluation, because
no-one would bid low enough on the construction costs.
3. There is typically a substantial difference between
cost figures obtained by adding up the known requirements for labor
and materials and the costs actually experienced. When there is
experience, contractors take it into account. When there isn't
experience, smart contractors bid high. Those who don't often go
out of business, or confront the buyer with a choice between
paying for cost over-runs or losing the project.
4. The less the knowledge, the larger the chance that a priori
cost computations will have large optimistic errors. The least
knowledge at present concerns costs of construction in space or
on the moon. The ccst of maintaining large osmotic barriers or
the cost of keeping OTEC systems unfouled or the cost of fixing
windmills whose blades break also require experience to estimate.
The nuclear industry has had experience with unanticipated ocsts,
and the alternative technologies can expect it.
5. The tolerable ratio of capital cost to annual production
depends on whether you imagine a growing economy or a static
economy. Utopians often design static economies and therefore
are willing to contemplate large capital costs. The actual ratio
for industrial projects in the U.S. around 1968 (I didn't see more
recent figures) was around 1.6, i.e. 1.6 dollars in capital cost
led to a production of $1 per year. For less developed countries,
perhaps contrary to intuition, the ratio is larger. For the
Soviet Union, the ratio was 2.6 and for Iran 3.2. For utilities,
the ratio has been larger than for other industries, and therefore
they have financing problems, which are especially acute when
interest rates are high.
6. High inflation rates make it difficult to justify long
term projects if the returns are in nominal dollars. However,
many countries with persistent high inflation have nevertheless
managed high investments, e.g. Japan in the 50s and 60s.
It won't work, however, in regulated industries in which the
return allowed is on nominal dollars. One is a fool to sell
a house or gold and invest in utility stocks when the return on
the investment will buy less house or gold than one started with.
There have been such fools, e.g. institutional investors like
pension funds looking for "conservative investmenst", but the
supply is running low. Some kind of real dollar return after taxes on
investment is required. With such a proviso interest rates would
be quite low. A few years ago, some bonds in West German Marks
sold with interest rates of four percent in preference for the
same bonds in dollars at ten percent.
∂10-Feb-81 1648 Robert Elton Maas <REM at MIT-MC> Re: campaign [collection of our remarks to President]
Date: 10 February 1981 19:30-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: Re: campaign [collection of our remarks to President]
To: MERRITT at USC-ISIB
cc: POURNE at MIT-MC, ENERGY at MIT-MC
Well, I'd find it hard to restrict myself to just one paragraph and
not have it be just a general statement that duplicates what everybody
else is saying in their general statements. I'm inclined to write a
point by point essay telling why each item is crucial for our survival
both as a species an as a developing society. But I gotta find a
time slot when I can sit here for 10 hours straight in a burst of brilliance.
If we're going to send one paragraph each, I suspect separate postcards
or telegrams would be more effective than a single mailing.
∂10-Feb-81 1658 ME E macros/variables
∂10-Feb-81 1641 JMC date and file macros for E
Would it be possible to have built-in macros that give the date
and the name of the file being edited. They would be especially
useful to include in other macros.
ME - Yes, that's in my plan for string variables, which E doesn't have
yet. Dates and names require strings. Currently E has only Numeric
readonly variables, like page and line number. Of course, date could
be returned in such form in three or four parts.
This stuff will exist when I get some time to work on E again.
∂10-Feb-81 2055 Robert Elton Maas <REM at MIT-MC>
Date: 10 February 1981 22:37-EST
From: Robert Elton Maas <REM at MIT-MC>
To: JMC at SU-AI
cc: ENERGY at MIT-MC
You quoted 1.6 as if it were a pure number, as if you put in 1.6
dollars and you get back 1 dollar ($1.6 / $1 = 1.6).
As far as I could tell from your later explanation, you should have
said 1.6 years instead of just 1.6, i.e. you put in 1.6 dollars ONCE
and you get back 1 dollar EACH year, ($1.6 / ($1/yr) = 1.6 yr).
I.e. what you were really trying to say was the repayment of investment
took 1.6 years, not 1.6 pure numbers.
If that figure of 1.6 years to repay investment is typical, I can
understand how investors balk at 30-year repay-investment or even
5-year repay-investment that is typical with solar energy, nuclear
fission, and space industrialization. (Although I guess it's
notunivorm, it might be 5 years to repay the original investment,
then 1.6 years extra for each additional repayment-copy?)
I think you understood it correctly. The capital cost of a plant
producing 1 dollar per year is 1.6 dollars. This doesn't mean that
the investment is repaid in 1.6 years, because there are other costs
involved in production besides capital investment. When someone
says that a machine will pay for itself in n months, he usually means
something different - such as that the money saved compared to the
old production method will pay for the machine in n months.
∂10-Feb-81 2222 AYRES at USC-ISIF Teloperators by Mike Hyson
Date: 10 Feb 1981 2220-PST
From: AYRES at USC-ISIF
Subject: Teloperators by Mike Hyson
To: MINSKY at AI, JMC at SAIL
I wrote this recently and Jerry Pournelle asked me to send it to you
for comment. He would also like to see you comments.
Thank you.
Mike Hyson
1035 North Glendale Ave #2
Glendale, CA 91206
213-243-4554
-----------------------------------------------
TELEFACTORS
INTRODUCTION
Telefactors or teleoperators are a very important technology for the
near term industrialization of space. Various tasks in space, such as materials
processing, construction, or repair will be accomplished by a mix of
methods which will be determined by the job at hand. Some will be done by
men, others by robots, and still others by teleoperation.
This paper discusses the uses and general state of the art of
teleoperator designs and suggests a plan for their development along with
milestones and estimated funding levels for the near term. While robotics
may ultimately replace most human activities in space, it appears to us
that a viable large scale space effort in the next several years must rely
heavily on teleoperated or manned operations. Precisely because the tasks
are complex and poorly understood, much experience must be gained before
it is possible to effectivly automate many operations. It is much better
to develop the needed automation after some experience rather than
attempting to automate everything first.
Teleoperated and manned experience will let us begin the
industrializaion of space NOW without waiting for complex robotics.
Indeed, we will show that much of the needed technology was well
developed 20 years ago. Current technology can improve the techniques
with minimal research leading to the greatest return possible for invested
effort and monies.
Teleoperator technology was fairly well developed by 1969 using the
descrete, analog electronics and the linear control theory available. They were
and are being used in the nuclear power industry with great success The
improvement of teleoperators with the aid of modern microcomputers and
communications technologies will be a straight forward process.
The result will be the capability of placing the ability and
intelligence of a man at a range of at least 10,000 miles, and with reasonable
extensions of the state of the art, we will permit us to operate space factoriesand processors in low earth orbit, geosynchronous orbit, and on the moon or
asteroids.
A telefactor or teleoperator is here defined as a device to allow
action at a distance by a human operator. In the broadest sense, even a
wrench is a telefactor in that it extends the reach of a man's hand.
For the present discussion, we mean a more complete form of teleoperator that
allows the operator to manipulate his environment at a great distance and
will give him a strong sense of being located at the remote site. This is
called " telepresence".
THE NEED FOR TELEOPERATION IN SPACE
Men in space require special life support equipment such as space
suits and other life support. As long as these remain expensive, other
alternatives will be considered to reduce the cost of space operations.
One option is robots, totally autonomous machines that carry out tasks
without human intervention. The problem with robots in their current state is
that they are incredibly stupid, and likely to remain so (in comparison to men) for many years to come. Programming them for complex tasks is very
expensive and time consuming. This is not meant to critisize current robotic
efforts, but merely to point out that the problems of making a general purpose
robot are enormous, and will not be solved in the near term.
A second major option is the use of specially designed machines for
each special task we encounter. This is possible but hardly cost effective.
If it were, building houses, plumbing, TV repair and other similar tasks would
have been automated long ago on earth. Because these jobs require many
decisions, great flexibility and an almost open ended set of possibilities,
they are well beyond the capability of special machines or even present robots.
The third option, teleoperators, are general purpose devices and can
perform a wide variety of jobs with the only development delay being the time ittakes the human operator to learn to do the task. With computer assistance and
general purpose end effectors (hands) along with special purpose tools, a man
with a teleoperator can do almost anything he could do in person. In
environments where a man cannot survive, he could obviously do much that he
could never do in person.
The man and his senses are "projected" to the remote site by use of
radio, laser or other communication link and performs his task through a machinethat duplicates his actions at the remote site and transmits sensory data back
to him. The machine is located in the harsh environment while the operator is ina comfortable "shirt sleeve" environment.
PAST TELEOPERATOR WORK
Teleoperators require Sensors and Displays, Effectors, Mobility systems and Communication Links. There are two general types of teleoperators, Man
Extenders and Man Amplifiers. Prototypes of all these types were made by 1969.
Sensors and Displays
A key concept in teleoperation is to make a man-machine interface that
is transparent to the user. It ideally would have no physical connections to theoperator but would duplicate all his sensory inputs in such a way that he would
not know he was at the remote site. We are far from this ideal.
However, the human mind is very flexible and has a wonderful capacity
for sensory projection and adjustment. For example, if a man has his visual
field turned upside down or reversed left to right, he adapts to it in a few
days and is able to function normally. People are able to adapt to odd sensory
inputs and use them to replace their ordinary senses. For example, blind
children are able to use a sonar device to "see" and deaf children are able
to use buzzers on the skin to substitute for hearing. Phantom limb phenomena,
where a person still feels an amputated limb also demonstrate the adaptability
of our nervous system to odd or distorted inputs. The result of this is that
sensory displays that are imperfect and lower bandwidth than our senses will
none-the-less invoke a strong feeling of telepresence.
Head Aimed Televisions (HATS)
About 1958, Hugh Upton, John Chatten and William Bradley developed
head-aimed television displays for use in helicopters. They have great potentialfor space teleoperation. A projection system is used where the view seen by
a remote camera was displayed to the operator. The head position of the operatorwas sensed and used to steer the remote camera. Thus the operator saw a view
of the remote area just as he would have if he had been there.
Bradley writes that the display was black and white and monocular,
without any
stereoscopic vision. Even so, he felt he was "present" at the remote site
after he had adapted to the display for only five minutes. Using a similar
display and associated remote control links, he drove a truck through
the desert at a range of 10 miles.
Bilateral Force Reflecting Manipulators
In the 50's Ray Goertz, Carl Flatteau and others developed
manipulators for use in nuclear reactors. The operator moved a master
arm and a slave arm of identical or similar geometry duplicated his
motions. The master was powered very much like the slave and errors in
the servos of either arm were corrected by both the master and the slave
arm. The effect of this bilateral servoing was that the operator felt
the forces operating on the slave arm because they resulted in similar
forces at the master.
After using such arms for about one hour one can
pick up dimes, pour liquids from one glass to another and many other
fairly complex tasks with both arms in coordination (much better than
existing robots). After two hours of use, one feels odd using his own arms!
It takes about a half hour to re-adapt to one's own arms. This happens
in spite of the fact that these arms had no touch sensors other than the
force reflection.
It was found that a delay of about a tenth second could be
tolerated by operators, giving a range of operation of at least 10,000
miles without use of special delay compensation methods. This is perfect
for operations in low earth orbit conducted from the ground and probably
adequate for operations in geosynchronous orbit.
The bandwidth required for each joint of the arm was about 100 Hz,
well within the capability of even the slowest microprocessors now
available. With 14 degrees of freedom per arm ( the maximum for a human arm)
the band width would be 2800 Hz, even if no encoding to reduce bandwidth
were used.
Bilateral force reflection with a man in the loop is a powerful
control technique. There is no ready solution to the control problems
for a 14 jointed manipulator that can be solved in real time by existing
computers, yet a human operator does this without conscious thought!
A General Electric master/slave system of the period, Handi-man, allowed
a man to twirl a Hula-Hoop on the end of the slave; try that with a robot!
Mobility Systems
General Electric developed a " Quadruped Transporter " in the
early 60's using bilateral force reflection and master slave techniques.
It was essentially a four legged walking truck. The legs duplicated the
arm and leg motions of the operator. It worked fairly well and was able
to walk into a mud puddle and kick a jeep free of the mud. It had
problems. The geometry of the legs was not sufficiently similar to
a man's and this lead to mental strain that exhausted the operator.
Legged locomotion has continued to be studied by Robert McGhee
at Ohio State. He has a prototype six-legged walker that can climb
stairs and adjust to uneven terrain under computer control. With
human supervision, perhaps by teleoperation, such devices could traverse
all terrains. Recently, McGhee and co-workers have studied a three
legged device specifically designed for use on large space structures
such as solar power satellites. They would also be useful for planetary
exploration.
Use of walkers or other mobile bases in conjunction with
manipulators, head aimed televisions and the like could result in a
"remote man" capable of going anywhere and performing any job a man
might at a range of at least 10,000 miles.
Man Extension
There are obvious extensions of this technology. If we scale the
teleoperator until it is much larger or smaller than a man, we can
use the same basic system for tasks as diverse as earth moving and
micro-surgery. The main problem will be scaling the sensory inputs to
the operator so that he feels he is working with something of human scale.
Man Amplification
General Electric also experimented with man amplifiers called
Hardi-Man. It was an exo-skeletal powered frame work controlled by a
man strapped into the frame. The power output of the man was increased
by a factor of about 25. Tests were done in which a man walked with
loads of 1500 pounds in his "arms". The technology could be used for
powered space suits, heavy construction in spce, movement of men on high
gravity planets or could allow a man to move easily even though he had
heavy radiation shielding or life support units.
For example, Apollo astronauts used up to four times the
normal effort to walk in their pressure suits. A powered assist from
a man-amplifier could extend the time spent on the lunar
surface or working in space. The increase in useful output would lower the
cost of using man in space.
Teletourism
With the increase of bandwidth provided by space based
communication satellites and perhaps laser links, tourism by teleoperation
becomes feasible. Ordinary people could "travel" anywhere and experience the
sights and sounds of the earth. They could have experiences either to
dangerous for them to do in person or perhaps which they are incapable of
doing without aid.
For example, invalids could experience hiking in the Sierras or
skin diving on the barrier reef or flying through sensory aids and
teloperators.
One could even have scientists operate their own Shuttle
experiments from the ground, gaining an intuitive knowledge of the space
environment through this kind of "hands-on" experience.
The most exciting possibility is that huge numbers of people could
experience space flight or a walk on the moon, or operate a space factory;
all through the development of such teleoperated proxys. The presence of
Hundreds of people on earth "working in orbit" would surely improve
the NASA image and provide a direct return to the average person without
the lift costs and other problems of actual space tourism.
SIGNIFICANCE OF TELEFACTORS
As stated above, the development of teleoperators for space use
will allow operations in space that minimize direct use of man in space,
which reduces costs, while letting use begin NOW with all manner of space
activities without need of autonomous, intelligent robots or a myriad
of special purpose machines. This means that the R&D and front end cost is
reduced for many programs. It should be obvious that such technology is
needed right now by the nuclear industry and many other industries where
men are still needed in extremely dangerous environments. Foundery work,
coal mining and many other industries could be remotely manned, as Minsky
among others has suggested.
RESEARCH NEEDS
The reasearch needs are minor for the first phases of the
teleoperator development. Most of the problems are applied engineering.
Some of the problems are discussed below.
Bandwidth
Teleoperators need wide bandwidths. This is not a problem if
cables or fiber optics are used but could definitly be a major problem if
operating many units from the ground. One alternative is to use the
earth orbital station to house the operators. The communications links could
then be local and carried by laser beams, perhaps. Range would be only from
the space station to the local work stations. This would minimize EVA's
by the crew and allow work to be carried out from inside the station.
Free-flying teleoperators present special problems of keeping
the communication link up despite changes in orientation.
Displays & Sensors
Although the displays described above are the best approach,
they need much improvement. For example, the displays were monocular
and black and white. Color and Stereoscopic vision could be included.
The displays used in the past have been bulky. Concepts using fiber
optics could make the visual display no more bulky than a pair of glasses.
For the ultimate in bandwidth reduction, the video portion of the
signal could be precisely matched to the characteristics of the eye.
If eye position tracking plus variable resolution were used, bandwidth
could be reduced by a factor of 1000. One might achieve even greater
reductions by the transmission of difference encoded, spatially filtered
images that were planned for the "Picture Phone" researched by Bell Labs.
Touch and temperature sensors for the hands and arms remain a
major problem. The conductive rubber technology now used is inadequate.
While force reflection is adequate for most tasks, touch would be a
very usefull addition and shoud increase dexterity markedly.
Controls
The advent of microprocessors and non-linear control theory
would allow better control of the manipulators and other systems.
Digital techniques permit the computation of the higher order non-
linearities that were not easily included when analog servo electronics
of the early 60's were used.
Manipulators
There are few arms with a good correspondence to the geometry
of the human arm. Since no system now approaches the overall power
output to weight ratio of the human arm, research into powerful,
fast, light manupulators is needed. Recent advances in biomechanics
such as are now being applied to athletes could be used to help design
better arms and legs for teleoperators and man amplifiers.
Hands
There needs to be a decent general purpose human like hand developed.
This could find wide applications in prosthetics and robotics as well
as teleoperation.
Walkers
Walkers will allow general mobility when it is needed in rough
terrain or complex structures or in weightlessness. The dynamic stability
problems that have plagued walkers are solved if a human operator is
coupled to the machine. The unconscious balance motions of the operator
will balance tha machine if the right sensory feel is provided.
Free Flyers
Free-flying teleoperators have been studied but usually have
not considered the master/slave concepts presented here. Such a device
could repair Shuttle tiles or communication satellites, etc. There
are special problems with catching tumbling bodies, or holding steady
positions under changes in CG etc. which probably require at least a
computer assist to a human operator.
MILESTONES
First System Integration
The major milestones involve the system integration of almost off-the-
shelf hardware into a complete teleoperator test bed. Noone to our
knowledge has combined in one system a moble base, a head aimed display,
bilateral force reflecting manipulators, and communication link. Certainly
these types of devices have never been space qualified, with the exception
of the Shuttle remote manipulator.
Use of Advanced Teleoperation with the Shuttle Arm
Consideration should be given to using master slave concepts and head
aimed display technology with the Shuttle arm. Sensory scalling woud be
required to permit a more human feel with such a long arm. It seems that
the increase in dexterity of the operators should increase the
effectiveness and safety of Suttle operations. have never been
Develop a ComSat Repair System
It has been said that many failed satellites could be repaired by a man
with a screwdriver. Teleoperators could be used, directed from the ground
or as free flyers from the Shuttle. A study by Marshall Space Flight Center
indicated a large market and interest by ComSat in an on orbit repair system
Teleoperators have been proposed and studied for theis purpose. They have
not been supported at a high level, despite their importance. For example,
the teleoperator that was to save Skylab was put on the back burner as
soon as it was decided to let Skylab fall.
Development of an Orbital Resupply System
Teleoperators should be considered as aids to docking of spacecraft, with
resupply of propellants. This would be a major need of a permanent manned
space station.
Teleoperation of Shuttle Experiments
In other near term projects, various remotely controlled experments
can be developed for the Shuttle that incorporate teleoperation to gain
experience in ground operation of orbital processes.
Teleoperation of the Shuttle
NASA should consider unmanned operation of the Shuttle. Since it is the
first vehicle ever to fly without an unmanned trial first, it is very
risky. There is no reason it could not be flown from the ground.
This is especially true of bulk cargo transport, where no expensive
man-rating of the Shuttle would be needed.
Ground Demonstration of Orbital and Lunar Factories
Bradley has suggested that factories using extraterrestrial material
could be totally remotely manned. Minsky has suggested just such a lunar
factory. We could begin immediately on a ground based demonstration of
such a factory, simulating the orbital and lunar time delays. The goal
would be a ground based demonstration of feasibility by July 1984 with
flight of the first system to orbit, the moon or both by 1990.
Studies of Teleoperation for Asteroid Mining
The use of computer assisted teleoperation must be developed. This is
necessary for asteroid mining or in other instances where time delays
are long. Perhaps total automation is required, but much flexibility
could be achieved by supervising or commanding operations by teleoperation
techniques. Early exploration could be supplemented by use of teleoperation
with later installation of propulsion devices for return of asteroidal
resources.
-------
My opinion is that tele-operation on the scale envisaged in the
Ayres message will surely be possible in the long run. It is even
possible that it can make major contributions in the next ten years. A
major tele-operation test bed should be developed on Earth - for example,
in mines. In a rational world tele-operation in space would then follow.
In the existing less rational world, immediate experimentation with
tele-operation based on the Shuttle might be in order, since the
organization that should be developing terrestrial tele-operation seem to
be asleep. I think Ayres may be mistaken about the state of
tele-operation in the nuclear industry. It was vigorously developed
during World War II, but the indications are that the capability has been
allowed to decay somewhat. Ayres is right about the necessity of
mechanical engineering development as well as control developments. If
tele-operation were more vigorously developed, research in automatic or
semi-automatic operation would be advanced.
∂11-Feb-81 0100 POURNE at MIT-MC (Jerry E. Pournelle) NASA payoff
Date: 11 FEB 1981 0342-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: NASA payoff
To: JNC at MIT-MC, ENERGY at MIT-MC, SPACE at MIT-MC
Precisely : I can't quarrel with your statement at all. And
not long ago, Larry and I rode from Boston to NYC on a
French-buiolt aircraft flown by Eastern Airlines...
Date: 10 FEB 1981 2109-EST
From: JNC at MIT-MC (J. Noel Chiappa)
One thing a lot of people have skirted the edge of, but never
gotten to exactly, is how much good NASA has done for us economically.
People are always saying "Oh, the technology pays for itslef several
times over",, but consider this: the heavy industry that brought this
country to industrial power is now no longer competitive in many cases.
Cars are a case in point. Light manufacturing (e.g. clothes, shoes)
is also far gone. The ONLY thing left that is REALLy keeping us
afloat economically is advanced technology, which was spurred
(almost entirely) by space and defence. The consumer electronics
followed where the others led. Thus, you might say that were it not
for the space program, we'd be broke today; not ten ot twenty
years down the road, NOW!
Noel
∂11-Feb-81 0127 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 11 February 1981 04:27-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI
Thanks. AYRES = Michael Hyson, who would like an account now
that he has net access. He isn't sure how to ask for one, or
where he ought to ask. H;'es using AYRES account just now.
∂11-Feb-81 0309 REM via SU-TIP
Still the ratio is 1.6 years, not 1.6 pure numbers.
∂11-Feb-81 0730 Darden@SUMEX-AIM PLANS
Date: 11 Feb 1981 0727-PST
From: Darden@SUMEX-AIM
Subject: PLANS
To: JMC@SAIL
NICE TO GET YOUR MESSAGE. A CAR WILL DEFINITELY MAKE THINGS EASIER.
YES, TUESDAY WOULD BE BETTER FOR ME SINCE I DON'T HAVE TO TEACH THAT DAY.
I'LL LOOK FORWARD TO TALKING TO YOU AND GETTING THE BOOK YOU ARE
SENDING. I WILL BE TEACHING ONE CLASS TODAY (WEDNESDAY) AND WILL BE
GONE FROM ABOUT 1:00 TO 4:00 (EST).
BYE, LINDLEY
-------
∂11-Feb-81 1112 FFL
To: JMC, FFL
Please call Les Dugan at 55 495 5669. This is the correct number. I
misunderstood him yesterday and gave you the wrong number.
∂11-Feb-81 1305 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Commencement
Date: 11 Feb 1981 1249-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Commencement
To: CSD-Faculty:
Commencement is Sunday, June 14.
H&S has requested a faculty member from our department serve as a
Marshal for the Main Ceremony, Frost Amphitheater.
"16 H&S faculty are needed to serve as marshals during the main ceremony
in Frost Amphitheater on Sunday morning, June 14. Specifically, ten are
needed to assist undergraduates and six to help with the graduate
procession and seating."
Is there a volunteer?
p.s. Brian Reid volunteered to be our Sigma Xi representative.
-------
∂11-Feb-81 1323 david at UTEXAS-11 Political messages
Date: 11 Feb 1981 at 1455-CST
From: david at UTEXAS-11
Subject: Political messages
To: energy at mit-mc, pourne at mit-mc, space at mit-mc
Using Arpanet to orchestrate a mass mailing for the purpose
of influencing a political decision strikes me as an impropriety.
-------
∂11-Feb-81 1339 MERRITT at USC-ISIB Re: Political messages
Date: 11 Feb 1981 1330-PST
From: MERRITT at USC-ISIB
Subject: Re: Political messages
To: david at UTEXAS-11, energy at MIT-MC, pourne at MIT-MC, space at MIT-MC
In-Reply-To: Your message of 11-Feb-81 1328-PST
I suppose, then, that using the telephone th orchestrate a mass mailing for
the purpose of influencing a telephone company decision is also improper?
-------
Although I support the mass mailing, I fear that David is right that the
ARPAnet shouldn't be used to organize it.
∂11-Feb-81 1415 FFL
To: JMC, FFL
Robert Pugh, a student writing a paper on Modeling Neural Circuitry on
the Brain wants to speak with you about computer design systems. He can
be reached at 9 327 9753 after 6 p.m. I suggested to him that he try
again later here as it might be difficult for you to call after 6.\
∂11-Feb-81 1416 FFL
To: JMC, FFL
Susan Newman of B. Cummings called. She would appreciate your calling
her at 9 854 6020. She may try calling you again.
∂11-Feb-81 1600 Janofsky.Tipi at RADC-Multics Re: Political messages
Date: 11 February 1981 16:52 est
From: Janofsky.Tipi at RADC-Multics
Subject: Re: Political messages
To: david at UTEXAS
cc: energy at MIT-MC, pourne at MIT-MC, space at MIT-MC
In-Reply-To: Msg of 02/11/81 16:26 from david
Using Arpanet to orchestrate a mass mailing for the purpose
of influencing a political decision strikes me as an impropriety.
-------
I personally don't think it is any more improper than the mailing lists
themselves. In my opinion, the importance of the space program
investments to the national economic well being is so great as to
overwhelm any possible minor impropriety. Furthermore, if I am
convinced that the issue is important to the country, I feel justified
in using any forum available to espouse the cause!'
∂11-Feb-81 1621 LEWIS at SRI-KL (Bil Lewis) Re: Political messages
Date: 11 Feb 1981 1406-PST
From: LEWIS at SRI-KL (Bil Lewis)
Subject: Re: Political messages
To: david at UTEXAS-11, energy at MIT-MC, pourne at MIT-MC, space at MIT-MC
cc: LEWIS at SRI-KL
In-Reply-To: Your message of 11-Feb-81 1319-PST
It seems to me that David has a very good point.
Much as we like to think of the net as "ours", it isn't.
Very unpleasent events could transpire should certain
persons get wind of some of the goings on. Private
mailings to known individuals is undoubtly safe as noone
is going to investigate all net-mail, but ENERGY & SPACE
are public lists.
In any case, this is the place for complaints --
For the sake of the ARPA-NET (Peace be upon it!) don't go
crying to your congressman that we've been "misusing" it.
For that matter I'd advise leaving any mention of the net
entirely out of political communications.
As for CC-ing the lettres, how about just sending
the CCs to Merrit/Pourne and not flooding the net any
futher (and not being any more obvious -- discretion &
valor & all)?
regards, Bil (friendly but nervous supporter)
-------
∂11-Feb-81 1635 KLH@MIT-AI Use of ARPAnet, and letter addresses
From: KLH@MIT-AI
Date: 02/11/81 19:01:21
Subject: Use of ARPAnet, and letter addresses
KLH@MIT-AI 02/11/81 19:01:21 Re: Use of ARPAnet, and letter addresses
To: space at MIT-MC, energy at MIT-MC
Whatever you do, do NOT under any circumstances involve the
ARPAnet. I don't have the time to explain -- ask Roger Duffey (Duffey@AI)
to send you a copy of some digests in which the issue is addressed.
The main thing is not whether the ARPAnet will be mis-used or
not -- by DCA's definition it already is -- but whether this "mis-use"
will be publicized, to the resulting detriment of many good things
(pure or otherwise). People who are compiling messages should remove
any trace of message headers; in fact it would be much better to
simply not send any messages at all. Just write your own letter
directly.
Along those lines, could someone take the simple step
of furnishing (I know, again) the names/addresses which should go
on the postcards/letters/mailgrams? The EXACT strings, please.
One other thing: there isn't much time, but it seems to
me that even if everyone on the SPACE list were to write a letter,
the response would still be small compared with that generated if
you ALSO informed a friend or two. McGrath's article is a
logical extension of this, I guess.
I am sure thre is a book smewhere listing all sorts of
ingenious lobbying techniques. I wonder if any of us have it.
∂11-Feb-81 1536 DHM
Mike Lowry tells me that you know where to get a manual called "Short WAITS",
which the usual documentation people don't know about. How can I get a copy?
By the way, I passed the EE qualifying exam. Apparently I did end up in the dis-
cretionary area, but passed thanks to Tom's intervention.
Dave.
∂11-Feb-81 2000 JMC*
1 bedroom apt. 2.5 miles, fireplace, private yard, $350, 326-2444
∂11-Feb-81 2011 MINSKY at MIT-AI (Marvin Minsky)
Date: 11 FEB 1981 2311-EST
From: MINSKY at MIT-AI (Marvin Minsky)
To: MINSKY at MIT-AI, jmc at SU-AI, ayres at USC-ISI, pourne at MIT-MC
From: AYRES at USC-ISIF
Subject: Teloperators by Mike Hyson
As far as I can see, the teleoperator field stopped developing in the
1950's. As a result, the uclear power industry is impoverishes in
this area. Note the recent message about repairs to TMI: there
is no submersible teleoperator, these days. If there were a decent
teleoperator at all, it could wear a rubber suit.
I agree off hand with all the Ayres remarks, although I would change
the weighting on the different research problems.
The horrible thing is that there seems to be nowhere in the
government to press for this, because it is a tool with
multiple applications. I'm planning to talk to Frank Press before long
about it.
∂11-Feb-81 2207 CSD.ULLMAN at SU-SCORE rumble at chez Lieberman
Date: 11 Feb 1981 2201-PST
From: CSD.ULLMAN at SU-SCORE
Subject: rumble at chez Lieberman
To: csl.crc.ejm at SU-SCORE, jmc at SU-AI, rwf at SU-AI, csd.dbrown at SU-SCORE
I am in the process of setting up a meeting between GJL and representatives
of CSD about the LOTS issue. I've been informed by his deanship that I'd
better bring plenty of muscle, so I've picked you guys. The dates so far
discussed are 9am Thurs. 2/19 and 9:30am Mon. 2/23. Please let me know
immediately if you cannot make either date. Also, should anyone else be
invited?
-------
I can make either date, and the invite list seems appropriate. At some point
we may want to request a meeting with yet higher authority. I prefer the
Monday, since it's later and I don't teach on that day.
∂11-Feb-81 2356 RMS at MIT-AI (Richard M. Stallman)
Date: 12 FEB 1981 0257-EST
From: RMS at MIT-AI (Richard M. Stallman)
To: JMC at MIT-AI
Do you know whether they want their letters typed?
Do they get turned off by hand-printed letters or do
they think "look, an ordinary citizen"?
I don't think that matters, since there isn't much special about typing,
and the clerks mainly count them.
∂12-Feb-81 0026 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 12 February 1981 03:26-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: MINSKY at MIT-AI
cc: POURNE at MIT-MC, jmc at SU-AI, ayres at USC-ISI
It may be that we can find somewhere in NSF to play with
waldoes, but my suspicion is that we may as well just go on and
ask for funds for NASA. It's possible we can get them, if we
have a sufficiently clever proposal. one difficulty is that
NASA has no expertise in the field (that I know of) and thus
wouldn't quite know how to run the project. It would have to be
hired out, and to whom?
If we can get our ducks in a row, we may well be able to
do something. At the least we can include the proposal in the
package that the Citizen's Advisory Council submits; that will
go to Congress as well as into administration, and maybe from
there we can catch someone's eye.
Jerry
∂12-Feb-81 0044 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 12 FEB 1981 0345-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Date: 11 Feb 1981 1358-PST
From: John McCarthy <JMC at SU-AI>
Although I support the mass mailing, I fear that David is right that the
ARPAnet shouldn't be used to organize it.
John, I agree, but I do notice that the draft resisters seem
never to get into trouble when they use the bboards to organize
their rallies...
Stay well,
Jerry
∂12-Feb-81 0115 Jerry E. Pournelle <POURNE at MIT-MC> Re: NASA payoff
Date: 12 February 1981 04:12-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Re: NASA payoff
To: Janofsky.Tipi at RADC-MULTICS
cc: POURNE at MIT-MC, JNC at MIT-MC, ENERGY at MIT-MC, SPACE at MIT-MC,
MERRITT at USC-ISIB
For your information, the House Space Committee is:
Chairman Congressman Ronnie Flippo, Democrat of Alabama
Cannon Office Building, Washington DC 20515 202-225-4801
Congressman Bill Nelson (d) Florida (Cape Kennedy)
Cannon 20515
Congressman George Brown, Jr. (D) California
Rayburn 20515 202-225-6161
Congresswoman Marilyn Lloyd Bouquard (D) Tennessee (Oak Ridge)
Rayburn 20515 202-225-3271
(Ms. Bouquard is new to the committe but not to congress)
Congressman Ralph Hall (D) Texas (rural area)
Longworth Offiec Building 20515 202-225-6673
(Newly elected congressman)
Congressman Harold C. Hollenbeck (Ranking Republican) New Jersey
Longworth 1526
202-225-5061
Congressman Raymond McGrath (R) New York - Long Island
Cannon 506 20515 202-225-5516
Congressman Bill Lowry (R) Calif. San Diego
Longworth 1331 20515 202-225-3201
For obvious reasons these people tend to be very friendly to
space programs. They are also one of the best sources of
information, especially if you happen to live in their district.
∂12-Feb-81 0230 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 12 February 1981 05:30-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: MINSKY at MIT-AI
cc: POURNE at MIT-MC, jmc at SU-AI, ayres at USC-ISI
One possibility: if we can get a decent waldo research plan
together, we can put it into several hoppers. My problem is
that I have no feel for the situation. How much money can be
used? (There is an upper limit to what yo can get, but also an
upper limit to what you ought to have for start-up research.)
Who would do the work, and where? And like that.
Given a decent R&D plan -- particularly one that is "seed money"
in tahtt it opens some holes and does some proof of principle
demonstrations, thus attracting private development capital for
further research -- we might can sell it; but I do not know
anything about the plan at the moment.
Help?
JEP
∂12-Feb-81 1312 FFL
To: JMC, FFL
Katherine Park of CBS-TV is doing a show on "The Future, What's Next."
She is calling tto follow up about Teleoperating, mentioned by Minsky
in a recent book. Supposedly we are working on this. She would like
to speak to you about it. Collect 213 473 964l
∂12-Feb-81 1518 FFL
To: JMC, FFL
Ray Reiter called to say, with reference to the philosophy conference
in Vancouver that the funding agency needs two things before it will
act: your Curriculum Vita and an abstract of your talk. These things
should be sent to Guy Parent, Social Sciences and Humanties Research
Council of Canada, 255 Albert St., PO Box l6l0, Ottawa, Canada, K1P6G4.
For it to be meaningful we must indicate the file reference;
443810062
Fran: The abstract is VANCOU.PRE[F80,JMC]. Please print a new copy,
since I've changed it. The C.V. is BIOJMC[JMC,PAT]. It needs pubbing.
∂12-Feb-81 1658 FFL
To: JMC, FFL
Les Dugan called. SAid he has inormation on time comparisons for power
plant construction. 55 495 5669
Miro conv. with Westinghouse contractor in France and Diablo canyon.
50 to 62 months in France, 7 plants in 1980, time from signing of
intent to build to delivery of power on a regular basis.
OHI-1, 62 month from signing to criticality, mandatory 15 months to
regular use on grid.
Dec. 66, application for certificate of public convenience and necessity.
∂12-Feb-81 2103 Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE> meeting re: comprehensive
Date: 12 Feb 1981 1832-PST
From: Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE>
Subject: meeting re: comprehensive
To: csd.schreiber at SU-SCORE, csl.ver.hlo at SU-SCORE, jeb at SU-AI,
csd.gischer at SU-SCORE, csd.clarkson at SU-SCORE, jmc at SU-AI,
csd.yao at SU-SCORE, csd.rwf at SU-SCORE, csl.lantz at SU-SCORE,
als at SU-AI, csl.sso.owicki at SU-SCORE
cc: csd.sonkowsky at SU-SCORE, csd.wiederhold at SU-SCORE
pSince I have only one time left to pass the comprehensive, I was
directed to set up a meeting with my advisor, Professor Wiederhold,
to discuss a "plan of action". He has requested that I see who from
the comp. committee would be also be able to attend. It will be at
11 A.M. on Tuesday (Feb. 17) in his office, I believe.
Thank you.
Paul Sonkowsky
-------
∂12-Feb-81 2106 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Policy on Report Distribution to Foreign Countries
Date: 12 Feb 1981 1008-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Policy on Report Distribution to Foreign Countries
To: CSD-Faculty:
cc: CSD.BScott at SU-SCORE,
CSD-Secretaries:
Excerpt from January 6, 1981, Faculty Meeting Minutes:
3) Report Distribution to Foreign Countries
Professors Knuth and Ullman presented the following proposed policy for
Report Distribution to Foreign Countries.
1. We shall resume sending abstracts to institutions in the USSR
and to all other countries. However, we shall include with abstracts
sent to the USSR, and in the future to any other country the
faculty may designate by majority vote, a statement to the effect
that certain reports may be withheld by the authors because of
abridgment of scientific freedom (or other cause in hypothetical
future cases) believed to take place in that country.
2. When reports are published in the future, the authors will be queried
whether they wish to withhold their report from any country, among
those designated by the faculty in I. If all authors of a report agree,
the report will be flagged in the stockroom, and persons charged
with filling requests will deny requests from countries forbidden by the
authors.
3. For reports published to date, it is up to the authors to take
the initiative to have their reports flagged. A report will not be
withheld unless all authors agree.
Ed McCluskey commented that he thought it was a reasonable compromise.
The suggestion was made that an addition be made to the policy.
Aplicable laws or contract provisions shall take precedence over the above.
John Hennessy moved that the proposed policy be accepted with the addition;
John McCarthy seconded the motion, and it passed unanimously.
-------
∂12-Feb-81 1453 RPG letter
Earlier I asked you to send a letter of reccomendation to
Illinois. Actually I am most interested in CMU and am attempting to
get a 1/2 position in robotics and 1/2 time in CSD. I have the former
offer in hand but need to get CSD. You said you could recommend on the basis
of LISP knowledge, hacking ability etc. Since the 1/2 time position is
in SPICE (Steele, Fahlman) this is appropriate. Please send it to:
Nico Habermann
Computer Science Dept.
Carnegie-Mellon U
Pittsburgh pa
Thanks.
-rpg-
I'll send the letter.
∂12-Feb-81 2145 Purger
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging.
If you don't delete some of your files, the purger will.
Your disk quota is: 4080
Your files occupy 4394
∂12-Feb-81 2130 BYY AN ABSTRACT TO ACL THAT MIGHT AMUSE YOU
SCASS[1,byy]
∂13-Feb-81 0530 Darden@SUMEX-AIM PLANS
Date: 13 Feb 1981 0522-PST
From: Darden@SUMEX-AIM
Subject: PLANS
To: JMC@SAIL
THIS IS A QUICK MESSAGE BEFORE I GET DUMPED OFF. TYMNET SEEMS TO BE
HAVING PROBLEMS SO I WILL TRY TO CALL YOU TODAY AT THE DEPARTMENT.
IF I DON7T REACH YOU, THEN PLEASE CALL ME OVER THE WEEKEND OR WHEN
YOU ARRIVE.
-------
∂13-Feb-81 0731 DISRAEL at BBND Paper request
Date: 13 Feb 1981 1029-EST
Sender: DISRAEL at BBND
Subject: Paper request
From: DISRAEL at BBND
To: JMC at SU-AI
Message-ID: <[BBND]13-Feb-81 10:29:33.DISRAEL>
Could you please send me a copy of "Representation of Recursive Programs
in First-Order Logic". Thank you
David Israel
Bolt Beranek and Newman
50 Moulton St.
Cambridge, Mass.
∂13-Feb-81 1056 FFL
To: JMC, FFL
Have reservations to Washington. Leave Monday, Feb. l6, at 9 a.m.and arrive
Dulles at 4:52. Leave Wednesday at 535 and arrive SF 8:25 p.m. TWA.
Ticket to be delivered here this afternoon.
∂13-Feb-81 1149 100 : frances larson
To: JMC, FFL
Is there a problem with the Ketonen salary? It was arranged last
fall that he would be carried 50% on 526.
∂13-Feb-81 1259 CSD.JEANIE at SU-SCORE Faculty Meeting
Date: 13 Feb 1981 1255-PST
From: CSD.JEANIE at SU-SCORE
Subject: Faculty Meeting
To: tenured-faculty: ;
There will be a meeting of the tenured faculty Feb. 24 at 4:00 pm. The
purpose of this meeting is to consider Gio Wiederhold's case for a tenured
position. It is important to read the letters we have received. Please
come by my office (MJH 214) to look at the folder.
Jeanie
-------
∂13-Feb-81 1741 Waldinger at SRI-KL huh?
Date: 13 Feb 1981 1741-PST
From: Waldinger at SRI-KL
Subject: huh?
To: jmc at SAIL
i never told He you would get him an office. as i understood it,
he didnt NEED an office. all he wanted was to go to classes, seminars,
etc. and to use the library. this i asked you about a month ago,
and inquired regularly, without ever getting a definite answer.
a simple yes or no would have sufficed.
rjw
-------
I am, alas, incapable of a simple yes or no, which is why I don't like
having supplicants referred to me. Sorry for complaining though. I
had a bad cold and was regretting the appointment with He that I couldn't
call off.
∂14-Feb-81 0250 POURNE at MIT-MC (Jerry E. Pournelle) Comments on space strategy
Date: 14 FEB 1981 0551-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Comments on space strategy
To: POURNELLE at MIT-MC, MINSKY at MIT-AI, JMC at SU-AI, LLW at SU-AI
mc:pourne;saving world has a column that will appear in late
may-early june.
It's on strategy when the budget gets cut.
Comments welcome if done soon.
JEP
∂14-Feb-81 0856 BYY ACL abstract
To: JMC
CC: TW
Did you receive my abstract? My outgo message file makes it look like you
didn't. Jon
I didn't receive it in your message, but the message contained all the
clues needed to find and print the file. I'll have comments. - John
∂14-Feb-81 1304 RWW
what does feferman log on as?
∂14-Feb-81 2007 JMC
Hotel Mayflower, 1127 Connecticut Ave. between M and N. 202 347-3000
11am, lunch
2 pm Science Comm. Don Fuqua
4 pm McClure
Ken Davis
∂15-Feb-81 0545 JMC
Dixiana Moon by William Price Fox, Viking
∂15-Feb-81 2351 Jerry E. Pournelle <POURNE at MIT-MC> Possible bad news
Date: 16 February 1981 02:46-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Possible bad news
To: REM at MIT-MC
cc: SPACE at MIT-MC, ENERGY at MIT-MC
The maximum effort last week was designed around precisely the
point REM makes: you must win before the cuts are formally
announced, because afterwared OMB will hold the line not with
the sword only, but with the battleax.
Given that cuts are inevitable, then the space community
has a decision. Do we try to direct the cuts? Do we use this
as an opportunity to get a new charter for NASA?
My view is that space is too important to be left to
government. Thus we want to move toward making profits in
spsace as soon as possible, thus moving us into space whether
government pays or not. This measn government has its historic
role of building roads to the new frontier and protecting/aiding
the early settlers.
This means if you buy the above that technologies are more
important than missions; that a manned industrial facility in
Low Earth Orbit is first priority, to be done by 1988 (inside
the presidential cycle) if possible; and that a manned Moon base
to use lunar resources before 1995-2000 is second priority. It
means that SPS technology is vital, because even if we never
beams down a watt to Earth from space, we will need power in
space for industries there. It means a "space industrial park"
with facilities for private investment and industrial modules to
be attached.
But that means also that if the budget is to be cut, and
technologies and optons preserved, then---then some missins will
have to go. Missions which exploit technology but don't develop
it. It puts lunar polar (search for polar ice) ahead of Halley
or Gailieo, and it measn VOIR just has to wait until we're in
space.
Of course if we go SPS for Earth power (not likely with a sane
fission reactor policy, but will SANE let us be sane?) then we
either step up the Moon mines and build it from lunar materials,
or we build a big fleet of recoverable heavy lift vehicles
(HLV). If we go HLV, then the moon base and science and
everything else is done n third shifts and weekends, and we can
go anywhere we want, damned cheap. If we go the other way, then
I fear we wait to see Jupiter closse up again about 1995. Or later.
This may interest you, Jerry or your student. - John
∂16-Feb-81 1113 REM via SU-TIP
The brochure/booklet on EIES (a computerized conferencing system developed
at NJIT and available thru TELENET) has a nice bibliography on papers about
EIES and computerized-conferencing in general. Have you seen it yet?
If not, would you like to sometime? JMCKENDREE@BBNB sent it to me.
-- REM Information Service
∂16-Feb-81 1324 NEUMANN at SRI-KL VERkshop II
Date: 16 Feb 1981 1306-PST
From: NEUMANN at SRI-KL
Subject: VERkshop II
To: VERKshop: ;
MESSAGE FROM STEVE WALKER TO VERkshoppers:
Sorry for the long delay in commenting on the agenda for VERkshop II.
There have been a number of developments in the past few weeks which
will have a significant impact on the verification community, and I
wanted to have things settle down a bit before commenting.
Many of you are familiar with my efforts over the past few years to
establish a capability for the evaluation of the integrity of computer
systems. On 1 Jan 1981, the Director of NSA was assigned responsibility
for a Computer Security Evaluation Center for the DoD. The Center
will be established as a Program Management Office reporting directly
to the Director and independent of other organizations at NSA. It
will be separately funded with both an R&D and an industry evaluation
role. Details of the staffing and organization of the Center are now
being worked out. I hope more definitive information will be
available by the time the Verkshop is held.
So what does all this mean to the verification community? One major
impact is that in order to be successful in the long term, the Center
is highly dependent upon the widespread availability of verification
tools and techniques. Previous R&D in this field was done for the
general benefit of software development. Now an organization exists
(or soon will) which has a clear charter and a vital interest in the
development of verification techniques. A second major impact which
will follow from the first is a strong interest in formal specification
and verification techniques by the computer manufacturing and system
development communities. This interest will offer significant new
opportunities for technical advances but will also require careful
planning and coordination to avoid chaos. It is my hope that the
Verkshop attendees can be instrumental in organizing an orderly
approach to this situation.
So what should we do at the Verkshop to begin this process? I have a
list of agenda items of my own, and I encourage you to suggest others
that you feel are important. I'm going to ask Bill Wilson of Mitre to
assemble the suggestions and come up with a rough-cut agenda for the
Verkshop. We shouldn't feel constrainded by a formal agenda but we
need a guideline to be sure we cover what is important.
First to get (re)acquainted and come up to speed, we should have
(very) brief status reports on the major activities of the past
year. I would like to discuss the Evaluation Center as part of
this time and get your reactions and concerns about the Center.
Then I propose we have a roundtable discussion of the best way(s)
to capitalize on the opportunities for advancing the state of the
art in verification technology being offered by the Center and
the increased industry interest in verification. What steps
should we take in the next year, five years and ten years? How
can we develop an effective transfer of technology to the
manufacturers and system developers? And perhaps most important,
how can we accomplish all this and protect the essential
research nature of the verification community?
Included among the topics of this discussion should be the status of
Ada and the possibilities of developing effective verification
capabilites focused on Ada. I suggest Ada here because of the
significant efforts underway by many manufacturers to develop Ada
facilities. If a verification capability can be developed around Ada,
then many of the problems of technology transfer and proliferation of
different techniques can be eased. However, if Ada is not a suitable
environment for verification systems, then we should realize this
early and begin developing alternative approaches. Many of you have
participated in the development of Ada or have studied its suitability
for verification. I hope we can have a good discussion of this area
during the Verkshop.
Protocol verification is an area which must be considered during
our meeting. I will leave it to others to suggest the best way
to approach this topic. Marti Branstad and others at NBS are
exploring less formal methods of verification and validation. I
hope we can spend some time discussing these efforts and how they
complement formal procedures both in the short and long term.
There is a major concern which is emphasized by the establishment
of the Evaluation Center. Some people have made the argument
that if verification techniques are used to somehow prove the
integrity of a system, then those same techniques could also be
used by an adversary to detect flaws in the system. This line of
reasoning quickly carries one to the conclusion that any
verification system used on a system which will protect
classified information must itself be protected by being
classified. I personally believe that while there is a concern
here, the negative impact of general application of this approach
would be so great as to undermine all progress in this field. We
must come up with reasonable ways to deal with this concern short
of resorting to such self-defeating means as broad scale
classification. I hope we can spend some time at the Verkshop
discussing this problem and can suggest some potential solutions.
I've probably rambled on enough. I firmly believe that verification
systems are going to play a major role in the future development of
major computer systems and that we have a good opportunity to
influence that role through meetings like the Verkshop. I encourage
each of you to prepare your thoughts on the areas I have mentioned and
any others you feel are important, and come prepared to work hard.
Send your agenda items and position papers if you wish to write any to
Peter and Bill Wilson. The meeting will be April 21-23 at NBS in
Gaithersburg, Md. More details on location and hotel reservations, etc.,
in a little while.
Steve
-=-=-=-=-=-=-=-=-=-=-=-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=
NOTE NET ADDRESSES FOR THOSE OF YOU WITH NET ACCESS:
Walker@ISI
Wilson@RADC-Multics
Neumann@SRI-KL (or in a pinch when KL down, SRI-CSL)
-=-=-=-=-=-=-=-=-=-=-=-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=
NOTE FROM PETER NEUMANN:
Please begin to formulate your contributions. I will distribute all
Verkshop statements received on-line by me prior to 16 April to those
of you with net access, and would hope to be able to hand out in
hard-copy everything received by the 16th on the 21st. The variety of
formats of last year is probably still appropriate, although if you
stick with basics it will help me. (See the July 80 ACM SIGSOFT
Software Engineering Notes [SEN], and the contributions that show
particularly clear evidence of my editing. Please avoid bold face,
underlines, and italics for the NET version, although you may send me
SCRIBE input for hard-copy and publication versions.) You should also
bring auxiliary material such as reports and relevant published papers
to hand out.
Peter
-------
∂17-Feb-81 1023 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Barbara Huberman
Date: 17 Feb 1981 1020-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Barbara Huberman
To: bboard at SU-SCORE, bboard at SU-AI, jmc at SU-AI
Barbara Huberman will be recognized at the Computer Science Conference
in St. Louis as the first woman to receive the PHD in Computer Science.
Raj Reddy lost out as the first man by 3 months.
Does anyone know the whereabout of Barbara? The last note in her folder
says Mitre Corporation in Bedford, Mass. If you know, please send msg
to CSD.Tajnai.
-------
Barbara Huberman is now Prof. Barbara Liskov of M.I.T.
∂17-Feb-81 1116 CSD.JEANIE at SU-SCORE Faculty Meeting
Date: 17 Feb 1981 1106-PST
From: CSD.JEANIE at SU-SCORE
Subject: Faculty Meeting
To: tenured-faculty: ;
A meeting has been scheduled for Tues. March 3 at 2:30. The agenda is
the Dalquist appointment. The meeting will be in MJH 252. The papers
will be in my office (MJH 214) for you to look at.
--jeanie
-------
∂17-Feb-81 1140 TOB task statement for NSF
We have a new task statement for NSF proposal,
a minor modification of the previous one that you
signed. If you give the ok, I will have Fran sign
for you. I am hoping to speed this off in advance
of your return, because each day delay delays receipt
of the money.
Regards,
Tom
I'm back, so either is ok with me.
∂17-Feb-81 1627 CSD.ULLMAN at SU-SCORE Lieberman meeting
Date: 17 Feb 1981 1624-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Lieberman meeting
To: csl.crc.ejm at SU-SCORE, csl.jlh at SU-SCORE, csd.dbrown at SU-SCORE,
rwf at SU-AI, jmc at SU-AI
The meeting is at 9am Thursday 2/19 in Bldg. 10. We've had several
comments that we ought to get our act together beforehand. Unfortunately,
I have to be away or at a CIS meeting all day tomorrow. I suggest that
we get together at my house about 8pm Wednesday. Please let me know
if you plan to attend.
-------
∂17-Feb-81 2008 VRP Wed meeting
To: "@SUN.DIS[P,DOC]" at SU-AI
I sent a message last week advertising a meeting for the morning of
Wednesday the 17th (tomorrow), but apparently it didn't make it, a fact
that I did not realize until now. Since there were two messages, one giving
the time, the other the place, I don't see how both got lost. Let's
try again, this time for nxt week, Wed. 25th, unless it turns out to be a
bad dream and you all know about this weeks meeting.
∂18-Feb-81 0904 FFL
To: JMC, FFL
Please call Prof. Rowen at 326-4916.
∂18-Feb-81 0920 REM via SU-TIP Crunch on microprocessor
To: JMC at SU-AI, TVR at SU-AI, UCL at SU-AI
Recently (severalmonths between other things) I have been
perfecting my adaptive model for minimum use of memory while
not losing significantly compared to IC2, and eventually
surpassing IC2 in ability to maintain adaption even after memory
fills up. It uses 8-bit instead of 18- or 36-bit samples in the
histograms, and is optimized arefully for that very restricted
range of values using integer arithmetic. Finally the past two
nights I've written a LISP program to do everything I've designed,
and it now works. (It took only two nights, using EMACS at 300
baud as the only text editor, no hardcopy yet at all, and MACLISP
with STEP package for running and debugging.)
For a demo of what I have so far, do this:
AL 1,REM
R PLISP
(REMLOAD 'CRU812)
(CRU-DEMO)
When it scales a histogram down due to overflow, or makes a clone of
a histogram to fork a context, it prints the original histogram and
the result after scaling. The trace of those histogams is PRIN1
which is crufty. The trace of context when it forks is in the wrong
order. Otherwise it seems to work.
∂18-Feb-81 1311 FFL
To: JMC, FFL
Please call Christopher Keith at 9 212 623 5100.
∂18-Feb-81 2145 TOB
John
Can you sign the task statement tomorrow morning?
Tom
∂18-Feb-81 2138 JMC
I'm back, so either is ok with me.
∂18-Feb-81 2311 TOB GM
Yes, that will help quite a bit.
∂19-Feb-81 0003 Brian K. Reid <CSL.BKR at SU-SCORE>
Date: 19 Feb 1981 0001-PST
From: Brian K. Reid <CSL.BKR at SU-SCORE>
To: JMC at SU-AI
In-Reply-To: Your message of 18-Feb-81 2112-PST
I am out of personal supply, but there are N dozen manuals in the bookstore.
-------
∂19-Feb-81 0416 REM via SU-TIP Update on new crunch program
To: JMC at SU-AI, RWG at SU-AI, UCL at SU-AI
CRU812.LSP[1,REM] -- Source is 2.1k words, compiled 560 words.
It has full model for syntax of file except no maximum context depth
and no keeping track of how full memory gets. It computes for each
character compressed what fraction of a bit or of bits will result, and
adds them all up. It doesn't actually produce compressed output.
Today's version has prettier output of histograms so you can actually
figure out what they are internally.
Global variable G:DEMO:FILENAME can be changed to test the algorithm
on some file other than the short test file I created. Typically it
gets 2:1 compression (8 to 4 bits per char) when it has seen about 5k
bytes of normal text or about 2k bytes of assembly-listing.
I plan to put limit on context-depth next, then later keeping track
of how many bytes of memory it would take on a microprocessor, then
purge when some specified amount of memory fills up.
∂19-Feb-81 0926 FFL
To: JMC, FFL
Manning of the Math Library called to ask if SUDS is a memo or operating note.
He would like to spool a copy if it is on line and permissible. Somehow he
needs one for the library. Do you know where it is available?
∂19-Feb-81 1230 JMC*
Please call Christopher Keith at 9 212 623 5100.
∂19-Feb-81 1300 JMC*
Rowen 326-4916
∂19-Feb-81 1328 REM via SU-TIP LOTS town meeting, expanding LOTS.
To: JPM, REG
CC: JMC
Curious -- whatever happened to JMC's plan of a couple years ago to
expand the awareness of good computer systems to students and the
general public, by both making general use of computers available to
all students at Stanford and also setting up some public terminals
so the general public could get some exposure to something beyond
the usual microprocessor system or IBM crock even if they couldn't
make general use of Stanford facilities?
(I may have misstated some of the intent of JMC's plan, but I do recall
something about public terminals in libraries and also a video mapper
using S-1 or Knight-terminal or somesuch for getting something cheaper than
Datadisk and better available for all regular users.)
∂19-Feb-81 1735 Raphael at SRI-KL Address Change
Date: 19 Feb 1981 1727-PST
From: Raphael at SRI-KL
Subject: Address Change
To: Distribution:
March 6 will be my last day at SRI, and the last day on which I will have
convenient access to the ARPAnet. Thereafter I will be able to receive
U.S. mail at my new office address:
-------
∂19-Feb-81 1756 TOB adjunct
I have a new publication:
T.O.Binford;
"Inferring Surfaces from Images";
to appear in the A I Journal, 1981.
∂20-Feb-81 2112 REM at MIT-MC (Robert Elton Maas)
Date: 20 FEB 1981 0759-EST
From: REM at MIT-MC (Robert Elton Maas)
To: JMC at SU-AI
Gee, that's interesting. Too bad you didn't supply
a subject line (featherbedding). Well, there's a good reason
to license doctors, when you go to one in an emergency you don't
have time to consult a consumer magazine and your life is at stake
if the doctor you go to isn't a doctor. But all the others, undertakers
etc. are indeed featherbedding. Pretty burial isn't necessary, a
hole in the back yard should do nicely, after it does nicely for
your pet cat that died. Real estate is merely infomation retrieval,
I'd rather have a computer do it. And although quality control is
necessary I too am opposed to formal licensing of computer programmers
except for military applications such as missile guidance.
∂20-Feb-81 2117 CSL.VER.HLO at SU-SCORE Programming Exam
Date: 20 Feb 1981 1201-PST
From: CSL.VER.HLO at SU-SCORE
Subject: Programming Exam
To: jmc at SU-AI
Extract from <csd.files>phdrequirements.doc:
On June 10, 1980, the Faculty approved the following new procedures for
the comprehensive examination programming problem.
``Students in the M.S. and Ph.D. programs (and Ph.D. Minor
students who have passed the written examination) in Computer
Science must prepare a programming project of sufficient
complexity and quality to demonstrate competence in computer
programming.
This project must be supervised and endorsed by a member of
the Computer Science Department faculty and submitted to the
Comprehensive Examination Committee for final approval. The
project must be written at Stanford by the student, working
independently.
The project must exhibit the use of sophisticated algorithms
and data structures and be well documented. Programs will be
judged on the basis of correctness, efficiency, clarity, and
style. The project may be the result of CS293 work, although it
need not be. The project should represent at least 3 units of
work."
See the charge to the comprehensive committee for further discussion.
-------
∂21-Feb-81 0027 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 21 FEB 1981 0328-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Date: 21 Feb 1981 0025-PST
From: John McCarthy <JMC at SU-AI>
I like the word "gneerously" with its overtone of "sneer".
Gee, a neologism that makes sense. I'll have to keep it.
∂21-Feb-81 0033 POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
From: POURNE@MIT-MC (Sent by COMSAT@MIT-MC)
Date: 02/21/81 03:34:32
POURNE@MIT-MC (Sent by COMSAT@MIT-MC) 02/21/81 03:34:32
To: JMC at MIT-MC
[COMSAT: This was a failing QSEND.]
POURNE@MIT-MC 02/21/81 03:34:29
Did you get the packet of conferene materials yet?
Got it today.
∂21-Feb-81 1014 REM via SU-TIP
To: JMC at SU-AI, TVR at SU-AI, UCL at SU-AI
By the way, if you want to see a transcript of a few test runs of CRU812,
both on ordinary (UPI) text and on a very-low-entropy test file (a couple
hundred E's and an end-of-file mark that also must be compressed), with
the in-context probability of each token (character) compressed and trace
of histogram rescalings to stay within 8-bit range and histogram clonings
when a new context is spawned (forked) -- look at CRUD1.TTY[1,REM]
(Lauren, can be FTP'd without password. About 9.1 k words in size.)
No secret of IRSM is given, but the on-the-fly adaptive stochastic model
is gloriously displayed in the detailed trace in that transcript file.
∂21-Feb-81 1446 CLT
I'm sorry, I seem to have insulted you again. I didn't mean to.
∂21-Feb-81 1840 BYY Help
I would like to send a message to everyone on the system telling them that
a campus house is for sail, with details in my file HOUSE. How should I
tell them to get to it? Jon
mail *
House on Stanford campus for sale. Details in HOUSE[1,BYY].
∂22-Feb-81 0628 REM via SU-TIP Best possible compression using adaptive model
To: JMC at SU-AI, TVR at SU-AI, UCL at SU-AI
CC: REM at SU-AI
I've calculated the compression obtained by my adaptive model
(CRU812) under conditions of completely-predictable data,i.e.
zero entropy if you have the trapdoor info that generates it.
When any context has zero entropy, i.e. the next character in that
context is always the same, my model gradually becomes more and mor
confident in that character until it reaches a fixed-cycle in which
each 8-bit (7-bit on pdp-10) character is compressed down to only
0.0067 bit, i.e. 8/.0067 = better than a thousand to one compression.
This is the absolute best that can be done with my method using 8-bit
histograms as would be probably used on a micro. I'm not at all displeased.
That's 150 times as good as a Huffman code that requires at least 1 bit
for any token explicitly compressed (CRU2 and CRU3.FAI used a similar
model except larger than 8-bit samples and non-adaptive, in 1975, but
used Huffman code).
Experiments with CRU812 (see the transcript mentionned earlier) show
that about 48 bits are generated total during the learning phase as
it first starts taking 8 bits to represent the first instance of
the predictable character and shrinks it down quickly until it's only
.0067 bit/char. Then so long as that context is predictable it gives
better than 1000 to 1 compression, and if the end-of-file occurs in
this context it takes just under 8 bits to escape from the pattern
and 8 bits to quote the eof token. Thus for a file of length one
million bytes all the same, it takes 48 + (1,000,000 * .0067) + 16
which is about 6800 bits (845 bytes). There are another 16 bits
at the end to flush the IRSM shift register, which isn't significant,
so the overall file compression ratio is greater than 1000 to 1.
I challange any run-length coding method to beat that and still be
able to compress more reasonable files as well as my method does.
∂22-Feb-81 1223 JK fred abramson
Abramson's 9 mo academic salary for 1980-81 is $25000 according
to the letter he sent me a couple of monthhs go.
∂22-Feb-81 2334 TVR DTN troubles
To: MRC at SU-AI
CC: JMC at SU-AI, REG at SU-AI
The problem is indeed in DISPLY.MID[1,MRC]; your macro DMCHAR clobbers AC
0, which is used quite frequently as a repeat count, hence the infinite
loop. I would like you to either fix it yourself, or incorporate a fix i
am willing to come up with, while still retaining responsibility for that
code. I'll remind you that i do not know MIDAS and at this point in time,
i have neither the time nor desire to change that.
--- Tovar
∂23-Feb-81 0348 Mark Crispin <Admin.MRC at SU-SCORE> Re: DTN troubles
Date: 23 Feb 1981 0346-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Postal-Address: 12155 Edgecliff Place; Los Altos Hills, CA 94022
Stanford-Phone: (415) 497-1407
Subject: Re: DTN troubles
To: TVR at SU-AI, JMC at SU-AI
In-Reply-To: Your message of 22-Feb-81 2334-PST
Maybe DTN is fixed now.
-------
∂23-Feb-81 0839 FFL
To: JMC, FFL
Judy phoned a reminder of the Information International Board meeting on
the 25th.
∂23-Feb-81 1004 FFL
To: JMC, FFL
Jeff Ullman stopped in and wishes to speak with you this afternoon about
the meeting tomorrow.
∂23-Feb-81 1051 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, February 24
SPEAKER: Richard Schwartz and P.M. Melliar-Smith, SRI International
TITLE: Temporal Logic Specification of Distributed Systems
ABSTRACT:
This talk presents our ongoing research applying temporal logic
specification and verification methods to distributed systems -- in
particular, to communication network protocols. We specify a protocol by
stating a Service Specification, describing only externally visible
behavior, and an Implementation Specification, describing the communication
interfaces between each node in the distributed system. Our approach states
the required properties of the protocol without any assumption of atomicity
or the use of sequences or history variables. In the talk we will discuss
the manner in which temporal logic is used for defining each level of the
specification and for verifying consistency between levels.
∂23-Feb-81 1107 FFL
To: JMC, FFL
Did you ever send your Vita and abstract for yur Vancouver talk as requested
by Reiter?
∂23-Feb-81 1106 FFL
To: JMC, FFL
Prof. Wagoner of Physics Dept. would like to have the title for your
seminar talk on April 5, please. His phone is 7-4561.
∂23-Feb-81 1104 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
Revival of the Knowledge and Action Seminar
Wednesday, March 4, 4:15 p.m.
Speaker: Professor Ray Reiter, University of British Columbia
on: Non-monotonic Reasoning
PLEASE NOTE the change of date.
NOTE ALSO that Professor Reiter is the speaker at the Computer Science Colloquium
on Tuesday, March 3. His talk is titled
"Towards a Logical Reconstruction of Database Theory"
-------------------
The following week we return to our regular time, Thursday, March 12 at 4:15 p.m.,
when Kurt Konolige will speak on "A First Order Formalization of Knowledge
and Action for a Multiagent Planning System."
∂23-Feb-81 1143 JMC*
Call Dorothy Bender.
∂23-Feb-81 1143 JMC*
Doug Scott
∂23-Feb-81 1314 CSD.JEANIE at SU-SCORE Meeting 2/24
Date: 23 Feb 1981 1308-PST
From: CSD.JEANIE at SU-SCORE
Subject: Meeting 2/24
To: tenured-faculty: ;
The meeting scheduled for Tues. 2/24 has been cancelled. The meeting has
been rescheduled for Tues. March 3 at 2:30. If you have not had a chance
to look over the papers, please do so.
Jeanie
-------
∂23-Feb-81 1408 RWW misc
1. If there is a data media around i'd like to borrow it until my imlac is fixed
2. I am about to code circumscription - any last suggestions?
richard
∂23-Feb-81 1535 PP talk by Peter Pepper at SRI
To: "@SEM.DIS[SEM,VER]" at SU-AI
CC: PP at SU-AI
Dr. Peter Pepper is visiting Stanford from the Institut fuer Informatik
of the Technical University Munich. He will talk about the
Munich transformational methodology for program construction.
That's Wednesday, Feb. 25, 10:45 am, in the small (map) CSL conference room
at SRI, in building 30 (on Ravenswood Avenue)
Program Development by Formal Transformation
The Munich Approach
We present a transformation method for program development.
This method begins with high-level specifications and proceeds
through several intermediate stages until an efficient low-level
program is obtained. The transitions between subsequent versions
are achieved by applying formally verified transformation rules.
The Munich approach is characterized by the following properties:
* The choice of which rule to apply next is left to
the user; i.e., no strategy or automated deduction
is built into the system. Hence, it may be viewed
as a small tool box.
* The development process takes place within the
framework of a "wide-spectrum language," which
comprises specification constructs as well as
low-level ("machine") constructs.
* Particular emphasis is given to formal notions
of correctness, including the semantic fundamentals
of both the language and the methodology.
∂24-Feb-81 0103 LLW Caught, White-Hatted!
To: JMC at SU-AI
CC: LLW at SU-AI, minsky at MIT-AI, pourne at MIT-MC
Dear John:
I was delighted to catch you, wearing a white hat, jumping up and down on
a U.S. Senator in his sanctus sanctorum last Tuesday afternoon on behalf
of the civilian nuclear power program. Who (least of all myself) would have
ever expected a respectable, well-mannered Stanford professor to be engaged
in such morally dubious activity as attempting to influence legislation
behind closed doors!
[I had infiltrated this session in Senator McClure's office in order to
drag Edward on to his appointment with Senator Warner, only to find that
the session was just breaking up and that you were still earnestly
lecturing your captive audience. Not wanting to cut in to the extent of a
single syllable, I quietly moved out with Edward, although I strongly
wanted to congratulate you on the spot on your obviously heightened
political consciousness.]
Congratulations, and welcome to the Washington Lobbyists Guild! May the
programs you advocate prosper mightily! Perhaps you'll be generous enough
to give Marvin and Jerry a few pointers on how to bring showers of Federal
loot down onto the heads of programs which you favor.
Lowell
∂24-Feb-81 0640 Darden@SUMEX-AIM bread and butter
Date: 24 Feb 1981 0637-PST
From: Darden@SUMEX-AIM
Subject: bread and butter
To: JMC@SAIL
Hi, John. It was a pleasure seeing you and resuming our, always
interesting ,conversations. Thanks for the dinner and book. Do
let me know when you will be in D.C. again.
Any advice you can give me as to how to make learning LISP easy
and fun will be appreciated.
I plan to spend this morning thinking some more about how to use
analogies to construct new theories. I am having an electronic
seminar on analogy with some of the HPP folks. Good luck with
unresolving ambiguities.
Bye, Lindley.
-------
∂24-Feb-81 0658 BYY Moving
Yes, I'm afraid we have decided to go back to Wisconsin. My family
just hasn't been very happy here. Jon
∂24-Feb-81 1300 JMC*
river ranch
∂24-Feb-81 1505 Hans Moravec at CMU-10A (R110HM60) Nuclear pumped X ray zapsat
Date: 24 February 1981 1801-EST (Tuesday)
From: Hans Moravec at CMU-10A (R110HM60)
To: energy at MIT-MC, space at MIT-MC
Subject: Nuclear pumped X ray zapsat
In the AWST article the satellite is small enough that "several"
could be launched in a shuttle flight. Or they could be put into
space in time of crisis on ICBM warheads. Each satellite has,
in the artist's depiction 32 rods of a "high density material",
with each rod being an individually steerable laser. When
the small nuclear device in the body of the satellite detonates
this porcipine satellite shoots 32 bursts in as many directions,
each one on the order of 10↑15 watts and 10↑-9 seconds, enough
to destroy targets by absorbed energy shockwave (and you can't
reflect X rays very well). Of course, each satellite can be used
only once. The rods look pretty thin in the picture, but presumably
their aperture is large enough to give adequate collimation at 10↑3
km range. The jubilation is because the concept was successfully
demonstarted in an underground test within the last few months.
I wish the people who read this list who were involved
with this could tell us more.
∂24-Feb-81 1725 Susan Hill <CSD.HILL at SU-SCORE> [ Solomon.Datanet at MIT-Multics: Re: Computer Facilities Charges]
Date: 24 Feb 1981 0853-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
Subject: [ Solomon.Datanet at MIT-Multics: Re: Computer Facilities Charges]
To: jmc at SU-AI
I was under the impression that Solomon was to pay for his facilities usage
himself. I sent him a message indicating this and the charges to date. This
is the response. He has had 2 aliquots on SAIL since October l980. Did you
authorize this account? Do you want to pay for it? Susan
---------------
Mail-from: ARPANET site MIT-MULTICS rcvd at 23-Feb-81 2015-PST
Date: 23 February 1981 23:15 est
From: Solomon.Datanet at MIT-Multics
Subject: Re: Computer Facilities Charges
To: Susan Hill <CSD.HILL at SU-SCORE>
In-Reply-To: Message of 23 February 1981 19:29 est from Susan Hill
I presume you are talking about SAIL. We never worked out a schedule of
charges, so i assumed that my account was as a guest. If I am expected
to pay, we would have to review the use of the machine. I am not visitng
anybody. i am in Cambridge. The account was authorized by john McCarthy.
Just what am I to be charged ofr? CPU, storage, etc.?
---------------
-------
How much does Solomon use amount to?
∂24-Feb-81 2318 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, March 3
SPEAKER: Vaughan R. Pratt, MIT, on sabbatical at Stanford University
TITLE: Actions = Functions for Distributed Computing
ABSTRACT:
All existing models of parallel computation are either fatally flawed or
conceptually uneconomical. We introduce a new model of parallelism, the
action. Actions extend functions, and action composition extends function
composition, in a natural way. The model survives the Brock-Ackerman test.
Despite its conceptual simplicity, a wide range of phenomena arising in
parallel computation may be modelled, including race conditions, and both
buffered and unbuffered (unqueued) input, permitting faithful modelling of
silicon technology, operating system processes, networks, and other sources
of distributed computation.
∂24-Feb-81 2358 Jerry E. Pournelle <POURNE at MIT-MC> Nuclear pumped X ray zapsat
Date: 25 February 1981 02:54-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Nuclear pumped X ray zapsat
To: Hans Moravec at CMU-10A
cc: ENERGY at MIT-MC, SPACE at MIT-MC
Come on, Hans, you know that no one is really building zapsts,
and no one is really interested in destabilizing the arms race
by engaging in research on stuff like that.
∂25-Feb-81 1343 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Betty Scott
Date: 25 Feb 1981 0928-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Betty Scott
To: CSD-Faculty:
cc: Admin.Gorin at SU-SCORE
We regret to inform you that Betty Scott's mother died last night.
The Department will send flowers to the funeral.
Many thought it would also be nice to send flowers to Betty. If you
would like to contribute to this, please see me today.
Carolyn
-------
∂25-Feb-81 1426 Susan Hill <CSD.HILL at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
Date: 25 Feb 1981 0823-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
Subject: [John McCarthy <JMC at SU-AI>:]
To: jmc at SU-AI
He uses 2 aliquots - 240K - $48.88/mo.
---------------
Mail-from: ARPANET site SU-AI rcvd at 24-Feb-81 1728-PST
Date: 24 Feb 1981 1727-PST
From: John McCarthy <JMC at SU-AI>
To: csd.hill at SU-SCORE
How much does Solomon use amount to?
---------------
-------
∂25-Feb-81 1447 TW
To: DPB at SU-AI, DBL at SU-AI, JMC at SU-AI,
buchanan at SUMEX-AIM, genesereth at SUMEX-AIM
I am thinking of showing the film on AI done by Nova a couple of years
ago (The Mind Machines) to my CS75 class this spring. It can be rented
for $65 or purchased for $700. If there is enough interest in showing
it for other classes (e.g. 101, 224) it might be worth getting it. I
estimate its viable lifetime on the order of 5 years. For those who
don't remember it, it includes interviews with Minsky, Arthur Clarke,
Schank, me (with SHRDLU demo), Pople and Myers, Weizenbaum, Dreyfus,
and a nuumber of others. It is a reasonably well done and balanced
presentation.
If you are interested in encouraging purchase, let Denny know.
--terry
When the Nova film was being made, Ed Feigenbaum advocated
boycotting the activity, and I didn't agree. Now I think he was right,
and I just refused co-operation with a film-maker who was hoping to make a
filem on AI for Nova. I was not especially put out by the fact that they
used the footage they took of me only in a trivial way, because I can
count and measure the ratio between what they took and what they could
include in the final film. The letter of thanks was what set me off.
Paula Apsell, the producer wrote that while they didn't use much, it was
what she needed for "a point she wanted to make". Who are we to serve as
props for points Paula Apsell wants to make? The matter was rubbed in by
the example of database i~which it was retrieved that Nova is the only
science documentary. If there is to be one, then the scientific community
should co-operate only as part of an agreement on the content. Anyway I
don't advocate that the Department buy the film, and it's a bit obsequious
of you Terry to rent it. Isn't your participation enough to get you a
free rental.
Also I don't understand the point about the film being balanced -
balanced between what and what?
∂25-Feb-81 1452 FFL
To: JMC, FFL
How long do you want the loan period to be on the paper put on reserve
in the math library for CS226? Richard now has it on 24 hour loan but
wonders if you need a shorter period so that everyone may read it.
24 hours is ok.
∂25-Feb-81 1636 ROY on TTY3 1636
John, the Imlac is picking a bit. I'm trying to determine at what point it is
doing it.
∂25-Feb-81 1802 RWW circumscription
I have not yet done it. I need to restore some files that were deleted
(purged) while I was away. This will take some time as There are quite
a few and the total number of available disk blocks on waits is not enough.
How soon do you need it?
richard
∂26-Feb-81 0814 Feigenbaum@SUMEX-AIM (Response to message)
Date: 26 Feb 1981 0804-PST
From: Feigenbaum@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI, DPB@SU-AI, DBL@SU-AI, buchanan@SUMEX-AIM
In response to the message sent 25 Feb 1981 2110-PST from JMC at SU-AI
I agree completely with John on all points. I had not heard about the
rental or purchase of the NOVA program, but think either a waste of
money.
Ed
-------
∂26-Feb-81 0919 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) CS226
Date: 26 Feb 1981 0918-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: CS226
To: JMC at SU-AI
The Dept. Sec. from Philosophy 326 called re CS226 for 1981/82.
I told her it was given alternate years and would not be given
again until 1982/83.
Is that correct?
Carolyn
-------
CS226 is indeed given in alternate years.
∂26-Feb-81 1051 FFL
To: JMC, FFL
Dorothy Bender called to remind you of your appointment at 1 p.m., or
slightly after, at the Faculty Club.
∂26-Feb-81 1553 Purger
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging.
If you don't delete some of your files, the purger will.
Your disk quota is: 4080
Your files occupy 4439
∂26-Feb-81 1608 TW
John,
I have no opinion on the appropriateness of Paula Apsell's correspondence
with you, but I do not agree that the subjects of a documentary film
should have a controlling say over what goes into it, whether they be
scientists, politicians, bureaucrats, businessmen, political activists or
whatever. I understood that I was agreeing to be a subject of filming,
and my agreement was based on discussions with Paula in which I judged her
to be of high intelligence and good faith (both of which I still feel).
As to my having to rent the film, I called WGBH to get in touch with
Paula, assuming as you did that I would be able to get it for free.
She no longer works for NOVA, and the rights to the film have
been sold to Time-Life films. I was referred to an office in New
Jersey where T-L films are distributed. I suppose I could write
them suggesting that it would be courteous to give it to me free,
but in general when courtesy and profits are at odds, the free
enterprise system provides clear priorities, and I suspect they
did not include in their contract with NOVA a provision that anyone appearing
in a film has free rights to it.
Finally, my comment on "balance" was based on the fact that the film
came across as neither "Gee whiz, look at all these wonderful things
scientists are creating to improve your lives" nor as "Computers
are going to dehumanize society and need to be stopped". It talked
about both the promise and the problems.
I don't really have strong feelings about buying it -- I am happy
to do my own thing for my own course indefinitely, and I'm sure
that within a few years it will be obsolete when someone does
a more up-to-date film. If you object strongly, it's fine with
me to just drop the issue.
--t
As for buying the film, I have no more to offer. My original impression
of Paula Apsell was like yours, but it changed. As for balance, I don't
think that a discussion of a technology should be balanced between the
view that regards technology as beneficial and that which regards
technology as harmful. I think there is more of a societal consensus than
that.
As for this issue of control, consider the fact that Nova has a monopoly
as a regularly appearing and government supported documentary on science.
Its monopoly character gives it greater responsibilities than if it were
in a written medium where there is more free competition. If Paula Apsell
were writing for (say) the New York Times, then I would not have been
offended by her remark "it helped make a point I wanted to make".
Finally, anyone is free to co-operate or not with a given attempt to make
films, and my opinion is that AI can get better treatment by bargaining
with information monopolies like Nova than by unconditional
co-operation.
∂27-Feb-81 0021 Jerry E. Pournelle <POURNE at MIT-MC> Caught, White-Hatted!
Date: 27 February 1981 03:18-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Caught, White-Hatted!
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI, JMC at SU-AI
all tips gratefully appreciated. White Hats work, do they?
∂27-Feb-81 0034 LLW White-Hatted Money Rustlers
To: pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI, minsky at MIT-AI
∂27-Feb-81 0021 Jerry E. Pournelle <POURNE at MIT-MC> Caught, White-Hatted!
Date: 27 February 1981 03:18-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Caught, White-Hatted!
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI, JMC at SU-AI
all tips gratefully appreciated. White Hats work, do they?
[Almost anything works, if you're advocating nuclear power or national
defense in that town these days. A white cowboy hat is said to be
especially effective if you happen to casually run into The Man In The
White House--his Commerce Secretary is said to have got his job by being
out roping cattle for rodeo practice when the White House called to
inquire as to his availability, as you may recall. A bearded university
professor eloquently advocating nuclear power is at least as rare as a
white-hatted cowboy on the Banks of the Potomac, however--and novelty
counts for a lot in the sales game.]
∂27-Feb-81 1105 FFL
To: JMC, FFL
E. Bloom's secretary called to remind you that you are to meet him at
11.30 today in his office and then go to Faculty Club for lunch.
∂27-Feb-81 1212 RPG Appointment
To: REG, JMC
As I understand it, I am scheduled to work as a system programmer
through March. This was predicated on 2 facts: 1) I would have had and
accepted an offer from wherever I would end up by then and 2) that place
would not be local.
Well, both are turning out to be near enough false that I would like to change
my plans, given your approval. First, though I have received many assurances
of offers in the works, I only have half of one in hand. Second, I am more
strongly leaning towards staying in the vicinity of Stanford - possibly
part time at Stanford.
So, I would like to extend my system programmer's appointment until the end of
June if possible.
In addition, I am currently highly receptive to suggestions concerning the
future of Lisp computing at Stanford if you would care to make any.
-rpg-
∂27-Feb-81 1225 RPG Proposal
My promised proposal for a grand and glorious Lisp future is going
quite slowly, it being behind the list: fix Lisp bugs, fix Lisp programmers'
bugs, install the E/MacLisp interface, write IJCAI and AIJ papers, consult
part time nights for RWW at HP, look for a job, get my thesis out as a report.
Since time grows short for me to make a decision, I think we should forget about
the large proposal and just start talking about the possibilities. Since you,
like Reagan, prefer shortened versions, I can probably tell you in a few
paragraphs the basic ideas, and then we can go around and around on them.
I am not adverse to having a part time appointment at Stanford and at, say,
LLL or with RWW, but I would like to have at least the following:
1. some more prestigious title than `system programmer'
2. the freedom to pursue some of my own ideas, even to the
extent of making proposals to NSF, ARPA, etc.
In any case, if there is any interest in my remaining at Stanford, I am
receptive to such suggestions at the moment.
-rpg-
A few paragraphs will suffice most likely. The title of research associate
is probably quite possible. While we can include you in a proposal that
will contain some of your ideas, the privilege of making proposals
independently to government agencies is available only to titles with
the word "professor", i.e. regular faculty or adjunct professor.
∂27-Feb-81 1540 CSD.AUBERY at SU-SCORE cs226
Date: 27 Feb 1981 1535-PST
From: CSD.AUBERY at SU-SCORE
Subject: cs226
To: jmc at SU-AI
cc: csd.mccarthy at SU-SCORE
I would like to change my grading option for cs226 to pass/no credit
if that is mutually agreeable.
Eric Aubery
-------
∂27-Feb-81 1540 CSD.AUBERY at SU-SCORE cs226
Date: 27 Feb 1981 1535-PST
From: CSD.AUBERY at SU-SCORE
Subject: cs226
To: jmc at SU-AI
cc: csd.mccarthy at SU-SCORE
I would like to change my grading option for cs226 to pass/no credit
if that is mutually agreeable.
Eric Aubery
-------
∂27-Feb-81 1548 TOB
To: JMC at SU-AI, csd.floyd at SU-SCORE, csl.lantz at SU-SCORE,
csd.golub at SU-SCORE, TW at SU-AI
Richard Vistnes has a project in mind for the programming
part of the comprehensives. Do you agree to his programming
in assembly language, if that is appropriate for what he
chooses to do?
It is a real-time collision avoidance algorith for the arms.
ok for Vistnes to use machine language provided writeup is good.
∂27-Feb-81 1726 CSD.GOLUB at SU-SCORE
Date: 27 Feb 1981 1725-PST
From: CSD.GOLUB at SU-SCORE
To: TOB at SU-AI, JMC at SU-AI, csl.lantz at SU-SCORE, TW at SU-AI
cc: CSD.GOLUB at SU-SCORE
In-Reply-To: Your message of 27-Feb-81 1548-PST
Sounds OK to me.GENE
-------
∂27-Feb-81 1735 Keith A. Lantz <CSL.LANTZ at SU-SCORE>
Date: 27 Feb 1981 1732-PST
From: Keith A. Lantz <CSL.LANTZ at SU-SCORE>
To: CSD.GOLUB at SU-SCORE, TOB at SU-AI, JMC at SU-AI, TW at SU-AI
cc: CSL.LANTZ at SU-SCORE
In-Reply-To: Your message of 27-Feb-81 1725-PST
It better be good. He should probably talk to MRC; whose assembly code
is some of the best possible.
Keith
-------
∂28-Feb-81 2217 TOB
To: "@COMPUT.DIS[DIS,TOB]" at SU-AI
The situation at SAIL has deteriorated drastically in the past two months.
There is now a wait queue at 9 am, and editing is very much slowed down most
of the day. Research is seriously degraded, it appears to me.
I propose that a committee be formed for the computer facility.
As a suggestion, I propose that alternative computing facilities
be sought for student computing to return SAIL and SCORE for research
usage. As a guess, since student computing is only newly on those
machines, most student computing is not tied to those systems. Research
is much more closely tied to those machines.
As a trial balloon,
I propose that we seek funding for a solution along these lines.
That there is an infinite sink of compute use which does not depend on
particular system. That a reasonable departmental system be sought,
that it will be overloaded. That machines which were research machines
be freed for research.
I think your perceptions about editing being slowed more than (say) ten
percent ar are mistaken, but you are not alone in having such perceptions.
I propose that an actual test be made of how fast people edit under
different load conditions. We have an excellent chance of getting a
2080 (5 times faster) in about 2 years (or an S-1, much faster than that),
but I think we'll be making a mistake to go after additional computer
facilities now. We will dissipate our resources for purchasing, housing
and operating computer facilities.
∂01-Mar-81 0004 TOB
To: JMC at SU-AI, csd.ullman at SU-SCORE
Message to Feigenbaum
Ed:
We do need lab space here at MJH, so we would need to find office space.
Members of my group are really in bad shape. They have been exceptionally
good about it. I am considering setting up an outpost with some offices
somewhere. We must be as close as possible to MJH; Welsh Road seems
reasonable for us. I don't know whether there is an incompatability or
not, and we should be prepared that there is an incompatibility. There
has been more and more interaction between the vision group and HPP at the
level of students and for me. I have enjoyed that, and it would hurt us
to be isolated from the rest of the department again, as at Powers Lab. I
will fight hard to maintain that contact for me and for the students.
There is an advantage in having a critical mass to avoid isolation in any
outpost. Also, if there are any overhead items, they might be shared.
I am gathering information now, to see what realities are and what would
be workable. If there is a reasonable alternative, I would act quickly.
But as I said, that would mean finding space for a small group of
unfortunate exiles.
I thought that as far as computing, we would have character-only terminals
there. Serious computing for vision requiring displays or robotics
laboratory work would be done at laboratory space here. That is
definitely a bad situation.
Since you raised that possibility of renting outside space a year ago at a
department meeting, I have not done any homework to see how much it costs
and what we could afford. I thought at that time that you said it could
be provided by the department, so please set me straight. If so, under
what conditions? I.e., does the University overhead pay for space? And
if so, what flexibility do we have in providing for student categories not
currently given desk assignments in CSD, because of limited space there.
I am just getting good projections of finances, and I don't know what the
financial situation is, but I think that we are tight for the short term.
Until I know that we are clear, I don't see how we can pay much for space.
A little later, I think that we will have more freedom, but with the new
administration, I feel uncertain.
For four years or more, a condition on any proposal has been that it not
involve additional space. We have eight requests to accomodate visitors,
some of whom are attractive researcher, which I can do nothing about. I
assume that your group has similar problems, so in this note I want to set
out our needs and possible actions. I plan to do something for a short
term unsatisfactory expedient; I seek wisdom in how to go ahead. I look
for a long term solution involving the department and will work toward
that. Robotics has a substantial number of people who are in advanced
stages of PhD programs but who do not have desks, because they are not in
the categories assigned space. I want room to accomodate them, to expand,
to accomodate our new projects and opportunities for growth and to take
advantage of qualified, promising visitors.
In summary, I am gathering information. Tom
∂01-Mar-81 0033 POURNE@MIT-MC Saving Civilization on the installment plan
From: POURNE@MIT-MC
Date: 03/01/81 03:32:48
Subject: Saving Civilization on the installment plan
POURNE@MIT-MC 03/01/81 03:32:48 Re: Saving Civilization on the installment plan
To: PDL at MIT-MC, JMC at MIT-MC, LEECH at MIT-MC, MINSKY at MIT-MC
To: LLW at MIT-MC, POURNELLE at MIT-MC
1. You will all be pleased to know that we are now on the
Request List; that is, not only are our space plans being read
at the Special Assistant to President level, but I have letters
requesting them along with very sensible questions.
2. Congressman Newt Gingrich (R, Georgia) is sufficiently
impressed that he has delegated an aide to monitor our
activites, and told me that he (Gingrich) has a dozen (the
number he used was 20!) Republican Congresspeople who will sign
our final report and he will carry it to Stockman. I'm still a
little dazed at this unsolicited proposal, but I'll take it.
3. The Citizen's Advisory Council is thus off the ground, now
only limited by the energies and abilities of your chairman; I'm
dancing as fast as I can...
4. I have asked LEECH, a Cal Tech student, to do a survey of
computer resources possibly available to L-5 and the Citizen
Council. Please cooperate with him. I have always had the
suspicion that we could get camera ready copy, if not for the
L-5 News, then at least for a newsletter, through net resources;
and that a good bit of the eiditing of the newsletter could be
done through this means. I may be wrong? Anyway, I am now
involved in (1) getting the final draft of the Council Report
together in time to do some good; (2) trying to coordinate a L-5
Membership drive; and (3) trying to keep enough troops working
with the trimble's to keep their Star Trek mail machine grinding
along. Thus I feel spread as thin as boarding house butter, and
any help from any of you will be apprecaited. I remember when I
wrote books before I took up saving the space program for a
living. Books are more lucrative...
Jerry
∂01-Mar-81 0155 TOB computing overload
To: EAF at SU-AI, JMC at SU-AI
Ed:
Thanks for your message. I understood your concern at the ARPA
meeting and at the SIGLUNCH meeting. By the way, I liked your
presentation very much. I am not coming as an adversary, nor putting
the responsibility to do something at your feet.
I realize I am preaching to the pope.
Is it worthwhile starting a committee on computing resources. If so,
I will serve. I need to debug my thinking about this problem.
I have appreciated several of your statements which have strongly supported
finding solutions including reserving resources for research.
I was surprised that the faculty approved the policy of excluding
course work from SAIL and SCORE. I did not support it, but I was
happy about the emphatic action.
I proposed that MS students not have accounts. That was recorded and voted
on "as no students have accounts". One objection is the mail system, but
MS students could have mail-only accounts.
For what do students need computing resources if they are not supported
and can't do course computing? Private non-supported research, but those
could be supported by the department on the basis of proposals.
I don't know the reason for increased
use of SAIL. We need that information, but I don't see reversing the
trend even if we have the information.
I think the edict is being interpreted as forbidding the use of these
machines for the large courses, not for others. I understood the
ban to be for all but a few courses with specialized needs such as
FOL and the robotics laboratory with explicity approval.
How hard can we be in enforcing such
a ruling, for how long, and how can we know? We would have to reiterate the policy
frequently and strongly for it remain alive. It's a situation which favors
people who don't obey.
If the problem is overselling the machine, then perhaps the appropriate
reseearch groups should buy out the machine at higher costs. Who are
the new users? If it is research use which is overloading these machines,
then what is the solution? Individual groups getting their own machines?
You and John certainly have taken the lead in providing facilities.
I am available, not anxious, to take part in seeking a facility for
student use. I am not sure that is good judgment for me to participate
in seeking essentially teaching facilities.
I would like to see those with heavy involvement in teaching organize an
NSF proposal with parallel contributions, equipment or discounts from
computer forum members (we might get lots of small computers).
Perhaps I might better work toward better facilities for research.
I am definitely available to work for equipment from ARPA or other
source. I did participate in the proposal which Ullman initiated;
I am making a plan for our group.
I am available for some work for the department.
But, I am sorry that I am not coming forward with a plan of action.
Personal energy is the most important resource required now.
I want to find out what seems feasible now, especially to you and
John, who are the wise heads about facilities.
I assume that you understood the content of the suggestion and that
your not understanding was about higher issues.
In case I was not clear, my suggestion
was that student work be segregated, that it does not require the
SAIL and SCORE environment, which research groups require.
Many systems would do.
Student use of computation is realistically unsatisfiable now;
such a system will saturate and thus should be segregated or it
will saturate all other systems.
The problem is, of course, the energy to get that system. I did not
see any energy to do anything at the last faculty meeting.
Of course, student course computing is not permitted on the machines.
This might be like use of newts in warfare in Capek's "War with the Newts";
since there was an agreement not to use newts in naval warfare, clearly
no nation was using newts in their navies, and nothing could be done to
control them.
John McCarthy thinks that we have to count on some non-research
student computing on research facilities.
I agree that there has to be some provision for that; I am not sure on the
same facilities because I am not sure we can limit that usage to an acceptable
fraction of research usage. I decided to take a position to exclude such use
from research machines to force the problem.
As long as research machines are used to solve
the problem, I don't think it will be solved. Only you and John and Jeff have
taken that initiative and for research purposes.
I think that non-research student computing will happen as long
as we have student research accounts. Also, research is a major part
of graduate education in CS. I fear that for the time being
unless there is a clear segregation, a separate but unequal two level system,
research will be squeezed out, because overloaded systems become
editors and microcomputors. I think that there is no other way to
count on limiting "some non-research computing" to "some". Otherwise it
will be most not some. Unless computers sit idle, researchers
sit idle. Unless we do so, we will be irresponsible about research.
Tom
j28-Feb-81 2234 Feigenbaum@SUMEX-AIM heavy computer useage
Date: 28 Feb 1981 2229-PST
From: Feigenbaum@SUMEX-AIM
Subject: heavy computer useage
To: tob@SAIL
cc: reg@SAIL, csd.dbrown@SAIL
Tom, I am as concerned (extremely) as you are at the incredible loads on
our machines (SAIL,SCORE, SUMEX). I honestly don't understand your message.
I had assumed that student work was kicked off the research machines
of the department, more precisely that the student work on the machines
was related to their reseach roles not their coursework. Has that not come
to pass?
In any event, the department just needs MORE computing (and More and More
and More; that's the kind of place we are). There is a massive leadership
job needed to get more. I did this re SCORE; and continue to do it
re SUMEX. Frankly I'm getting tired of that sort of thing. Jeff took such
a role re the recent VAX proposal; John re the original SAIL machines.
Who next????
I am in total sympathy with your frustration.
Ed
-------
∂01-Mar-81 0515 Admin.Kanef at SU-SCORE (Bob Kanefsky) advice taker in SPHERE
Date: 1 Mar 1981 0513-PST
From: Admin.Kanef at SU-SCORE (Bob Kanefsky)
Subject: advice taker in SPHERE
To: JMC at SU-AI
Here's the story so far:
I talked to Steve Gadol the SPHERE project manager, and he says that
in principle SPHERE should be able to EVAL the final SHOULD assertion
as TRUE, and probably even find all of the SHOULDs that are true in
S0. However, as you suspected, there's a bug or two in SPHERE which
only showed up when it was used by someone who has not had a hand in
developing it. The problem had to do with your HOLD predicate; they'd
never tested SPHERE with a predicate taking a function as an argument.
Also, the axioms have to be restated slightly. The proof checker
likes existential quantifiers better than universal quantifiers, so
ATTRANS should be
(AXIOM ATTRANS (FORALL X (FORALL Z (FORALL S
(IMP (EXISTS Y (AND (HOLDS (AT X Y) S)
(HOLDS (AT Y Z) S)))
(HOLDS (AT X Z) S))))))
and in order to know what it can plug in when trying to prove the
EXISTS statement, it has to know the extension of the sort of Y:
(EXTENSION PLACE (I SUITCASE DESK HOME CAR AIRPORT COUNTY))
Apparently it's not yet smart enough to use the DECLAREd variables as
the extension of that sort.
Steve also said that the proper way to handle the problem using SPHERE
is to use contexts to represent situations. He's going to show me how
to do that next week.
-------
∂01-Mar-81 1413 TOB computing
John
I find that queues for terminals are a great problem. I have to
wait considerable time at 9 am recently. I assume that's a portent
of the future. I find E very slow in the daytime.
Yes, it slows me down greatly. Try editing a mail file which involves
deleting pages and file switching.
The load average has been high quite a lot recently. That may
be Ethernet related. I don't like the prospect of putting up
with the current bad service for two years. It may take that
long to do anything. As for the S-1, that is an unknown, without
much system, and with no other user community. I have reservations
about that one.
VAX is not bad. Today is Sunday and load average is 5.3.
We could put the computing problem in a framework such as, "others
would be delighted with these resources".
We have often disagreed on this issue. Still, I listen carefully
to what you say about solutions to computing problems.
Tom
∂01-Mar-81 1345 JMC
I think your perceptions about editing being slowed more than (say) ten
percent ar are mistaken, but you are not alone in having such perceptions.
I propose that an actual test be made of how fast people edit under
different load conditions. We have an excellent chance of getting a
2080 (5 times faster) in about 2 years (or an S-1, much faster than that),
but I think we'll be making a mistake to go after additional computer
facilities now. We will dissipate our resources for purchasing, housing
and operating computer facilities.
I agree that editing large mail files is slow. The actual computer time
used is substantial, but not for any good reason that I have been able
to fathom. I have talked to Marty about it, but he seems defensive and
even evasive and suggests using monthly rather than quarterly mail files,
which I may be driven to do. It should not take 2 seconds of actual
computer time to delete a page from a file even if there are 400 pages.
∂01-Mar-81 1423 TOB
Thanks Martin. Has the high load average been Ethernet related
or real load?
ME - A couple of weeks ago, there were some high loads due to a
misfeature of the Ethernet that was running wild. However, that
problem was fixed then and no longer occurs. I haven't noticed
many times of high load lately, although there have been brief ones
caused by lots of TTY (non-DD) I/O, especially from the Canon
printer (which is connected to the system like a TTY).
∂01-Mar-81 1428 TOB mail files, etc
I have gone to keeping my mail file to a few pages because of that
problem. It's still not fast. And that's true of other big files
I edit. I do a lot of page switching and transferring text.
∂01-Mar-81 1419 JMC
I agree that editing large mail files is slow. The actual computer time
used is substantial, but not for any good reason that I have been able
to fathom. I have talked to Marty about it, but he seems defensive and
even evasive and suggests using monthly rather than quarterly mail files,
which I may be driven to do. It should not take 2 seconds of actual
computer time to delete a page from a file even if there are 400 pages.
∂01-Mar-81 1755 RWW CIRCUM
I AM IN THE MIDDLE OF CODING. There are several questions to ask?
How fancy do you want me to be about the choice of bound variables?
The problem is what sort should they be? The most straight forward
thing to do is to see if there are any variables of sort universal
e.g. U1, and to produce the variable lists U1, U2, U3, U4, etc.
If there is no variable of sort UNIVERSAL I'll choose one that's
not otherwise used and do the same. If this is OK let me know.
I haven't thought about how circumscription would interact with the
sort mechanism. One alternative is to ask the user to give the
required variables.
∂01-Mar-81 1831 Keith A. Lantz <CSL.LANTZ at SU-SCORE> owicki
Date: 1 Mar 1981 1830-PST
From: Keith A. Lantz <CSL.LANTZ at SU-SCORE>
Subject: owicki
To: jmc at SU-AI
Please add Susan Owicki to the Comp mailing list; she will be representing
CSL this quarter (with me in the background.) -- CSL.SSO.OWICKI
Keith
-------
∂01-Mar-81 1831 MINSKY@MIT-AI Hydrogen
From: MINSKY@MIT-AI
Date: 03/01/81 21:23:42
Subject: Hydrogen
MINSKY@MIT-AI 03/01/81 21:23:42 Re: Hydrogen
To: ENERGY at MIT-AI
Probably hydrogen should be burned in fuel cells, to avoid Carnot
losses. Even when burned for heat, this is true, to get best heat
pump performance. And I doubt that fuel cells are energetic enough to
oxidize nitrogen. Please stop trying to find more things wrong with
hydrogen than it has already.
By the way, the best chemical fuel would be cryogenic atomic hydrogen,
I believe, for it has much more energy per gram than plain hydrogen.
Its major defects are non-existence or proneness to spontaneous
explosion should it happen to exist. There are also conjectures that
metallic hydrogen might be metastable.
Here is a conjecture for physicists: the problem with fusion is that
the electrons dissipate the energy needed for making protons fuse.
When you throw a proton into some matter, the chance of hitting
another nucleus head on is so small that you need enormous energy to
get enough such chances -- hence all those plasma and laser fusion
heating schemes. However, if you just aim carefully enough,
exploiting symmetries, then you need only a hundred kilovolts or so.
If we can design appropriate control and focussing devices, then, we
should be able to make good micro-fusion generators. R. Feynman
thought this was a good idea, but saw no way to obtain the required
precision, but suggested that if we rumored that someone had such a
scheme, then maybe other jealous physicists would invent one. Lowell
Wood points out that the thermodynamics of positioning the target to
such precision is most unfavorable and, even if feasible, would not
remain so after some reaction ensues. It is perhaps worthwhile, though,
to remind ourselves that "cold" fusion is not yet provably impossible.
∂01-Mar-81 2356 LWE NS TAPE
To: REG
CC: JMC
THANK YOU VERY MUCH INDEED FOR THE DUMPER TAPE WHICH ARRIVED OK!!!
∂02-Mar-81 0040 POURNE at MIT-MC (Jerry E. Pournelle) CHarge!!
Date: 2 MAR 1981 0334-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: CHarge!!
To: JMC at MIT-MC, MINSKY at MIT-MC, LEECH at MIT-MC
To: POURNELLE at MIT-MC, LLW at SU-AI
Re LLW's plaint, as far as I know our stuff is REALLY
being read at the Special Assistant level; at least the
questions I get are intelligent and show that some thought has
been given to what was said in the documents.
Jerry Gray of AIAA was wondering what hit Stockman; he
told Mark Hopkins that "something must have happened"..when told
of the 4500 letters and telegrams generated by L-5, he told
Hopkins, "Gee, that must explain were we got the extra 142
million... I thought we'd lost that."
As to Gingrich etc his aide is indeed carrying a
pro-space petition around to all the other congresscritters, and
they will indeed attempt to enlist everyone in the cause; and I
suspect that he's set to try to get some publicity. What he's
waiting for is the final report. What the report is waiting for
is me to get back to work. I have folded in the comments of
about 7 people into the draft, now I have to make it sound like
a human wrote it and not a committee.
Another 25 or so have endorsed it; including a number
who were not present at the weekend meeting.
Salisbury is taking the report to the big OMNI meeting
this week, and Harry Stine is also going up to NYC for that
meeting. It appears that our strategy and report will dominate
the OMNI effort since Sheffield, Stine, and Hubbard are all on
the OMNI team to draft their position paper, and they're all
essentially going to take ours and put it in as the Citizen's
Advisory Council report; it may be that OMNI will simply endorse
us but that's problematicl given that we've yet to hear from
Sagan and Murray.
Ah well, back to work.
∂02-Mar-81 0043 POURNE at MIT-MC (Jerry E. Pournelle) Re: Helping the Boss
Date: 2 MAR 1981 0343-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Re: Helping the Boss
To: JPM at SU-AI
CC: ELLEN at MIT-MC, PDL at MIT-MC, MINSKY at MIT-MC
CC: POURNELLE at MIT-MC, jmc at SU-AI, llw at SU-AI
My note on electronic office stff has been carried to
Mr. Meese after favorable comment by a Special Assistant. What
all that means is not clear to me; I'm just telling you what
they tell me.
I have NOT heard what progress if any was made in
putting in computer net stuff in EOP, except that the questions
asked of me in clarification of my memo make it a fair inference
that at least two Special Assistants in the Office of Policy
Develoopment have never even heard of computer nets, much
lesshave any access to them, and their comments re my memo (and
taking it over to Meese, who is said to be a gadget fan) support
that inference.
Thus there's a chance that they'll ask help in setting
up something; but I have no guarantees. I forward a note sent
me, with request for comments from any of you. My suspicion is
that next time I'm in DC they'll have me come in to talk about
possibilities of electronifying things, but that it won't be a
high prioirity item.
JEP Date: 01 Mar 1981 1220-PST
From: Jim McGrath <JPM at SU-AI>
I was just put on this mailing list, so please forgive me if my
inquiry is out of date (I have read OAF's files containing previous
messages).
Jerry (POURNE) mentioned the possibility of the people in the off
White House installing a mail system for internal use (with the
eventual possibility of linking it up to any net they could shake
a stick at). Has any action been taken on this? Because if it
has not...
Stanford is installing a University wide text processing and mailing
system that is eventually designed to link all staff, faculty,
administration, and students. This system will also be linked to
other facilities on campus (LOTS, the undergraduate facility, SCORE,
SAIL, some 3033's, etc...). It uses state of the art technology,
and is not trying for anything fancy.
Essentially you take a DEC 2060, add 64 ports, put on the system
MM (a very good mail reader), TVEDIT (display editor developed
at IMSSS here at Stanford), EMACS (standard MIT emacs), EDIT
(DECs supported line oriented editor), PUB (document formatter
developed at SAIL), SCRIBE (another document formatter, developed
at CMU I think), TEX (document formatter that works well on
material requiring strange fonts, like technical papers), SPELL
(the standard spelling correction program), and XSEARCH (allows
you to search for key words in multiple files). Oh, also add about
1 megabuck.
This supports 64 users at any single time. Want more? Add another
2060. They can all be linked via a DECNET connection. For the boss
we might throw in a NS type program to prepare news summaries from
the news wires, and fancy printers (like the DOVER I think the White
House already has). Also ZORK, ADVENTURE, EMPIRE, BKG, DND, your
favorite chess program, and a NET connection so Ronnie can read
SF-LOVERS before the NY Times every morning.
Seriously, all the specs (both technical and financial) exist for
a system like this. Give us the money, 3 staff (1 hardware, 1
experienced in TOPS 20, and another to support the above mentioned
user programs), and you can have it up and running as fast as
you can get the hardware from vendors plus 2 weeks for installation.
Any takers? Contact me via the NET, and I can get hardcopy material
to you this week.
Jim
∂02-Mar-81 0101 POURNE at MIT-MC (Jerry E. Pournelle) helping the boss
Date: 2 MAR 1981 0401-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: helping the boss
To: JMC at SU-AI
Right. I'll let you know more when I know more. I have been
asked by several special assistant types about this, but I have
NO OFFICIAL REQUESTS at the moment. More when I know more.
∂02-Mar-81 0104 LLW The Referenced LLW Plaint. . .
To: JMC at SU-AI, minsky at MIT-MC, leech at MIT-MC
CC: LLW at SU-AI, pourne at MIT-MC
∂01-Mar-81 1814 LLW Sustaining The Charge
To: pourne at MIT-MC
CC: LLW at SU-AI
∂01-Mar-81 0038 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 1 MAR 1981 0338-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: llw at SU-AI
COMSAT@MIT-MC 03/01/81 03:32:55 Re: Msg of Sunday, 1 March 1981 03:32-EST
A copy of your message is being returned, because:
"LLW" at MIT-MC is an unknown recipient.
Will try sending to the file "GUEST3;LLW MAIL".
Failed message follows:
-------
[Dear Jerry:]
POURNE@MIT-MC 03/01/81 03:32:48 Re: Saving Civilization on the installment plan
1. You will all be pleased to know that we are now on the
Request List; that is, not only are our space plans being read
at the Special Assistant to President level, but I have letters
requesting them along with very sensible questions.
[I'm skeptical--consider how magnificently the new crew is interacting with
the Congress and the press; we're probably get getting the Standard Reagan
Treatment. How they're stroking everyone simultaneously with such a small,
new crew is beyond me, though.]
2. Congressman Newt Gingrich (R, Georgia) is sufficiently
impressed that he has delegated an aide to monitor our
activites, and told me that he (Gingrich) has a dozen (the
number he used was 20!) Republican Congresspeople who will sign
our final report and he will carry it to Stockman. I'm still a
little dazed at this unsolicited proposal, but I'll take it.
[Great! *That's* the kind of stuff that really counts!!! Where's our
legislative liaison types--why aren't they carrying Gingrich's covering
letter around to *every* Congressional office, asking for other
co-signers??? Congressmen will sign *anything* that's not wildly
controversial, if they have just a bit of constituent encouragement to do
so. Backing the space program and high technology is presently only
slightly more controversial than endorsing motherhood. Why isn't Mrs.
Trimble running a `Support the Gingrich Initiative' mail campaign? If
there's anything a Congressman loves more than notice by the media, it's
the attention of his colleages; we can get Gingrich both, and turn him
into a hard-charger for us. Let's try to really snowball this one, and
then have it presented to Stockman with enormous press bally-ho. Remember
The Mouse Who Roared! (If our aerospace friends won't foot the bill for
at least the campaign's mailing expenses, they *deserve* to go to
Stockman's guillotine!)]
3. The Citizen's Advisory Council is thus off the ground, now
only limited by the energies and abilities of your chairman; I'm
dancing as fast as I can...
[Just pretend that you have 5000 North Koreans charging your tiny outpost,
and that you're again driving the men of your platoon forward to battle at
pistol-point. I'm completely confident that you have *at least* one good
campaign left in you--if you protest that you aren't quite as physically
igorous as you were at 18, just remember that you're a whole lot
sneakier! As one of my favorite posters in my office continually
reassures me, `Old Age and Treachery Will Always Triumph Over Youth and
Skill!' Charge!]
4. I have asked LEECH, a Cal Tech student, to do a survey of
computer resources possibly available to L-5 and the Citizen
Council. Please cooperate with him. I have always had the
suspicion that we could get camera ready copy, if not for the
L-5 News, then at least for a newsletter, through net resources;
and that a good bit of the eiditing of the newsletter could be
done through this means. I may be wrong? Anyway, I am now
involved in (1) getting the final draft of the Council Report
together in time to do some good; (2) trying to coordinate a L-5
Membership drive; and (3) trying to keep enough troops working
with the trimble's to keep their Star Trek mail machine grinding
along. Thus I feel spread as thin as boarding house butter, and
any help from any of you will be apprecaited. I remember when I
wrote books before I took up saving the space program for a
living. Books are more lucrative...
[I'm loathe to advertise it, but we can generate arbitrarily large
quantities of camera-ready copy locally, and routinely do so for
programmatic purposes. The problem is that we're a very Thin Red Line in
my area, and someone (e.g., OTA@su-ai,mc,ml, etc, an L-5 fanatic if there
even was one) would have to be torn away from his work to juggle the text
through a document compiler, push it through the copy generator, debug the
source after getting output, etc.--it's typically a fairly good-sized job,
time-wise. In other words, we can--and are willing to--do special,
high-priority jobs every now and then, but we can't handle routine stuff
in any quantity. (If someone elsewhere could be trained to prepare input
semi-flawlessly for our document compilers--POX or TEX, both of which are
well-known on the ARPANET--we might be able to generate camera-ready copy
more routinely.)]
Jerry
[Lowell]
∂02-Mar-81 0101 LLW Plaint-ively Speaking. . .
To: JMC at SU-AI
CC: LLW at SU-AI, pourne at MIT-MC
∂02-Mar-81 0043 JMC
what is the afore-mentioned LLW plaint?
[John: From context, I deduce that you didn't get a copy of my note to Jerry.
One is on the way. Lowell]
∂02-Mar-81 0128 POURNE at MIT-MC (Jerry E. Pournelle) Towards A System For The EOP
Date: 2 MAR 1981 0427-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: Towards A System For The EOP
To: PDL at MIT-MC, ELLEN at MIT-MC, JMC at MIT-MC, MINSKY at MIT-MC
To: POURNELLE at MIT-MC, LLW at SU-AI
Lowell brings up the problem of security for EOP.
My prediliection is for a series of micros, each with its own
disk system, connected by something like CP/net. The only way
you can get data from my file under that is to have physicl
access to my disks (or for me to WANT you to have it and thus
send it]. The very primitive nature of the operating system
makes the security problem less severe. For example, there is
no way any of you can get at my disk files--especially since
most of them are not even in my machine now! So even if you
could send instructions to my micro without my knowing you did
(and Zeke would tell me, I think) and get Zeke to send out files
from his disks without my knowing (and the clickm,ing and
clunking and light flashing would tell me something was strange)
I still don't see you getting my hypnotized into putting my
files in the machine..
I realize this is primitive by everyone's standards, but
for around $5,000 a stataion (including software) I can install
it easily, and its a system I understand. You can then use
these as terminals for a more complex net system if you like;
but the file access problem is solved, it seems to me. Not
optimum use of equipment, but better than paper and typewriters
(and they get text editors and all that sort of thing as a
matter of course; in fact they get dupes of my system.←
∂02-Mar-81 0134 JPM Re: Late
To: JMC at SU-AI
CC: pourne at MIT-MC
Thanks for the info (or lack thereof) on the status of a computer
system for the White House people. A couple of additional points:
TOPS 20 provides for file security in a nice and easy to understand
fashion. It also allows you to set protections for your entire
directory. Finally, you can set individual protections for 'user
groups,' i.e. arbitrary groups of directories. In all, the
protection afforded is extremely flexiable.
In regard to the actual security of the system, therein lies the
beauty of a limited system. Since users are only running text
editors, document formatters, mailing systems, canned programs, etc...
(i.e. they are not actually writing programs), they do not need
access to such things as DDT. Without the ability to poke around
the guts of the monitor, the security cannot be broken ( except
by giving away passwords and such, which is a problem with ANY
system, computer based or otherwise). There is only one exception -
your systems people will be able to break security.
But so can file clerks in a normal office. Since you need only
a handful of systems people (like I said, you can do with 3 for
one machine, 5 for 2, and then add a person for each additional
machine you hook up), your monitoring of them can be MUCH tighter.
Do deep personal checks, allow access to the system only via
certain terminals which are physically secure (i.e. no one
should be able to dial in and break system security - they would
have to do it in person), make everything like tape drives physically
secure as well, and you should have no problems.
[actually, in theory only ONE person for the whole system has to
have the ability to break security. However, for system work you
frequently NEED to break some security, so one is probably too
small a number to deal with easily]
Only difficulty I can see is finding hackers for your staff who are
good security risks. But I am sure they can be found.
A side matter - you could like this machine up to CongressNet,
with suitable security between systems. In practice this
probably means just the ability to send mail between machines
and a limited FTP capacity. Add on a NET for the Supreme Court,
hook them all onto the ARPANET, and then we will REALLY get some
input into the decision making process!
Anyway, I'd like to keep informed about this. I can see no
technical problems implementing a secure system with all the
capabilities I mentioned for text processing and electronic
mail. People problems yes - but those come with ANY system
you can imagine.
Jim
∂02-Mar-81 0150 JPM Re: The Boss
To: JMC at SU-AI
CC: pourne at MIT-MC
Some financial figures -
Cost of 2060 = 750,000
300 terminals and lines - 750,000 (2500 each)
physical security - 200,000
addons to the system - 250,000 (you will eventually want
more disk, tapes, etc...)
TOTAL 1,950,000
Staff 3 full time at 40,000 each
3 more trouble shooters (terminal problems, etc...) at 20,000 each
DEC app 100,000 per year.
TOTAL 180,000
Savings - 30 minutes of time per manager per day - 600,000 per year
cutting app 50 clerks, etc... - 750,000 per year
TOTAL 1,350,000 per year.
So the system pays for itself in under 2 years, with increased
productivity.
Now, the cost figures are pretty firm, but the savings are not.
However, assume you need 1 staff per 3 managers (probably a
VERY low figure). The AI Lab use to use 3 per 100. That is
a sayings of an order of magnitude in personal costs. I only
cut the figure in half (assuming we start with 50 staff). I
can only think the entire calculation is extremely conservative,
and yet the system still pays for itself in under 2 years!
So if the administration is serious about saving money, they
can save a min of 2 million over 4 years here alone (and
the real savings are probably closer to 10 million). That
AND increase productivity.
Jim
∂02-Mar-81 0736 BYY LUNCH THURSDAY
YES, lets have lunch Thursday. I teach at 1. Is 11:45 or 12 ok?
Alas, that's too tight a squeeze. I teach till 12:15 Tuesday and
Thursday. How about Friday or the following Monday or Wednesday?
∂02-Mar-81 0910 FFL
To: EJM at SU-AI, JMC at SU-AI, FFL at SU-AI
In reply to your question Prof. Reiter is a Professor of Computer Science at
The University of British Columbia in Vancouver and will be the speaker at
the department colloquium on Tuesday, March 3, and at the Knowledge and Action
Seminar on Wednesday, March 4.\
\
∂02-Mar-81 0928 FFL
To: JMC, FFL
Caroline Tajnai asks me to tell you that there will be "mingling" in the
Lounge on the 3rd floor at 3:45 p.m., preceding the Colloquium, and that
she is counting on you to introduce Prof. Reiter at the Colloquium.
∂02-Mar-81 1531 FFL
To: "@FAC.LIS[1,FFL]" at SU-AI
To clarify the arrangements for the Wednesday luncheon
with Professor Reiter, the luncheon is on Wednesday at
12:20 p.m. but his Department Colloquium talk is on
Tuesday at 4:15 pm.
John McCarthy
∂02-Mar-81 1557 CSD.ULLMAN at SU-SCORE reiter
Date: 2 Mar 1981 1538-PST
From: CSD.ULLMAN at SU-SCORE
Subject: reiter
To: jmc at SU-AI
count me in for lunch on the 4th.
-------
∂02-Mar-81 1642 RPG
∂02-Mar-81 1142 JMC
What do you presently use for spindling and crunching?
I use PACK and Don Wood's packager, called BUNDLE.
-rpg-
∂02-Mar-81 1650 RPG
∂27-Feb-81 1625 JMC
A few paragraphs will suffice most likely. The title of research associate
is probably quite possible. While we can include you in a proposal that
will contain some of your ideas, the privilege of making proposals
independently to government agencies is available only to titles with
the word "professor", i.e. regular faculty or adjunct professor.
Ok. I meant being able to make proposals under the auspices of some
`sponsor', not on my own.
I will send you the idea I was thinking of, though only a small
preliminary study of it would fit into any proposal that didn't
intend to have this as part of it.
You may also be interested to know that I am doing a study of
LISP timings on as many LISPs as are available. The idea is
to get n benchmarks from the various places. Then there are m
volunteers who will do the timings based on some methodology
I will invent (including translations). Then it's all gathered
together (along with experiences with translating with the hopes
of gathering the tools to do so in general) and published in
some convenient place, like CACM. Now, this sort of thing is
of marginal scientific interest, but of large practical or
engineering interest. In addition, it isn't much of my time
and it helps put Stanford back on the Lisp map a little.
I already have a number of volunteers for doing the tests
and supplying programs. If you want to supply some program,
send me a message. You are already on my distribution list
for this project, though I haven't sent anything out on it yet.
-rpg-
∂02-Mar-81 1652 KGK via SRI-KL third exercise
John, I've written up the third exercise; it's EXER3.TEX[1,kgk] and .PRE
in [1,kgk]. I think the Dover is broken, so you may not be able to list
the press file. If you need to re-TEX it, I use DEF.TEX and BASIC.TEX on
my area, and also have defined "@" rather than "\" as the control char.
I didn't include a semantic attachment part. --kk
∂02-Mar-81 1650 RPG
As most of you know, there will be an attempt made to do a
series of Lisp timings in which various benchmarks submitted by the
Lisp community are tested on a variety of different Lisp systems.
Since there will need to be some translations done in order to run
these benchmarks in systems for which they were not intended, there
is the secondary (!) problem of learning what is really needed to do
these translations more readily in the future.
I will be co-ordinating this effort and will be distributing
the results when they are in. For this purpose I have set up 3
mailing lists at Stanford:
LISPTIMING the list of people interested in this topic
LISPTRANSLATORS, the list of people who have volunteered
to do the timing tests (and translations)
at the various sites
LISPSOURCES the list of people who will be supplying
benchmarks
You can MAIL to these entities at SAIL (e.g. MAIL
LISPTIMING@SAIL...) and thus avoid swamping the mailing lists we
have beenusing so far.
If you care to be on one of these lists, please send me
(rpg@sail) your login name and site exactly as your mailer will
understand it along with which list you wish to be on. If you are
supplying programs or talent, please let me know which Lisp, which
machine, and which OS you are representing.
In addition, a list of all messages pertaining to this
extravaganza will be on TIMING.MSG[TIM,LSP] at SAIL (you can
FTP from SAIL without logging in). In general, this area will
contain all of the information, programs, and results for this
project.
If you know of anyone who is not on the net and who may be
able to help, send me a message and a method for getting in touch
with him/her. Over the next few days I hope to establish some of the
methodological considerations (such as GC times) for the project.
Dick Gabriel (RPG@SAIL)
Sorry I misunderstood your instructions in haste and mailed my request
to the list.
∂02-Mar-81 2116 TOB 2080
On thinking about it, a 2080 two years from now
will probably be much too underpowered. It might do now
but usage will grow a lot by then. As things stand,
students have to use LOTS for course work.
∂02-Mar-81 2153 BYY LUNCH WED.
GOOD, LUNCH WEDNESDAY.. JON
∂03-Mar-81 1319 MINSKY at MIT-ML (Marvin Minsky) Thanks for pre-text
Date: 3 MAR 1981 1617-EST
From: MINSKY at MIT-ML (Marvin Minsky)
Subject: Thanks for pre-text
To: MINSKY at MIT-ML, rah at SU-AI, llw at SU-AI, jmc at SU-AI
Thanks for draft copy. I'll rework some of it next week after I see
what happens when I give a talk on it.
That underground engine sounds great. I take it that we can't make
the rings superconducting because the AC flux changes cause heating
losses? Where can I find good explanation of superconductor
limitations?
I was considering the possibility of very sparse structure for the
upper parts of the tower. But I guess it makes little difference
because of the small effective G.
The anchor seemed remarkably heavy until I realized that the effective
G-force out there is only about .005 of what it is here. When you
call it the "anchor", you must have a variety of uses in mind. It
occurs to me that it can be used to deal with the DC stability from
the top, by constructing some of it in complicated cable-linked
dumbbells that rotate; then the skyhook-top can pull itself sideways
in various directions for lateral corrections from the top. I realize
that the dumbbells will exhibit diurnal precessions, so there must be
at least two of them, or maybe three, and the thing would be quite a
circus up there.
Did you get a note mentioning that the ground station needs only 450,
not 540 degrees? Indeed, if we considered two towers a diameter
apart, then we get both for just 360 degrees. Proposal: these two
towers are passive pipes, leaning together for 50Km or so, then merge
and deflect the beams upward the required degree or two. This saves a
lot of ground engine. (Note that the deflection engine here is your
efficient lateral type, not longitudinal, though a small longitudinal
component is desired here for lift.) I admit this would certainly
complicate (a) building the tower and (b) elevating payloads.
In any case, there might some gain in making more of the lower part
passive, where G is large even paying the cost of mildly tapered
cable. I imagine that 100 kilometers or so might be feasible, and
somehow combine to make the atmosphere tube as well, so that we don't
have to combine that with the motors.
-- marvin
∂03-Mar-81 1332 FFL
To: JMC at SU-AI, DEK at SU-AI, EAF at SU-AI, FFL at SU-AI
Overseas telegram received addressed to McCarthy, Knuth and Feigenbaum;
Planning attend San Diego March 9-12 Software Conference. Wonder about
weekend visit to Stanford. Please cable reaction. Andrei Ershov.
from Novosivirsk. Cable reference C2354-03
I will send Ershov a telegram inviting him for weekend visit unless
there are very strong objections.
∂03-Mar-81 1348 FFL
To: JMC, FFL
Scott Palmer of INTERFACE, a manufacturing and engineering magazine, called
to ask you questions about AI. Had spoken with H. Moravec who suggested that
he speak with you. Will call again.
∂03-Mar-81 1514 RPG Lisp Timing
To: JMC
CC: RPG
I talked to Guy Steele this morning, and he is quite
enthused about this Lisp timing project. So far I have 22 volunteers
to help with it all over the Arpanet. People in Europe are also
being solicited. Steele thought that ARPA money could easily be thrown
at this thing, especially if it can also isolate issues of transportability
as well as performance and expressive power. So far, it looks as though
some non-trivial amount of resources at Stanford will be used up
in this effort, so I am willing to `head' this project here, at least
for the time being. My guess is a year's worth of work, while Steele thinks
3 years is more like it.
There are a number of methodological questions about how to
accurately quantify all the things that go into a Lisp system, so
I think there are some interesting, non-managerial problems to work on.
Plus, I think that ARPA would be eager to throw some money at Stanford
for it. I also think it's good for Stanford to get this publicity.
Do you want to proceed on this, or would you rather I do it
as a hobby?
-rpg-
I would like to proceed. I will phone Bob Engelmore, who is now out
of town, on Thursday, and I can mention it then. However, if you want to
talk to him yourself, that's also fine. Let me know what you decide.
Ikuo Takeuchi of the Japanese National Telecommunications Laboratory
did a similar study (probably less extensive) a few years ago, and
you should find out what he did.
∂03-Mar-81 1616 RPG Methodology considerations:
To: "@LSPTIM.DIS[P,DOC]" at SU-AI
1. GC time is critical. Every timing should include CPU time
as measured by the CPU clock plus GC time. If GC time is not
accounted in the LISP, we should include a standard test, such
as a program that creates a large, standard structure (balanced
tree of some sort?) and then count CPU time on a forced GC, resulting
in a seconds/cell figure for each machine. Maybe we should do this
in addition to the benchmarks? [In fact, measuring GC time in a meaningful
way is que difficult due to different algorithms. Perhaps a range of
tree structures? Maybe not all algorithms are symmetric on car/cdr?]
2. Translating non-standard control structures can be a problem.
What about non-local goto's ala catch/throw? These can be simulated
with ERROR/ERRSET or with spaghetti hackery in InterLisp. These questions
should be addressed by having each translator propose various techniques
and having the source decide on which to use. Or maybe we should use
all such methods?
3. All non-LISP syntax must be pre-expanded (i.e. CLISP) to allow
the local system to optimize as appropriate.
4. Both interpreted and compiled code will be timed.
All code will have macros pre-expanded (at local sites?) so that
efficiencies due to incremental destructive expansion can be
eliminated.
5. Numeric code should have all types announced to the translators by the
sources so that optimizations can be made without deductions.
All other such information must be provided.
6. The size of such programs can be arbitrary, though translating
MACSYMA may take a while to do.
7. All tools developed to aid translation should be forwarded to
RPG so that they may be evaluated and distributed if appropriate.
8. Programs that are useful to a source but impossible (in a
practical sense) to translate should merit special attention to
decide if there is a useful feature involved.
9. (from GLR)
Timing various programs is a good idea, but the data will
be a little hard to extrapolate. Is anyone going to measure
parameters such as CONS rate, time to invoke a procedure,
and add times? [Not only that, but number CONSing and its
effect on numeric calculations should be measured as well. Here
RPG will appoint some experts (like JONL) to make up some
good numeric testing code to isolate implementational problems
with specific aspects of Lisp).
10. People should supply some estimate of the runtime and the results
of their benchmarks. Such things as 2 minutes of CPU on a KL running
TOPS-10 is OK, but for unfamiliar machines/Lisps this may not be good enough.
Try to aim at some estimate in terms of the number of CONSes or function
call depth.
11. Every candidate system should have a detailed description of that
description (CPU architecture, memory size, address size, paging algorithm...)
∂03-Mar-81 1524 RPG Lisp Timing Mailing List
To: "@LSPTIM.DIS[P,DOC]" at SU-AI
Welcome to the Lisp Timing mailing list. As you may have
already guessed, the scope of the Lisp Timing Evaluation project is
very large in scope, and if we are to make a contribution to the
undertanding of how to evaluate such an elusive thing as an entire
computing environment we will need to consider many methodological
issues. Since I am no expert on such evaluations I am going to require
a good deal of help, and so far more than 20 people have volunteered.
The problems we face are not just how to measure the performance
of these Lisp systems, but how to take a diverse set of benchmark
programs and get them to run on systems very different than those they
were written for.
I hope at the end of this project to be able to report not
only times for programs, but descriptions of systems, translation
problems, and a general guide to the world of Lisp computing.
The first substantive mailing will be a quick list of
methodological points we need to consider. This list is not complete,
but aims at the directions we need to go before actual timing runs
can be performed.
Thank you for your help in this project.
Dick Gabriel (RPG@SAIL)
∂03-Mar-81 2118 CSD.ULLMAN at SU-SCORE
Date: 3 Mar 1981 1809-PST
From: CSD.ULLMAN at SU-SCORE
To: CSD-Faculty: ;
A few words of report about how i am spending your money are in order.
1. I have agreed with the CIS directorate that the nominal share of CIS
that would "belong" to CSD is 20%. This means we are liable for 20% of
the net deficit, which next year should be about $25K, decreasing, we hope,
to 0 in 3 years. On the bright side, we shall be entitled to 20% of the
space in the building, which should be in the 5-10K sq. ft. range.
I think that's a good investment. We could "tough it out" and insist on the
space anyway, but I feel that we are better off this way. Any objections?
support??
2. EE has announced it will give 8% of the NRO revenues to the instructors
of TV courses. We are effectively forced to go along, or else cross-listed
courses will be taught only in EE, with the department losing substantial
revenue. There is a great deal to be said for the policy anyway, as teaching
on TV is a substantial chore. The policy will cost us about $10K/year.
Again--comments anyone??
-------
∂03-Mar-81 2256 DEK ershov
I want very much to see him, since we are co-editing the proceedings of
the Urgench conference and it is crucial that we discuss things instead
of waiting for 6-week turnaround in the mail!
Let me pay half the travel expenses.
Why don't we get him to pay his own travel expenses? I am broke and
really have only personal reasons for wanting to see him. If you have
a justifiable reason and we can't get him to pay his own, it would seem
that you should pay the full expenses.
∂04-Mar-81 0935 FFL
To: JMC, FFL
David Kaplan of Wiley publishers called. He would like you to return
his call collect. 9 Operator 212 850 6533.
∂04-Mar-81 1116 FFL
To: JMC, FFL
Prof. Colmaureur called. He will call again tomorrow at your home
between 8 and l0:30 a.m. He left the message that he would like you
to come as soon as you can.
∂04-Mar-81 1117 FFL
To: JMC, FFL
Lester Dugan called. 55 495 5669.
∂04-Mar-81 1121 CSD.JEANIE at SU-SCORE Faculty Meeting
Date: 4 Mar 1981 0859-PST
From: CSD.JEANIE at SU-SCORE
Subject: Faculty Meeting
To: tenured-faculty: ;
cc: csd.bscott at SU-SCORE
Another faculty meeting has been scheduled for March 10 at 2:30 pm. It will
be in MJH 252.
-------
∂04-Mar-81 1358 FFL
To: JMC, FFL
Prof. Ljung, visiting professor in engineering, called and asked that
you call him. 7-2873.
∂04-Mar-81 1534 CLT Calendar items
∂04-Mar-81 1622 FFL
To: JMC, FFL
Your bill at the Faculty Club for the Reiter lunch was $5l.65 for the food.
∂04-Mar-81 1639 FFL
To: JMC, FFL
Bpb $Bob Grant called re the estate bill. 9 328 656l.
∂04-Mar-81 2216 TOB
Progress
ACRONYM looks at an image of four aircraft. The edge results and ribbons
it starts with are not very good, and one aircraft it does not make sense
of. The second it says is not a 747 but could be an L1011. The third
and fourth it says could be aircraft, but they are not 747's.
ACRONYM has models only for L1011 and 747. Rod thought
they were 747's. When he went back to the original image, they turned
out to be 707's.
New theoretical developments show a clean way of implementing directional
edge operators [Binford 81; AI Journal].
New theoretical results outline a theory of interpretation of
interpretation of line drawings which promises to include curved surfaces
with no restrictions on number of surfaces at a vertex, including
paper (non-solids), wires, and transparency.
[Binford 81; AI Journal] These results appear to have definite application
for stereo and photointerpretation. See below.
This work is being extended.
Experimental results show use of interpretation of a single image
of an L1011 to build a three space model. They use occlusion to build up
a layered relative depth model: the fuselage is above the wing which
is above the engine pod which is above the shadow which is on the ground;
the shadow is over a white line which must be on a smooth surface.
The techniques are quite general, they do not depend at all on knowing
the objects in question, and they use cues which are universal available.
The work uses shadows to build accurate geometrical models of heights of
surfaces. Again, the techniques are widely applicable.
This work is being extended.
∂04-Mar-81 2347 BIS london,request
To: JMC
CC: TOB
I have been asked to teach a 4-day short course on Robotics, the week
after next (17-20 March 1981) in London, England.
Since I think this can be classified as `consulting,' it falls within the
scope of the undated, unsigned memorandum, `Consulting Policy for Academic
Staff,' which was distributed throughout the Department last May.
I hereby request permission to use those 4 days for consulting purposes.
I have used no days for consulting within this quarter.
I've sent Tom a message explaining how my teaching this short course does
not adversely affect progress on our grants, and I hope that he will send
you a message confirming this.
∂04-Mar-81 2348 BIS london,reasons
To: TOB
CC: JMC
The week of the 16th is clear, as far as I can see. ROD's orals are the
week before. I have no appointments on my calendar for that period. The
proposal for computer-aided mechanical design will be out of our hands by
next Friday.
Why?
(0) It will force me to wrestle with teaching robotics to a group
of non-experts. This should help my understanding of the field.
(1) It will give me a chance to meet industrial people and listen to their
problems.
(2) I haven't taken a day off work since 01 January.
(3) I have not yet set foot on British soil.
(4) There are some terrific paintings in London by Winterhalter and Ingres.
Could you send JMC a message approving my proposed action?
He is technically required to be able to vouch that my consulting
will not adversely affect work on the grants.
Thanks.
∂05-Mar-81 0037 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 5 MAR 1981 0336-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: MINSKY at MIT-MC, POURNELLE at MIT-MC, JMC at SU-AI, llw at SU-AI
The Space Council report is mostly done; anyway I have a draft
that is respectable and has been sent off to various places in
Washington. Now it may really be time to look at the telefators
problem and the artifical hands and all that.
If we have a decent proposal--=that is, an idea of what might be
done, who ought to be doing it, and what funding level--I might
be able to sell it. As with space, if you are to have a horse
race you must, to win, have a horse...
∂05-Mar-81 0107 JMC
1p Doyle money
∂05-Mar-81 0557 RWG
To: ROY at SU-AI, JMC at SU-AI, MLB at SU-AI, HPM at SU-AI,
es at MIT-MC, REM at SU-AI
yow! i cn log in from home for the first time
since before the iran-iraq war. Roy, if that was you out front in the head-
phones, thanks.
∂05-Mar-81 0924 FFL
To: JMC, FFL
Les Dugan called. He would like to speak with you as soon as convenient.
55 495 5669.
∂05-Mar-81 1043 CSD.ULLMAN at SU-SCORE Reiter
Date: 5 Mar 1981 1002-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Reiter
To: jmc at SU-AI
I spoke with him about his research interests, at some length. There is
no doubt that the sort of approach to database issues he takes is mathematically
sound; indeed he can, in principle, solve many of the connundra that trouble
me in practice. However, without trying to judge in advance whether this
logic-based approach will prove useful, there is no doubt that he has
never considered the very serious implementation problems needed to make his
ideas useful. Moreover, I infer that he has no intention of doing so.
Thus, without judging the man or his abilities in general, I am not thrilled
at the idea of bringing him here AS A DATABASE GUY.
-------
∂05-Mar-81 1100 JMC*
Call Fred Abramson 219 283-7245
∂05-Mar-81 1143 CSD.ADAMS at SU-SCORE (lia) Comprehensive exam committee
Mail-from: ARPANET site SU-SCORE rcvd at 5-Mar-81 1108-PST
Date: 5 Mar 1981 1107-PST
From: CSD.ADAMS at SU-SCORE (lia)
Subject: Comprehensive exam committee
To: csd.mccarthy at SU-SCORE
cc: csd.adams at SU-SCORE
I got a letter in the mail a while back saying that the Comp
Committee would arrage a meeting with me to discuss strategy
for my last chance at the comp. Harold Ossher tells me that
meeting with you will serve the same purpose. Do you have
office hours today, when we could talk? I am free all this
afternoon and all of tomorrow except for 1:15 - 2:05. Would
you please send me a message giving a time to meet that is
convenient for you? Thank you.
Lia Adams
-------
∂05-Mar-81 1301 JMC*
call cate
∂05-Mar-81 1403 BIS london?
Tom sent me the attached message.
Will you formally authorise me to accept?
∂05-Mar-81 1056 TOB london,go
Barry:
Enjoy the trip. Sure, do it.
A point about consulting. It is on
your time, not on Stanford's time. I.e. faculty have 13 days per
quarter consulting, which means that is time in addition to
their 40 hrs per week. The situation is unclear for research
associates and generally left to the PI. Les interpreted it
as the same as faculty.
tom
∂05-Mar-81 1503 CSD.ADAMS at SU-SCORE (lia)
Date: 5 Mar 1981 1459-PST
From: CSD.ADAMS at SU-SCORE (lia)
To: JMC at SU-AI
cc: csd.adams at SU-SCORE
In-Reply-To: Your message of 5-Mar-81 1233-PST
OK, I'll come find you tomorrow afternoon. Thanks.
-------
∂05-Mar-81 2200 Buchanan@SUMEX-AIM J.King's application to move from Mathematics to CSD
Date: 5 Mar 1981 1726-PST
From: Buchanan@SUMEX-AIM
Subject: J.King's application to move from Mathematics to CSD
To: jmc@SU-AI
cc: csd.dbrown@SCORE
John,
When the Admissions Committee discussed Jonathan L.F. King,
for whom you wrote a letter, there seemed to be no compelling
reason to put him into one of the 20 CSD slots for next year.
We felt that if he wants to work with you, or someone else in
the CSD, on a thesis then he could either do it from within
the Math Dept or set up an interdisciplinary grad special
program. Getting a minor in CS would legitimize his interest
in the eyes of potential employers.
Does this sound reasonable to you? There were many people
whose applications were stronger, and we're trying to optimize
as best we can.
thanks,
bgb
-------
∂05-Mar-81 2304 Wiederhold@SUMEX-AIM Some notes on research and distributed data
Date: 5 Mar 1981 2302-PST
From: Wiederhold@SUMEX-AIM
Subject: Some notes on research and distributed data
To: jmc@SAIL
Once we have a simple distributed datasystem with named access ( work but
not research ) then a number of true research topics could be investigated:
1. Automatic domain mapping simplest case is 1:1 unique representation
transforms, but in many cases there would be
loss of information during such transforms.
(Happrns now in trivial case between SAIL and
SUMEX ascii )
2. Recognition of attribute communality perhaps operationally
3. Bias compensation of stored values collected from different underlying
populations.
I find it difficult to encourage investigations of this type without the
underlying support systems. I had a grant request in to HEW for some work
applied to a medical domain, but was refused. A well working ethernet would be
a big step and bringing a common database ( interface ) up on our heterogenous
machines a second step.
Poor Erik Gilbert is still doing his multiprocessor thesis by simulating
a multiprocessor S-1 environment.
Gio
-------
∂06-Mar-81 0036 POURNE at MIT-MC (Jerry E. Pournelle) a sort of strange thought...
Date: 6 MAR 1981 0334-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: a sort of strange thought...
To: MINSKY at MIT-MC, ELLEN at MIT-MC, PDL at MIT-MC
To: POURNELLE at MIT-MC, LLW at SU-AI, JMC at SU-AI
1. There is no science advisor. This isn't my preference, but I
understand there are problems I wot not of.
2. On the other hand, there has to be a science policy,
and the outfit charged with that happens to be receptive to
ideas, and answers phone calls.
3.The Space Policy we put together seems to be
winning by default; it may not be all that good, but it is
consistent, and reasonable, and there doesn't seem to be a lot
of competition.
4. Could we put together a reasonable
science policy? I know how to get it read at a level that is
decisive--provided, of course, there is something worth
recommending. What would we do if we were appointed "Lords
Commissioners for executing the office of Special Assistant to
the President for Science and technology"?
Is this interesting? If not, forget it.
JEP
∂06-Mar-81 0756 Mark Crispin <Admin.MRC at SU-SCORE> new TOPS-20 TELNET supports SUnet!
Date: 6 Mar 1981 0756-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Postal-Address: 12155 Edgecliff Place; Los Altos Hills, CA 94022
Stanford-Phone: (415) 497-1407
Subject: new TOPS-20 TELNET supports SUnet!
To: System at SU-SCORE
cc: @SUN at SU-AI
A new version of TOPS-20 TELNET has been installed at SCORE
that supports TELNETing to SUnet hosts. For hosts on the MJH
subnet (50) it uses the gateway server at SAIL; otherwise it uses
SUMEX (presently there isn't a gateway server on SUMEX but
hopefully there will be one soon).
What this means is that "TELNET SHASTA", "TELNET DIABLO",
"TELNET HELENS", and "TELNET IFS" all work now; and hopefully
"TELNET SUMEX-2020" will work soon. To TELNET to Shasta over the
TTY link you can still run the old SHASTA program, or say "TELNET
TTY SHASTA".
Also, MM and XMAILR have supported SUnet for a while now, so
whenever SAIL or SUMEX support a mail gateway it will be possible
to send mail to these systems.
-- Mark --
-------
∂06-Mar-81 0801 FFL Trip to France
To: JMC, FFL
You can have reservations leaving SFO at 8:40 am on the 18th and arriving
Marseilles 10:50 am the l9th. Leave Marseilles at 7:40 am and arrive,
via London, at Dulles 6:10 pm. Leave Washington 5:35 p.m. and arrive
SFO 8:25 p.m. If these connections are OK, I will confirm.\
\
∂06-Mar-81 0905 Raphael at SRI-KL Party!
Date: 6 Mar 1981 0905-PST
From: Raphael at SRI-KL
Subject: Party!
To: Distribution:
The RAPHAELs are having a Welcome-to-Spring-in-California
PARTY -- OPEN HOUSE
Saturday, March 14
Games -- Conversation -- Spaghetti -- Pistachios
Please drop in anytime: 2:00PM to Midnight
12820 Viscaino Road, Los Altos Hills, CA 94022. (415)941-0102
Directions: From the Los Altos Hills Town Hall on FREEMONT, take
CONCEPCION to VISCAINO to the fifth driveway on the right.
Hope you can come!
-Bert 'n Anne
[RAPHAEL@SRI-KL is now going off the air!]
-------
∂06-Mar-81 0937 CSD.ULLMAN at SU-SCORE David Cheriton
Date: 6 Mar 1981 0935-PST
From: CSD.ULLMAN at SU-SCORE
Subject: David Cheriton
To: CSD-Faculty: ;
The speaker at the colloquium March 10, David Cheriton, is under consideration
for employment by the CSD in some capacity. I would appreciate it therefore
if all who had interest in the matter and could mamange to do so, would
listen to the talk.
-------
∂06-Mar-81 0951 FFL
To: JMC, FFL
E. Bloom cannot meet with you today as scheduled.
∂06-Mar-81 1000 JMC*
phone Ljung 7-2873
∂06-Mar-81 1053 AVB Intel iAPX 432
To: "@SUN.DIS[P,DOC]" at SU-AI
Intel will give a technical seminar on their new 432 32-bit microprocessor
on March 24, 1981, from 8 to 12, at 3375 Scott Blvd, Suite 100, Santa Clara.
Call Linda Gale at 408-987-8086 to make reservations.
∂06-Mar-81 1128 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Courses and Degrees 1981/82
Date: 6 Mar 1981 1116-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Courses and Degrees 1981/82
To: CSD-Faculty:
cc: CSD.Tajnai at SU-SCORE
All updates have been recorded in our on-line version of the
Courses and Degrees.
<CSD.Files>Bulletin
Carolyn
-------
∂06-Mar-81 1652 FFL mail jmc,ffl
To: JMC, FFL
mail jmc,ffl
Prof. Suppes called.
∂07-Mar-81 0017 Jerry E. Pournelle <POURNE at MIT-MC> So you think our letters did some good...
Date: 7 March 1981 03:13-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: So you think our letters did some good...
To: KATZ at USC-ISIF
cc: ENERGY at MIT-MC, SPACE at MIT-MC
1. They have tabulated 8,000 pro-space letters. These will
be subpoenaed by the Space Committee when the time comes.
2. Many L-5 People who wrote to the White House have
received answers to their letters. Moreover, two special
assistants to the president are L-5 members and look out for,
and call attention to, pro-space mail.
3. Congressmen Newt Gingrich and Tribble have
sent circular letters to all Congresspeople which begin "Dear
Colleague, Have you noticed a lot of mail in favor of NASA and
the space program? We have..."
4. The Report of the Citizen's Advisory
Council on National Space Policy (created by joint action of the
L-5 Society and American Astronautical Society) is completed,
and will be signed by at least 20 Republican Congresspeople
before delivery to Stockman.
5. Therefore, don't give up; now IS THE TIME TO INCREASE
THE MAIL AND COMMUNICATIONS. We are definitely having an
effect. Jerry Grey, Administrator for Public Policy of AIAA,
told Mark Hopkins that L-5 letters and mailgrams and telephone
calls over the weekend following the Chicago Sun Times "Black
Book" (deliberate) leak saved at least $150 million for the
space program, possibly a lot more.
6. Ad astra...
∂07-Mar-81 0058 POURNE at MIT-MC (Jerry E. Pournelle) more policy development
Date: 7 MAR 1981 0358-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: more policy development
To: JMC at MIT-MC, PDL at MIT-MC, ELLEN at MIT-MC, MINSKY at MIT-MC
To: llw at SU-AI
CC: POURNELLE at MIT-MC
I have been thinking. I suppose I should cure that. Anyway,
the idea is, Reagan has a real problem: many of the people he
needs to for policy development don't want to work for him.
Make that, don't want to work for the gov't. People like Haig
and Reagan and such can afford it because they are rich; but you
and I can't live on what the gov't pays and we don't want to
interrupt our lives to go to Foogy Bottom anyway.
This creates some real problems at the level between the
cabinet and junior officers. They recognize they have those
problems, and tend to use Citizen's Advisory Groups like the one
on Space that I created out of thin air.
It occurs to me that we could show how this could be
regularized: tht is, for the cost of a terminal ($2000 surely
not a lot more); about $1000 for transportation for a single
annual face-to-face meeting; another $2000 for living expenses
while at the meeting (perhaps that's low, I'm thinking of the
Pajaro Dunes plan); and maybe another $1000 in phone bills and
incidentals, making about $6000 per person per year in all (with
$2000 of that being a one-time expense) you have a
"commissioner". Multiply by 25 and you have a commission. Add
now about 100,000 for a single professional level staff member
and a single clerical level employee plus overhead, and a
computer account (maybe I'm too low here? ) paid up and for a
quarter million dollars you have a high-level commission. Set
up several. One on military policy (my daughter is a career
army officer, tough as nails, and she's unhappy about a lot of
stuff that seems to defy common sense.
But more than that: you have a group that can do Peer
Review of NSF grant proposals--I am going to propose that "peer"
be redefined as including a couple of qualified people OUTSIDE
the relevent grant field; if you can't convince them that your
proposal makes sense and ought to be done at taxpayer money
borrowed at 20% interest, then...
and indeed a mechanism for getting a lot of very high
wuality advice from people who will do it for expense and the
chance to influence policy--provided of course that htey believe
they have a chance to influence policy. But since that's what
you want them for...
I mean a lot of us pound on this computer net for
semi-social reasons anyway. My scheme ha the problem that
nothing sensitive can work on it since you will get gratuitous
interlopers and such like, but so what? Grant proposals are
stripped of names anyway (aren't they)? And commissions are not
the appropriate mechanism for studying highly classified
projects anyway. But for overall policy development, a couple
of such commissions, provided with terminals (many already have
them of course)and a secretariat that can be a grad student at
Stanford or MIT or wherever...
Could do a lot of valuable work for the government cheap?
Commentrs appreciated.
∂07-Mar-81 0130 POURNE at MIT-MC (Jerry E. Pournelle) this sounds silly, but think about it...
Date: 7 MAR 1981 0430-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: this sounds silly, but think about it...
To: MINSKY at MIT-MC, JMC at MIT-MC, llw at SU-AI
CC: POURNELLE at MIT-MC
I was talking about space last night to some people, and a chap
asks me "How much would it cost to build a system on the moon
that could send tonne lots to Earth and land them withing, say,
500 feet of the aim point. They could be hard landed..."
I thought about it.
"Could they do it for $40 billion?"
"I expect so," says I. "In fact, yeah, I'm, sure of it."
"Wouldn't that be better than MX? And the cost is what
they say they'll have to spend minimum for MX. The difference
is this gets us on the Moon."
The more I thnk about that, the better the damn thing
looks. Before I make an idiot of myself and propose this to my
buddy in advanced system plans in the five-sided funny farm TELL
ME WHAT IS WRONG WITH IT!!
AFTER ALL, throwing rocks from the Moon, beloved of
science fiction, would certainly do narsty thngs to Soviet
missiles and cities and mines and anything else. And the
ammunition is cheap...
Why do I continue to think of this as science fiction
and not real? Isn't it real? Note I have NOT put this out to
any general mail list on the net.
Jerry
I think you are mistaken in supposing it could be done for 40 billion
dollars. Anyway the conventional wisdom has always been that it is
cheaper to shoot from one place on the earth to another than to keep
the missiles in space or on the moon. Perhaps for this reason, it has
been possible to come to an agreement internationally that "weapons of
mass destruction" would not be kept in space or on the moon or other
bodies in the solar system. I believe that this treaty was what carried
along with it all the genuinely bad stuff about all activity in space
being under government control.
Perhaps the computations "showing" that it was cheaper to shoot from
one place on the earth to another were part of the wishful thinking
of the arms controllers or perhaps they are no longer valid. My guess
is that they are still valid and will remain so for some years.
∂07-Mar-81 0148 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 7 March 1981 04:47-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI, LLW at SU-AI
Yeah, but are you sure? I mean--accounting is strange. IN the
air force planning stuff once I was costing a system and asked
what price was the warhead.
"Don't bother. They're free."
"Don't be ridiculous. Thermonuclear weaopns costs money
and lots of it."
"They don't cost the Air Force anything. They come out
of the AEC budget. Don't rock the goddam boat, cost the system
and assume the warhead's free."
Of course in my Moon Shoot system the warheads are free.
And can you say this is a "weapon of mass destructon"? It's a
mass driver, presumably, or a slinger; and useful for commercial
things; and only if reprogrammed right is it a weapon of mass
dectruction.
∂07-Mar-81 0148 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 7 March 1981 04:48-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI, LLW at SU-AI
I gather the idea of trying to do a science policy isn;t intewresting?
∂07-Mar-81 0153 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 7 March 1981 04:53-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-AI, LLW at SU-AI
BTW, the next meeting of the space advisory council will
probably be in September, about when the budget is pencil but
not ink so we can do things to it. It will be a bit larger
group since lots of people are demanding "seats"...I'll also
trim some off.
One topic for certain: development policy on robotics
and telefactors, with a meaningful research plan, and a
well-written committee report on (1) what we ought to do and (2)
why this is important. I'll assign good writers to help draft a
really persuasive document.
Be thinking about it and maybe doing some draft stuff
for distribution?
I gather by silence that you've no "this is stupid"
objectios to the draft report so that I can take that as my
brief to write the final report?
No, I have no objections to the draft report,
though I have a feeling that if only I thought
about it long enough, it would gel differently. However, I don't have
any real intuition about what your addressees respond to.
Pushing tele-operators is worthwhile, but it needs to be done tactfully.
∂07-Mar-81 0206 Jerry E. Pournelle <POURNE at MIT-MC> science advice
Date: 7 March 1981 05:05-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: science advice
To: JMC at SU-AI
cc: POURNE at MIT-MC, POURNELLE at MIT-MC, MINSKY at MIT-MC,
LLW at SU-AI
1. I completely agree that if we want to have some effect we
should not get involved in the present budget hearings and
consequent mess.
2. I don't know what happened to the TELLER- RAMO axis; my
suspeicion is that they never managed to reduce their advice to
short, readable, comprehensible short documents short enough to
be read shortly. But I don't know.
3. All I do know is that there is a vacuum, and some of the
people in EOP realize there is a vacuum; and for a while at
least we probably have a chance to insert short readable
comprehensible short documents into the system in such a way and
place that we "can be persuasive where persuasion is useful" to
quote the language I used in convening the space council.
I don't guarantee no results, but we have a chance, if
we have some interest. The real questin is, do we have a
science policy to recommend?
∂07-Mar-81 1000 JMC*
Franklin H.
∂07-Mar-81 1442 QCC Kurt
Could you please tell me what Kurt's account name is? I want to
send him a message about the logic exercise.
Colleen Crangle
∂07-Mar-81 2122 NEUMANN at SRI-KL [dkb at nbs-unix: Motels near NBS]
Date: 7 Mar 1981 2057-PST
From: NEUMANN at SRI-KL
Subject: [dkb at nbs-unix: Motels near NBS]
To: VERKshop: ;
Mail-from: ARPAnet host NBS-UNIX rcvd at 6-Mar-81 1214-PST
Date: 6 Mar 1981 at 1513-EST
From: dkb at nbs-unix
Subject: Motels near NBS
To: neumann@sri-kl
-----
As per Steve Walker's request, the following list of Motels are
suggested for your meeting in April. As they fill up often, urge
the participants to get resrvations early.
Ramada Inn; 2 exits south of NBS on Rt. 270; 30l-424-4940.
Holiday Inn; NBS exit on Rt. 270 and 1 mile East; 301-948-8900.
Washingtonian; 1 exit south of NBS on Rt. 270; 301-948-2200.
Potomac Sheraton; 1 exit south of NBS on Rt. 270; 301-840-0200.
The rates all run close to $50 per night plus 10% tax. The Washingtonian
is somewhat older and cheaper.
-----
---------------
-------
∂08-Mar-81 0753 PDL at MIT-DMS (P. David Lebling) [Re: more policy development]
Date: 8 Mar 1981 1052-EST
From: PDL at MIT-DMS (P. David Lebling)
To: POURNE at MIT-MC
Cc: JMC at MIT-MC, ELLEN at MIT-MC, MINSKY at MIT-MC,
llw at SU-AI, POURNELLE at MIT-MC
In-reply-to: Message of 07 Mar 81 at 0358 EST by POURNE@MIT-MC
Subject: [Re: more policy development]
Message-id: <[MIT-DMS].189068>
$10,000 for an account is probably high. Not long long ago BBN was
selling an unlimited-time account for $30,000/yr. Some TOPS-20
time is sold for $200/cpu-hour.
∂08-Mar-81 0900 JMC*
Colmerauer
∂08-Mar-81 1442 CLT
I will be at HP for a while
∂08-Mar-81 1745 DEK via SU-TIP datamedia
The TEX project has ordered two datamedias with the SAIL keyboard, but
we were told that there would be a 2-3 months' waiting time.
Rumor has it that one of your projects has at least one spare. We have an
urgent need for one, and would make good use of two, so we would like
to rent yours until ours arrive. Ideally we want one to be
available right away upon my return a week from Monday. Please let me know
if you have one (or two) available.
∂08-Mar-81 1909 LLW Swift (If Not Prompt) Deliveries
To: pourne at MIT-MC
CC: LLW at SU-AI, minsky at MIT-MC, JMC at SU-AI
∂07-Mar-81 0130 POURNE at MIT-MC (Jerry E. Pournelle) this sounds silly, but think about it...
Date: 7 MAR 1981 0430-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
Subject: this sounds silly, but think about it...
To: MINSKY at MIT-MC, JMC at MIT-MC, llw at SU-AI
CC: POURNELLE at MIT-MC
I was talking about space last night to some people, and a chap
asks me "How much would it cost to build a system on the moon
that could send tonne lots to Earth and land them withing, say,
500 feet of the aim point. They could be hard landed..."
I thought about it.
"Could they do it for $40 billion?"
"I expect so," says I. "In fact, yeah, I'm, sure of it."
"Wouldn't that be better than MX? And the cost is what
they say they'll have to spend minimum for MX. The difference
is this gets us on the Moon."
The more I thnk about that, the better the damn thing
looks. Before I make an idiot of myself and propose this to my
buddy in advanced system plans in the five-sided funny farm TELL
ME WHAT IS WRONG WITH IT!!
AFTER ALL, throwing rocks from the Moon, beloved of
science fiction, would certainly do narsty thngs to Soviet
missiles and cities and mines and anything else. And the
ammunition is cheap...
Why do I continue to think of this as science fiction
and not real? Isn't it real? Note I have NOT put this out to
any general mail list on the net.
Jerry
[There are only two reasons why it's not a great idea: it has a day-scale
latency from lunar launch to earth landing for any reasonable launch
velocity (over minimum)--which is awkward in those situations in which
sub-hour delivery carries a stiff premium--and you have to ask for ten
thousand tonnes coming in at earth escape velocity if you're going to
accept 500 foot miss distances and still expect to be paid for delivery
(assuming that you can somehow convince your customer to accept the goods
a day after the party has started); a one tonne packet would need
bulls-eye level terminal homing, which is slightly tricky at present
technological levels. However, I'd love to see the kinks ironed out of
such a proposal, as it would be great to convince somebody to pick up the
tab for its implementation. Alternatively, we could market the metal
thread Squirt Gun for related objectives, and then use it for providing
the necessary materials to our space colonies--including the lunar ones.
As OTA and others continue to remind me, merely sending a copy of your
note to yourself and myself is completely equivalent to putting it out
to any and all interested parties on the Net.]
∂08-Mar-81 1920 LLW Frank Press As Science Adviser
To: pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI
∂07-Mar-81 0140 JMC science advice
To: pourne at MIT-MC
CC: LLW at SU-AI
There is a long established collection of advisers to all administrations
on science policy. This establishment, in my opinion, has greatly damaged
its pull by allowing itself to be co-opted by the liberals. Appointing
Electing Frank Press, Carter's Scientific Adviser, President of the National
Academy of Sciences, would have been dumb even if Carter had been re-elected,
since NAS is supposed to give non-partisan advice when this is requested.
Now it will be disastrous for the "scientific advice establishment", because
it seems to make the NAS a part of the Democratic "government in exile".
Even if the present Administration didn't already distrust scientists perhaps
even more that the Nixon Administration did, they would have good reason to
fear that if they asked for advice on any but the most purely technical issues,
what they got in return might be a part of the next or last Democratic platform.
During the campaign and the transition, there apparently arose a Republican
advisory establishment with people like Teller and Ramo. However, even
this group seems to no longer have any inside track. Perhaps the suspicion
of scientists was too deep or perhaps the group used up its credibility
in order to save some programs or whatever. Perhaps I'm mistaken in my
belief that this group was shunted aside, but one of its prominent members
told me that "It doesn't seem to have amounted to much".
In spite of all this, issues with a large scientific component will continue
to arise in the Government, and the Government will need scientific advice.
If your contacts will facilitate their getting it, this will be good. It might
be best to start after the budget cuts have been made effective. My own opinion
is that most of the NSF cuts are in the right areas, but social science should
not be quite so heavily cut. Anyway it might be best for some group to stay
away from budgetary issues.
[I'm afraid that John has spoken to several of the basic science policy matters
very accurately. I agree with his high-level conclusions quite completely.]
∂08-Mar-81 1932 LLW Really Heavy Lifting
To: minsky at MIT-ML
CC: LLW at SU-AI, pourne at MIT-MC, JMC at SU-AI
∂07-Mar-81 0900 minsky at MIT-ML (Marvin Minsky)
Date: 7 MAR 1981 1153-EST
From: minsky at MIT-ML (Marvin Minsky)
Sent-by: ←←←020 at MIT-ML
To: MINSKY at MIT-ML, llw at SU-AI, jcm at SU-AI, pourne at MIT-MC
LLW has worked out low-cost schemes for propelling huge payloads
using nuclear cannons, at presumably low cost. I gather
the details of the schemes are secret.
[Actually, Rod Hyde and Mike Roberts worked out many of the pertinent
details--I merely suggested that they do so. The details of how the
nuclear explosive generates the cannon's hot gas are probably classified
(I never bothered to ask, as I didn't want to know), but all the rest
of the scheme set features are quite certainly unclassified.]
Since nuclear applications are against current social policies, such ideas
could be used only in an "exceptional case". For example, if there was a
generally exciting reason to build a large moon base -- say for the
purpose of getting lumar meterials for some SPS-like energy application,
then the public might accet the idea that we could save 100,000,000,000 by
hard-landing all the terrestrial supplies need for that factory, perhaps
in one shot. A single such event, landing, say, 20,000 tons of supplies
on a moon base, might be socially acceptable at the cost of one negligably
radioactive earth-site. Possibly it would save all the cost of developing
a heavy launch lift technology.
[Exactly! Unfortunately, the savings get really huge in absolute dollars
only when the scales get very large, and only huge absolute savings will
crush the political barriers.]
While the political climate is quite unfavorable to nuclear applications,
a turn-about is possible - e.g. in connection with crash energy programs
caused by an oil cut-off or even through the development of the climate of
opinion, possibly helped by white hat propaganda. Our time will come.
∂08-Mar-81 1924 LLW Discussion
To: JMC
CC: LLW
∂07-Mar-81 0202 JMC conversation
When convenient I would like to talk with you about 1) the application I
sent you, 2) whether you think Jerry Pournelle can do good things or is
likely to merely spin wheels.
[I just got back from being in the East since last Monday. I'll call you
tomorrow. ]
∂08-Mar-81 1923 LLW Tactful Teleoperators, Et Al
To: pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI, MM at SU-AI
∂07-Mar-81 0200 JMC
To: pourne at MIT-MC
CC: minsky at MIT-AI, LLW at SU-AI
No, I have no objections to the draft report,
though I have a feeling that if only I thought
about it long enough, it would gel differently. However, I don't have
any real intuition about what your addressees respond to.
Pushing tele-operators is worthwhile, but it needs to be done tactfully.
[With the emphasis on tact--my impression is that axe-grinding in such
contexts must be kept to a minimum, if the grinding wheel is not to stop
abruptly, and not re-start.]
∂09-Mar-81 0106 Jerry E. Pournelle <POURNE at MIT-MC> Really Heavy Lifting
Date: 9 March 1981 04:06-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Really Heavy Lifting
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-ML, JMC at SU-AI
I gather this nuclear cannon scheme is real; more real than the
tower? Because easier than that tower I can sell a case of
influenza.
Given that the cannon could put a base on the Moon in
one whack, what does that do to the total cost of a system able
to hard-land tonne lots with 500 foot CEP?
∂09-Mar-81 0112 Jerry E. Pournelle <POURNE at MIT-MC> I'm going to say this one more time--
Date: 9 March 1981 04:12-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: I'm going to say this one more time--
To: LLW at SU-AI
cc: POURNE at MIT-MC, JMC at SU-AI, MM at SU-AI
We are not in the business to do self-censorship. That
is--I'll do the censorship of ideas before I pass them on with
any endorsement of mine, and I do in fact use acceptability as a
criterion (one of them) in deciding what to do here. We have at
the moment some good ins at high levels; each time you exercise
one of those information channels you either gain credibility
(anbd thus make it easier to push more into the slot at another
time) or you lose credibility and get a reputation as a time
waster, making the slots close.
It's easier to blow a slot up than to expand it.
BUT: telefactor equipment isn't YOUR idea in the sense
of YOUR proprietorship and thus self-seeking, if I am the one
pushing it. I have no stake in nothing; and it seems to me that
having a waldo inside the 3-mile island containment would be
worth a small sum nowadays...
If we have a good research plan, I can see about pushing
it. If we have none--that is, if I don't know how much money
and effort is involved and who should do the work and how long
it takes (which is the present situation of my knowledge) then I
at least am removed as a factor in trying to get support for
that kind of reserch.
Self-censorship is a brutal thing. The NASA types who
tell me we cannot do wsomething often mean only that they think
we won't be allowed to, and thus they don't want to ask, and
that means their prophecy is 100% correct. We sure won't...if
we don't try.
But note the success of our council report.
∂09-Mar-81 0624 ROD Thesis draft
To: TOB at SU-AI, JMC at SU-AI, csd.lenat at SU-SCORE
There is a draft of my thesis in your MJH mailbox.
It is not complete, but I think it is more than a solid skeleton.
Chapters 2 and 10, sections 5.5 and 9.2 and appendix 3 are essentially unwritten.
Chapters 7 and 8 need a lot of work yet.
The oral exam is on Friday.
Rod
∂09-Mar-81 1000 JMC*
dentist
∂09-Mar-81 1027 FFL
To: JMC, FFL
Lester Dugan called to suggest you see the editorial page of today's
S.F. Chronicle.
∂09-Mar-81 1027 FFL
To: JMC, FFL
Do you want me to make up the announcement of the Thursday seminar?
I have Kurt's abstract.
Yes, please.
∂09-Mar-81 1100 JMC*
Call Mr. He.
∂09-Mar-81 1117 LGC Appointment?
Will you have a half-hour or so to talk with me this afternoon? If so, what
time would be best? -- Lew
I can talk for about 15 min at 4:30.
∂09-Mar-81 1206 Konolige at SRI-KL exercise #3
Date: 9 Mar 1981 1207-PST
From: Konolige at SRI-KL
Subject: exercise #3
To: JJF at SAIL, JAK at SAIL, JXK at SAIL, JMM at SAIL,
To: JSR at SAIL, QST at SAIL, QHY at SAIL, YUE at SAIL,
To: JPM at SAIL, QCB at SAIL, QCC at SAIL, QEH at SAIL,
To: QBK at SAIL, QDM at SAIL, QHP at SAIL, QEP at SAIL,
To: QLP at SAIL, QJP at SAIL, QHS at SAIL, QCT at SAIL,
To: ML at SAIL
cc: konolige, jmc at SAIL
Well, I goofed a little on the first problem, as a number of you
noticed. First, the definition of K as a conjunction of B and "true"
should be an equivalence, not just an implication. Second, in the first
problem, you should assume "-K(John,Z)" and try to prove "-K(John,X)",
where I have used "-" for the negation sign (I'm typing on a Datamedia
terminal). Finally, try to solve the last problem using FOL. I may
give a hint later in the week, if people are having problems. --kk
-------
∂09-Mar-81 1220 LGC Appointment Time
As things now stand, I'll be able able to make it after 4 pm; would 4:30
be a good time?
∂09-Mar-81 1327 RWW
COMMENT |
The syntax for this command is:
CIRCUMSCRIBE NAME < ... <predsym> <predpar> <varlist> ... > IN <vl>;
The result is to produce an axiom
Below is a working example
This is now on the system.
PS THIS MESSAGE CAN BE FETCHED THE PREVIOUS ONE CAN'T!!! |
DECLARE INDVAR x;
DECLARE INDCONST a b;
DECLARE PREDCONST B 1;
DECLARE PREDPAR P 1;
ASSUME B(a)∧B(b);
CIRCUMSCRIBE AX B P x IN ↑;
∧I AX[P←λx.(x=a∨x=b)];
∂09-Mar-81 1431 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) MTC Qual, Sat. May 23
Date: 9 Mar 1981 1414-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: MTC Qual, Sat. May 23
To: PHD-Distribution-list:
cc: ZM at SU-AI, JMC at SU-AI
The MTC Qual has been set for Saturday, May 23.
Syllabus is available in Jacks 210.
Please send message to CSD.Tajnai to sign up.
-------
∂09-Mar-81 1530 LGC Papers
I've just XSPOOLed copies of my IJCAI paper (Epistemology of Commonsense
Reasoning) and one on Causal Explanation, in case you'd like to look them
over before we talk. If not, I'll pick them up when I arrive. -- Lew
∂09-Mar-81 1559 CSD.ULLMAN at SU-SCORE
Date: 9 Mar 1981 1557-PST
From: CSD.ULLMAN at SU-SCORE
To: CSD-Faculty: ;
There will be some dudes from Swedish radio here on March 23 to investigate
the future of computers. Anyone interested in being interviewed please
contact Prof. Young at 7-2873.
-------
That is Professor Ljung.
∂09-Mar-81 1644 CSD.GRIFFITHS at SU-SCORE Re: GRADE
Date: 9 Mar 1981 1641-PST
From: CSD.GRIFFITHS at SU-SCORE
Subject: Re: GRADE
To: JMC at SU-AI
In-Reply-To: Your message of 7-Mar-81 1311-PST
The quarter was Spring 1980. Thanks very much. -Bill Griffiths
-------
∂09-Mar-81 1809 Vaughan Pratt <CSD.PRATT at SU-SCORE> supply and demand
Date: 9 Mar 1981 1806-PST
From: Vaughan Pratt <CSD.PRATT at SU-SCORE>
Subject: supply and demand
To: @sun at SU-AI
I've started a Shasta file, /usr/sun/qstore (for Quartermaster Store)
logging loans and needs of Sun-related equipment. Please let me know
what equipment you could profitably use, for what, starting when. To some
extent this information may sooner or later impact budgets, but for now
it will be used only to help juggle priorities. If you're not sure what
you need, then for now it is very safe to say you need nothing; this will
change when the supply picture improves. Once qstore stabilizes it may
be possible to make commitments with a reasonable degree of confidence.
But prepared for a "no commitment yet" response.
-------
∂09-Mar-81 2045 MINSKY at MIT-ML (Marvin Minsky)
Date: 9 MAR 1981 2338-EST
From: MINSKY at MIT-ML (Marvin Minsky)
To: MINSKY at MIT-ML, llw at SU-AI, JMC at SU-AI, pourne at MIT-MC
Teleoperators.
I really don't know whether to push teleoperators right now..
1. It would be good to have a good one for disaster work, e.g.,
3-mile island. It would take a lot of engineering to make one that
would work in that bad environment, but I have an idea of how to do
it. (Building all the motors into one unit that is sealed, except for
pull-cables, that pull through bellows-diaphragms.) I now believe
that 3-mile will not be fixed for several years, the costs will be
larger than they think -- e.g., if twice, that's 4 billion -- and that
they might even give up and consider pouring concrete in, or something
-- just to save money.
The problem is that disaster preparation is unpopular.
2. Better teleoperators for space is obviously, I think, valuable,
but I have the feeling that right now we have a problem of proving
space is of value in the first place. Perhaps it would confuse and
disturb people to try to think this new way. For SPS-like operations
it could be decisive, I think.
By the way, one can't argue that telepresence reduces SPS costs
decisively by reducing life-support costs for man in space, since the
cost of lifting life-support for a lot of people in space is small
when one talks of 100000 tons for a space-manufacturing facility of
bunch of SPS's. I would argue instead that the usual estimates of
manned productivity around large space structures are poor; that
people will always be getting into trouble, etc., that one
teleoperator will do the work of perhaps 20 men (because of 24-hour
workload, and better access).
3. Teleoperators for mining and industry are also valuable, I'm sure,
but perhaps the time isn't ripe for this right now because it is so
sure to get involved with (wrong and irrelevant) employment issues.
Their use for safety and hazard avoidance would be, I think, decisive
-- but again, perhaps not this year.
So, except for the Nuclear application, perhaps we shouldn't try to
push this one in our hopeful role as Science Policy inciters -- except
to propose a healthy budget for it, when the opportunity arises, in
connection with energy and productivity research. I hope that
opportunity arises
In any case, JEP and I plan to write a popular book on telepresence,
and to get the idea generally well-known that way. It may be that
this will have the desired effect anyway. For we can expect a lot of
people to read the whole 200-odd page thing for fun and excitment, who
would never get through a 10 page technical suggestion document. We
expect this book to be pretty popular, and will be asking for comments
and suggestions when we get into it. This might be an excellent
medium for ideas and hobby-horses that you don't want to kill by
prematurely trying to find federal slots for. Our book will include a
variety of other futuristic ideas.
-- marvin
Well I don't think you'll have too much trouble about tele-operators
from the workers. I think the miners will like the idea of operating
the mine from the outside.
There are a lot of military applications of tele-operators and other
forms of telepresence.
∂10-Mar-81 0634 Darden@SUMEX-AIM Re: Washington trip
Date: 10 Mar 1981 0631-PST
From: Darden@SUMEX-AIM
Subject: Re: Washington trip
To: JMC@SU-AI
In response to your message sent 08 Mar 1981 0009-PST
Nice to hear from you John. Bruce sent me a copy of the INTERLISP manual
and I have been slowly working my way through Weissman's LISP 1.5 PRIMER.
No, I didn't receive your book.
How exciting to be going to France. I haven't been since 1971.
How would you like to come out to visit me on Sunday afternoon,
March 21? We could use my terminal and if the weather is nice go for
a walk at Greenbelt lake. Then we could plan to have dinner here and
if you don't mind, I could invite some friends to join us.
I hope by the 21st to have a paper finished that I am working
on for the Boston colloquium in philosophy of science in April. If it is
indeed done, this would be a little celebration. Do you like chicken
with tarragon?
A tout a l'heure. Bon voyage.
-------
C'est un tres bon plan. Je pense que je serai a l'hotel Mayflower.
∂10-Mar-81 0848 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Date: 10 Mar 1981 0844-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
To: CSD-Faculty:
As Chairman of the Department Library and Publications Committee, I
received a memo from Harry Llull, Librarian of the Mathematical
and Computer Science Library, concerning the heat and ventilation
problems in the library.
He is asking our support in attempting to solve the problem of excessive
heat.
"If you have limited your use of the library or avoided using the
library because of the heat, such information would be most helpful."
Send comments to CSD.Tajnai.
-------
∂10-Mar-81 1259 SJF Reference on Lispki's work on nulls
"On Semantic Issues Connected with Incomplete Information" by Witold Lipski,
in the September 1979 issues of ACM Transactions on Database Systems
(Volume 4, Number 3), pages 262-297.
Using modal logic as a basis, Lipski deals with a generalization of nulls,
allowing specification of incomplete information--that is, specification
that values come from particular subsets of their domains. External and
internal interpretations are provided, as is an axiom system for
transformations of queries. Answers to queries are defined in this
framework.
Shel
∂10-Mar-81 1300 JMC*
meindl
∂10-Mar-81 1300 JMC*
change grades Griffiths and other.
∂10-Mar-81 1305 CSD.ADAMS at SU-SCORE (lia) comprehensive exam
Date: 10 Mar 1981 1305-PST
From: CSD.ADAMS at SU-SCORE (lia)
Subject: comprehensive exam
To: jmc at SU-AI
cc: csd.adams at SU-SCORE
I spoke with Carolyn Tajnai today about the comp, and she tells me
that there is no rule for permitting extensions on the comp -- at
least none that applies to me. So, my two choices seem to be
(1) taking the exam in May, and leaving my fate to the mercy of
the faculty on Black Friday if I don't pass, or
(2) taking a leave of absence for Spring term, thus granting myself
a de facto extension (and endangering the tenure of my fellowship).
Did your talk with Denny bring out any other options or possibilities?
If so, I would be interested to hear about them. Thank you.
Lia Adams
-------
∂10-Mar-81 1343 KAPLAN at SRI-KL new DARPA proposal
Date: 10 Mar 1981 1327-PST
From: KAPLAN at SRI-KL
Subject: new DARPA proposal
To: jmc at SU-AI
cc: wiederhold at SUMEX-AIM, kaplan at SRI-KL
We have completed a draft (with estimated budget) of our portion of the new
DARPA proposal. If you are going to visit Washington, do you want to take
a copy of the draft?
We would like to forward it to Bob Englemore in advance for comments, at
an appropriate time (soon). Gio is planning to stop by DARPA next week.
Jerry Kaplan
-------
∂11-Mar-81 0035 SAM
there is a dog right outside m.j.hall .wonder it is the same dog
∂11-Mar-81 0202 SJF Lipski
The Jan 81 JACM has another related paper by Lipski, "On Databases with
Incomplete Information".
Shel
∂11-Mar-81 0530 Darden@SUMEX-AIM (Response to message)
Date: 11 Mar 1981 0524-PST
From: Darden@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI
In response to your message sent 10 Mar 1981 1033-PST
Bon. J'attendrai votre appel telephonique a Dimanche. Encore,
mon number est 474-0037. Avoir, Lindley.
(Do French terminals have accent marks?)
-------
C'est le dimanche 22 mars. Je regarderai des terminals francais, et
je vous informerai.
∂11-Mar-81 1059 ZM
To: JMC, FFL
Robert Grant called. 9 328 6561.
∂11-Mar-81 1155 Wiederhold@SUMEX-AIM Visit to washington and discussion about KBMS
Date: 11 Mar 1981 1149-PST
From: Wiederhold@SUMEX-AIM
Subject: Visit to washington and discussion about KBMS
To: engelmore@ISI
cc: jmc@SAIL, kaplan@SRI-KL
The workload in the Health care Technology study section, which brings
me to Washington, has increased to the extent that I will
also have to be there Wednesday morning ( March 18). That would make seeing
you slightly easier on Wednesday, but if your schedule is set I can
see you at the arranged time (Tuesday, 17 3:30 pm ) as well.
It is for me a tradeoff of inconvenience ( getting back and forth ) versus
boredom ( 2 1/2 days of proposal evaluation is really too much).
If I don't hear from you I'll stick to our earlier schedule.
Thanks Gio
-------
∂11-Mar-81 1303 KAPLAN at SRI-KL KBMS portion of new DARPA proposal
Date: 11 Mar 1981 1300-PST
From: KAPLAN at SRI-KL
Subject: KBMS portion of new DARPA proposal
To: englemore at USC-ISI
cc: Wiederhold at SUMEX-AIM, kaplan at SRI-KL, jmc at SU-AI
Dear Bob:
We have completed a draft of our portion of the new DARPA proposal, on
management of distributed knowledge, along with a rough budget. As Gio
will be coming out next week to discuss this with you, I have forwarded
to you a hardcopy via regular mail, and will also send a netmail copy (in
the next message).
Please let me know if you would like any additional information in
advance of Gio's visit.
Jerry
-------
∂11-Mar-81 1555 RWW fol
One reason that a new system is needed is that MACLISP (or any other
LISP ) seems to me to be unsatisfactory to use as a basis of a
full fledged mathematical theory of programming. It is not LISP
that is at fault. LISP is fine. It is the feeling that the
existing implementations are not very faithful to a reasonable
theoretic treatment. This is what led CLT and myself to think
about programming language design. The importance of this cannot
be underestimated. A reasoning system that will reason about
programs needs a corresponding programming language implementation
that is not ad hoc but has a plausable MTC. Jussi's system will
record proofs ok and will appear more programmable than FOL, but
its dependence on MACLISP means that verifying its own activities
will be almost impossible. It is this problem that I want to avoid.
Richard
∂11-Mar-81 2140 LLW Teleoperators For Dirty Businesses
To: minsky at MIT-ML, pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI
∂09-Mar-81 2101 JMC
To: minsky at MIT-ML, LLW at SU-AI, pourne at MIT-MC
Well I don't think you'll have too much trouble about tele-operators
from the workers. I think the miners will like the idea of operating
the mine from the outside.
There are a lot of military applications of tele-operators and other
forms of telepresence.
[Again, I strongly suspect that John is right. Coal mines are generally
nasty places, as judged even by coal miners (try Orwell's `Homage to
Catalonia' for a gripping description). Indeed, teleoperators would
be highly valued in many military operations--their parents don't vote,
they don't get tired or terrified, they don't come home in body bags, etc.]
"The Road to Wigan Pier" is the one with the description of coal mining.
∂11-Mar-81 2226 LLW Peddling Blue Sky
To: pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI, minsky at MIT-ML, RAH at SU-AI
∂09-Mar-81 0106 Jerry E. Pournelle <POURNE at MIT-MC> Really Heavy Lifting
Date: 9 March 1981 04:06-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Really Heavy Lifting
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-ML, JMC at SU-AI
I gather this nuclear cannon scheme is real; more real than the
tower? Because easier than that tower I can sell a case of
influenza.
Given that the cannon could put a base on the Moon in
one whack, what does that do to the total cost of a system able
to hard-land tonne lots with 500 foot CEP?
[The nuclear cannon scheme is as real as a calculated-but-untested
idea can be; I scaled the numbers in the write-up which you have
from detailed calculations done for more modest (and somewhat more
military) payloads by Hyde and Roberts, using self-evidently
correct scaling laws.
Don't despair of becoming enthusiastic about the Star Bridge--you'll
find that its potentialities will sell it to you quite compellingly,
and you'll come to realize that it'll be cheaper and easier to do
than Apollo was!]
∂11-Mar-81 2220 LLW Naming Spades
To: pourne at MIT-MC
CC: LLW at SU-AI, JMC at SU-AI, minsky at MIT-ML
∂09-Mar-81 0112 Jerry E. Pournelle <POURNE at MIT-MC> I'm going to say this one more time--
Date: 9 March 1981 04:12-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: I'm going to say this one more time--
To: LLW at SU-AI
cc: POURNE at MIT-MC, JMC at SU-AI, MM at SU-AI
We are not in the business to do self-censorship. That
is--I'll do the censorship of ideas before I pass them on with
any endorsement of mine, and I do in fact use acceptability as a
criterion (one of them) in deciding what to do here. We have at
the moment some good ins at high levels; each time you exercise
one of those information channels you either gain credibility
(anbd thus make it easier to push more into the slot at another
time) or you lose credibility and get a reputation as a time
waster, making the slots close.
It's easier to blow a slot up than to expand it.
BUT: telefactor equipment isn't YOUR idea in the sense
of YOUR proprietorship and thus self-seeking, if I am the one
pushing it. I have no stake in nothing; and it seems to me that
having a waldo inside the 3-mile island containment would be
worth a small sum nowadays...
If we have a good research plan, I can see about pushing
it. If we have none--that is, if I don't know how much money
and effort is involved and who should do the work and how long
it takes (which is the present situation of my knowledge) then I
at least am removed as a factor in trying to get support for
that kind of reserch.
Self-censorship is a brutal thing. The NASA types who
tell me we cannot do wsomething often mean only that they think
we won't be allowed to, and thus they don't want to ask, and
that means their prophecy is 100% correct. We sure won't...if
we don't try.
[The NASA types of whom you speak have exact analogs in DoD, DoE,
and several other Government operations which I pass through from
time to time--their common property is known in the vulgate as
"cowardice," and it is *the* occupational disease of bureaucrats.]
But note the success of our council report.
∂11-Mar-81 2149 LLW Teleoperating With Public Funds
To: minsky at MIT-ML
CC: LLW at SU-AI, pourne at MIT-MC, JMC at SU-AI
∂09-Mar-81 2045 MINSKY at MIT-ML (Marvin Minsky)
Date: 9 MAR 1981 2338-EST
From: MINSKY at MIT-ML (Marvin Minsky)
To: MINSKY at MIT-ML, llw at SU-AI, JMC at SU-AI, pourne at MIT-MC
Teleoperators.
I really don't know whether to push teleoperators right now..
1. It would be good to have a good one for disaster work, e.g.,
3-mile island. It would take a lot of engineering to make one that
would work in that bad environment, but I have an idea of how to do
it. (Building all the motors into one unit that is sealed, except for
pull-cables, that pull through bellows-diaphragms.) I now believe
that 3-mile will not be fixed for several years, the costs will be
larger than they think -- e.g., if twice, that's 4 billion -- and that
they might even give up and consider pouring concrete in, or something
-- just to save money.
The problem is that disaster preparation is unpopular.
[More importantly, unlike other unpopular social activities (e.g., war),
it's also very poorly funded.]
2. Better teleoperators for space is obviously, I think, valuable,
but I have the feeling that right now we have a problem of proving
space is of value in the first place. Perhaps it would confuse and
disturb people to try to think this new way. For SPS-like operations
it could be decisive, I think.
[Indeed!]
By the way, one can't argue that telepresence reduces SPS costs
decisively by reducing life-support costs for man in space, since the
cost of lifting life-support for a lot of people in space is small
when one talks of 100000 tons for a space-manufacturing facility of
bunch of SPS's. I would argue instead that the usual estimates of
manned productivity around large space structures are poor; that
people will always be getting into trouble, etc., that one
teleoperator will do the work of perhaps 20 men (because of 24-hour
workload, and better access).
[Your first point is quite correct, but very little realized. Your
latter point won't sell against NASA's extensive studies and simulations
in this area and its institutional biases. The incomplete human-ness
of teleoperators will be the probable response of the more thoughtful
of those disagreeing with you re their cost-effectiveness in man-units.]
3. Teleoperators for mining and industry are also valuable, I'm sure,
but perhaps the time isn't ripe for this right now because it is so
sure to get involved with (wrong and irrelevant) employment issues.
Their use for safety and hazard avoidance would be, I think, decisive
-- but again, perhaps not this year.
So, except for the Nuclear application, perhaps we shouldn't try to
push this one in our hopeful role as Science Policy inciters -- except
to propose a healthy budget for it, when the opportunity arises, in
connection with energy and productivity research. I hope that
opportunity arises
In any case, JEP and I plan to write a popular book on telepresence,
and to get the idea generally well-known that way. It may be that
this will have the desired effect anyway. For we can expect a lot of
people to read the whole 200-odd page thing for fun and excitment, who
would never get through a 10 page technical suggestion document. We
expect this book to be pretty popular, and will be asking for comments
and suggestions when we get into it. This might be an excellent
medium for ideas and hobby-horses that you don't want to kill by
prematurely trying to find federal slots for. Our book will include a
variety of other futuristic ideas.
[I'm pleased that the concept will get a public hearing. Why don't you
also consider a SciAm article as a distilled-down companion piece?--you'll
get a much wider readership of the type of people who you will want to
get thinking about the subject.]
-- marvin
∂12-Mar-81 0114 Jerry E. Pournelle <POURNE at MIT-MC> Teleoperators For Dirty Businesses
Date: 12 March 1981 04:13-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Teleoperators For Dirty Businesses
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-ML, JMC at SU-AI
SEE, WE MAKE PROGRESS. Now what I need is a horse: namely, what
would be a rational plan for studying teleoperations: who gets
how much money, for what institutions; who is the monitoring
agency, and should NSF or Aerospace or what be the parent
funding group. And like that.
The more I understand about what ought to be done, the
better chance of telling someone who ought to know...
∂12-Mar-81 0126 POURNE at MIT-MC (Jerry E. Pournelle)
Date: 12 MAR 1981 0419-EST
From: POURNE at MIT-MC (Jerry E. Pournelle)
To: JMC at SU-AI
Date: 09 Mar 1981 2312-PST
From: John McCarthy <JMC at SU-AI>
I will bet that the current state of the art of interfacing brains to
computers is not up to storing even one bit in the brain for an hour
and retrieving it later other than by some gross form of cheating that
doesn't use the brain's own memory storage mechanisms - like putting
high currents between pairs of electrodes and using the bit of brain
in between as a fuse. While I am not up on the neurophysiology
literature, I read Science and Science News, and anyone who had done
even that much would excite a lot of interest.
------------------------------
--------
spoilsport
∂12-Mar-81 0838 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
There will be talk on Monday, March 16, in 402 Jacks Hall, at 4 p.m.
Speaker: Andrei P. Ershov
of the Computation Center, Novosibirsk, U.S.S.R.
Subject: Abstract Computability in Arbitrary Domains and Bases
∂12-Mar-81 0910 FFL
To: JMC, FFL
Jergen Retz of IBM Germany called. Is coming with his manager, Dr. Endres,
to a meeting in San Jose next week. Would like to see you on Tuesday,
3 p.m., March l7. Will check on this when he arrives in San Jose.
∂12-Mar-81 0936 FFL
To: JMC, FFL
Leave SFO via TWA 842 at 8:40 a.m. Arrive Marseilles (via Paris(
on Air Inter 6263 at ll a.m. on the l9th. Leaave Marseilles on Air Inter 6564
on the 21st, change in Paris to Air France 8l2, to London. Leave London
on Pan Am 107 at 2:50 p.m. and arrive in Washington at 6:10 p.m. the 21st.
Leave Washington the 23rd on TWA 63 and arrive SFO at 8:25 p.m. You have
your reservations at the Mayflower. Your ticket and TWA boarding passes will
be delivered today or tomorrow.
∂12-Mar-81 0941 FFL
To: JMC, FFL
I am going to a class this morning and will not be back in the oofice
until l p.m.
∂12-Mar-81 0945 Konolige at SRI-KL seminar today
Date: 12 Mar 1981 0945-PST
From: Konolige at SRI-KL
Subject: seminar today
To: jmc at SAIL
cc: konolige
John, I haven't seen any notices about the startup of the
knowledge/action seminar on the distribution lists I'm on. Is it
still set for today at 4:15? Being as its dead week, I'm willing to
hold off until the start of next quarter. --kk
-------
∂12-Mar-81 1012 Konolige at SRI-KL (Kurt Konolige) Re: seminar notice
Date: 12 Mar 1981 1001-PST
From: Konolige at SRI-KL (Kurt Konolige)
Subject: Re: seminar notice
To: JMC at SU-AI
In-Reply-To: Your message of 12-Mar-81 0955-PST
Richard Waldinger remembers it being announced with Reiter's
talk last week; but there was no abstract (I didn't get it to Fran
until Friday). If it's still on, you might send out a notice today
on the system, along with the abstract. --kk
-------
∂12-Mar-81 1346 LLW The Simple Joys of Coal Mining
To: JMC
CC: LLW
∂11-Mar-81 2233 JMC
"The Road to Wigan Pier" is the one with the description of coal mining.
[You're absolutely right, of course--I have been unmasked as an Orwell
anthology reader, and one suffering from senile memory failure, at that.
My only exculpatory plea is that I read the volume in question a quarter
century ago, under quite non-leisurely circumstances (Orwell wasn't
politically fashionable in those days, and volumes of his works were
only transiently accessible to the high school crowd of which I was
then a member).]
∂12-Mar-81 1352 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
The speaker for the Knowledge and Action Seminar this week, on Thursday,
March l2, at 4:15 p.m., in Room 301, Jacks Hall, will be Kurt Konolige.
TOWARDS A LOGICAL RECONSTRUCTION OF DATABASE THEORY
Abstract
I am interested in constructing a computer agent whose behavior will be
intelligent enough to perform cooperative tasks involving other agents
like itself. One of the key tasks such an agent must perform is to form
plans to carry out its intentions in a complex world in which other
planning agents also exist. To construct such agents, it will be necessary
to address a number of issues that concern the interaction of knowledge,
actions, and planning. Briefly stated, an agent at planning time must take
into account what his future states of knowledge will be if he is to form
plans that he can execute; and if he must incorporate the plans of other
agents into his own, then he must also be able to reason about the
knowledge and plans of other agents in an appropriate way.
Formalisms for both knowledge and action separately have been examined in
some depth, but there have been few attempts at a synthesis. The exception
to this is Bob Moore's thesis on reasoning about knowledge and action, for
which Doug Appelt has recently written a planning system. Moore shows how
a formalism based on possible-world semantics can be usedd to reason about
the interaction of knowledge and action. In this talk, I will present an
alternative formalism for reasoning about knowledge, belief, and action that
is based on a proof-theoretic, rather than possible-world formalism; it is
similar to McCarthy's "Individual Concepts and Propositions" approach. I
will show how this formalism can be integrated with a planning formalism,
for use by a plan-constructing system.
∂12-Mar-81 1359 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
Konolige's topic today will be
"A First Order Formalization of Knowledge and Action for a Multiagent
Planning System"
∂12-Mar-81 1354 QCC Paper for Epistemelogigal Problems in AI course
I don't think I will have my paper finished before you leave on Wednesday.
May I have an "incomplete" to allow me to give you the paper on your return
the following Monday?
Colleen Crangle
Yes, I'll give you the an incomplete.
∂12-Mar-81 1408 LLW Fitting The Punishment To The Crime
To: pourne at MIT-MC
CC: LLW at SU-AI, minsky at MIT-ML, JMC at SU-AI
∂12-Mar-81 0114 Jerry E. Pournelle <POURNE at MIT-MC> Teleoperators For Dirty Businesses
Date: 12 March 1981 04:13-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Teleoperators For Dirty Businesses
To: LLW at SU-AI
cc: POURNE at MIT-MC, minsky at MIT-ML, JMC at SU-AI
SEE, WE MAKE PROGRESS. Now what I need is a horse: namely, what
would be a rational plan for studying teleoperations: who gets
how much money, for what institutions; who is the monitoring
agency, and should NSF or Aerospace or what be the parent
funding group. And like that.
The more I understand about what ought to be done, the
better chance of telling someone who ought to know...
[Clearly, Marvin should be required to pay his own executioner, just like
the Good Ole Days. Make him cough up the details!]
∂12-Mar-81 1634 Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE> directed reading
Date: 12 Mar 1981 1633-PST
From: Paul Sonkowsky <CSD.SONKOWSKY at SU-SCORE>
Subject: directed reading
To: csd.mccarthy at SU-SCORE, csd.schreiber at SU-SCORE, csl.lantz at SU-SCORE
cc: csd.sonkowsky at SU-SCORE
Dear Sirs:
In attempting to pass the comprehensive next quarter, I am signing up
for directed reading (CS 390) next quarter. You were recommended to
me as the ones to sign up under for the areas of Math. Theory of Comp.,
Num. Analysis, and Hardware/Software Systems. Please let me know it
there is something specific I should do, or if I should come talk with
you, etc.
Thank you,
Paul Sonkowsky
-------
I am not in a position to take students for directed reading who merely
want to pass an exam.
∂12-Mar-81 2042 TOB japan trip
John
I am making a brief trip to Japan. I would welcome
any suggestions you have about where to visit
and what objectives while I am there.
Tom
∂12-Mar-81 2159 JMC
To: "@FRLIST.[1,JMC]"
meeting 2pm Monday Mar 16 to discuss ARPA proposal
∂12-Mar-81 2225 TOB
I am going May 21 until June 10 or so.
∂12-Mar-81 2117 JMC
When are you going? I have no special suggestions at present.
∂12-Mar-81 2242 CLT
∂12-Mar-81 2159 JMC
To: "@FRLIST.[1,JMC]"
meeting 2pm Monday Mar 16 to discuss ARPA proposal
I have a dental appt. at 2:30 and won't be able to attend.
If you wish me to do something in advance, I'll be glad to.
OK, thanks. Maybe there'll be something.
∂12-Mar-81 2244 CLT q/x
i noticed FFL left SCOTT.XGP[LET,JMC] in the xgp que
don't know if you wish it flushed in case the xgp comes alive
at some inconvenient time.
∂12-Mar-81 2256 LGC ARPA Meeting
Is the meeting on Monday just of Stanford people, or will an employee of ARPA
be present? If the latter, will it be Engelmore? Since this is the first
I`ve heard of the meeting, I'd like to know more in detail about the agenda,
and what I might be expected to contribute. -- Lew
Only Stanford people. We will discuss informally what might be proposed
to ARPA in the formal reasoning and basic AI area.
∂13-Mar-81 0506 Darden@SUMEX-AIM (Response to message)
Date: 13 Mar 1981 0422-PST
From: Darden@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI
In response to your message sent 11 Mar 1981 1114-PST
D'accord, le dimanche 22 Mars. Encore, bon voyage.
-------
∂13-Mar-81 0900 JMC*
Call levinthal and curtis
∂13-Mar-81 0902 MERRITT at USC-ISIB Re: duplicated messages
Date: 13 Mar 1981 0858-PST
From: MERRITT at USC-ISIB
Subject: Re: duplicated messages
To: POURNE at MIT-MC, VaughanW at HI-MULTICS
cc: ENERGY at MIT-MC, SPACE at MIT-MC
In-Reply-To: Your message of 13-Mar-81 0035-PST
I agree that there is a bit of a pain with duplicated mail, however
I, like most others (I hope), have a mail reader which makes duplications
pretty obvious, and I need not ever read the second copy; just delete it.
I also support the clipping service, and would like more information on
where that is coming from. (I assume that nobody is typing that stuff in)
<>IHM<>
-------
∂13-Mar-81 1023 FFL
To: JMC, FFL
Betty Scott and I have been reviewing the grants and contracts which I
monitor. Your 5l4 grant expires in June. If you wish to do so, a new
proposal should be sent in as quickly as possible. There is insufficient
money to carry beyond June.
755 will end in September, so it is necessary to present a new proposal
for it now if you wish it to continue.
A proposal to continue 514 was submitted in the Fall, and continuing
755 is in the works too.
∂13-Mar-81 1155 FFL
To: JMC, FFL
Prof. Ershov called. He arrives P
PSA 821 at SFO at 10:18 p.m. He is very anxious to be able to identify
the person who is going to pick him up, as he understands it will be a
stranger. He has left a telephone for me to call him to tell him the
name and presumably the description of his chauffeur. Who is it to be,
please?
∂13-Mar-81 1157 FFL
To: JMC, FFL
R. D. Arnold called. You are on his Reading Committee for his thesis.
He wishes to set a date for his oral and suggests May 1. Is this
satisfactory with you? I am to let him know.
∂13-Mar-81 1317 FFL
To: RWW, JMC, FFL
Betty Scott has indicated that the 526 grant will end its second year
August 1981. She tells me that if you do not wish any lapse in funding,
it would be wise to have the progress report and budget for the third
year ready no later than May. The budget projection for the third year
on the original budget projection came to $58,581.
∂13-Mar-81 1506 FFL
To: JMC, FFL
E. Levinthal returned your call. We assumed that you had called about
seeing him when you are in Washington. We said he would be there on the
21st and 22nd. He also is coming to Stanford and would be able to see
you on Monday afternoon, March 30th, if that is satisfactory. He left
his home phone number: 202 733 3360.
∂14-Mar-81 0744 JRA msg
just got msg that you wanted to talk with me? is now convenient?
∂14-Mar-81 0934 QLP how to enter a saved Fol proof in a file ?
After I saved my proof I typed thhe command to enter it to the monitor
but the file entered consisted of a series of αβ⊗ etc. What did I wrong,
please .
∂14-Mar-81 1623 JPM Concerned Scientists Meeting
Yes, I got a notice of the meeting via ID mail. I plan on coming, with
copies of the draft for those articles I told you about concerning
relative energy risks that will appear in the Daily. The first two
articles should be done, with the others in draft or outline stage
(I will not be finalizing them until a week or so before publication,
since I might (probably will!) have to respond to other editorials and
letters on the same topic).
I hope to get some feedback from the people there, as well as to just
attend the meeting to find out what is up from the organizations point
of view.
See you there.
Jim
∂14-Mar-81 1652 Feigenbaum@SUMEX-AIM Re: Andrei Ershov
Date: 14 Mar 1981 1647-PST
From: Feigenbaum@SUMEX-AIM
Subject: Re: Andrei Ershov
To: JMC@SU-AI
In response to your message sent 14 Mar 1981 0735-PST
I just got back from LA myself and have family visiting. Will see Andrei
tomorrow. I'll call Tiki or your house....Ed
-------
∂14-Mar-81 1946 CLT
ive gone to a recital in annenberg
∂15-Mar-81 1139 RPG Talk
Let's get together sometimes this week and discuss my possibilities
at Stanford. Wednesday is bad since I'll be at LLL all day.
-rpg-
I will be away from Wednesday thru the following Monday. Let's talk
between today and Tuesday, because I will be talking to ARPA in
Washington next Monday.
∂15-Mar-81 1158 RPG
∂15-Mar-81 1157 JMC
I will be away from Wednesday thru the following Monday. Let's talk
between today and Tuesday, because I will be talking to ARPA in
Washington next Monday.
Ok, I'm free all day tomorrow and tuesday.
-rpg-
In that case, please come to my office at 1pm tomorrow.
∂15-Mar-81 1204 RPG
∂15-Mar-81 1203 JMC
In that case, please come to my office at 1pm tomorrow.
ok.
∂15-Mar-81 1526 TOB
This is more news about the budget problem as I am told it.
There is $51,000 which was withheld from Image Understanding on what
seems to be a misunderstanding about timing. We thought that we had it,
and just found out that it is not so.
For the first year we were supposed to get 412,719 (including JMC's funds).
We only received 305,000. Engelmore gave a reason for this, which is a
misunderstanding, namely that we started spending later (January 80 instead
of October 79). They would give us the rest (107,000?) at the end of the
contract period. This will be October 81. However our problem is up to
Sept 31, for now, at least.
Tom
I suggest you talk to Engelmore and Druffel about this and make a trip
to Washington if necessary. I guess I can talk about it with them
on March 23 if necessary.
∂15-Mar-81 1548 TOB
John
I have sent a message to Druffel and will call him Monday.
Then we will know whether it will be useful for you to talk with
them on your trip.
Tom
∂15-Mar-81 1543 JMC
I suggest you talk to Engelmore and Druffel about this and make a trip
to Washington if necessary. I guess I can talk about it with them
on March 23 if necessary.
∂16-Mar-81 1052 LGC Pre-meeting Info
∂12-Mar-81 2159 JMC
To: "@FRLIST.[1,JMC]"
meeting 2pm Monday Mar 16 to discuss ARPA proposal
∂12-Mar-81 2256 LGC ARPA Meeting
To: JMC
Is the meeting on Monday just of Stanford people, or will an employee of ARPA
be present? If the latter, will it be Engelmore? Since this is the first
I`ve heard of the meeting, I'd like to know more in detail about the agenda,
and what I might be expected to contribute. -- Lew
∂16-Mar-81 1100 LGC Pre-meeting Info [CORRECTED MSG]
A new section of my paper on commonsense reasoning may be relevant to our
meeting today; you might wish to take a look at it before the meeting; it's in
REAS.TXT[EP,LGC]. Also, I'd like to know when a document for ARPA must be
produced as a result of today's meeting, and what the time frame of our
discussions is likely to be.
∂16-Mar-81 1226 RPG Solicitation
To: lispsources at SU-AI
Now is the time to start thinking of benchmarks. Since I know that
everyone is very busy we ought to start now and with nothing too
difficult: there are still many methodological questions.
First, please send me a description of your LISP, your OS, and
your machine. For well known machines and OS' please state their
names. Please include memory size, speed, cache speed and size,
and disk specs (manufacturer, speed). For your LISP, include
binding type (shallow dynamic, e.g.), ancestry, compiler technology
(tail recursive...), macro style, and anything else you can think of.
Also, please think of and send me some benchmark possibilities. I don't
care that they are well thought out, I need some idea of what people
have in mind. I will collect these ideas and try to get them all running
at SAIL to help debug the benchmarks as benchmarks so that you will
not have to do all this work alone. Any sort of program will do and I
will help prune them down if necessary.
Thank you for your help.
-rpg-
∂16-Mar-81 1307 RPG
∂15-Mar-81 1203 JMC
In that case, please come to my office at 1pm tomorrow.
You were off with Ershov at 1300. I'll be around all day and will wait
for you.
-rpg-
∂16-Mar-81 1404 TOB arpa proposal
John
How are you coming with your ARPA proposal?
Drafts were to be finished Mar 15, to enable
interacting with program managers. Final versions
by April first.
Tom
∂16-Mar-81 1429 HEDRICK at RUTGERS Re: Solicitation
Date: 16 Mar 1981 1725-EST
From: HEDRICK at RUTGERS
Subject: Re: Solicitation
To: RPG at SU-AI
cc: lispsources at SU-AI
In-Reply-To: Your message of 16-Mar-81 1526-EST
ELISP: extended R/UCI lisp. This will be a reimplementation of
Rutgers/UCI lisp for Tops-20 using extended (30-bit) addressing. It is
implemented using typed pointers and a copying GC, but will otherwise be
almost exactly the same as R/UCI lisp (unless you are accustomed to
CDR'ing into the innards of strings, etc.).
hardware - Model B KL processor or Jupiter. I am not clear whether
a 2020 has extended addressing. If so that would also be
usable.
OS - Tops-20, release 5 or later (release 4 useable with minimal
patching)
binding type- shallow dynamic, with same stack mechanisms as
UCI Lisp
compiler - Utah standard lisp transported to our environment
At the moment performance appears to be the same as R/UCI Lisp, except
that the GC takes about twice as long for a given number of CONS cells
in use. The time per CONS may be less for substantial programs, since
we can afford to run with lots of free space, whereas our big programs
are pushing address space, and may not be able to have much free space,
hence GC a lot.
At the moment I have an interpreter that does a substantial part of Lisp
1.6. I hope to finish Lisp 1.6 by the beginning of the summer. I also
hope to have a compiler by then. I am doing the interpreter personally,
and one of my staff is doing the compiler. I am implementing R/UCI
lisp roughly in historical order, i.e. Lisp 1.6 first, then UCI lisp,
then Rutgers changes, though a few later features are slipping in (and
I am not doing anything I will have to undo).
Note that I have little if any interest in performance. I want to match
R/UCI lisp, since users may complain if things suddenly slow down, but
that is about it. I am more concerned about reliability (since I will
have little time to maintain it) and how long it takes to write it
(since I have little time to write it). Our users are doing completely
traditional Lisp work, and have little or no interest in more flexible
binding or control semantics (we supplied a version of R/UCI lisp with
Scheme semantics, and no one was interested), nor in speed in
arithmetic. The system is designed to be modular enough that
improvements can be done as needed. I am giving some thought to
transportability, though not as much as the Utah folks. I think we
should be able to transport it to a system with at least 16 AC's and a
reasonable instruction set (e.g. VAX) with 2 man-months or less.
As far as the hardware we have available for testing, we will shortly
have 1M of MOS memory, 4 RP06's on 2 channel, and a model B KL processor
(the model matters since the model B is faster than the model A. Note
that the processor model number is almost the only variable you care
about in a 20, but it is not derivable from the DEC marketing
designation, since a 2050 or 2040 may be either model. However a 2060
is always model B).
-------
∂16-Mar-81 1516 FWH PV+A Seminar
To: "@SEM.DIS[SEM,VER]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
PLACE: ERL 237
TIME: 2:30 p.m., Tuesday, March 17
SPEAKER: Gerard Guiho, University of Paris (South), France
TITLE: Program Construction (and Validation) from Abstract Data
Type Decompostion.
ABSTRACT:
I present a method of program construction based on "top-down
programming." Each data type relevant to the problem is defined in a
recursive way to obtain decomposition patterns. These patterns
introduce new subproblems, which are solved by using type constructors
or by a new decomposition. This method gives a first program, which is
then refined using classical techniques: program transformation,
representation of one data type by another, etc. A proof of the program
is obtained as the program is being constructed.
I will show how the method applies to one interesting example:
the problem of finding ascending subsequences of maximum length within a
given sequence.
∂16-Mar-81 1521 GFS Ethernet article
To: "@SUN.DIS[P,DOC]" at SU-AI
A friend just told me that last Friday's Wall Street Journal (March 13) had
a front page article on the Ethernet. I would like to borrow that issue from
someone if it's convenient.
∂16-Mar-81 1829 CSD.DOYLE at SU-SCORE proposal revised
Date: 16 Mar 1981 1821-PST
From: CSD.DOYLE at SU-SCORE
Subject: proposal revised
To: jmc at SU-AI
I have modified the proposal i gave you, and it should be
in the Dover D box. You may pick it up there today if you
wish, else I will place it in your mailbox sometime tomorrow.
-------
∂17-Mar-81 0346 DCL ARPA
To: JMC at SU-AI, TOB at SU-AI, GIO at SU-AI
Hows everybody progressing on the amalgamated proposal?
Are we going to meet the first deadline of March 20th for
preliminary draft?
-David
No, we won't meet it. I am seeing Engelmore next Monday to discuss
what work he can support, and our part will be ready shortly thereafter.
∂17-Mar-81 0651 BYY
∂17-Mar-81 0000 JMC*
lunch with jmc at faculty club at noon
Thanks for the reminder. I'll be there. Jon
∂17-Mar-81 1013 NEUMANN at SRI-KL VERkshop II
Date: 17 Mar 1981 0959-PST
From: NEUMANN at SRI-KL
Subject: VERkshop II
To: VERKshop: ;
OK, gang, the time is drawing near to 21-23 April. I hope you are all
seriously engaged in writing up your contributions. I imagine that
the agenda would be somewhat driven by what you submit or bring along,
although oral contributions are of course also appropriate.
Nevertheless, we would like to have some sort of advanced warning as
to what you might like to discuss, in an effort to make the VERkshop
as useful to you as possible.
Bill Wilson has supposedly been collecting your suggestions, although
(1) he has received very little from you, (2) his net connection has
been down much of the time, and (3) we didn't push you very hard. (I
was away for most of the last three weeks.) So please give us an
early idea of whether you plan on attending, what you might expect to
contribute on-line, and even some idea of topics you might wish to
have discussed orally -- if you can think that far ahead.
[Note: Some of you are receiving this although you have already
indicated you would probably not be able to attend. You may happily
ignore our requests, although your comments are of course also
welcome. However, I would appreciate your letting me hear from you,
even if to say that you definitely cannot come, or that you do not
wish to receive the increasingly heavy flurry of mail that can be
expected in the next five weeks. If you cannot come, the default will
be to send you the final preVERkshop distribution of contributions by
by US Dogsledmail, plus any summary material that we might subsequently
produce for the SIGSOFT SoftwareEngineeringNotes.]
Several possible broad agenda items currently contemplated include the
following.
Presentation and discussion of the CSEC charter (Computer
Security Evaluation Center).
Summary progress reports of the last twelve months.
Practical results and user experience with verification systems.
Implementation experience in verification systems.
Policy issues concerning the current state of the art and the future
of verification. What are the visible successes? Are they adequately
represented in the literature? Are the [to-be-distributed] VERkshop II
contributions reflective of the progress? Is there an urgent need
for more solid demonstrations of the practical applicability in the
near future? Is there expected to be adequate funding? Is
verification oversold? undersold? Have DeLiptIs had any
real impact? If that impact is negative, is there any need for a
rebuttal to that impact (other than the implicit rebuttal of some
dramatic success stories)?
To keep things simple, please address all responses to Neumann@SRI-KL.
You might CC: Wilson@RADC-Multics (if you can get through!) on
suggested agenda items, although I can forward to him if you forget.
I will keep Steve informed.
A note to West Coasters: remember TWA's new fare, announced yesterday,
effective 20 April: $298.
[A side note, FYI: As of today, the AI people at SRI are now @SRI-AI
(but residually still also @SRI-KL for a while). The CSL people are
either @SRI-KL or @SRI-CSL (one of our Foonlys, formerly @SRI-F2).
But mail to SRI-KL will be forwarded as necessary in any case.]
See you soon? Peter
-------
∂17-Mar-81 1035 Stan at SRI-AI Common Business Communication Language
Date: 17 Mar 1981 1035-PST
From: Stan at SRI-AI
Subject: Common Business Communication Language
To: JMC at SAIL
cc: nilsson, stan
A few of us at SRI have been having some discussions recently on the
subject of your memo "The Common Business Communication Language."
Among those participating have been Nils Nilsson, Kurt Konolige, Bob
Moore, Doug Appelt, and myself. We have been looking at issues having
to do with how distributed business machines might model one another's
"cognitive state" with the goal of carrying on successful business
transactions, negotiations, etc. This is being done in the context of
SRI's Distributed AI project, which I am leading; as you know both Kurt
and Doug are doing dissertations on related issues.
Nils suggested that you might be interested in becoming involved,
either informally or formally, in this activity. If you were, perhaps
you, Nils and I could get together for a preliminary meeting. (We can
agree on time and place in due course.)
--Stan Rosenschein
-------
Naturally, I am interested and even in getting formally (financially)
involved. I will be away till next Tuesday, so let's arrange a time
for after that.
∂17-Mar-81 1204 RPG Proposal
When we discussed my possible stay at Stanford earlier, I was
under the impression that you had some Lisp based task in mind, so
that the `opus' I was in the middle of was for a global architecture
for a CSD computing system focussing on its Lisp applicability, but
also combing the best from personal and timseshared machines. Hence
it was quit a surprise when yesterday we talked of scientific research
instead of system development research. I've mentioned the grand scheme
to a number of people, including some of the LLL people and they think
it might work out.
In regard to the advice taker, I believe it will take some
time to determine whether our a priori differences of approach will
sink the effort. I found it interesting to be considered for working on
this given that I don't think you know what I did for a thesis.
What I did was basically write a natural language generation
system, except that it was done as an example of a style of large
system organization rather than as a new generation theory. What I mean
by this is that in any real, large AI program there will be millions of
line of code (or millions of sentences if you prefer) and there must be
some way to control the activity of the system, even within itself so that
it has a chance of applying appropriate abilities (or information( at the
right time.
So what I have has 2 basic components: 1. an observational
facility which allows the orgranizer of the system to separate out
facts about simple doing things from facts about how to modify those
simpler things. In the generation system this meant that there were
facts about how to say simple sentences (in English) about things and
there were observers of this activity that proposed collapsings and other
stylist techniques to reduce ambiguity etc. Not that there is some deep
implementational or power advantage, but simply an expressability
advantage. 2. a descriptive facility on top of the representation
system. Thus I had a system for representing facts and strategies (in a
KRL-like manner) and separate from that was a description in some less
structured language (this is important) that described the applicability
in a Planner style. While running a description of the current situation
would be maintained, which was a focus or context mechanism, and when
the equivalent of search was needed, the descriptions of the packages
of information would be matched against the situation to isolate relevant
sub-areas of concern. However, the interesting point was that there
were several other sources of modification on this scheme, namely what I
call ``influences'' and ``soft-constraints''. What these are are user
or system supplied pieces of advice about how a match was to be rated (matches
are rated numerically). What I had in mind was being able to say: ``write about
situation x, but tell me about these things in this order of preference
and here are some English stylistic things to keep in mind (don't use
too many adjectives, don't use words with sexual connotations...)''
Hence, there is a kind of advice that is given about how to
do things or how to judge what is important.
In terms of what I'm willing to give up from this approach, the
fact of KRL-like representation is irrelevant and is a little difficult
to handle, though I'd like to be able to package information in some
interesting way. The numeric results of matching probably should be
evenutally symbolic, though EVALable at some point to provide an explicit
rating. I allow inconsistencies in my descriptive language in order to
allow a continuous change from one point of view (or area of concern)
to another. In English plans there is often no such thing as stating
conclusively some things, but they must be reinforced by re-iteration.
Since the descriptive system is a control system and not a reasoning
system I don't think this is bad. The observational facility is
exactly Richard's concept of meta in FOL (we've discussed it at
length).
So it's interesting that you have advice taking in mind when
that's what my thesis is concerned with to a large extent. However,
can we negotiate or styles? I'm not sure. My educational background
is very formal (mathematical logic, recursive function theorym decision
theory in groups...) but I've strayed educationally and in attitude.
What do you think?
-rpg-
I can't react to this immediately, probably not till I return
next Tuesday. I didn't think at all about the possibility that your
thesis work was connected with the Advice Taker.
∂17-Mar-81 1155 Masinter at PARC-MAXC Re: GC
Date: 17 Mar 1981 11:54 PST
From: Masinter at PARC-MAXC
Subject: Re: GC
In-reply-to: RPG's message of 16 Mar 1981 1234-PST
To: Dick Gabriel <RPG at SU-AI>
cc: LispTiming@su-ai, LispTranslators at SU-AI
Interlisp-D uses a reference-count garbage collection scheme. Thus, "garbage
collection" overhead is distributed to those functions which can modify reference
counts (CONS, RPLACA, etc.) with the following important exceptions:
no reference counts are maintained for small numbers or literal atoms
references from the stack are not counted
Reference counts are maintained in a separate table from the data being counted.
The table can be thought of as a hash table. In addition, the "default" entry in
the table is reference count = 1, so that in the "normal" case, there is no table
entry for a particular datum.
"Garbage collection" then consists of (a) sweeping the stack, marking data with a
"referenced from the stack" bit in the reference count table if necessary, (b)
sweeping the reference count table, collecting those data whose reference counts
are 0 and which are not referenced from the stack.
--------------
Because of this scheme, it is very difficult to measure performance of Interlisp-D
independent of garbage collection, because the overhead for garbage collection is
distributed widely (although the timing for the sweep phase can be separated
out).
Secondly, the choice of a reference count scheme over the traditional
chase-and-mark scheme used by most Lisps was conditioned by the belief that
with very large virtual address spaces, it was unreasonable to require touching
all active storage before any garbage could be collected.
This would indicate that any timings should take into consideration paging
performance as well as garbage collection overhead, if they are to accurately
consider the overall performance picture.
Larry
∂17-Mar-81 1218 RPG Bureaucracy
To: lisptiming at SU-AI
In sending mesages around, the following facts are useful:
RPG is on LISPSOURCES which is equal to
LISPTRANSLATORS, which is a subset of LISPTIMING.
So there is no need to send me a copy of everything, nor
is it necessary to have LISPTIMING and LISPSOURCES on the same
header, for example. Thanks.
-rpg-
∂17-Mar-81 1456 Nilsson at SRI-AI Firdman
Date: 17 Mar 1981 1456-PST
From: Nilsson at SRI-AI
Subject: Firdman
To: JMC at SU-AI, Bobrow at PARC
cc: Nilsson
I just rcvd a phone call from Henry Firdman's mother in Philadelphia.
She said that Henry had just rcvd permission to emigrate and will be
leaving on April 6!
-------
∂17-Mar-81 1529 CSD.ULLMAN at SU-SCORE Re: IBM grant for proving
Date: 17 Mar 1981 1525-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Re: IBM grant for proving
To: JMC at SU-AI
In-Reply-To: Your message of 16-Mar-81 1149-PST
I assume you are presenting me with a bill for about $7K. The TA matter
is something the department should pay out of its usual sources.
Anyway, go do it!
-------
Done. Malik's RA will be charged to the IBM grant for Spring and Summer. Fran
will arrange it with Betty or whomever.
∂17-Mar-81 1628 FFL
To: "@SEMINA.[1,FFL]" at SU-AI
Professor Nachum Dershowitz
Computer Science Department, University of Illinois
will speak on
"Termination of Term-rewriting Systems"
on Thursday, March 19, 2 p.m., 402 Jacks Hall
I'll be away, but I would like a written version of Nachum's talk
if there is one.
∂17-Mar-81 1648 FFL Oral
To: RDA, JMC, FFL
John expects to be in France from around the end of April until July 1.
This makes it difficult for him to schedule the orals meeting. Perhaps
it would be easier to find a replacement for him?
∂17-Mar-81 1921 Bernard S. Greenberg <Greenberg at MIT-Multics> Re: Solicitation
Date: 17 March 1981 2142-est
From: Bernard S. Greenberg <Greenberg at MIT-Multics>
Subject: Re: Solicitation
To: lispsources at SU-AI
Cc: Multics-Lisp-people at MIT-MC
Well, Multics MacLisp, letsee:
Multics Maclisp, consisting of an interpreter, compiler, LAP (not used
by the compiler, tho), runtime, and utilities, was developed by
MIT Lab for Computer Science (LCS) in 1973 with the aim of exporting
the Macsyma math system to Multics (of which MIT-Multics was the only
one at the time). Dave Reed (now at LCS) and Dave Moon (now at MIT-AI
and Symbolics, Inc.) were the principal implementors then, and
Alex Sunguroff (don't know where he is now) to a lesser degree.
Reed and Moon maintained it to 1976, I maintained it until now.
Its maintenance/support status since my flushance of Honeywell
(December 1980) is now up in the air, although Peter Krupp
at Honeywell is now nominally maintainer.
The interpreter and general scheme of things were developed partly
on the experience of PDP-10 Maclisp, visavis running out of space,
and an earlier Multics Lisp by Reed, visavis better ways to do this
on Multics. Multics MacLisp features virtually infinite address
space (limited by the size of a Multics Process directory, which
is virtually unlimited), a relocating/copying garbage collector,
strings, bignums and other MacLisp features, general compatibility
with (ITS) MacLisp, and very significantly, the facility to interface
to procedures in other languages (including Multics System routines)
on Multics.
With the notable exception of the compiler, which is a large (and
understandable, as well as effective) Lisp program of two large
source files, the system is in PL/I and Multics assembler: the
assembler portions, including notably the evaluator, are that
way for speed. The language was designed to be as close to
ITS Maclisp as possible at the time (1973), but has diverged some.
The compiler was developed as two modules, a semantics pass
reworked from the then-current version of the fearsome ITS
COMPLR/NCOMPLR (1973), and the code generator was written anew
by Reed (1973), although it uses NCOMPLR-like strategies
(I have a paper on this subject).
Although used in the support of Macsyma, the largest and most important
use of Multics Maclisp is as the implementation and extension language
of the Multics Emacs "text processing and video process management"
system. Other large subsystems in Multics Maclisp over the years
have included a Multics crash and problem analysis subsystem and
a management-data modeling system (KOMS, about which I know little).
Pointers in Multics Maclisp are 72-bit, which includes a 9-bit
type field. Non-bignum numbers (fixna and flona) are directly
encoded in the pointer, and do not require allocation, or the
hirsute "PDLNMK" scheme of ITS MacLisp. Symbols and strings are
allocated contiguously, and relocated at garbage-collect time.
Binding is the standard MacLisp shallow-binding (old values
saved on PDL, symbol contains "current" value). Other Maclisp
language accoutrements (property lists, functional properties,
MacLisp macros, etc.) exist.
"A description of my OS:"
Well, the Multics Operating System enjoys/suffers a paged,
segmented virtual memory, implementing virtual storage and virtual
file access in a unified fashion. The paradigm is so well-known
that I cannot bear to belabor it any more. The net effect
on Lisp is a huge address space, and heavy interaction
between the GC algorithm and performance. Multics will run
in any size memory between 256K words and 16 million (36 bit
words) The Multics at MIT (there are about three dozen multices
all over the world now) has 3 million words of memory,
which I believe is 1 microsecond MOS. The MIT configuration runs
3 cpus - other sites vary between 1 and 5. The cache per
CPU is 2k words, and is "very fast", but the system gets CPU limited,
and can rarely exceed 1 MIP per cpu (highly asynchrounous processor),
although powerful character and bit string handling instructions
can do a lot faster work than a 1 mip load/store chain. You
wanted to know a bout disks:
Date: 16 March 1981 22:54 est
From: Sibert (W. Olin Sibert)
An MSU0451 has 814 cylinders, of 47 records each. Its average seek time
is 25 ms. (I don't know whether that's track-to-track, 10 percent, or
half platter -- I'll bet it's track-to-track, though). Its average
rotational latency is 8.33 ms. Its transfer rate is about 690K 8bit
bytes (614K 9bit bytes) per second, or 6.7 ms. per Multics record.
[1024 words]
I cannot really think of benchmark possibilities that would
show the performance of Multics MacLisp to great advantage.
For all its virtual memory, the antiquated basic architecture
of the Honeywell 6000 series (from the GE600) provides a
hostile environment to the Lisp implementor. Only one register
(AQ) capable of holding a full Lisp pointer exists, and this
same register is the only one you can calculate in, either.
Thus, the compiler can't do useful register optimization
or store-aviodance, and comes nowhere near NCOMPLR, which
is using the same techniques to implement the same language,
in the performance of its object code.
MacLisp type and array declarations are supported, and utilized
in the straightforward way by the compiler to improve generated code,
but in no way could it be claimed that what it generates is
competitive.
Multics MacLisp is "owned by MIT. It is distributed by MIT to anyone
who wants. It is part of some Honeywell products [Emacs], and is
supported by Honeywell to the extent and only to the extent necessary
to keep these products operative. Honeywell can distribute it,
but may not charge for it, but may charge for products written it it".
Although its support is a current hot potato, interest in using
Multics Maclisp is continually growing, and interesting subsystems
in it are being developed as of this writing.
Anything else?
∂17-Mar-81 2200 JMC*
reiter again and remind again. Also take phon.
∂19-Mar-81 0816 FFL
To: JMC, FFL
R. D. Arnold would prefer not to replace you on his Orals committee as he
has already lost one of the originals. He asks is you can make it on
April 27 thru 30? April 20 is also a possibility, altho there is soome
question whether Tom will be available then. We should let him know
as soon as possible.∞
∂19-Mar-81 1818 Denny Brown <CSD.DBROWN at SU-SCORE> use of CSD-CF for classwork
Date: 19 Mar 1981 1814-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: use of CSD-CF for classwork
To: CSD-Faculty: ;
Purpose of this memo is to clarify the position on use of SAIL, SCORE,
and the ALTOS for CS classes. As a rule, we wish to have NO courses
use our systems. In some cases, facilities exist on our systems
that are unavailable elsewhere, and that are central to the goals of
a course. In these cases, we plan to allow such use. If
we allow use of our systems, we must provide access for *all* students
in the class. Since this means a commitment of departmental funds
and facilities, we will require that all such cases be approved by the
department chairman.
Any faculty member who plans to allow use of any of our systems for
Spring quarter should contact me immediately. I will maintain a list
of approved courses in the file <CSD.FILES>CF-COURSE-USAGE.TXT.
-Denny
-------
∂19-Mar-81 1824 Denny Brown <CSD.DBROWN at SU-SCORE> approved courses for using CSD-CF
Date: 19 Mar 1981 1819-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: approved courses for using CSD-CF
To: guibas at PARC-MAXC, dek at SU-AI, jmc at SU-AI, tob at SU-AI
Here is the initial contents of <CSD.FILES>CF-COURSE-USAGE.TXT, just so
you don't have to contact me. Send me mods to the apologia if you feel
like it. -Denny
This is a list of courses approved for using SAIL, SCORE, and/or the
ALTOS for class homework.
Quarter Course Instructor System(s) Expected enrollment
-----------------------------------------------------------------------------
Spring 80-81 CS442B Leo Guibas ALTOS 25 (restricted)
The course is a graphics course for the ALTOS. Enrollment
is limited to limit the ALTO load.
Fall 81-82 CS204 All 25 (restricted)
The course often uses graphics, either on SAIL or the ALTOS.
One purpose of the course is to acquaint new PhD students with
the dept. and its facilities.
Fall 81-82 CS206 John McCarthy SAIL 20
JMC plans to use a new proof-checker based on FOL as an integral
part of the course. Software available only on SAIL.
Fall 81-82 CS227 Tom Binford SAIL 15
Robotics course needs the Hand-Eye lab for experiments.
-------
∂19-Mar-81 2203 Wiederhold@SUMEX-AIM visits
Date: 19 Mar 1981 2200-PST
From: Wiederhold@SUMEX-AIM
Subject: visits
To: engelmore@ISI
cc: sagalowicz@SRI-KL, jmc@SAIL, kaplan@SRI-KL
I am sorry that we couldn't get together. I hope you feel better soon.
I dod talk a bit with Duane Adams, but would still like to discuss KBMS with you.
I left some drafts of the proposal, please msg or call me about any comments.
I understand you will be at Stanford April 7th - could we set up some time on
that day ? I'll keep my day free.
Sincerly Gio
-------
∂20-Mar-81 0954 CSD.JEANIE at SU-SCORE Faculty Meeting
Date: 20 Mar 1981 0950-PST
From: CSD.JEANIE at SU-SCORE
Subject: Faculty Meeting
To: CSD-Faculty: ;
cc: csd.bscott at SU-SCORE
A faculty meeting has been scheduled for Tues. March 31 at 2:30. Approval of
degrees is on the agenda and also John Linvill will talk about CIS. The meeting
will be in MJH 252.
-------
∂20-Mar-81 1555 QEH exercise
To: KGK
CC: JMC
I am just beginning to work on the exercise on the machine. I believe that
I have done the preliminary work on paper, and hope that I can finish in time
to give them to you this afternoon, but frankly the possibilty is not too
great. In any case I will work on it tonight and tommorow until finished.
I understand that you both need to get grades in and would like to get an
incomplete if my tardiness necessitates that.
I did find the exercise to be stimulating, and look forward to completing it
in the near future. eric hamilton.
∂20-Mar-81 1729 MMD Documentation for my TEX macros is finally complete, and comes in two flavours:
To: "@MACMAI.L[TEX,MMD]"
txtdoc.pre[tex,mmd] (48pp. text only)
docmac.pre[tex,mmd] (90pp. includes macro listings)
Your comments are appreciated. It has lots of examples (and anti-examples).
Any minute now, new macros will replace old ones (which will be preserved
as *.old[tex,sys]). There ARE incompatible changes (not many, though);
please see p7 of macros.log[tex,mmd]. As the manual now proves (?),
everything works...
I'm also documenting ALL changes from now on in file log. Better forget
about READ MAXTEX for the moment. It is badly outdated. A short on-line
reference guide will be prepared.
Now the bad news. The index macros are not ready yet. They will someday
sometime soon.
∂20-Mar-81 2328 GREEP at RAND-AI CS226 term paper
Date: 20 Mar 1981 2323-PST
From: GREEP at RAND-AI
Subject: CS226 term paper
To: JMC at SU-AI
I left it in your mailbox in Jacks.
-------
∂21-Mar-81 0030 Robert S. Boyer <BOYER at SRI-CSL> Verkshop
Date: 21 March 1981 00:30-PST (Saturday)
From: Robert S. Boyer <BOYER at SRI-CSL>
To: JMC at SU-AI
cc: Bledsoe at UTexas, Boyer at SRI-CSL
Subject: Verkshop
Are you still thinking of going to the Verkshop? It may be
a definitive political meeting, though I'm less optimistic
about the technical aspects.
One issue is how NSA should pursue its Computer Security
Evaluation Center charter. As I understand it, NSA is
setting up a semi-public facility for evaluatiing computer
systems. It might dictate an interest in proving facts
about computer programs.
I hope that you and Woody Bledsoe show up.
Yes, I still plan to attend.
∂21-Mar-81 1144 KGK on TTY162 1144 via SRI-AI exercise 3
To: QEH
CC: JMC
Please leave it in the box by McCarthy's door; I'll pick it up
today (Saturday) or tomorrow. --kk
∂21-Mar-81 1155 Konolige at SRI-AI Answers to exercise 3
Date: 21 Mar 1981 1156-PST
From: Konolige at SRI-AI
Subject: Answers to exercise 3
To: JJF at SAIL, JAK at SAIL, JXK at SAIL, JMM at SAIL,
To: JSR at SAIL, QST at SAIL, QHY at SAIL, YUE at SAIL,
To: JPM at SAIL, QCB at SAIL, QCC at SAIL, QEH at SAIL,
To: QBK at SAIL, QDM at SAIL, QHP at SAIL, QEP at SAIL,
To: QLP at SAIL, QJP at SAIL, QHS at SAIL, QCT at SAIL,
To: ML at SAIL
cc: jmc at SAIL
A commented set of axioms, as well as proofs of the three problems
in the exercise, are in exer3.ax, exer31.prf, exer32.prf, and exer33.prf
in [1,kgk]. You can peruse these with a TV screen, or get hardcopy from
the Dover. I'll put the corrected exercises in the box in front of JMC's
door. --kk
-------
∂21-Mar-81 1539 MINSKY at MIT-AI (Marvin Minsky)
Date: 21 MAR 1981 1840-EST
From: MINSKY at MIT-AI (Marvin Minsky)
To: MINSKY at MIT-AI, rah at SU-AI, llw at SU-AI, jmc at SU-AI
That last draft was pretty good. I have a number of additional
comments and points it should include. Perhaps we should now run the draft
through each of us for revisions and try to finish it. The dynamics
of the reference model seem pretty clean, and we could put what we know of the
stability in an appendix.
I mentioned it to Frank Press, who understood and liked it. But
because of disaster-proneness, he didn't think we could be serious.
What we should do is have a section on security:
meteor cross-section
failure modes
worst-case groundsite damage
Then, one can argue that the idea shows a certain feasibility, but presumably
others can now invent variants that are more practical.
(cross-section of catching random space satellites?)
(possibility of dynamic moving aside, or ABM countermeasures)
∂23-Mar-81 0000 JMC*
call reiter about visiting
∂23-Mar-81 0000 JMC*
cbcl to Loebner after call?
∂23-Mar-81 0942 FFL Visit of Bob Grafton of ONR
To: JMC, FFL
Zohar asked me to tell you that Bob Grafton is coming to Stanford
and that Zohar arranged that he should talk with you between
ll and 2:30 p.m. on April 3. Pls. let Zohar know if not OK.
∂23-Mar-81 1048 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) Grades are Due
Date: 23 Mar 1981 1048-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: Grades are Due
To: CSD.Siegel at SU-SCORE, SJF at SU-AI, CSL.DRA at SU-SCORE,
CSD.Buchanan at SU-SCORE, CSD.Genesereth at SU-SCORE
cc: LTP at SU-AI, RWF at SU-AI, CSD.DBrown at SU-SCORE, TOB at SU-AI,
JMC at SU-AI, CSL.Lantz at SU-SCORE, DEK at SU-AI, CSL.EJM at SU-SCORE,
CSL.BKR at SU-SCORE, CSD.Reges at SU-SCORE, CSL.SSO at SU-SCORE, PHY at SU-AI
A reminder that grades are due in the Registrar's Office by noon tomorrow.
Please try to have them to me by 5 this afternoon or early tomorrow morning.
Please do not send them by ID mail.
Thanks,
Carolyn
-------
∂23-Mar-81 1424 FFL
To: LGC, JMC, JD, CG, CLT, BCM, ZM, FFL
I understand that all of the department auditrons are to be collected for
alteration or replacement soon. In the meantime if you find your auditron
does not work, please see me or Mary Texeira for assistance in getting a
temporary or new key.
∂23-Mar-81 1535 FFL
To: JMC, FFL
∂23-Mar-81 0942 FFL Visit of Bob Grafton of ONR
To: JMC, FFL
Zohar asked me to tell you that Bob Grafton is coming to Stanford
and that Zohar arranged that he should talk with you between
ll and 2:30 p.m. on April 3. Pls. let Zohar know if not OK.
Zohar has now asked that you have lunch that day with Bob Grafton and
Zohar at noon.
lunch and conversation with Bob Grafton ok
∂23-Mar-81 1657 SCH greetings
extended to you from Richard Feynman in Pasadena.
∂23-Mar-81 2000 JMC*
CS226 grades.
∂23-Mar-81 2018 RPG
John, here's what I think CSD should be doing about facilities
for the future:
I don't believe in personal computing at any level, in the
traditional sense of the phrase, for research use. I feel, also,
trivial tasks take up too much of other peoples' time on a timeshared
machine. So, I propose to combine them.
The idea is a star configuration with something like an S1 as the
main machine with as many smaller machines with the same logical
architecture as possible in a pool closely coupled to the large
machine (some shared memory or dma between the small and large
machines).
The small machines are behind a video switch or the current technology
accomplishing the same effect. When you log in, you get the first available
personal machine. While you edit or debug you run in the small machine,
possibly paged to local and/or global disk storage. When you are cycle
intensive the large machine begins execution, paging out of the small
machine into itself. When you start to interact, or when there is a speed
gain associated with it, you page back into the small machine. Thus you are
logically always running on the large machine, but the process migrates
around to the best place. Since there are only 2 such places (or a total
order of them if there are classes of personal machines) there is not
a complex problem associated with deciding where the process should be.
Another common mode is for the personal machine to run your standard editor
with the large machine running the computing engine, communication being
very fast between the two (ala the E/MacLisp setup I'm working on now).
This is made possible because the interaction is via channels and not ethernets.
When no small machines are available you simply log in on the large
machine.
Another possible addition to this is to make the entire system a `LISP' machine
in that there is a small kernal of io routines with the entire OS built on top
in LISP. Each job is essentially a closure in the LISP sense, and several jobs
may be subfunctions of others. Each process also needs to have a garbage
collector associated with it (to avoid a big gc!) or an incremental gc needs to
be used.
This last part is not too well thought out, but the first part (the
architecture) is generally favored by those I've discussed it with. I would
not object to heading a project investigating this in conjunction with any
Advice Taker work you have in mind for me.
-rpg-
∂23-Mar-81 2055 ML cs226
To: JMC
CC: KGK
If possible, I would like to take an incomplete for CS226 so that I can
finish a paper I would like you to read. I have done the first 2 problem
sets, and I am handing the 3 one in Monday night. I've been busy the last
few weeks working for TW as a TA.
Mike Lowry
ok about incomplete in cS226
∂23-Mar-81 2140 JK
There is a complete proof of Ramsey's theorem in ramsey.doc[ekl,jk] -
there are practically no comments on it; it was pretty-printed by ekl.
An internal form of the proof (suitable to be read in) can be found in
ramsey.prf[ekl,jk].
Some statistics: The main proof (called "ramsey") is 24 lines long
and relies on two tecnical lemmas, called lem1 (5 lines) lem2 (13 lines) and
a diagonalisation lemma (21 lines) . The rest (16+12+27+19) are general
"set theoretic" facts. Of the total set of lines 32 are declarations - these
have to be counted as lines because of locality of lines unlike in FOL.
The main thing in my opinion is the fact that the proof is readable. I did not
use the programmability of EKL since that could be regarded as cheating in
a test case like this - after all, one could write a program that now proves
it in precisely one step: execute all the commands!
Current problems: I found the goal structuring machinery unusable -
it was way too fancy for the purpose it was intended; this will be replaced
by simple commands allowing one to link and bypass parts of proofs. EKL is
currently quite egregious in the amount of space used; this is fairly easy
to fix, too. The current decision procedure is a bit too slow - it does a
a great deal of unnecessary recomputation; this is because I ended up coding
it quite conservatively - again an easy fix.
The main problem was the lack of higher order unification algorithms -
this forced me to do most applications of induction by hand. Ofcourse, the
problem in its full generality is unsolvable, but one should be able to code
simple subcases efficiently.
∂24-Mar-81 0729 Darden@SUMEX-AIM regards
Date: 24 Mar 1981 0714-PST
From: Darden@SUMEX-AIM
Subject: regards
To: JMC@SAIL
Hi, John. I presume you arrived safely at your hotel, had a productive
meeting and are now back at Stanford. Thanks for the book and the LISP
instruction. I wish I had more time to spend on learning LISP now; maybe
I will have a chance to take you course in the next coupleof years if
I can find grant money to support my learning a new field. Everyone
at the dinner very much enjoyed talking to you. Let me know when
you will be in Washington again.
Keep in touch. Bye, Lindley.
-------
∂24-Mar-81 0833 KAPLAN at SRI-AI ARPA proposal
Date: 24 Mar 1981 0824-PST
From: KAPLAN at SRI-AI
Subject: ARPA proposal
To: DCL at SU-AI
cc: wiederhold at SUMEX-AIM, tob at SU-AI, jmc at SU-AI, kaplan at SRI-AI
David:
A draft of our part of the proposal, is on KBMS.DOC[1,SJK] at SAIL. A
copy was sent to Bob Englemore of ARPA for comments, and he said he will
get back to Gio quite soon.
Jerry
-------
∂24-Mar-81 0916 NEUMANN at SRI-KL VERkshop agenda, first cut
Date: 24 Mar 1981 0843-PST
From: NEUMANN at SRI-KL
Subject: VERkshop agenda, first cut
To: VERKshop: ;
Well, Now that the SRI-KL is up again (yesterday there was no power in
the new building!), I see that I have had only a few responses.
(Note: our Foonly is much more reliable than the KL, so try me at SRI-CSL
[a.k.a. SRI-F2] if the KL won't respond.) But Steve, Don Good, and I
have been exchanging net notes. Here then is a summary of current
thinking for your comments.
Don Good has reiterated the primary goal, namely, to address the issue
of how to attain "widespread availability of of verification tools and
techniques." We should cover
1. Where are we now?
2. Where do we need go?
3. How do we get there?
Your contributions (oral or written) should address whichever of these
topics you see fit. Those of us who have discussed it seem to be in
general agreement that the vital issues at this time are Items 2 and
3, and it is hoped that those vital issues will be the primary focus
of this VERkshop -- following a brief summary of progress.
Following introductory status reports, Don suggests that we turn to
the security community and address the issues of what do they think is
needed, what do they really need, what can the verification community
reasonably provide, what can't it provide, what are reasonable goals
for the future, and how do we attain them. The main question there is
how can we provide effective verification support for building
verified, secure systems. On the other hand, I think that Boyer and
Moore feel that the near-term successes are going to come not from
secure systems verification, but from verification of small but VERY
REAL programs [in their case, currently Fortran is receiving their
attention!]. I think it is clear that we should work in both
directions. We need all the help we can get!
FROM DON GOOD:
If several major discussion issues can be identified, I think
it also would be a good idea to solicit some discussion organizers
in advance to concentrate on rounding up the people who have
something to say on that issue. Discussions are usually productive
in direct proportion to the amount of thought people have put into
the topic in advance. If people want to support these discussions
with written position papers, so much the better.
FROM STEVE WALKER:
Topics I feel we should include in an agenda are:
Status reports from the major verification efforts - SRI,
UTexas,ISI,Stanford, IPSharp ...
Status of DoD verif research support - Walker - Darr, Lubbes,
Marmor-Squires
Status of Computer Security Initiative - Walker - Computer
Security Evaluation Center
General discussion
- where are we now? - where do we want to be? - where can we
hope to get? - what will it take to get there?
Also to be included:
Ada - what role can/should it play (what about Euclid, Gypsy,
etc)
and Protocol Verification - where do we stand
FROM DON GOOD:
Let me propose the following top-level design of an agenda:
Tuesday, April 21. Status Reports
I think it would be good if we could get these all in on the first
day. It would discourage lengthy advertising campaigns, and it also
would give people one extra evening of barroom discussion with all
of the past years progress reports out in the open. If there really
isn't enough time, there would be no major problem with carrying
over to Wednesday.
Wednesday, April 21. Assessment of Outstanding Problems.
The paper I have been preparing has been directed specifically at this
question. The outline I have so far is below. For purposes of general
discussion, I would think the focus should be on obtaining a community
assessment of these problems (and whatever other ones folks might think
are important), and on discussing what degree of solution we need in
order to make verification practical. I would be happy to organize
a day's worth (or whatever amount is appropriate) around these issues.
My draft presently is really drafty, but if any of you would like to
see it, I would be happy to circulate it around.
1. Verification Issues
My theme here is that verification currently is impractical
because it is just too expensive. We need drastic cost reductions
to be effective.
1.1 Transforming Specified Programs into Theorems
This provides a forum for discussing proof methods for
protocols, concurrency, exception handling, generics, ....
and other outstanding verification problems and techniques.
1.2 Languages.
This provides the forum for discussion of both programming
languages and specification languages. The ADA issue is
one of the important ones that needs to be discussed.
1.3 Verification Systems.
Here we need to discuss what state our systems are in
presently, how for we can reasonably expect them to
be pushed, what should future systems look like....?
I would include theorem provers as a part of this discussion.
1.4 Absence of High Level Theories.
(Or as Sue Gerhart calls it, the "knowledge gap."). I think
the issues here whether other people think this is as serious
a problem as Sue and I do, and if so, what we can reasonably
expect to attain in this area.
2. Security Issues.
My focal issue here is that if we really want to build verified,
secure systems, the verification community and the security
community have to join forces. We have to USE verification
throughout the entire system building process. We need to move
past building secure systems with verification "in mind", and
we need to seriously consider alternatives to security kernels
which are essentially verification retrofit efforts. From near
the beginning, we verification folks have said that retrofitting
won't work.
Thursday, April 23. How to Solve Them.
This is the really difficult question, and I put it here just to be
sure there would be some time for discussion before people begin
catching planes. I think we definitely need to get started on it
first thing Thursday am.
I really don't have any specific answers to propose here. It clearly
would be better if someone had some specific proposal to serve as
a focal point for a discussion. Maybe the Government has some plans
to lay on the table for discussion.
FROM PETER: SO, PLEASE LET US HAVE YOUR COMMENTS. RESPONSES TO ME
@SRI-KL (or @SRI-CSL if KL down). I WILL REDISTRIBUTE AS APPROPRIATE.
I ALSO NEED TO KNOW FAIRLY SOON FROM EACH OF YOU WHO WOULD LIKE TO
COME. WE MAY HAVE TO HAVE A CUTOFF ON ATTENDANCE IF MORE PEOPLE WOULD
LIKE TO COME THAN THE FACILITIES WILL HOLD. (HOWEVER, I HOPE THAT
WILL NOT BE NECESSARY.)
-------
∂24-Mar-81 1002 FRG Fran`s sick
Fran called in to tell you she will not be in today. She is ill. Jake
∂24-Mar-81 1040 Stan at SRI-AI distributed business machines
Date: 24 Mar 1981 1041-PST
From: Stan at SRI-AI
Subject: distributed business machines
To: jmc at SAIL
cc: nilsson, stan
Are you available to meet with Nils & me sometime this week to discuss
the background of the project, some questions that might be pursued,
and the nature and scope of your involvement? We are available
tomorrow morning (3/25), Thurs. afternoon (3/26), or anytime Friday
(3/27). We can meet at Stanford, if that's convenient.
--Stan
-------
∂24-Mar-81 1120 CSD.TAJNAI at SU-SCORE (Carolyn Tajnai) grades
Date: 24 Mar 1981 1117-PST
From: CSD.TAJNAI at SU-SCORE (Carolyn Tajnai)
Subject: grades
To: JMC at SU-AI
Can you get the grades to me by 1:30?
Carolyn
-------
∂24-Mar-81 1135 CSD.ULLMAN at SU-SCORE
Date: 24 Mar 1981 1134-PST
From: CSD.ULLMAN at SU-SCORE
To: JMC at SU-AI
In-Reply-To: Your message of 23-Mar-81 2259-PST
I noticed that in the newspaper (with glee). Thanks.
-------
∂24-Mar-81 1220 Walker at SRI-AI net address change for SRI AI Center
Date: 24 Mar 1981 1221-PST
From: Walker at SRI-AI
Subject: net address change for SRI AI Center
To: balzer at ISIE, bobrow at PARC, buchanan at SUMEX-AIM,
To: erman at ISIB, feigenbaum at SUMEX-AIM,
To: feldman at SUMEX-AIM, rick at RAND-AI, jmc at SU-AI,
To: minsky at MIT-AI, newell at CMUA, nilsson at SRI-AI,
To: reddy at CMUB, sacerdoti at SRI-AI, simon at CMUA,
To: tenenbaum at SRI-AI, athompson at USC-ECL, walker at SRI-AI,
To: tw at SU-AI, phw at MIT-AI, robinson at SRI-AI
The SRI AI Center has its own computer and a new net address: SRI-AI.
Mail addressed to SRI or to SRI-Kl will be forwarded, but since that
machine is highly unreliable, we would all appreciate you updating
your records accordingly (particularly those little address files that
make our lives a little bit easier). Those most directly involved in
AAAI activities in addition to me are Nilsson and Robinson. However,
Sacerdoti and Tenenbaum will probably continue to receive mail here
for a while.
Don
-------
∂24-Mar-81 1442 RDA phd oral
I am one of Tom Binford's students, working on stereo vision.
Since you are on my thesis reading committee, I would like to
arrange a time for my orals so you will be able to attend. I
wa hoping to schedule it for late April, but I know you are
leaving for Europe around then. Do you know your schedule,
and do you have any suggestions?
I am hoping to finish by June. Other members of my
committee are Tom Binford, Thomas Cover, and Elliot Levinthal
(who probably won't be available).
Dave Arnold
∂24-Mar-81 1553 RPG Results
How did your talks with Englemore go? Jonl White is planning to
be out for the Lisp Honcho meeting, and he has agreed to help
me out with the E/MacLisp interface and with some of the theoretical
(GC evaluation) problems in the Lisp Timing project. Can Stanford
help with some of his expenses?
Lew Creary has been talking to me about the Advice Taker, and
perhaps some guidelines about what you want should be laid down
to all concerned so that everyones' ideas about what is going on
won't be upset - I think he thinks I will be working with him or
vice versa.
-rpg-
∂24-Mar-81 1604 CSL.JLH.MOGUL at SU-SCORE (Jeffrey Mogul) Comments on the Intel iAPX 432
Date: 24 Mar 1981 1555-PST
From: CSL.JLH.MOGUL at SU-SCORE (Jeffrey Mogul)
Subject: Comments on the Intel iAPX 432
To: @sun at SU-AI
Several of us went today to a seminar on the Intel 432, their
newest microprocessor. It all looks pretty impressive, and I
thought it might be useful to describe some of its features.
The basic architecture seems elegant and well thought-out. The
buzzwords are: capability-based object-oriented environment,
40 bit virtual/24 bit physical addressing, on-chip memory
management, transparent multi-processor dispatching, and
on-chip message send/receive primitives.
The hardware comes in several pieces: a 2-chip General Data Processor,
and a 1-chip I/O Processor. They communicate over a synchronous
packet-protocol bus, and up to 255 processors may be put onto a
bus. The GDPs are designed so that the number of processors on
the bus is ENTIRELY irrelevant to the software. All processors
share the physical memory on the bus; other peripherals are
connected on your favorite I/O bus (e.g., multibus) via
I/O Processors.
Virtual address translation is done by the processors; each
"object" in the system (processor, process, message port,
memory segment, etc.) is identified by an object id (essentially
a capability) which points to an object descriptor; therefore,
the hardware checks all references for type and protection.
There are 2↑32 possible objects; memory pages are 64kb long.
The hardware does physical memory allocation, your choice
of stack or heap (each activation record is a new segment,
a la Pilot.)
Hardware also does multi-processor dispatching; processes
are put into priority queues, and each processor can get
a new task when it is done with the previous one. There are
no interrupts per se, but processors can cause each other
to pre-emptively re-dispatch (to a higher-priority process).
There are several hardware synchronization primitives.
The hardware provides message send/receive primitives (done
apparently by transferring ownership of objects) - 80 uSec
per send.
Software: Intel is pushing this as an Ada machine, although
they intend to support other languages (include Pascal
and Cobol.) Their current Ada compiler runs on a Vax (there
is some confusion as to whether they use Unix in-house, but the
compiler available in June is for VMS.) They are going to
provide some O/S software (but no filesystem until the
second release, a year from now!) However, this looks
like a good Perseus machine.
There will not be an assembler.
The only hardware currently available is a single board
thing that needs an Intel development system to supply
memory and I/O. They are supplying this with a interpretive
language only. This "Object Programming Language" is
meant only to introduce hardware engineer types to new
software concepts, but we accidentally found out that OPL
is really SmallTalk! In fact, this machine is at least as
good a match for SmallTalk as it is for Ada, although they
don't intend to make a real SmallTalk product.
In summary, for ca. $4k you get a 500 Kips machine, heavily
influenced by Multics, C.mmp/Hydra, Pilot, SmallTalk, and
(unfortunately) Ada.
Andy managed to talk them into giving us a set of three manuals
for which they were charging everyone else $90 (a major fraud,
it appears). These (and preliminary data sheets) are available
for anyone who wants to read them.
-------
∂24-Mar-81 1719 csl.jlh at SU-SCORE (John Hennessy) Re: Comments on the Intel iAPX 432
Date: 24 Mar 1981 1720-PST
From: csl.jlh at SU-SCORE (John Hennessy)
Subject: Re: Comments on the Intel iAPX 432
To: CSL.JLH.MOGUL at SU-SCORE, @sun at SU-AI
In-Reply-To: Your message of 24-Mar-81 1555-PST
Just a data point: the time to execute the instruction A := B + C,
on the first 432 to be released is approximately 50uS (yes, I said
50 microseconds). Needless to say this is no where near 500Kips.
-------
∂24-Mar-81 1748 LGC
I'm at home, and just read your msg saying that you would be in with news.
Is it important that we talk today, or would tomorrow be soon enough?
∂24-Mar-81 2329 TOB brief discussion
I would like to spend about 10 minutes in a brief
discussion with you on Wednesday.
Tom
∂25-Mar-81 0622 JRA cs institute
john,
the good news is that the posters should be in the mail by now.
the bad news is that they want all course materials ready for duplicating
by may 1! do you want to try to meet that deadline? i'm putting a
rough draft of the schedule together this week, and will be in reasonable
shape, i hope by tuesday afternoon if you'd like to see it.
∂25-Mar-81 1333 FFL
To: JMC, FFL
∂24-Mar-81 1650 JMC
It seems that Reiter as more money coming to him.
It was my understanding that Reiter was to receive $350 in total. This
he received. $l50 for honorarium and $100 for expenses. As I recall,
his expenses came to $96. So does this not make it come out correctly?
∂26-Mar-81 1138 Feinler at SRI-KL Re: EOP
Date: 26 Mar 1981 1134-PST
From: Feinler at SRI-KL
Subject: Re: EOP
To: JMC at SU-AI
cc: pourne at MIT-MC, FEINLER
John,
His name was given to us by someone else (possibly ARPA - not sure).
I have not had any new contacts at the Executive Office since the
new administration came online. Would suggest you call EOP directly
or if you want to be a little more discreet, try calling Pete
McCloskey's local office. They are good about such things. We
have a service which gives the heads of various government agencies
but it has not sent any updates to us since last October...guess
they can't keep up either!
Regards,
Jake
-------
∂26-Mar-81 1507 FFL
To: JMC, FFL
Bob Stone of Data Systems Marketing called. 9 94l-0240
I suspect I don't want to talk to him, but please find out what
he wants.
∂26-Mar-81 1520 FFL
To: JMC, FFL
∂26-Mar-81 1507 FFL
To: JMC, FFL
Bob Stone of Data Systems Marketing called. 9 94l-0240
He is not in his office just now. When he called, he did not seem
ready to tell me what he wanted. I have asked tha he call me back.
∂26-Mar-81 1928 Kanerva@SUMEX-AIM DEC-20 High Hurdles
Date: 26 Mar 1981 1627-PST
From: Kanerva@SUMEX-AIM
To: Danforth@SUMEX-AIM, Feigenbaum@SUMEX-AIM, Gorin@SAIL,
To: CSD.Hofstadter@SCORE, Knuth@SAIL, Lederberg@SUMEX-AIM,
To: LTP@SAIL, McCarthy@SAIL, Rindfleisch@SUMEX-AIM,
To: RSmith@RUTGERS, Wiederhold@SUMEX-AIM, Wilcox@SUMEX-AIM
cc: Sandelin@SUMEX-AIM, Siegman@SUMEX-AIM, Kanerva@SUMEX-AIM
Subject: DEC-20 High Hurdles
We are now over the highest hurdle. The purchase of the DEC-20
computer for CONTEXT (text) service to the University has been approved
by the vice-presidents. The machine is due to arrive at Stanford by
mid April (it's already built). It will be located in Forsythe Hall,
and service will start in the summer.
Getting here took over two years of preparation. Thank you for
your help along the way. - Pentti
-------
∂27-Mar-81 0638 Darden@SUMEX-AIM reference on 3-valued logic
Date: 27 Mar 1981 0636-PST
From: Darden@SUMEX-AIM
Subject: reference on 3-valued logic
To: JMC@SAIL
Hi, John. I have a friend that is trying to track down a reference
of yours. The passage says "TKP-M is implemented for the 3-valued
logic proposed by J. McCarthy ["Predicate calculus with 'undefined'
as a truth value", especially pp. 56-60, Stanford Artificial
Intelligence Project Memo. No. 1, Stanford University, Stanford
California, 1963]." He wrote to csd but was told the memo is no
longer available. Was it published somewhere? Was that material
published in some other paper? Can you help?
Sorry to have to bother you with this but I do think, in general,
the exchange of ideas is important. Keep me posted on the new
ideas you have; they are unfailingly fascinating, though common
sense (whatever that is) interests me more than plastic bags.
Warn regards, Lindley.
-------
Many thanks for the dinner. Had I known that xxx (What's the
name of that very smart fellow) was being coerced, I'd have volunteered
much sooner. Perhaps my manners would have recovered sooner had I been
less tired. Sorry about the plastic bags.
The memo you mention wasn't published, but I recently had it
entered into the computer, and a note to FFL@SU-AI referring to
AIM1.PUB[W79,JMC] PREDICATE CALCULUS WITH "UNDEFINED" AS A TRUTH VALUE
will induce my secretary to print a copy and send it to whatever
address you say.
It looks like I'll be in Washington again April 21 to 23 at
a meeting at NBS on proving properties of programs. How about dinner
on the evening of the 20th?
∂27-Mar-81 0816 LGC Discussion Today
I have some specific ideas to propose concerning an initial advice-taker
implementation, and am prepared for a focussed discussion of them, and
of related matters. Will you have some time for such a discussion today?
If so, when would it be best to meet?
∂27-Mar-81 0902 FFL
To: JMC, FFL
Please call Doris Hyde at III. 9 213 390 8611.
∂27-Mar-81 1104 Hans Moravec at CMU-10A (R110HM60) Pellet streams
Date: 27 March 1981 1337-EST (Friday)
From: Hans Moravec at CMU-10A (R110HM60)
To: llw at SU-AI, rah at SU-AI, minsky at MIT-AI, jmc at SU-AI
Subject: Pellet streams
CC: Hans Moravec at CMU-10A
I suspect some of you are already aware of this -
Interstellar Propulsion Using a Pellet Stream for Momentum Transfer,
C.E. Singer, JBIS 33 # 3, March 1980 pp 107-115.
and Notes on above by same author JBIS 34 #3 March 1981 pp 117-119.
discusses periodic collimation because of electromagnetic and gravitation
perturbations, launcher anchoring and other relevant (or is it
irrelevant - can't seem to get those words straight!) issues.
∂27-Mar-81 1130 FFL
To: JMC, FFL
A woman called to inquire the progress of Mr. Kasahara's request to do
research here. He is an employee of NEC Systems Lab in Tokyo. She is
calling back on Tuesday for an answer.
∂27-Mar-81 1244 CSD.ULLMAN at SU-SCORE Faculty meeting
Date: 27 Mar 1981 1241-PST
From: CSD.ULLMAN at SU-SCORE
Subject: Faculty meeting
To: CSD-Faculty: ;
cc: icl.linvill at SU-SCORE
We have had a small change of plans. We shall meet at 1:30pm on Tuesday,
3/31 in MJH 252 for a round table discussion of CIS with John Linvill.
At 2:30 we shall break for approval of degrees, with the CIS discussion
resuming if appropriate.
-------
∂28-Mar-81 0837 JRA wics
did you get my msg about the summer institute?
(material due by may 1).
btw posters are out now.
∂28-Mar-81 1026 REM at MIT-MC (Robert Elton Maas)
Date: 28 MAR 1981 1325-EST
From: REM at MIT-MC (Robert Elton Maas)
To: BLUE at MIT-MC
CC: FFM at MIT-MC, LAUREN at UCLA-SECURITY, JMC at SU-AI
Of course there is another way to look at the whole thing, which I
often do to help get rid of that kind of feeling... Just think, if you
weren't lonely (read:occupied with companionship of some form) you would
not be as free to do the things you obviously like to do.. design,
create etc..
Incorrect: in 1978 during a short period of having a sexual partner who
was living with me and playing with my penis to get me turned on when I had
been too tired initially, I came up with my brilliant inventin that
is revoluntinizing data compression. During the period when Betsy and I
were communicating intently and I was worrying much about her, I finally
got the initiative to spend 2 nights writing lisp code to implement the
latest refinement of the markovian adaptive algorithm that is used in
conjunctin with my major invention. Being not alone helps not hurts
my inventive genius.
I often miss my family life (I used to be married) but now realize if
I still had to go thru all that, (not that I'd mind) I would far far
less productive than I am now. It's all just a matter of what you really
want to do, I guess.
We two people are different.
∂28-Mar-81 1333 Darden@SUMEX-AIM (Response to message)
Date: 28 Mar 1981 1329-PST
From: Darden@SUMEX-AIM
Subject: (Response to message)
To: JMC@SU-AI
In response to your message sent 27 Mar 1981 1409-PST
Thanks for the reference. Yes, I would enjoy having dinner with
you on Monday, April 20. Would you be interested in coming out
to the University of Maryland campus in College Park to hear
a talk by Hugh Dewitt of Lawrence Livermore Lab on "Nuclear
Secrecy, the Atomic Energy Act, and Open-Scientific Concepts" ?
He will be speaking at 4:15 in the coloquium series of the
Committee on the History and Philosophy of Science. I should
probably go; we usually go to the Faculty Club for drinks
afterward. If you don't come, I would be able to get away
by 6:30 or so. I will be on campus teaching that day; phone:
454-2850.
Today is a beutiful spring day. After working on income tax
this morning (ugh), I treated myself to work in my garden
planting sugar snap peas (the new mutant that is sort of like
snow peas and sort of like green peas) and lettuce.
Bye, Lindley.
-------
I'll check, but I think that planes from the West Coast don't get in
on time to make the seminar. It's a beautiful day here too.
∂28-Mar-81 1353 Darden@SUMEX-AIM Request for copy of McCarthy paper
Date: 28 Mar 1981 1349-PST
From: Darden@SUMEX-AIM
Subject: Request for copy of McCarthy paper
To: FLL@SAIL
cc: JMC@SAIL
Hello, I am a friend of John's who has a friend who needs a paper
of John's. (Goodness, what a complicated sentence that became!)
John said I should ask you to send a copy of AIMI.PUB[W79,JMC]
entitled "Predicate Calculus with 'undefined' as a Truth Value"
to Joel Berman, c/o Natalie Schmitt, 1253 Westgate Terrace,
Chicago, Illinois 60607.
Thanks very much. Lindley Darden
-------
That's AIM1.PUB[W79,JMC], and it needs to be PUBbed first.
∂28-Mar-81 2036 RPG Advice
Soon I will need to make some decisions about what I will be doing,
at least in terms of this summer. What I propose is to work for LLL on
the S-1 NIL and to then switch allegiance to the appropriate place
in september. An appropriate place would be Stanford, and I'd like
to talk to you about that. I assume that LLL summer work would meet
with your approval since we (stanford) hope to obtain an S-1 someday.
About the Advice Taker and Creary. I would like to work on the Advice
Taker since it is quite an interesting project, has deep historical
significance, and since I feel the time is ripe for it. Creary has
been spending a lot of my time telling me about his ideas on the
subject. I let him read some parts of my thesis, which have convinced
him of two things: 1, that my system is an advice taker (in his sense
of the words) and 2, it's the best thing since white bread. This has prompted
him to hatch grandiose schemes about tagging along in my effort to make
my system into a real advice taker. So far I've humored him without any
specific committments.
One of things I don't want to do is to alienate your point of view (too
much) from such a project. In other words, I'd prefer to have you
taking a reasonable interest in any approach I'd pursue and not have
you in the role of incidental patron of my work.
However, I do feel that there can be an amalgam of your point of view
and mine on this project, though I'm not sure what Creary would contribute
(and I don't mean this in a pejorative sense in that he has shown some
good insights, but backed up with slim details, though the context may
have made such details un desirable).
In any event, I'd like to discuss some of this, along with my work
plans, with you fairly soon, in particular, before Creary assumes I'm
working with him.
-rpg-
How about getting together tomorrow, i.e. Sunday? 3pm at my office would
be convenient, but any other time not too early is also possible.
∂29-Mar-81 1409 RPG
∂28-Mar-81 2116 JMC
How about getting together tomorrow, i.e. Sunday? 3pm at my office would
be convenient, but any other time not too early is also possible.
I'm getting a bit of a cold. How about monday afternoon?
-rpg-
4pm is better Monday.
∂29-Mar-81 1455 KDO via SU-SCORE What REM said
Hi. You know me because I was in your 206 class last winter
and haven't finished my final project yet...well my girlfriend
(that I went to LA with because she was sick) is Betsy Lasarow,
and I hear from REM that he had talked to you about her and I
wanted to find out what he said and make sure that you haven't
gotten misinformed about her, since REM seems to have quite a
bit of misinformation.
Thanks,
Ken
Since I have no present intention of doing anything, it doesn't much
matter if I am misinformed. However, REM said she was in continuous
pain to which I replied that in that case she should go into the hospital
right away rather than wait for an appointment for an operation. I
decided against offering this advice directly, because I suspected
that REM might not be well informed. What got me involved is that
REM was again emitting suicidal ARPAnet messages, and I wanted to
do my bit to reduce the probability that he would commit suicide.
While almost all suicide threats are false alarms, people who make
them often eventually do commit suicide. Of course, if there is any
way in which I can actually be helpful I would be glad to.
∂29-Mar-81 1554 RPG
∂29-Mar-81 1433 JMC
4pm is better Monday.
Ok. I'll be there. Thanks for your time.
-rpg-
∂30-Mar-81 0054 POURNE@MIT-MC HELPING THE BOSS
From: POURNE@MIT-MC
Date: 03/30/81 03:46:11
Subject: HELPING THE BOSS
POURNE@MIT-MC 03/30/81 03:46:11 Re: HELPING THE BOSS
To: JMC at MIT-MC, MINSKY at MIT-MC, ELLEN at MIT-MC
I find that EOP has "word processors", not well understood by
the executives, and USED ONLY BY SECRETARIES to produce
documents. No consoles for executives, at least not at policy
analysis level; if there is much of an intertie between the
word processors the executives don't know it.
sigh.
∂30-Mar-81 0156 Jerry E. Pournelle <POURNE at MIT-MC> Re: EOP
Date: 30 March 1981 04:56-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Re: EOP
To: Feinler at SRI-KL
cc: POURNE at MIT-MC, FEINLER at MIT-MC, JMC at SU-AI
EOP has "wqord processors" used by secretaries but nothing for
the policy analysts and exectuives and if the word processors
are connected to each other the officer level types do not know
it (or at least my two contacts don't).
∂30-Mar-81 1216 NEUMANN at SRI-KL VERkshop II: Formal Application for Participation Needed!
Date: 30 Mar 1981 1200-PST
From: NEUMANN at SRI-KL
Subject: VERkshop II: Formal Application for Participation Needed!
To: VERKshop: ;
Subject: VERkshop II
If everyone who has indicated he might come or would like to come were
actually to come to VERkshop II, we would have something more like 90
people instead of the 50 that we should really have if we are to have
any meaningful discussion. On the other hand, we haven't heard from
most of you as to whether you really are planning on coming. Thus,
please respond quickly, even if your answer is NO, and we will try to
turn it around and decide who actually can come among the various
requestees. (Perhaps there will be enough NOs that we won't even have
to turn people away.) A quick reply will do.
For those of you who have replied in the last week, or who consider
yourself on this list for information purposes only, you may ignore
this message. Thanks. Peter
-------
I will definitely come to the Verkshop.
∂30-Mar-81 1229 ELLEN@MIT-MC HELPING THE BOSS
From: ELLEN@MIT-MC
Date: 03/30/81 15:29:11
Subject: HELPING THE BOSS
ELLEN@MIT-MC 03/30/81 15:29:11 Re: HELPING THE BOSS
To: POURNE at MIT-MC
CC: JMC at MIT-MC, MINSKY at MIT-MC, ELLEN at MIT-MC
I'm not surprised... executives are well known for not sullying
their hands with typing, which is defined as "secretary's work".
However, my experience is that if you get the secretaries well
indoctrinated into the marvels of electronic communication and
word processing, gradually the bosses get lured into trying it.
∂30-Mar-81 1241 RPG
To: JMC at SU-AI, TW at SU-AI, green at SCI-ICS
∂30-Mar-81 1236 Denny Brown <CSD.DBROWN at SU-SCORE> Re: Report
Date: 30 Mar 1981 1233-PST
From: Denny Brown <CSD.DBROWN at SU-SCORE>
Stanford-Phone: (415) 497-2274
Subject: Re: Report
To: RPG at SU-AI
cc: csd.ullman at SU-SCORE, csd.bscott at SU-SCORE, csd.tajnai at SU-SCORE,
ddy at SU-AI
In-Reply-To: Your message of 27-Mar-81 1547-PST
Finally, we have a yes. General policy will be to publish all theses
in the CS report series. (That probably already was policy but nobody
seemed to know.) If there is not a source of funds to pay for printing
costs, we'll advertise it as "microfiche only", and print some minimal
number of paper copies for local distribution, libraries, etc. With
support, whoever pays the bill gets to decide how many paper copies to
print. Talk to Dawn about the details, like report number, copies
needed etc. -Denny
-------
The above is the final word on making my thesis a report. Terry doesn't have
much in the way of cash for this sort of thing, but I have had a lot
of requests for my thesis so far. I don't know of any alternatives
but to either hog all the `minimal' number for myself and give them out or
to print it myself and charge to people who want it. Any ideas?
-rpg-
If you print me a copy, I will try to decide whether to subsidize the
printing of more copies. Howver, I don't have a lot of money.
∂30-Mar-81 1657 DEK self-proposed visitors
No room!
∂30-Mar-81 1743 YM new course, abstract data types
To: "@SEMINA.[1,FFL]" at SU-AI, "@SEM.DIS[SEM,VER]" at SU-AI
SPECIAL COURSE FOR SPRING QUARTER
CS453
ABSTRACT DATA TYPES
Dr. Peter Pepper
2 units
Thursday, 2:15-4:15 51R
Theoretical foundations and practical applications of abstract data types.
Many-sorted algebras and morphisms. Equational theories. Formal notions
of implementations of abstract data types, also in terms of machine
constructs. Operations on abstract data types that lead to an algebraization
of the process of program development. Semantics of programming languages
by abstract data types; functional programming.
-------
∂31-Mar-81 1022 FFL
To: JMC, FFL
Can you let me know if you intend to go to the CIS Executive meeting
tomorrow at 12:15? They need to order a sandwich for you if you are
going to be there.
I won't be at the CIS Executive meeting tomorrow, since I have another
appointment.
∂31-Mar-81 1108 Buchanan@SUMEX-AIM AAAI Council
Date: 31 Mar 1981 1105-PST
From: Buchanan@SUMEX-AIM
Subject: AAAI Council
To: jmc@SU-AI
John,
The nominating committee for AAAI would like your name on the ballot
for the August, 1981 election for AAAI Council. Since you are now on
the Council, you know that there is little actual work between annual
meetings. But we feel the policy decisions that are discussed deserve
attention from senior-level people.
Will you agree? If so, is the following correct for the ballot:
John McCarthy
PhD, Princeton Univ., 1951
Currently at Stanford Univ.
thanks,
bgb
-------
I will be glad to be nominated to continue on the AAAI Council.
∂31-Mar-81 1326 FFL
To: JMC, FFL
Harold Ossher asks if it is possible to have comprehensive exam committee
meeting on this Friday, April 3, or on Monday, April 6. If OK, any
preference about the hour?
∂31-Mar-81 1509 ENGELMORE at USC-ISI Re: LISP conference
Date: 31 Mar 1981 1452-PST
Sender: ENGELMORE at USC-ISI
Subject: Re: LISP conference
From: ENGELMORE at USC-ISI
To: JMC at SU-AI
Message-ID: <[USC-ISI]31-Mar-81 14:52:23.ENGELMORE>
In-Reply-To: Your message of 31 Mar 1981 1444-PST
John,
Please consider yourself invited. The initial announcement
is appended here in case you don't have a copy already.
Knowing your general distaste for meetings, I didn't think you'd
be interested in this one; there was never any intention to exclude you.
Bob
Call for a Discussion of Lisp Options
IPTO recognizes both the critical need of our research community
for modern computer resources and a responsibility to provide the
resources necessary to maintain a high quality of research. This
message focusses on the AI community, most of which uses Lisp as
its primary programming language. Our current effort to meet the
need for more computing power (both CPU cycles and address space)
is confounded by the current multitude of options facing us in both
hardware and software. Our budget, of course, is finite, and
necessitates our choosing the best possible investment strategy.
In order to formulate that strategy and a management plan to
implement it, we need to discuss the options with you.
My primary concern here is not hardware, but software. The
long-term hardware issues will be dealt with once the software
question is resolved, but some discussion of hardware is relevant
(see below). There are now several respectable Lisp dialects in
use, and others under development. The efficiency,
transportability and programming environment varies significantly
from one to the other. Although this pluralism will probably
continue indefinitely, perhaps we can identify a single "community
standard" that can be maintained, documented and distributed in a
professional way, as was done with Interlisp for many years.
Here are some of the issues that need to be sorted out:
- Language Development: There are now a very large set of
Lisp dialects and sub dialects -- Interlisp, MacLisp,
CADR-Lisp, Spice-Lisp, Franz-Lisp, NIL, UCI-Lisp,
"Standard Lisp", MDL, etc. What are their relative
merits and significant differences? Is there an
opportunity to combine any of them as variants of a
common base, supported by a single implementation? How
much compatibility is needed between dialects?
- Programming environments: There are two main Lisp
programming environments: Interlisp and Maclisp. These
environments comprise a set of useful functions, a set of
conventions and a philosophy of programming. How
independent are these features from their respective
language dialects? Can both environments be supported
within a single system? Where lies the future with
respect to networking, or to utilizing the capabilities
of displays?
- Portability: Should we be investing more vigorously in
the development of a highly portable programming language
(and environment) so we can be less concerned about
hardware choices? What work needs to be done to minimize
the effort of transporting Lisp to the many
microprogrammable personal machines that are appearing
(or will soon appear) on the market?
- Other issues: Although this meeting is about software,
there are some machine-specific concerns that we can't
ignore. For example, the Vaxen are and will probably
continue to be a very widely used line of machines.
What's the future of Lisp for these machines? More
specifically, what are the pros and cons of Franz Lisp as
a near term solution to running Lisp programs on Vaxen?
If the Vax Interlisp and/or NIL effort fail to produce a
useful product, how big an effort would it be for their
users to translate their programs to Franz Lisp? How
essential is the use of microcode on the Vax for
efficient Lisp execution? How should the Lisp executions
change for use on a single user Vax? What about
exploiting the the large address space under TOPS-20 as a
near-term alternative for Interlisp or other Lisp
dialects?
I would like to propose a panel of users, implementers and IPTO
program managers to address these issues with the objective of
developing a plan for future Lisp development, maintenance and
support. The two main items on the agenda are 1) examining the
alternatives, and 2) formulating a plan of attack.
This meeting needs to be held soon, preferably within the next 4 to
6 weeks. Unless there are strong objections, I would like to
schedule the meeting on Wednesday, April 8, at SRI (Gary Hendrix
has kindly agreed to be the host). I think we can complete the
discussion in one day, but it may require an evening session as
well.
Please let me know as soon as possible if you can be there. I
think this will be a meeting that's well worth attending, and I
hope all the recipients of this message can participate. If you
can't make it, feel free to suggest an alternate attendee.
Distribution list:
DARPA: Kahn,Adams,Cerf,Druffel
BBN: Brachman,Yonke
CMU: Fahlman,Steele
ISI: Balzer,Crocker
MIT: JonL White,Greenblatt,Moon,Reeve,Vezza
Rand: Hearn,Sowizrel
Rochester: Feldman
Rutgers: Hedrick
SCI: Green
SRI: Hendrix,Shostak
Stanford: Weyhrauch,Genesereth,VanMelle
UCB: Fateman
Utah: Griss
Xerox PARC: Deutsch,Masinter
Yale: Riesbeck,McDermott
∂31-Mar-81 1553 Stan at SRI-AI Meeting to discuss business communication language
Date: 31 Mar 1981 1554-PST
From: Stan at SRI-AI
Subject: Meeting to discuss business communication language
To: jmc at SAIL
cc: nilsson, stan
Perhaps my last message was lost in transmission. Are you available
anytime this week to meet with Nils and me to discuss the business
communication language idea in more detail, especially the possibility
of your being a consultant on our project?
--Stan
-------
How about Thursday morning at 11 here?
∂31-Mar-81 1608 Stan at SRI-AI Meeting
Date: 31 Mar 1981 1609-PST
From: Stan at SRI-AI
Subject: Meeting
To: jmc at SU-AI
cc: nilsson, stan
Nils and I have a conflicting appointment at 11 on Thursday.
We can make it any time this week EXCEPT:
Wed 11-12, after 3
Thurs 10-1
Fri 10-1.
Having the meeting at SRI would be fine.
-------
∂31-Mar-81 2343 CSD.GENESERETH at SU-SCORE lisp project reports
Date: 31 Mar 1981 2341-PST
From: CSD.GENESERETH at SU-SCORE
Subject: lisp project reports
To: jmc at SU-AI
I stuffed a lptr listing of lisp project reports in your mailbox.
That's all I've received to date. Have you told Engelmore of
your interest?
mrg
-------
I told Engelmore I wanted to come to the meeting. Thanks for the
reports.
∂01-Apr-81 0800 JMC*
Call Okner about IRA.
∂01-Apr-81 1145 DCL kowalski
John,
when you've finished Kowalski's book, please let me know
your opinion as to whether a development of PROLOG would
be a candidate for "the next programming language"
-David
I would be interested in seeing Robinson's combination of PROLOG
and LISP in that connection.