perm filename W79.IN[LET,JMC]1 blob
sn#430487 filedate 1979-04-01 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00397 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00039 00002 ∂02-Jan-79 1043 Gaschnig at SRI-KL Donald Michie's address in Oxford
C00047 00003 ∂02-Jan-79 1040 HVA Vera's Tapes
C00049 00004 ∂02-Jan-79 1316 Hazen at SUMEX-AIM Siglunch this Friday, Jan. 5 in Chemistry Department Gazebo
C00051 00005 ∂02-Jan-79 1554 ARK via SU-TIP Mailing Lists
C00053 00006 ∂02-Jan-79 2219 JB YOUR SLOT IN COMPUTER FORUM.
C00054 00007 ∂03-Jan-79 0123 RWW
C00062 00008 ∂03-Jan-79 0321 MRC front end
C00063 00009 ∂03-Jan-79 1356 Jrobinson at SRI-KL (Jane Robinson) eat first
C00064 00010 ∂03-Jan-79 1454 CARLSON at USC-ISI Your Comments on RTSC Meeting
C00065 00011 ∂03-Jan-79 1520 JP INTERLISP manuals
C00066 00012 ∂03-Jan-79 1537 JB COMPUTER FORUM.
C00067 00013 ∂04-Jan-79 0057 Feigenbaum at SUMEX-AIM LISP and programs on 8080
C00068 00014 ∂04-Jan-79 1155 ARK S1 Meeting
C00069 00015 ∂04-Jan-79 1603 Wiederhold at SUMEX-AIM Re: S1 Meeting
C00072 00016 ∂04-Jan-79 2031 ARK S-1 Meeting Rescheduled
C00073 00017 ∂05-Jan-79 1615 100 : Steve Robbins (STEVER@MC)∩∩∩ via NBS-TIP Dialnet
C00076 00018 ∂05-Jan-79 1659 Hart at SRI-KL (Peter Hart) Joint publication
C00078 00019 ∂05-Jan-79 1708 SSO Seminars and Parties
C00080 00020 ∂05-Jan-79 2200 REM via AMES-TIP Losing MacLisp also...
C00082 00021 ∂05-Jan-79 2259 RSP via Easy Street EFIND default file
C00083 00022 ∂06-Jan-79 1803 ME digest
C00084 00023 ∂06-Jan-79 1858 REM via AMES-TIP Partial success with MacLisp
C00085 00024 ∂07-Jan-79 1014 REM Success with NOTICE.[UP,DOC] survey for compression contexts
C00089 00025 ∂07-Jan-79 1331 DCO
C00091 00026 ∂08-Jan-79 0201 REM via SU-TIP Dialnet, PCNet, et al
C00093 00027 ∂08-Jan-79 0314 REM via SU-TIP
C00094 00028 ∂08-Jan-79 0620 REM via SU-TIP Overview of data-compression methods
C00099 00029 ∂08-Jan-79 1143 PN Agenda for next meeting
C00100 00030 ∂09-Jan-79 0204 REM via SU-TIP Mainsail loses.
C00102 00031 ∂09-Jan-79 0808 S1 DISK quota
C00103 00032 ∂09-Jan-79 1331 SSO Party
C00105 00033 ∂09-Jan-79 1542 RPG MacLisp matcher
C00108 00034 ∂09-Jan-79 1548 RPG
C00109 00035 ∂09-Jan-79 1604 RPG
C00112 00036 ∂09-Jan-79 1607 HEDRICK at RUTGERS comments on Dialnet and Sumex protocols
C00114 00037 ∂09-Jan-79 2301 MRC Hedrick's thing
C00115 00038 ∂09-Jan-79 2330 REM via AMES-TIP
C00116 00039 ∂10-Jan-79 0019 LES
C00117 00040 ∂10-Jan-79 1259 MRC via SU-TIP reply to protocol critique
C00129 00041 ∂10-Jan-79 1311 SAL Program for Computer Forum Annual Meeting
C00134 00042 ∂10-Jan-79 1346 DPB forum talk
C00135 00043 ∂10-Jan-79 1449 BCM via SU-TIP call-by-name
C00141 00044 ∂10-Jan-79 1509 HEDRICK at RUTGERS Re: reply to protocol critique
C00146 00045 ∂10-Jan-79 2003 MRC protocols
C00150 00046 ∂10-Jan-79 2138 SMG paper
C00151 00047 ∂11-Jan-79 0531 RWG at MIT-MC (Bill Gosper)
C00152 00048 ∂11-Jan-79 1320 S1
C00153 00049 ∂11-Jan-79 1401 ALS
C00154 00050 ∂11-Jan-79 1410 Jrobinson at SRI-KL (Jane Robinson)
C00155 00051 ∂12-Jan-79 0953 Bmoore at SRI-KL (Bob Moore) meeting to discuss my thesis
C00157 00052 ∂12-Jan-79 1056 AFB
C00158 00053 ∂12-Jan-79 1607 PAT reminder
C00160 00054 ∂12-Jan-79 1652 GHG Requirements for the degree
C00163 00055 ∂12-Jan-79 1700 BCM derived functions
C00164 00056 ∂12-Jan-79 2118 REF
C00165 00057 ∂14-Jan-79 1345 DWW Gene Golub's proposal
C00169 00058 ∂14-Jan-79 1511 DCO
C00171 00059 ∂15-Jan-79 0000 JMC*
C00172 00060 ∂15-Jan-79 0800 JMC*
C00173 00061 ∂15-Jan-79 0924 Bmoore at SRI-KL (Bob Moore) An index of symbols...
C00174 00062 ∂15-Jan-79 1100 JMC*
C00175 00063 ∂15-Jan-79 1300 JMC*
C00176 00064 ∂15-Jan-79 1438 TW
C00177 00065 ∂15-Jan-79 1522 SAL Computer Forum Brochure
C00178 00066 ∂16-Jan-79 1140 100 : patte MA
C00179 00067 ∂16-Jan-79 1404 CLT
C00180 00068 ∂16-Jan-79 1936 Elmasri at SUMEX-AIM Serra House Key
C00181 00069 ∂17-Jan-79 1546 RPG UTIL.>
C00182 00070 ∂17-Jan-79 1806 Gardner at SUMEX-AIM Legal reasoning paper
C00183 00071 ∂17-Jan-79 2149 MRC
C00184 00072 ∂17-Jan-79 2201 MRC
C00185 00073 ∂17-Jan-79 2202 MRC modem STILL loses!
C00186 00074 ∂18-Jan-79 0742 RYLAND at RUTGERS Re: comments on Dialnet and Sumex protocols
C00189 00075 ∂18-Jan-79 1242 PAT MA part two (or is it three)
C00190 00076 ∂18-Jan-79 1322 100 : patte proofreading
C00191 00077 ∂18-Jan-79 1401 DBL filman
C00193 00078 ∂18-Jan-79 1458 REF
C00196 00079 ∂18-Jan-79 1609 LJH Trip to Los Angeles
C00197 00080 ∂18-Jan-79 1615 CET
C00199 00081 ∂18-Jan-79 1653 Wiederhold at SUMEX-AIM Last S1 meeting notes
C00205 00082 ∂18-Jan-79 1729 MINSKY at MIT-AI (Marvin Minsky)
C00206 00083 ∂18-Jan-79 2338 ME
C00207 00084 ∂18-Jan-79 2346 ME NYT
C00208 00085 ∂19-Jan-79 1117 DEW computer forum
C00210 00086 ∂19-Jan-79 1125 DEW
C00211 00087 ∂19-Jan-79 1344 CPP cs 206 grades fall quarter 1978
C00213 00088 ∂19-Jan-79 1353 Feldman at SUMEX-AIM My course
C00215 00089 ∂19-Jan-79 1558 RWW
C00216 00090 ∂19-Jan-79 1627 SAL FORUM PROGRAM
C00218 00091 ∂19-Jan-79 2341 DEW forum
C00219 00092 ∂20-Jan-79 1410 FC How to execute PCFORT
C00220 00093 ∂20-Jan-79 1441 DCO
C00222 00094 ∂21-Jan-79 1321 Rindfleisch at SUMEX-AIM (Response to message)
C00224 00095 ∂21-Jan-79 1348 DCL via SU-TIP Modem and phone line
C00225 00096 ∂21-Jan-79 1353 DCL via SU-TIP remote users
C00226 00097 ∂21-Jan-79 2216 PEG C1 UDP
C00227 00098 ∂22-Jan-79 0758 Feigenbaum at SUMEX-AIM need to talk
C00228 00099 ∂22-Jan-79 1106 HVA Vera's Tapes
C00229 00100 ∂22-Jan-79 1134 Mcgoveran at SUMEX-AIM Changes in Short-Term Disability Plans (SDI & VDI)
C00232 00101 ∂22-Jan-79 1142 REF
C00233 00102 ∂22-Jan-79 1350 Bmoore at SRI-KL (Bob Moore)
C00234 00103 ∂22-Jan-79 1508 REF
C00235 00104 ∂22-Jan-79 1623 ARK via SUMEX-AIM S-1 Meeting Notice
C00236 00105 ∂23-Jan-79 1104 PAT ringle
C00238 00106 ∂23-Jan-79 1440 ME
C00239 00107 ∂23-Jan-79 1439 CET Change of grade for Chuck Paulson
C00240 00108 ∂23-Jan-79 1552 BOBROW at PARC-MAXC2 A special issue of the AI journal on non-monotonic logics
C00243 00109 ∂23-Jan-79 1601 PAT Texas
C00244 00110 ∂23-Jan-79 1653 PAT Filman letters
C00253 00111 ∂23-Jan-79 2228 Jrobinson at SRI-KL (Jane Robinson) YMCA Fitness classes
C00256 00112 ∂24-Jan-79 0844 REF
C00257 00113 ∂24-Jan-79 0946 PN Compiling and running S-1 Fortran programs
C00258 00114 ∂24-Jan-79 1524 ZENON PYLYSHYN at CMU-10A (X320ZP51) Center plans for next yeAR
C00262 00115 ∂24-Jan-79 1720 ME
C00272 00116 ∂25-Jan-79 0010 NS
C00273 00117 ∂25-Jan-79 0507 MRC Dialnet blather
C00279 00118 ∂25-Jan-79 0930 HVA PATTE'S BIRTHDAY DINNER
C00280 00119 ∂25-Jan-79 1007 Bmoore at SRI-KL (Bob Moore) suggested participants for AI/Phil meeting
C00283 00120 ∂25-Jan-79 1034 LES Miami reporter
C00284 00121 ∂25-Jan-79 1117 Mcgoveran at SUMEX-AIM Meeting
C00286 00122 ∂25-Jan-79 1714 100 : patte Texas
C00287 00123 ∂25-Jan-79 1918 HENRY at MIT-AI (Henry Lieberman) quasar
C00289 00124 ∂25-Jan-79 1927 VENKAT at RUTGERS Request for a copy of your paper
C00290 00125 ∂25-Jan-79 1927 VENKAT at RUTGERS Request for a copy of your paper
C00292 00126 ∂26-Jan-79 0152 JBR
C00293 00127 ∂26-Jan-79 0856 Bmoore at SRI-KL (Bob Moore) another name to consider...
C00294 00128 ∂26-Jan-79 1154 HVA Bon Voyage
C00295 00129 ∂26-Jan-79 1322 VENKAT at RUTGERS
C00297 00130 ∂27-Jan-79 1627 LES
C00298 00131 ∂27-Jan-79 1800 BTH SMA3
C00299 00132 ∂27-Jan-79 2259 GEOFF at MIT-AI (Geoffrey S. Goodfellow)
C00301 00133 ∂28-Jan-79 1956 REF Dissertation and such
C00302 00134 ∂29-Jan-79 1303 Bobrow at PARC-MAXC Re:
C00303 00135 ∂29-Jan-79 1509 DCL
C00306 00136 ∂29-Jan-79 1559 SAL FORUM MEALS
C00308 00137 ∂29-Jan-79 1635 DCL
C00309 00138 ∂29-Jan-79 1815 TOB picture interpretation
C00310 00139 ∂30-Jan-79 1326 PEG PASMAC.TXT[1,ALL]
C00311 00140 ∂30-Jan-79 1433 PEG S1 UDP backup tapes
C00312 00141 ∂30-Jan-79 1657 LES SUDS licensing
C00313 00142 ∂30-Jan-79 1731 Feigenbaum at SUMEX-AIM conversations with Jerry Lieberman
C00314 00143 ∂31-Jan-79 1343 ECOOPER at BBN-TENEXD Lisp and first order logic
C00316 00144 ∂31-Jan-79 1845 MPG Thanks
C00317 00145 ∂31-Jan-79 2104 AFB
C00320 00146 ∂01-Feb-79 1102 HVA Postponement of Patte's Party
C00322 00147 ∂01-Feb-79 1849 DCO
C00323 00148 ∂02-Feb-79 0012 NS
C00324 00149 ∂02-Feb-79 1331 JLH MEETING
C00325 00150 ∂02-Feb-79 1345 LCW
C00326 00151 ∂02-Feb-79 1341 CET Summer Visitation Program at Hughes Aircraft Co.
C00329 00152 ∂02-Feb-79 1612 TOB
C00330 00153 ∂05-Feb-79 0013 LGC via AMES-TIP My IJCAI Paper
C00331 00154 ∂05-Feb-79 0854 JLH via SU-TIP Meeting,...
C00332 00155 ∂05-Feb-79 1241 TOB
C00333 00156 ∂05-Feb-79 1339 100 : patte meeting regarding yao and winograd
C00334 00157 ∂05-Feb-79 1441 PAT expenses
C00335 00158 ∂05-Feb-79 1545 Mcgoveran at SUMEX-AIM Dr. Ramakoti Sadamanda
C00337 00159 ∂05-Feb-79 1806 Feigenbaum at SUMEX-AIM re IJCAI travel, for your information
C00340 00160 ∂05-Feb-79 2345 Feigenbaum at SUMEX-AIM musicianswinograd
C00342 00161 ∂06-Feb-79 1059 TOB
C00343 00162 ∂06-Feb-79 1356 PAT AIM 1
C00344 00163 ∂07-Feb-79 1008 CVW Instructor's Statements for Autumn Quarter Courses
C00346 00164 ∂07-Feb-79 1235 HVA Terminal at Home
C00348 00165 ∂07-Feb-79 1524 CLT
C00349 00166 ∂08-Feb-79 0809 RAJ.REDDY(A610RR29) at CMU-10B IJCAI TRAVEL SUPPORT FLAP
C00353 00167 ∂08-Feb-79 0913 BALZER at USC-ISIB Re: IJCAI TRAVEL SUPPORT FLAP
C00356 00168 ∂08-Feb-79 1037 LES FIRST Pub bug
C00357 00169 ∂08-Feb-79 1030 DCL Auto logout
C00358 00170 ∂08-Feb-79 1037 TW
C00372 00171 ∂08-Feb-79 1040 PAT phone call
C00373 00172 ∂08-Feb-79 1150 ME
C00374 00173 ∂08-Feb-79 1420 PAT
C00375 00174 ∂09-Feb-79 0713 Scott at SRI-KL (Scott J. Kramer) LISP MACHINES
C00376 00175 ∂09-Feb-79 0957 Scott at SRI-KL (Scott J. Kramer)
C00377 00176 ∂09-Feb-79 1023 CET
C00378 00177 ∂09-Feb-79 1302 Bmoore at SRI-KL (Bob Moore)
C00379 00178 ∂09-Feb-79 1502 RPG ...
C00380 00179 ∂09-Feb-79 1457 RPG Dislocated open file helper
C00382 00180 ∂09-Feb-79 1512 Feigenbaum at SUMEX-AIM what Hennessy really feels...
C00383 00181 ∂09-Feb-79 1528 Bobrow at PARC-MAXC Re:
C00384 00182 ∂09-Feb-79 2351 Kanerva at SUMEX-AIM Today's meeting
C00385 00183 ∂10-Feb-79 1316 ME
C00386 00184 ∂12-Feb-79 0922 HVA Some thoughts....
C00387 00185 ∂12-Feb-79 1243 Ijcai79 at SUMEX-AIM IJCAI-79 Computers and Thought Award
C00390 00186 ∂12-Feb-79 1557 DEW
C00391 00187 ∂12-Feb-79 1723 TOB steering committee meeting
C00392 00188 ∂12-Feb-79 1745 ZENON PYLYSHYN at CMU-10A (X320ZP51) Forthcoming visit & housing
C00394 00189 ∂12-Feb-79 1753 ZENON PYLYSHYN at CMU-10A (X320ZP51) Housing required for 1980-81 term
C00396 00190 ∂13-Feb-79 0937 DEW my oral exam
C00397 00191 ∂13-Feb-79 0948 RWG annihilation
C00399 00192 ∂13-Feb-79 1036 RWG
C00401 00193 ∂13-Feb-79 1115 Feigenbaum at SUMEX-AIM deal with musicians
C00402 00194 ∂13-Feb-79 1408 HAYDEN at USC-ISIE LNI Assembly Language Interface
C00403 00195 ∂13-Feb-79 1528 TW Promotion materials
C00407 00196 ∂13-Feb-79 1548 Levinthal at SUMEX-AIM AIL Agreements
C00412 00197 ∂13-Feb-79 1839 TOB offer to research associate
C00413 00198 ∂13-Feb-79 1933 TOB space and Levinthal agreement
C00414 00199 ∂13-Feb-79 1953 HVA Tapes
C00415 00200 ∂13-Feb-79 2208 Feigenbaum at SUMEX-AIM Re: Levinthal memo
C00418 00201 ∂14-Feb-79 0921 TOB steering committee
C00419 00202 ∂14-Feb-79 1033 PAT letters
C00420 00203 ∂14-Feb-79 1103 Bmoore at SRI-KL (Bob Moore)
C00421 00204 ∂15-Feb-79 0831 EF at MIT-AI (Edward Fredkin)
C00422 00205 ∂16-Feb-79 1912 HVA Progress...some...
C00424 00206 ∂17-Feb-79 0359 ME KA system
C00426 00207 ∂19-Feb-79 1235 TOB staff
C00427 00208 ∂19-Feb-79 1405 HAMMER at MIT-XX kudos
C00430 00209 ∂19-Feb-79 1559 TOB remind
C00434 00210 ∂20-Feb-79 0215 REM via AMES-TIP New writeup by me on my data-compression
C00436 00211 ∂20-Feb-79 0856 Feigenbaum at SUMEX-AIM need for sail account for professor kedes
C00438 00212 ∂20-Feb-79 0858 Nilsson at SRI-KL (Nils Nilsson) Interview
C00440 00213 ∂20-Feb-79 0902 Feigenbaum at SUMEX-AIM musicians/mj hall space
C00441 00214 ∂20-Feb-79 1054 DEW
C00442 00215 ∂20-Feb-79 1134 PAT airline reservations
C00443 00216 ∂20-Feb-79 1547 RPG New STEP feature
C00445 00217 ∂20-Feb-79 2036 JMC
C00446 00218 ∂20-Feb-79 2209 Feigenbaum at SUMEX-AIM (Response to message)
C00447 00219 ∂20-Feb-79 2209 Feigenbaum at SUMEX-AIM winograd papers
C00448 00220 ∂20-Feb-79 2220 Feigenbaum at SUMEX-AIM 2060
C00450 00221 ∂20-Feb-79 2230 Feigenbaum at SUMEX-AIM (Response to message)
C00451 00222 ∂20-Feb-79 2349 ME Xreply in E
C00453 00223 ∂21-Feb-79 0838 CET Administrative Vacancy
C00455 00224 ∂21-Feb-79 0941 Nilsson at SRI-KL (Nils Nilsson) Richard Weyhrauch's Course
C00456 00225 ∂21-Feb-79 1004 CVW Numbers and Course Evaluation Committee
C00459 00226 ∂21-Feb-79 1055 CET Mathematical Sciences Sherry Party
C00460 00227 ∂21-Feb-79 1326 CET
C00462 00228 ∂21-Feb-79 1448 PB Noticing your letter on the b-board, I thought you
C00464 00229 ∂21-Feb-79 1734 HVA Tapes
C00465 00230 ∂22-Feb-79 0946 CET SPRING TIME SCHEDULE
C00466 00231 ∂23-Feb-79 1334 Mcgoveran at SUMEX-AIM Faculty Meeting
C00467 00232 ∂23-Feb-79 1729 Bmoore at SRI-KL (Bob Moore)
C00468 00233 ∂25-Feb-79 0134 Kanerva at SUMEX-AIM Editing vs. compiling
C00475 00234 ∂26-Feb-79 0240 Feigenbaum at SUMEX-AIM winograd case
C00476 00235 ∂26-Feb-79 0243 Feigenbaum at SUMEX-AIM brahms requiem Friday
C00477 00236 ∂26-Feb-79 1050 Bmoore at SRI-KL (Bob Moore) meeting
C00478 00237 ∂26-Feb-79 1058 EF at MIT-AI (Edward Fredkin)
C00479 00238 ∂26-Feb-79 1135 DPB suggestions for instructors needed.
C00480 00239 ∂26-Feb-79 1256 DPB AI curriculum
C00482 00240 ∂26-Feb-79 1257 DPB Gray Tuesday Meeting
C00484 00241 ∂26-Feb-79 1323 TOB soroka
C00485 00242 ∂26-Feb-79 1354 PAT scheinman's address
C00486 00243 ∂26-Feb-79 1338 TOB video projection unit
C00487 00244 ∂27-Feb-79 0308 RWG
C00488 00245 ∂27-Feb-79 0932 Levinthal at SUMEX-AIM ai-lab agreements
C00489 00246 ∂27-Feb-79 1101 Bmoore at SRI-KL (Bob Moore) Aaron Sloman
C00490 00247 ∂27-Feb-79 1546 Tajnai at SUMEX-AIM Winter Quarter Final Exams
C00491 00248 ∂27-Feb-79 1727 Bmoore at SRI-KL (Bob Moore) AI/Phil workshop
C00493 00249 ∂27-Feb-79 1733 Mcgoveran at SUMEX-AIM Agenda
C00495 00250 ∂27-Feb-79 2157 RWG
C00496 00251 ∂27-Feb-79 2157 REM via AMES-TIP Data-compressed machine language
C00500 00252 ∂27-Feb-79 2234 Wilcox at SUMEX-AIM data-compressed machine language
C00503 00253 ∂27-Feb-79 2246 DPB Undergraduate advising
C00509 00254 ∂28-Feb-79 0827 BS Creary
C00510 00255 ∂28-Feb-79 0853 Mcgoveran at SUMEX-AIM faculty meeting
C00511 00256 ∂28-Feb-79 1212 DEK Chinese visitor
C00512 00257 ∂28-Feb-79 1450 DEW my oral committee
C00513 00258 ∂28-Feb-79 1646 Bmoore at SRI-KL (Bob Moore) Help us celebrate!
C00515 00259 ∂28-Feb-79 1647 PAT
C00516 00260 ∂28-Feb-79 1733 FB dovers,altos,ethernets,grinnells,networks -- news
C00518 00261 ∂28-Feb-79 2112 REF
C00519 00262 ∂01-Mar-79 0949 Mcgoveran at SUMEX-AIM Visitors from Tsing Hwa University
C00520 00263 ∂01-Mar-79 0954 DPB Re: Visitors from Tsing Hwa
C00522 00264 ∂01-Mar-79 1130 PAT
C00523 00265 ∂01-Mar-79 1135 PAT
C00524 00266 ∂01-Mar-79 1157 PAT expenses
C00525 00267 ∂01-Mar-79 1225 PAT copies of first
C00526 00268 ∂01-Mar-79 1322 ME
C00527 00269 ∂01-Mar-79 1357 Kanerva at SUMEX-AIM Editing and compiling
C00530 00270 ∂01-Mar-79 1446 HVA DIALNET
C00531 00271 ∂01-Mar-79 1627 TOB visit
C00532 00272 ∂01-Mar-79 1702 HVA Ignacio Zabala
C00533 00273 ∂01-Mar-79 1903 FB Elephant
C00534 00274 ∂02-Mar-79 0315 REM
C00536 00275 ∂02-Mar-79 1010 PRATT at MIT-AI (Vaughan Pratt)
C00541 00276 ∂02-Mar-79 1101 DEW reference
C00542 00277 ∂02-Mar-79 1140 ARK S-1 Meeting
C00543 00278 ∂02-Mar-79 1501 Susan Owicki <SSO at SU-AI> Why use modal logic
C00544 00279 ∂02-Mar-79 1503 Susan Owicki <SSO at SU-AI> mailing list
C00545 00280 ∂02-Mar-79 1505 PRATT at MIT-AI (Vaughan Pratt)
C00548 00281 ∂02-Mar-79 1628 HVA Available Funds in Unrestricted Account 2 FCZ 154
C00549 00282 ∂02-Mar-79 1849 ME
C00552 00283 ∂03-Mar-79 1026 FB Elephant and Lucid
C00553 00284 ∂03-Mar-79 1808 DCO
C00554 00285 ∂03-Mar-79 1929 PRATT at MIT-AI (Vaughan Pratt)
C00556 00286 ∂04-Mar-79 0815 PRATT at MIT-AI (Vaughan Pratt)
C00558 00287 ∂04-Mar-79 0845 SSO Net-mediated discussion on modal logic
C00560 00288 ∂04-Mar-79 2149 BPM More bureaucracy, less $$$
C00567 00289 ∂05-Mar-79 1031 PAT
C00568 00290 ∂05-Mar-79 1205 CLT
C00569 00291 ∂05-Mar-79 1528 DCO
C00572 00292 ∂05-Mar-79 1834 RPG Book
C00573 00293 ∂05-Mar-79 2044 JBROWN at MIT-MC (Jordan Brown) ACKERMAN'S FUNCTION
C00576 00294 ∂05-Mar-79 2258 LLW Final ABOX Review
C00577 00295 ∂05-Mar-79 2316 LLW ABOX Review Reading
C00578 00296 ∂06-Mar-79 1027 Bmoore at SRI-KL (Bob Moore) Sloman's article...
C00582 00297 ∂06-Mar-79 1524 Bmoore at SRI-KL (Bob Moore) Modal logic vs. its possible-world translation
C00590 00298 ∂06-Mar-79 1559 Mcgoveran at SUMEX-AIM Chinese visitors
C00592 00299 ∂06-Mar-79 1657 LES Ethernet and friends
C00593 00300 ∂06-Mar-79 1705 PAM Petit channel
C00594 00301 ∂06-Mar-79 1727 PAM
C00595 00302 ∂07-Mar-79 0905 Feigenbaum at SUMEX-AIM Re: Ethernet and friends
C00596 00303 ∂07-Mar-79 0917 Feigenbaum at SUMEX-AIM thought you might be interested
C00598 00304 ∂07-Mar-79 0917 Feigenbaum at SUMEX-AIM letter to chowning
C00599 00305 ∂07-Mar-79 1349 SWB Phil Agre
C00600 00306 ∂07-Mar-79 1429 ES
C00604 00307 ∂07-Mar-79 1444 Bmoore at SRI-KL (Bob Moore) letter to study group participants
C00610 00308 ∂07-Mar-79 1856 LLW LEE S. PARKS--SU/CS APPLICANT
C00612 00309 ∂07-Mar-79 1904 KENNETH KELLER at CMU-10B school
C00620 00310 ∂07-Mar-79 1927 PRATT at MIT-AI (Vaughan Pratt)
C00621 00311 ∂07-Mar-79 1928 PRATT at MIT-AI (Vaughan Pratt)
C00624 00312 ∂07-Mar-79 1937 JMC
C00625 00313 ∂08-Mar-79 0844 DPB
C00626 00314 ∂08-Mar-79 1016 PHY Vanity plates
C00627 00315 ∂08-Mar-79 1219 Tajnai at SUMEX-AIM MACLISP Manual
C00628 00316 ∂08-Mar-79 1234 DCO
C00632 00317 ∂08-Mar-79 1405 LES Foonly channel
C00633 00318 ∂08-Mar-79 1451 TOB Sid Liebes
C00634 00319 ∂08-Mar-79 1530 Jrobinson at SRI-KL (Jane Robinson) Bob Moore's graduation present
C00635 00320 ∂08-Mar-79 2306 GIO Trip to LIvermore and S-1
C00637 00321 ∂09-Mar-79 0933 HVA National Geographic
C00638 00322 ∂09-Mar-79 1442 MRC
C00639 00323 ∂09-Mar-79 1615 Ijcai79 at SUMEX-AIM IJCAI Reviews
C00640 00324 ∂09-Mar-79 1900 JMC*
C00641 00325 ∂10-Mar-79 0044 REM via AMES-TIP
C00642 00326 I think I sent it but please check.
C00643 00327 ∂11-Mar-79 1748 HVA
C00644 00328 ∂12-Mar-79 1151 PAT
C00645 00329 ∂12-Mar-79 1357 BCM
C00646 00330 ∂13-Mar-79 0949 100 : patte
C00647 00331 ∂13-Mar-79 1043 PAT change of plans
C00648 00332 ∂13-Mar-79 1155 PAT Sarah's tuition
C00649 00333 ∂13-Mar-79 1419 REP Summer Support
C00651 00334 ∂13-Mar-79 2120 FML theory of computation quals
C00653 00335 ∂14-Mar-79 0024 RP
C00654 00336 ∂14-Mar-79 0814 Tajnai at SUMEX-AIM ( Forwarded Mail )
C00656 00337 ∂14-Mar-79 1317 DCL
C00657 00338 ∂14-Mar-79 1333 ME character sets
C00660 00339 ∂14-Mar-79 1342 DCL
C00663 00340 ∂14-Mar-79 1743 Amarel at SUMEX-AIM meeting
C00664 00341 ∂15-Mar-79 0931 Bmoore at SRI-KL (Bob Moore) Held over by popular(?) demand
C00667 00342 ∂15-Mar-79 1555 HVA Sat. Mar. 17th - 7 p.m.
C00668 00343 ∂15-Mar-79 1739 PEAGRE at MIT-AI (Philip E. Agre) meeting
C00669 00344 ∂15-Mar-79 2243 RAK Dear staffperson
C00670 00345 ∂16-Mar-79 1444 Gifford at PARC-MAXC Meeting to discuss CS258 Problems
C00671 00346 ∂16-Mar-79 1730 ME
C00674 00347 ∂17-Mar-79 0347 MRC new format packet
C00676 00348 ∂18-Mar-79 1049 CLT
C00677 00349 ∂19-Mar-79 0944 DPB Help!
C00679 00350 ∂19-Mar-79 1423 DPB Meeting re Jacks space and teaching plans next year
C00680 00351 ∂19-Mar-79 1458 DCO
C00681 00352 ∂19-Mar-79 1830 MRC Dialnet at SAIL
C00682 00353 ∂19-Mar-79 1941 100 : MRC
C00683 00354 ∂19-Mar-79 2210 DPB
C00697 00355 ∂19-Mar-79 2216 MRC grimlacs
C00698 00356 ∂19-Mar-79 2235 TOB your course
C00699 00357 ∂20-Mar-79 0233 LLW
C00701 00358 ∂20-Mar-79 0955 PAT aimemos
C00703 00359 ∂20-Mar-79 1109 Boyer at SRI-KL (Bob Boyer) A Talk on Automatic Theorem Proving
C00705 00360 ∂20-Mar-79 1116 PAT lib.lst
C00706 00361 ∂20-Mar-79 1440 DON
C00707 00362 ∂21-Mar-79 0307 MRC Forwarded without comment
C00710 00363 ∂21-Mar-79 1314 TOB soroka appointment
C00711 00364 ∂21-Mar-79 1928 REM
C00715 00365 ∂21-Mar-79 2003 ME
C00716 00366 ∂21-Mar-79 2040 REM via AMES-TIP POX and your file
C00718 00367 ∂22-Mar-79 0858 BPM
C00724 00368 ∂23-Mar-79 0218 RWG
C00726 00369 ∂23-Mar-79 0701 PJH via LONDON CASBS
C00727 00370 ∂23-Mar-79 1019 FRG CS258
C00728 00371 ∂23-Mar-79 1044 FRG CS390 Sec19
C00729 00372 ∂23-Mar-79 1047 DEW Oral exam
C00730 00373 ∂23-Mar-79 1312 PAT Jeff Ullman
C00731 00374 ∂24-Mar-79 2144 REM via AMES-TIP Second draft of my data-compression paper ready to review
C00733 00375 ∂25-Mar-79 0302 POURNE at MIT-DMS (Jerry E. Pournelle) Mostly hello
C00735 00376 ∂26-Mar-79 0813 BS Telephone Message
C00736 00377 ∂26-Mar-79 1011 PAT phone call
C00737 00378 ∂26-Mar-79 1533 DEW My oral exam
C00738 00379 ∂26-Mar-79 1648 DPB
C00739 00380 ∂27-Mar-79 1039 FRG my whereabouts
C00740 00381 ∂27-Mar-79 1709 ME
C00744 00382 ∂28-Mar-79 0940 Mcgoveran at SUMEX-AIM Faculty Meeting to Vote on Degree Candidates
C00745 00383 ∂28-Mar-79 1005 PAT whereabouts day 2
C00746 00384 ∂28-Mar-79 1413 REM via AMES-TIP Arpanet protocol violation by SU-AI
C00748 00385 ∂28-Mar-79 1439 REM via AMES-TIP Draft version of paper on my data-compression results.
C00750 00386 ∂29-Mar-79 2028 Jrobinson at SRI-KL (Jane Robinson) supper
C00752 00387 ∂30-Mar-79 0710 Feigenbaum at SUMEX-AIM Chowning and your letter
C00754 00388 ∂30-Mar-79 0811 Jrobinson at SRI-KL (Jane Robinson) Understand the problem
C00755 00389 ∂30-Mar-79 0913 Bmoore at SRI-KL (Bob Moore) Reminder
C00756 00390 ∂30-Mar-79 1108 FRG grading procedures
C00757 00391 ∂30-Mar-79 1649 RPG LISP reminder
C00758 00392 ∂30-Mar-79 2101 Jrobinson at SRI-KL (Jane Robinson) Tuesday's fine.
C00759 00393 ∂31-Mar-79 1245 SSO Elephant
C00760 00394 ∂31-Mar-79 1531 DCO corky
C00761 00395 ∂31-Mar-79 2039 ME disk drives
C00762 00396 ∂01-Apr-79 1149 Mark Crispin <Admin.MRC at SU-SCORE> Dialnet progress
C00764 00397 ∂01-Apr-79 1150 Mark Crispin <Admin.MRC at SU-SCORE> Sumex ''Dialnet''
C00766 ENDMK
C⊗;
∂02-Jan-79 1043 Gaschnig at SRI-KL Donald Michie's address in Oxford
Date: 2 Jan 1979 1040-PST
From: Gaschnig at SRI-KL
Subject: Donald Michie's address in Oxford
To: mccarthy at SAIL
Just a reminder...If you have the address in your files, I'd appreciate it.
Thanks.
Cheers (and happy new year),
John
-------
∂02-Jan-79 0352 MRC the future (long message)
By the way, I am going to try to get my application in for the 2060 job
later today. You may be assured from me personally that if I get the
job, I will continue to be interested in Dialnet. In fact, in that
position I will have a need for Dialnet in order to exchange Tops-20
files with LOTS. Since the WAITS code is essentially complete and
anybody (myself included) could write the user code without much
difficulty, my going to the 2060 should not harm Dialnet.
It is presently unclear what will happen if I do not get the 2060 job.
As you know, I consider my present salary and position at SAIL to be
inadequate. So far I've heard a lot of promises but haven't seen
anything that I can deposit in the bank, however I certainly cannot
leave SAIL unless I have someplace to leave to.
Even if I was given a $2000 raise, that would only equal what I was
offered as a starting salary in Cambridge two years ago. At the time I
accepted the job here, the pay was unimportant compared to being able to
live in California; however now that I am here I have grown accustomed
to things like eating at Louis'. It is also frankly demoralizing to be
one of the few System Programmer I's at Stanford. In the minds of most
people a "one" implies both inexperience and that my services are worth
somewhat less than industry standards. I have been doing machine
language programming on the PDP-10 for over five years now, and am one
of the few people who is familiar with all of the PDP-10 monitors in use
today. I think I deserve more than that.
One problem with the AI Lab is that the present environment discourages
creativity. If you have an idea that BH (or ME, or whomever is numero
uno at the time) disagrees with, you are forced to resort to subterfuge
if you want your idea to see the light of day. I do not like other
people having such veto power over me.
I believe that I do have good ideas that do not necessarily agree with
numero uno's ideas. I also believe that I can responsibly decide which
ideas I can act upon. If afterwards I decide it was a bad idea I would
undo it. I should never be FORCED to do so; in a month everybody might
fnd up agreeing with me, or perhaps I am convinced I was wrong.
This kind of environment does not presently exist at SAIL, and I feel
its absense has hurt SAIL severely both in the past and in the present.
It depresses me both that it is happening and that I am powerless to do
anything about it. My hope is that if and when I manage a computer
installation that I will do so in the way SAIL has.
Mark
∂02-Jan-79 0250 MRC What we need from DEC (reminder)
Here is a complete (I hope) list of what I (or somebody else for that
matter) will need from DEC to play with the front end starting from
ground zero.
(1) RSX-20F (ie, TOPS-20 front end) sources. Preferably a TOPS-20
format disk pack, although a magtape (in TOPS-20 DUMPER format)
or RSX-format disk pack would be alright I guess. This is in
order of how difficult it would be for me. If it is an RSX disk
pack, I will need instructions on how to use it on a TOPS-20
hardware configuration.
(2) PDP-11/45 processor handbook.
(3) PDP-11 software handbook.
(4) PDP-11 peripherals handbook.
(5) As much RSX-11 documentation as they are willing to part with.
Obviously I don't need the FORTRAN compiler user's guide, but
I do want the obvious user and wizard documentation. I don't
have a list of what's available, but a DEC front end wizard
should know what I'd want.
I already do have the "RSX-20F System Specifications" manual,
which was distributed by DEC at DECUS in San Francisco. While
this manual is very useful, it does not eliminate my need for
everything else. Hopefully DEC will not send this figuring it
is enough.
There is nothing I can do until I get this stuff, other than perhaps
playing with a PDP-11 to learn the instruction set. I can't even
work on fixing the existing bugs in the WAITS Dialnet service due to
this KA-10 mucking around.
Mark
∂02-Jan-79 1040 HVA Vera's Tapes
Before the day becomes more hectic here, I wanted to let you know that I have
made progress - be it ever so slow - and finished transcribing Tape # 1 at
home this week-end. The original time estimate (about 5-6 hours for all 4)
proved to be a bit optimistic - there are many names of places I do not
know (my spelling, therefore, is highly original), and there are some great
gaps where things were not at all clear to me. Thus, I think the first draft
on paper is a more realistic way for me to have begun - at home, I can turn
the volume to maximum, there are no interruptions to distract, and I am able
to spend more time (e.g., an hour or two in the evening and more on week-ends),
than would be possible at the Lab. I do know that you wanted me to do this
on the system and I will type it in when the unknowns are filled in. Mean-
while, I wanted to let you know how it was going.
∂02-Jan-79 1316 Hazen at SUMEX-AIM Siglunch this Friday, Jan. 5 in Chemistry Department Gazebo
Date: 2 Jan 1979 1315-PST
From: Hazen at SUMEX-AIM
Subject: Siglunch this Friday, Jan. 5 in Chemistry Department Gazebo
To: siglunch:
cc: Hart at SRI-KL, JMC at SAIL, Winograd at PARC,
cc: Deutsch at PARC, Teitelman at PARC, CCG at SAIL
All,
This Friday, January 5 at noon, we will be having our first siglunch of this year.
It will be held in the Chemistry Department Gazebo as before. Professor Ed Feigenbaum
will be the speaker for this Siglunch. He will give a report on the workshop on
Symbolic Computing Machines. Professor Feigenbaum will summarize the discussions at the
recent MIT workshop on LISP machines and other symbolic computing alternatives. An
important discussion of the LISP machine concept will be held.
Anyone wishing to attend this Siglunch but does not know where the Chemistry Department
Gazebo is located, please call me, Marion Hazen, at 497-4878, for the directions.
REMEMBER! January 5, at noon, in Chemistry Dept. Gazebo.
Marion
-------
∂02-Jan-79 1554 ARK via SU-TIP Mailing Lists
To: S1 at SU-AI
New external names have been set up. Mail about the runtime may be sent
to S1RUNTIME which sends it to those on the mailing list RUNTIM[DIS,S1]
and mail about the LINKER (and I suppose loader as well) may be sent to
S1LINKER which sends it to those on the mailing list LINK[DIS,S1].
If you alter one of these mailing lists, remember that people who prefer
that they receive their mail elsewhere even though they have local
accounts should have their remote account (with full ARPANET address)
on the list even though their mail is forwarded.
Arthur
∂02-Jan-79 2219 JB YOUR SLOT IN COMPUTER FORUM.
I would be interested in presenting GOAL at the COMPUTER FORUM.
Fine. You're on. Please talk to whoever is in charge.
∂03-Jan-79 0123 RWW
This course will be given. The time is 99% certain. The place is to be
determined. It starts Thursday Jan 4 and I will more info as to the place.
please feel free to call me and ask about it.
Title: Philosophy and AI
Winter Quarter
TTH 2:30 - 3:45
3 units
Instructor: Richard Weyhrauch
No Prerequisites. This course is for graduates and undergraduates with an
interest in AI. Text will consist of lecture notes and assigned readings.
Students will be expected to listen to the lectures, participate actively
in discussions and read the assigned readings. Course projects will be
available for anyone interested but they are not a requirement.
In this course I intend to address the question: what kind of thing should
a researcher in artificial intelligence be trying to build? I'll
introduce the idea of a "computer individual" (an electro-mechanical
associate) and ask what properties such an individual should have.
The course will have three parts.
In the first part you will be asked to suspend your judgement for a while
and learn some formal logic. This part of the course will be entirely
self contained and will not require any prior knowledge of logic. In fact
previous knowledge will probably get in your way for a while. There are
two purposes to this part of the course. The first is that the rest of
the course depends on your your technical understanding of the ideas I
will present. The second is that logic has been used only in a very
simple way in AI up to now and I hope to expand your perspective as to how
it might be used. To do this you need to know some technical things and
it is these I intend to teach. This will not be a logic course. It will
not cover the standard material in a logic course. It is a practical
course in building artificial intelligence systems. We will sometimes use
logic as a tool in way that it hasn't been used before. I hope to
substantially alter your view of what logic is good for. If I don't by
the end of this course I will consider it a failure. The points of logic
will be illustrated by the use of the FOL reasoning system, which embodies
these ideas.
The second part of the course will be taken up with relating the notions
of logic introduced in the first part to the question of building AI
systems capable of making intelligent conclusions from the data it has.
One section will be taken up with looking at problems that have been dealt
with using the McCarthy situation calculus style of representation. We
will introduce a new notion of situation and use blocks world problems as
an illustration of how to use this idea. We will work out D.Michie's keys
and boxes problem in detail. The discussion will include remarks about
STRIPS and SHRDLU. We will then use the meta-theoretical structures we
learned earlier in the course to explore the idea of machine learning and
problem solving in mathematical theory of computation. We will finish by
discussing reasoning about knowledge (one's own and other peoples),
"non-monotonic logic", and in general the problems of reasoning using
modalities. This is just a brief sketch of the topics that will be
mentioned. The purpose of this part of the course is to gain facility
with the use of the notions introduced at the start of the course and to
relate them to the kinds of reasoning systems that people currently find
interesting.
The third part of the course will directly address the question of
constructing electro-mechanical individuals - NOT the question of AI
programs. This shift of viewpoint will be central to our discussion of
the question: how is it that such a device can build a functional
understanding of his environment? This part of the course will discuss
many aspects of the problem of artificial cognition. I am interested in
several questions here. 1) how is perception possible, that is, how do we
get from sense impressions to theories of our world? In particular I view
natural language understanding as a perception problem, i.e., how do we
get from the stream of auditory input to an understanding of its meaning?
2) What is the origin of natural language? 3) What is a conversation? A
lot of this course will make remarks about the QUALITY of the conversation
we want to be able to have with our computer friend. In AI we talk about
common sense reasoning. I want to look at all of the above problems as
practical questions about what our common sense understanding of our world
is like and how we acquire and use it. These later questions will take us
far afield, with readings from psychology, philosophy, mysticism, politics,
etc.
In the end I hope that we will have had a good time exploring what a wide
range of contemorary ideas tell us about AI. In addition the technical
tools presented in my lectures are meant to describe practical and
implementable ways of embodying some of these ideas into an artificial
intelligence system.
∂03-Jan-79 0321 MRC front end
Apparently Ralph seems to be trying to get front end sources for us
too. Are your efforts coordinated with his?
∂03-Jan-79 1356 Jrobinson at SRI-KL (Jane Robinson) eat first
Date: 3 Jan 1979 1355-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: eat first
To: jmc at SU-AI
at my house, ok? Probably pot roast. Bill Paxton will also be with
us. I thought we would gather around 5:30, plan to sit down to eat about
6 pm, and get to Physics Lecture Hall 100 around 8 pm for doubtless crowded
8:30 lecture. Then the 5 of us will have a chance to sit together and,
as Bill says, be mystified together.
Jane
-------
Carolyn forgot that she would be in Idaho, but anyway 5:30 at your place
sounds fine.
∂03-Jan-79 1454 CARLSON at USC-ISI Your Comments on RTSC Meeting
Date: 3 JAN 1979 1447-PST
From: CARLSON at USC-ISI
Subject: Your Comments on RTSC Meeting
To: mccarthy at SAIL
cc: carlson
Dear John:
Thanks for you comments on the workshop. Your "History of
LISP" paper was most instructive. I have given the four
criteria you lay down for the ADA programming environment
to Dave Fisher. He will be putting together a technical
summary of ADA environment issues during the next few months.
I'll make sure you get copies of the material as it
is available.
Bill
-------
∂03-Jan-79 1520 JP INTERLISP manuals
To: "@ILISP.DIS[1,JP]" at SU-AI
They have arrived. Check your box. People with offices in Polya or Serra
should be getting it tomorrow via ID mail.
∂03-Jan-79 1537 JB COMPUTER FORUM.
Thank you very much! I'll proceed to make arrangements.
∂04-Jan-79 0057 Feigenbaum at SUMEX-AIM LISP and programs on 8080
Date: 3 Jan 1979 2240-PST
From: Feigenbaum at SUMEX-AIM
Subject: LISP and programs on 8080
To: siglunch:
Prof. Dave Stoutemeyer from the Univ. of Hawaii will be giving a demonstration
of a symbolic algebra system operating on an 8080 in the conference
room of Serra House at noon on THURSDAY.
He will demo his LISP system for the 8080 in the Serra House Conf. Room
on FRIDAY, beginning about 1:30 PM.
Should be entertaining.
Ed Feigenbaum
-------
∂04-Jan-79 1155 ARK S1 Meeting
To: S1 at SU-AI
There will be an S-1 Meeting in the Serra Conference Room on Wednesday,
10 January at 1:15 pm. I hope that this time is convenient to all.
Arthur
∂04-Jan-79 1603 Wiederhold at SUMEX-AIM Re: S1 Meeting
To: S1 at SU-AI
Date: 4 Jan 1979 1558-PST
From: Wiederhold at SUMEX-AIM
Subject: Re: S1 Meeting
To: Arthur Keller <ARK at SAIL, S1 at SU-AI
cc: ark at SAIL
In response to the message sent 4 Jan 1979 1200-PST from Arthur Keller <ARK at SU-AI>
OOPS, please chegk with Jayne before setting up meetings - I teach then.
3pm would be ok. Please RSVP if you cant make it.
ARK - why not a general S-1 meeting ?
gio
-------
∂04-Jan-79 2031 ARK S-1 Meeting Rescheduled
To: S1 at SU-AI
The S-1 Meeting has been rescheduled for 2:30 pm on Wednesday, 10 January
in the Serra House Conference Room. I hope that this rescheduling has not
caused any inconvenience and that the new time is acceptable. If the new
time is not convenient, send me a message suggesting alternate times which
will be considered at the meeting.
Arthur
∂05-Jan-79 1615 100 : Steve Robbins (STEVER@MC)∩∩∩ via NBS-TIP Dialnet
?
Excus eme, I am very intrested in joining Dialnet, where may I obtain
the requirements, protocols and/or host phone numbers? What language
s have the protocols already been written in? I guess what I am asking for
is general info. Please reply to STEVER@MC...
Steve Robbins
Ask MRC@SU-AI.
∂05-Jan-79 1659 Hart at SRI-KL (Peter Hart) Joint publication
Date: 5 Jan 1979 1658-PST
From: Hart at SRI-KL (Peter Hart)
Subject: Joint publication
To: Feigenbaum at SUMEX-AIM, JMC at SAIL
cc: Hart, BMOORE, SACERDOTI
Ed,
When discussing graduate students at SRI a while back, we
agreed that theses ought to be published simultaneously under SU and
SRI covers, with each cover mentioning, e.g., "also published as xxx."
JMC has indicated an interest in having Bob Moore's thesis come
out as an SU report as well as an SRI one. While Bob does not precisely
come under the "standard arrangement," I think we might as well start
with his.
Comments?
Peter
-------
∂05-Jan-79 1708 SSO Seminars and Parties
To: "@MODAL.DIS[1,SSO]" at SU-AI
First, I would like to have a party for modal logic seminar participants and
hangers-on. How does next Saturday (the 13th) sound to you? (that will
be the date unless it conflicts with something major.) It will be an
informal potluck (not dinner, just munchies) at my house. Directions later.
Second, the modal logic seminar will not exist this quarter; it is metamorphosing
back into a concurrent programming seminar. It will meet at 4:00 Thursday in
Polya 204. Jerry Feldman will be telling us about his language and system
for distributed computing. He hopes to stir up arguments and lively discussion.
I have a paper on the language PLITS, which is recommended reading for the next
meeting.
The modal logic seminar may or may not revive in the spring.
Sue
∂05-Jan-79 2200 REM via AMES-TIP Losing MacLisp also...
To: JIRAK at SUMEX-AIM, RWW at SU-AI, JMC at SU-AI,
WILBER at SRI-KL
Well, after giving up on Mainsail for my data-compression work, and finding
FAIL (PDP-10 assembly-language) too much work for any serious programs, I
tried converting my most difficult program to MacLisp. I'm getting a
strange error, (UNBOUND) undefined function, called from inside a system
routine LSH (Left-shift). Unless somebody fixes MacLisp for me I guess I
have no decent way to proceed with data-compression research. I'll give
MacLisp a few days, and if no fix I'll give up on it and spend my hours
writing in FAIL, bletch...
∂05-Jan-79 2259 RSP via Easy Street EFIND default file
To: LES, JMC
I am looking at satisfying a long ignored request for a default file
feature in EFIND. A default file could be specified in OPTION.TXT as
currently with FIND. Is there some file which would be suitable as a
system-wide default for people who do not put an EFIND: entry in their
OPTION.TXT? I believe the best candidate is PEOPLE.DAT[PER,CSD]. If you
know of a better choice, please let me know.
∂06-Jan-79 1803 ME digest
∂06-Jan-79 1534 JMC
NS hasn't picked up any digests since the 2nd.
ME - Looks like there has been some garbaging at just the times when
the digest has been moving during the last few days, and so NS didn't
detect those digests that actually were sent. It shouldn't keep going
like that unless AP is actually having some trouble transmitting them.
∂06-Jan-79 1858 REM via AMES-TIP Partial success with MacLisp
To: "@CRUNCH.DIS[1,REM]" at SU-AI, "@CRUPUB.DIS[1,REM]" at SU-AI,
RPG at SU-AI, GLS at SU-AI, JIRAK at SUMEX-AIM,
GSB at MIT-ML
I gave up for time-being getting Mainsail or MacLisp to do what I wanted,
but managed to coax MacLisp into doing half of what I want. MacLisp
program combined with Mainsail version of SSORT can now produce IRSMXX.XX*
files apparantly moderately efficiently, to be read into IC4B program,
but don't have time now to test on large file to see how long it takes.
More later....
∂07-Jan-79 1014 REM Success with NOTICE.[UP,DOC] survey for compression contexts
To: "@CRUNCH.DIS[1,REM]" at SU-AI, "@CRUPUB.DIS[1,REM]" at SU-AI,
RPG at SU-AI, GLS at SU-AI, GSB at MIT-ML
Using my MacLisp program for (1) scanning a file according to a set of
5702 left-contexts in order to generate histogram batches (only about
8k words can be surveyed in one core-image, after which that data must
be dumped to disk to make room for next batch), and (after SSORT written
in Mainsail has sorted the batches to bring all parts of a given histogram
together and to sort the histograms according to length) (2) merging all
pieces of each histogram into one and computing entropy and writing the
context+count+entropy+overhead figures out to disk files -- total time
for MacLisp and SSORT when surveying a 93824-word file NOTICE.[UP,DOC]
is 2 hours 30 minutes -- result is then fed into IC4B (Mainsail program)
to compare entropy of each context against its parent, to select those
contexts which yield significant improvement (net reduction of entropy),
and SSORTing the result to get good contexts in ascending order by worth
(best at the end), time 0 hours 6 minutes -- result is NOTICE.CXT[1,REM]
which has all 1752 worthwhile contexts. There are 500 which obtain at
least 214 bits of reduction each, 250 more which obtain between 133 and 214
bits reduction each, 250 more which obtain between 88 and 133 bits red.,
and the remaining 752 which obtain between 10 and 88 bits reduction.
Normally the best 500 (214 bits or more each) would be used, but in a
large 370 perhaps more would be used. On a microprocessor, much fewer
contexts would be used, just the very best. -- It will now take some time
to analyze these results in various ways to decide how best to optimize
my programs. Note, all contexts to length 20 with sufficient raw count
of occurances (I think I cut it at count=20) were among the 5702 trial
contexts. Only 1752 of them were at all reasonably useful, many just
marginally, thus only about one quarter were reasonably useful. This
shows the advantage of the new optimizing method over the method used
2 years ago, but more study will be needed to get a better picture of
how the various cutoffs (thresholds) relate to each other. More (to
those on CRUNCH or CRUPUB mailing list) after I analyze these results....
∂07-Jan-79 1331 DCO
To: "@GROUP.DIS[VCG,DCL]" at SU-AI
***********************************************************************
NO SEMINAR ON TUESDAY 5TH DEC.
***********************************************************************
VERIFICATION GROUP SEMINAR
TUESDAY 9th JANUARY
PLACE: ERL 237
TIME: 2:30 p.m.
TITLE: Data Type Definitions in TYPED LISP
SPEAKER: R. S. Cartwright
ABSTRACT
TYPED LISP is an applicative programming language designed to
simplify the task of reasoning about programs. The most important
component of the language is a comprehensive data type definition facility.
Despite the currect popularity of the algebraic specification approach to
defining data types, TYPED LISP rejects this methodology. Instead,
TYPED LISP includes a small set of mechanisms for constructing new
types from existing ones. The talk will concentrate on motivating
and describing the constructive approach to data type definition
employed ty TYPED LISP.
∂08-Jan-79 0201 REM via SU-TIP Dialnet, PCNet, et al
Many small businesses could be based on network (dial-up) access as
a way of doing business, once these networks become operational. One
that comes to mind is making labels. It's a pain on large system to
load line-printer with blank labels, lock out the spooler, adjust
everything, make a listing, get disgusted over horrible misalignment
of crufty printer, load normal paper back in, and verify that spooler
is working and paper is alighed correctly. For small batches of labels
(one to a few hundred) it would be more convenient to simply auto-dial
to a Dialnet or PCNet node that provided label-making service, send the
text of the labels and a Rivest-coded contract for payment, and then
either pick up the labels when a return message announces they are done
or wait for them to appear in US-mail or ID-mail. One small printer,
narrow (only 40 columns are needed, not 132 or 140), and well-maintained,
would suffice to put a node into business.
∂08-Jan-79 0314 REM via SU-TIP
See LABEL.WRU[1,REM] for more on that idea, feedback????????
∂08-Jan-79 0620 REM via SU-TIP Overview of data-compression methods
To: "@CRUNCH.DIS[1,REM]" at SU-AI, "@CRUPUB.DIS[1,REM]" at SU-AI
TASK 1 -- Get by some crude means a list of left-contexts to consider.
Presently it takes one pass across the file. A FAIL (PDP-10 assembly
language) program (neither Mainsail nor MacLisp can do this part) grows
trees in core and produces a whole bunch of output. A Mainsail program
consisting of two normal parts and two calls to sort/merge module combines
this data and selects the 100000 largest-count contexts or all above a
threshold whichever is fewest. The result is file X.5
TASK 2 -- For each proposed context, compute the actual entropy of the
characters in that context in the file, and select those contexts which
make a significant reduction in entropy over their parents. This is the
most time-consuming task! A MacLisp program loads up X.5 into a tree in
memory, then scans the file to build up chunks of histograms that don't
quite fill up all of memory (this is dangerous, if the 128k limit is
exceeded, MacLisp crashes and the run is lost, but if only 20000 histogram
characters are kept in core at any one time the program only usus about
110-120 k usually so it doesn't crash, usually), dumping these chunks to
disk. A Mainsail sort/merge program then sorts this file to bring together
all pieces of the histogram for each left-context. (The SAIL program SSORT
can't handle this task, it thrashes badly due to algorithm used in 1971
when it was written.) MacLisp is called again to collate and add up all
those histogram segments, compute entropy, and write IRSMXX.XX* files.
Finally a Mainsail program compares father vs. son entropy and generates
a file of improvement-in-entropy, context, and raw-count. It calls the
sort/merge routine one last time to sort according to improvement in
entropy. The user renames this to *.CXT where * tells how this survey
was obtained.
TASK 3 -- Given these useful contexts, the file is scanned one last time
(before actual crunch/uncrunch) according to longest-matching-context,
and the true entropy and Huffman-code are computed. This part is
incomplete due to a couple months of research trying to get TASK 2 to
run moderately efficiently. Eventually it will decide whether to use
an abbreviated Huffman code or an interval-refinement code, and will
generate the polish that specifies the code.
TASK 4A -- Crunch the file according to the code. (not written, although
for abbreviated Huffman-code there is 1976 software that still works)
TASK 4B -- Uncrunch the file according to the code. (ditto)
∂08-Jan-79 1143 PN Agenda for next meeting
To: S1 at SU-AI
A reminder: agenda items for the next meeting are supposed to be put in
AGENDA[DOC,S1]. So far, it looks like the only thing we have to talk about
is Fortran. Is this really the case?
∂09-Jan-79 0204 REM via SU-TIP Mainsail loses.
To: RWG at SU-AI, RWW at SU-AI, JMC at SU-AI, AJT at SU-AI,
PRATT at MIT-AI
Not only are the Mainsail people unable and unwilling to make it
possible for a user-level program to find out how close it is to
running out of memory, they can't even understand how such a feature
could possibly be used by any program even though I've already written
several FAIL programs that use that feature manually implemented on
bare-machine essentially. Bletch. At least the MacLisp people sort-of
understand how it might be useful although can't tell how to make
MacLisp do that for me.
∂09-Jan-79 0808 S1 DISK quota
To: "@S1[DIS,S1]" at SU-AI
S1 is over quota by 3,019,152 and is in danger of being purged!!!!!
∂09-Jan-79 1331 SSO Party
To: "@MODAL.DIS[1,SSO]" at SU-AI
MODAL LOGIC (and whatever) group party -- Sat. Jan. 13, 8:30
at Owicki's -- 35 Pearce Mitchell Place on the Stanford Campus
Guests and spice welcome.
Please bring some kind of munchies or drink.
Pearce Mitchell Housing is on the Southeast corner of the intersection
of Campus Drive and Mayfield Avenue.
If you enter Pearce Mitchell at the second entrance, you will be right
at #35 (it's on the first floor, on the left side of the entrance walk).
Tresidder Union
X Mayfield Avenue
X
X
X
X
X
X
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX Campus Drive XXXXXX
X
X I----------------------I
X I I
XXXXX I Pearce Mitchell I
X I Housing I
X I I
X I I
X I I
XXXXX I #35 I
X I I
X I I
X I I
X I----------------------I
X
X
X
X
∂09-Jan-79 1542 RPG MacLisp matcher
To: "@USERS.DIS[AID,RPG]" at SU-AI
Here's some changes to %MATCH you matcher fans:
;;;;;;;;;; the matching function ;;;;;;;;;;
;;;
;;; (arg 1) - p - pattern
;;; (arg 2) - d - data
;;; (arg 3) - alist - optional list of variables (* or ?) whose values
;;; are to be retained during the match, much like the
;;; = variables below.
;;; elements of a pattern:
;;; ? - matches anything
;;; * - matches one or more expressions
;;; ?<atom> - like "?", but sets ?<atom> to thing matched
;;; *<atom> - like "*", but sets *<atom> to list of things matched
;;; =<atom> - matched against value of <atom>
;;; (restrict <one of above ?-variables> <pred1> <pred2> .....)
;;; - the predi must eval to non-nil
;;; $r, ⊗r - same as RESTRICT
;;; (restrict <one of above *-variables> <pred1> <pred2> .....)
;;; - the predi must eval to non-nil when given the list
;;; that is being considered for that variable as its argument
;;; (irestrict <one of above *-variables> <pred1> <pred2> .....)
;;; - the predi must eval to non-nil when given each element of the list
;;; that is being considered for that variable as its argument
;;; (done incrementally). So %MATCH will apply these predicates as
;;; it scans the input.
;;; $ir,⊗ir - same as irestrict
;;;
;;; (%match p d <variables to retain>) attempts to match p against d
;;; (%continue-match p d <variables to retain>) attempts to get the next
;;; possible match between p and d (by different *-variable
;;; bindings.
;;*PAGE
What is the basic reference on MATCH?
∂09-Jan-79 1548 RPG
∂09-Jan-79 1545 JMC
What is the basic reference on MATCH?
It is the pattern matcher that 1. the MacLisp editor uses and 2. the
various macro definition primitives at the Lab use. It is a general
purpose matcher for trees and lists. The code is in MATCH.>[aid,rpg]
(you can actually say ET MATCH.>[AID,RPG] and get the right thing. It has
a numerical extension).
What else would you like to know?
-rpg-
Is the matcher suitable for general use as part of a lisp program,
and is it worth studying from the point of view of including material
on matching in CS206 and the book?
∂09-Jan-79 1604 RPG
∂09-Jan-79 1550 JMC
Is the matcher suitable for general use as part of a lisp program,
and is it worth studying from the point of view of including material
on matching in CS206 and the book?
It certainly is designed to be of general use in programs, with the documentation
I sent about all there is. This matcher is somewhat similar in its
simpler parts to that used by Winston in his course (book?), though
this is not considered a feature. Basically its best features are:
1. * variables, which match sublists of 0 or more elements, when
associated with a variable (*x for instance rather than just *)
must match the same thing wherever it occurs in the treem as well
as satisfying any predicates that are associated with it wherever they
appear. This, of course, means that normal recursion cannot be used
and so a continuation style is used. Also, it is written explicitly
in a tail recursive manner.
2. The match can be restarted after any sucessful match, when so set up,
so that the next possible assignment to the * variables will be done.
explicit (and non-obvious) state saving is used.
The main routine is about 400 lines of code (!!!) and is not exactly written
for ease of understanding. I suggest you look at it and the documentation.
Some of the techniques might be of interest to your class and book, but in
a simpler form.
-rpg-
∂09-Jan-79 1607 HEDRICK at RUTGERS comments on Dialnet and Sumex protocols
Date: 9 Jan 1979 1907-EST
From: HEDRICK at RUTGERS
To: jmc at SAIL, les at SAIL, mrc at SAIL, rindFLEISCH at SUMEX-AIM,
RYLAND, yeager at SUMEX-AIM
Cc: LEVY, ROBERTSON
Subject: comments on Dialnet and Sumex protocols
Please see the documents <hedrick>nets.doc at Rutgers or Sumex.
It is the result of some conversations between Chris Ryland of
Columbia and me. Be warned that Chris has not seen the document,
so some of the opinions therein may not be his, but it is mostly
a concensus. the major points of concensus are
- that we do not want to see two separate protocols for the same
task (Dialnet and Sumex)
- that we want an implementation that runs in user mode, and
particularly that does not involve front end changes.
But please see the document for our full comments.
The copy at Sumex is because we are up only for 8 hours a day today
and tomorrow.
Regards,
C. Hedrick
-------
∂09-Jan-79 2301 MRC Hedrick's thing
To: JMC, LES
It's on NETS.DOC[DLN,MRC].
∂09-Jan-79 2330 REM via AMES-TIP
To: RWW, JMC
By the way, my random (expertly-so) opinion currently is that the greatest
use for interval-refinement data compression is not so much natural
language text files (English) but special database formats and reports
(in the sense of COBOL/RPG, the pretty output they make, in machine-readable
form). English can get only a little better than 3:1 even with large files
(NOTICE[UP,DOC]) whereas many special report formats and database formats
can get much better I think.
∂10-Jan-79 0019 LES
∂09-Jan-79 2308 JMC
Can you look at the Hedrick criticism of Dialnet?
Yes, but it will take some time.
∂10-Jan-79 1259 MRC via SU-TIP reply to protocol critique
To: JMC at SU-AI, LES at SU-AI, Hedrick at RUTGERS,
Ryland at RUTGERS, Rindfleisch at SUMEX-AIM,
Yeager at SUMEX-AIM
This note is an attempt to reply to some of the issues raised in
Hedrick's and Ryland's paper. Hopefully it will make some sense
without too much flaming.
Dialnet is an attempt to create the functionality of the Arpanet
on ordinary TelCo lines. Because of this, we haven't spent much
time or energy on trying to win with anything other than the
standard Dialnet modems. However, nothing in my protocols forces
use of the Vadic modems; you could use the Dialnet protocols with
other equipment and win just the same.
As for monitor vs. user mode implementation, I consider a user
mode implementation to be impractical for a serious communication
effort. On low-performance operating systems such as WAITS (the
SU-AI monitor) and Tops-10, the problem is simply that the monitor
cannot handle the terminal I/O, even at a level such as image mode.
WAITS falls apart completely if a 9600 baud line runs open, and
any 1200 baud line doing so severely impacts the system (to be
fair, I must admit that WAITS uses a line scanner and not a front
end pdp11).
I feel that the idea of a user mode implementation is fine for
something that has to be "sneaked" onto a system, but it has no
place in an environment where functionality and performance are
important. The idea itself is based on the notion that the
monitor is some piece of magic that nobody can understand anyway
and anybody who goes near it will break it so let's leave it to
DEC to break it for us. Hopefully people have advanced in their
thinking beyond that; a monitor is merely a set of system
subroutines.
I will admit that a user mode implementation could be done on
Tops-20 with better success than on WAITS or Tops-10. However,
the performance cost was too much for me to stomach. In addition,
you lose an enormous deal of functionality by not having it in
the system: instead of having a DLN: device you have to do
I/O to a TTY line, or some such horrible kludge (IPCF!!). If
you do that you sacrifice any chance of accounting for usage
or ease in changing user programs if the low level stuff changes.
I wanted an implementation which a user program such as FTP
would not know about. FTP doesn't have the slightest idea what
packet framing or checksums are. It does a SIN and SOUT of a
byte string and occasional MTOPRs.
The low level protocol was designed with simplicity in mind. There
was strong opposition to multiplexing several users from the
management here and I had to fight to get multiplexed channels in.
Nothing prevents somebody USING the protocol to multiplex users
but I don't feel it is worth it. The overhead in multiplexing
probably isn't worth it compared to doing one user at a time.
I think my feelings about having a Dialnet port as a TTY line
which can be zapped by a ↑C are obvious and need not be repeated
here. I do not believe that a Dialnet port should be a TTY line
which a user could dial up as a TTY, log in to, etc., and I think
that anybody who does an implementation in that way deserves to
lose in all the ways s/he is guaranteed to lose. 'nuff said.
Dialnet does not have ACK loops. There is no such thing as an
ACK packet. The reason for sending a NOP every 5 seconds is to
guarantee that the other has the most up to date acknowledgement
information (of course, the top of the pending list should be
send instead of a NOP if there is a pending list). I really
don't understand what your point is on this. Since NOPs are
uncontrolled packets there should never be any trouble with
looping (I have verified this in fact). I hope Sumex isn't really
wasting a whole packet to acknowledge instead of using a byte
in the packet header (shades of RFNMs!).
Approximately .1-.5% of all Dialnet I/O errors are due to a
bad checksum in our test. Most are due to improper framing;
ie, noise or data missed corrupts a byte. A paranoid file
transfer program could have a redundant checksum; I don't think
extra hair in this is necessary for things like MAIL.
If the user ↑C's out of his program, the system will send either
a NOP or the top of the pending queue every 5 seconds for him/her.
The 5 second (actually it is up to the implementor) clock
stuff is the responsibility of the Dialnet NCP, not the user. This
is one reason why a system implementation is the way to win and
while I'm not saying you can't do a user mode program with Dialnet
I am saying you can't do it and win optimally.
About having more than 16 channels; come on! If you need that much
bandwidth you need more than one port. Talking about having more
than one "user" on a 1200 baud line is like the claims I heard in
my undergraduate years at Stevens that model 33 TTY's were more
than enough for any need (300 baud lines were FAST lines).
My feelings about records are that they should be implemented at
a high level in FTP, not in the low level, to avoid burdening
non-IBM machines with this sort of uselessness.
I would hope that a more sophisticated protocol would be used for
any high speed data link (50kbd is passee in these days of the
megabaud link). My protocol was designed to be reasonable at
1200 baud, perhaps up to 9600 baud, and nice for pdp10s and pdp11s,
especially small pdp11s.
One of the reasons I am doing a monitor implementation on Tops-20
is that that is the only way I can turn off XON/XOFF (please don't
ask about TTY PAGE; look at the code first). A problem with this
is that to do it right requires hacking the front end, which is a
royal pain in the ass itself. I am contemplating writing a "kludge
DLNTSK" which would really run on the -10 as a hook in TTFEDV but
would make DIALNE think it is talking to an -11 task, but the
performance problems are mind-boggling. We really do want to run
this thing at LOTS. Performance is CRITICAL. We can NOT sacrifice
performance to make somebody else happy. In my opinion, what other
people do is their own business and I certainly don't feel that I
am in the business of writing Dialnet implementations for the world.
I intend to make the Stanford Dialnet implementations for Tops-20,
WAITS (and a modified WAITS version for Tops-10) available to outside
people, and they must evaluate their suitability on their own systems.
Anybody is welcome to run a user mode Tops-20 implementation of
Dialnet, however, I don't want to get complaints that it eats up
15% of their machine if my version only eats up 2% (both numbers are
completely random). My feeling is that the Dialnet protocols are
such that it is COMPLETELY up to the implementor how s/he does it.
NOTHING prevents him or her from winning or losing as much as s/he
wants to. However, I have made recommendations based on my 18 months
of looking at the problem and I feel that a potential implementor
would be well advised to pay attention to my recommendations.
Well, that's it. Sorry for flaming off towards the end, but I get
a bit hot when I get the impression that somebody seems to think
that Dialnet can only be implemented by hairy monitor code (or that
I should write MY implementation in any way other than the way I
want to). I guess I've gotten too many people wanting me to write a
version for a BASIC-only microprocessor...(sigh).
Cheers,
-- Mark
∂10-Jan-79 1311 SAL Program for Computer Forum Annual Meeting
To: "@FAC.DIS[1,DPB]" at SU-AI
CC: SAL at SU-AI
Colleagues:
As I indicated in my memo to you of 7 November l978, the
Annual Meeting of the Computer Forum will be held on Thursday
and Friday, 1 and 2 February l979.
The Technical Sessions have been organized as indicated
at the end of this message. The Forum Committee wishes to
request that first-year faculty speak for themselves and
that other faculty select one of their students who has
not made a presentation in the past to speak for them.
Please get in touch with your Session Chairman to give him
the details concerning the topic and speaker. Do this
IMMEDIATELY , please, as our Program is VERY LATE in
getting disseminated to our affiliate members.
Twenty minutes (generally) have been allotted for
each talk. This includes a brief period for discussion
at the end. Only one speaker per faculty slot, please.
Session Chairmen: please send me the particulars
of your session by Friday, l2 January. Our program must
be typed and distributed to our Forum attendees right away.
This can be accomplished by sending a message to SAL
(Sally Burns) at CSL or by forwarding "hard copy" to
me at ERL 232.
Please ask your speakers to submit clear xerox copies
of the viewgraphs for their talks to me by 26 January 79 so
that they may be compiled into a booklet for distribution
at the Meeting. These must be limited to 10 per speaker.
The title of the talk and the name of the speaker should
appear on the first viewgraph.
"Dry runs" of the presentations are strongly encouraged
to enhance the clarity of the talks and to eliminate redundancies.
If Session Chairmen wish to use the ERL 237 conference room
for their practice sessions, would you please call Sally
Burns at 7-l458.
I enlist your wholehearted support to make this Annual
Meeting the very best that we have enjoyed to date.
Sincerely yours,
Mike Flynn, Acting Director
Stanford Computer Forum
Tentative Program:
Session I (Thursday, l Feb: 9:30 - l0:30)
D. Knuth, Chairman
R. Tarjan
F. Yao
A. Yao
Session II (Thursday, l Feb: ll:00 - l2:00)
G. Golub, Chairman
J. Herriot
J. Oliger
J. Wilkinson
Session III )Thursday, l Feb: l:30 - 3:10)
***Highlight Session***
D. Luckham, Chairman
S. Owicki
Program verification researchers
Session IV )Thursday, l Feb: 3:40 - 5:00)
F. Baskett, Chairman
F. Tobagi
J. Hennessy
G. Wiederhold
Session V (Friday, 2 Feb: 9:00 - l0:20)
T. Binford, Chairman
J. MCCarthy
T. Winograd
C. Green
Session VI (Friday, 2 Feb: l0:45 - l2:l5)
B. Kumar, Chairman
A. Peterson
E. McCluskey
M. Flynn
W. vanCleemput
Session VII (Friday, 2 Feb: l:l0- 2:45)
*** Highlight Session ***
E. Feigenbaum, Chairman
B. Buchanan
D. Lenat
Applied AI researchers
∂10-Jan-79 1346 DPB forum talk
To: SAL
CC: JB, JMC, TOB, DCL
Juan Bulnes recently talked to John McCarthy about doing
the forum talk in JMC's slot. Since Juan's research deals
more with verification than with AI, he would probably fit
better in the Luckham session than in the Binford session.
Comments/objections etc. to SAL. -Denny
∂10-Jan-79 1449 BCM via SU-TIP call-by-name
I think that derived functions involving call-by-name are done by
adding new parameters to the cost functions. Each new parameter
computes the cost of one of the original parameters. Whenever in the
original function a variable is accessed, in the cost function its cost
is used. The costs are passed since they can not in general be determined
from static analysis
The costs can
be passed by name. They can also be passed by value if the corresponding
expressions don't diverge so as to make the costs undefined.
For example:
f[x] ← if x>100 then x-10 else f[f[x+11]]
cf[x,cx] ← 1 + cx + (if x>100 then cx
else cf[f[x+11],cf[x+11,cx]]
Here, cf computes the number of function calls used. For example,
cf[101,0]=1.
This method may have bugs but see what you think.
As sometimes happens, giving you the problem made me think about it,
and I solved it also last night. Moreover, Corky solved it by a
different method from mine which seems to be the same as yours.
I will illustrate by the Takeuchi function which is
tak(x,y,z) ← if x ≤ y then y else tak(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y))
The function terminates for all integer x, y, z. (It probably also
terminates for real x, y, and z, but I haven't been able to prove it).
However, the number of recursions differs depending on whether the
computation is by name or by value.
My solution works as follows:
In the call-by-value case we have
ctak(x,y,z) ← if x ≤ y then 0 else 2 + ctak(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y))
+ctak(x-1,y,z) + ctak(y-1,z,x) + ctak(z-1,x,y).
This is along the lines discussed a year ago. In the call-by-name case, we have
ctak(x,y,z) ← if x ≤ y then 0 else 1 + ctak(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y))
+ (if use1(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) then 1 + ctak(x-1,y,z) else 0)
+ (if use2(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) then 1 + ctak(y-1,z,x) else 0)
+ (if use3(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) then 1 + ctak(z-1,x,y) else 0)
where
use1(x,y,z) ← T
use2(x,y,z) ← T
use3(x,y,z) ← if x ≤ y then F else
(use1(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) ∧ use3(x-1,y,z))
∨ (use2(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) ∧ use2(y-1,z,x))
∨ (use3(tak(x-1,y,z),tak(y-1,z,x),tak(z-1,x,y)) ∧ use1(z-1,x,y))
The "usen" predicate tells whether the nth argument of the function is actually
evaluated. In the above case, use3 simplifies to
use3(x,y,z) ← if x ≤ y then F else T,
because the second disjunct is the conjuction of two use2's which are identically
true. The evaluation of the call-by-name ctak must be call-by-name in cases
when call-by-value fails for the original function. For example, we have
cadiou(x,y) ← if x = 0 then 0 else cadiou(x-1,cadiou(x+1,y))
and
ccadiou(x,y) ← if x = 0 then 0 else 1 + ccadiou(x-1,cadiou(x+1,y))
+ (if use2(x-1,cadiou(x+1,y)) then 1 + ccadiou(x+1,y) else 0)
where
use2(x,y) ← if x = 0 then F else use2(x-1,cadiou(x+1,y)) ∧ use2(x+1,y).
use2(x,y) always evaluates to F, but call-by-name must be used to insure
termination.
Corky's solution to the same problem, which I think is the
same as yours, would write
ccadiou(x,y) = ccadiou1(x,0,y,0)
where
ccadiou1(x,y) ← if x = 0 then c else ccadiou(x-1,c,cadiou(x+1,y),ccadiou(x+1,c,y,d))
Well I'm not sure this is quite correct, although it gives the right answer,
and I'm also not sure what you intend with two or more variables.
Could you work out yours for the Takeuchi and Cadiou functions?
∂10-Jan-79 1509 HEDRICK at RUTGERS Re: reply to protocol critique
Date: 10 Jan 1979 1807-EST
From: HEDRICK at RUTGERS
To: MRC at SU-AI, JMC at SU-AI, LES at SU-AI, Hedrick at RUTGERS, Ryland at RUTGERS,
Rindfleisch at SUMEX-AIM, Yeager at SUMEX-AIM
Subject: Re: reply to protocol critique
In-reply-to: Your message of 10-Jan-79 1559-EST
MRC correctly points out that one of my comments was based on an
insufficiently careful reading of the Dailnet protocol: Their
NOP (which is basically equivalent to the Sumex ACK) is "uncontrolled".
This means that no acknowledgement is required. Thus my proposal
is essentially that the Sumex low level protocol be brought into line
with Dialnet in that regard. The actual Dialnet implementation is
that a message which is not to be acknowleged uses a packet number
of zero, but NOP will always have a packet number of zero. This
is probably a slightly nicer way of doing it than my proposal
that ACK never needs to be acknowleged.
Note however that the major thrust of my comment is not its
specific technical content. Mostly I am concerned that we not
have two different protocols for essentially the same task
(both from Stanford even!). I was just trying to start discussion
aimed at resolving the differences. I guess if we do things as
user mode programs, we can tolerate two protocols (if someone
else does the implementation - I am not about to do two myself).
But I would like to preserve our freedom to put it in the monitor
(though I am not as enthusiastic about doing this as MRC is), and
I pale at the thought of putting two such schemes in the monitor
(along with X.25, DECnet, ...).
It does not seem useful to continue the discussion about
monitor mode vs. user mode, since I don't think anybody is going
to convince anybody else. I attempted to advocate a neutral
stance on this: that the protocols should specifically be designed
with both in mind. As I see it, one of the three main differences
between the Sumex and Dialnet level 0 protocols is that Sumex has
a few extra op codes which are useful in the user mode case and
harmless otherwise (except for adding code). [The other two
differences are 1) acknowlegement strategy, where it now becomes
clear that I basically agree with Dialnet, and 2) addition of
the AER op code to make the system "complete" in a formal sense.]
By the way, I understand that the ARPAnet people are now working
on a new protocol on their own, which is designed to replace
NCP. It is specifically designed to be independent of the low
level protocols, so that it can be used over any network.
The idea is that it can be passed through gateways between
networks with a minimum of processing. The effort is apparently
fairly well along. Does anybody know anything about that? It
appears that this protocol is going to replace NCP within the
next year or so. If so, that more or less guarantees that
FTP, mail, etc., will have to be implemented for it on all our
machines. So possibly we should consider supplying a simple
low level protocol to use over TTY lines and using the new
arpanet utilities.
-------
∂10-Jan-79 2003 MRC protocols
To: Hedrick at RUTGERS, JMC at SU-AI, LES at SU-AI,
Ryland at RUTGERS, Rindfleisch at SUMEX-AIM,
Yeager at SUMEX-AIM
On the Arpanet protocol I believe you are thinking of TCP, which is
allegedly the "right" way to do things. It is also incredibly hairy.
I doubt very much that TCP will replace NCP, just as long leaders will
never replace short leaders (and I HAVE implemented long leaders at
SAIL...the first pdp10 implementation in fact). It is unclear whether
or not TCP is going to take over the world, but judging from BBN's
past performance on things like FTP I seriously doubt that if I write
a TCP that it will be for any other reason other than for the challenge
and fun of doing it (like with long leaders).
Moon at MIT could probably compare TCP with simpler protocols like Dialnet
or Chaos net better than I could. MIT has two local (megabaud) networks,
the Chaos net (hacker designed) and the LCS network (officially sponsored
by ARPA etc). The Chaos net is up and running and has been for a while
now, while the LCS net is two years (at least) behind schedule and of
course uses TCP. This is, of course, rather slanted; but not all that
far from the truth.
I heard of the Sumex thing first around last summer I think, when I'd
been working on Dialnet for about a year. I'm not sure what went on,
but I think they gobbled down the Dialnet protocols and munged them
for their own concepts of the world. I really don't know as there has
been little communication and what communication there has been has
been me sending pointers of my stuff to them. I know it sounds losing
but I've been sort of committed to my design for a while now and am
in the last stages of getting code running on Tops-20 (having had it
running on SAIL for several months) & all that.
I'm not sure; but I'm afraid there will end up being a third or
fourth (since the IBM world here is doing something with DECnet)
"Stanford network" before this is all through. We do need a megabaud
network here since the Arpanet is clearly way too slow, but unlike
MIT we don't have many real network hackers (I probably know more
about the guts of the Arpanet than anybody else currently working at
Stanford) and all that. Eventually we'll get our act together, in
some way that saves face for all concerned...
-- m
∂10-Jan-79 2138 SMG paper
Is there some version of the paper you are writing with Corky that I could
read now?
steve
Yes. It is the Representation of Recursive Programs in First Order Logic
that I gave at the Kyoto conference. There are copies in my office so ask
me in daytime. There will be quite a few changes in the new version including
a much better set of axioms not confused by a half-assed use of sorts.
∂11-Jan-79 0531 RWG at MIT-MC (Bill Gosper)
Date: 11 JAN 1979 0832-EST
From: RWG at MIT-MC (Bill Gosper)
To: JMC at MIT-MC
product(stirl(n)/n!,n,1,inf) = .9973195-,
much nearer 1 than i expected. stirl(1) = .99898+. initial (only!)
convergence is fast: product of 1st nine = .997667-. (recall that
stirl(n) = sqrt(2*pi*n)(n/e)↑n (1+1/12n) .)
∂11-Jan-79 1320 S1
To: "@S1[DIS,S1]" at SU-AI
S1 IS STILL OVER QUOTA BY 3,146,052!!!!!!!!
∂11-Jan-79 1401 ALS
To: "@S1[DIS,S1]" at SU-AI
S1 disk quota:
We really should do something about the S1 files. If every one would look
at the files he or she is responsible for and delete the obsolete ones,
then maybe we could get back under quota. It would be a great help, since
a purge will certainly cause us a lot of trouble.
∂11-Jan-79 1410 Jrobinson at SRI-KL (Jane Robinson)
Date: 11 Jan 1979 1409-PST
From: Jrobinson at SRI-KL (Jane Robinson)
To: jmc at SU-AI
So we heard. Am going early, with Anne Robinson and Bob Moore. Anne and
I will be going on to dinner somewhere afterwards. Maybe you'd like to
join us. No reply necessary.
Jane
-------
∂12-Jan-79 0953 Bmoore at SRI-KL (Bob Moore) meeting to discuss my thesis
Date: 12 Jan 1979 0951-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: meeting to discuss my thesis
To: jmc at SAIL
Yesterday you asked me what my schedule was as regards my thesis, and
we got distracted before I gave you an answer. I am tentatively planning
to go to Boston in 2 or 3 weeks, but I haven't set an exact date yet.
Perhaps we could get together sometime the week of the 22nd. That would
give you some more time to read and still leave me a few days to correct
any serious mistakes or problems that you catch before going on to Boston.
How does that sound?
Bob
-------
In any case by then. If I can get done earlier, then we should meet
earlier, so you could make larger changes if there were some.
∂12-Jan-79 1056 AFB
Dear Professor McCarthy,
Why is the grade for the 390 I took from you
'N' I finished up the work I wanted to do for now
and have other plans for this quarter. As such it is not
a "continuing course" as the N would indicate. I would
appreciate it if you could clear up this small administrative problem.
Thanks,
Anne Beetem
In that case, please come and see me and tell me what you did for 390.
∂12-Jan-79 1607 PAT reminder
this is a reminder...have you sent the memo to Feigenbaum regarding his memo
to the deans regarding space in the new building????
Dear Ed:
To commit to writing what I said about the musicians, I agree
that we should try to get their office space moved to Cedar or Serra.
However, it isn't true that we don't owe them anything. While they
have mortgaged back the KA-10, they currently own one of our two
disk controllers at $34K, one disk drive at $14K and the Foonly disk
channel at $20K. While one can imagine buying them out, justice and
our commitments require a solution in which they can continue to
make music. Their van mounted 2020, which I haven't heard of recently,
is not a solution, because it will be dedicated to a particular
application, the 100th anniversary of the Boston Symphony.
I think the best solution would be office space for them
in some compact space on campus with wired connection to the computer
together with space for the Sampson box and such like equipment in
the computer room.
John
∂12-Jan-79 1652 GHG Requirements for the degree
To: JED at SU-AI, DWW at SU-AI, "@FAC.DIS[1,DPB]" at SU-AI
It seems to me that it might be a good idea to require our Ph.D.
students to take courses outside of the department. I am concerned that
our students may be receiving a too narrow training and that it would be
beneficial for them to take some courses outside of the department.
Therefore, I'd like to recommend that we have a requirement that all
Ph.D. candidates, beginning with the class of students that is being
admitted in 1979, be required to take at least three technical courses
outside the department.
Of course, we could require the students to take a minor in
another department. This would mean that they would have to focus
their studies, but I don't believe we should make this a requirement.
Many years ago after the department was first formed, George
Forsythe made an observation which was something like the following:
He said, "After we admit students we should have them take their
entire program outside of the Department. After all, they'll probably
learn most of what they need from the other students in the Department."
Our field has grown enormously since then, but I feel it would be
beneficial to our students to broaden their horizons.
Please send me your comments on how you feel about this, and
perhaps we could discuss it in the near future at a faculty meeting.
Gene Golub
∂12-Jan-79 1700 BCM derived functions
I looked at mosz.lsp[w79,jmc] it looks basically ok but since the system
is going down, I don't have time to say more.
∂12-Jan-79 2118 REF
∂12-Jan-79 1302 JMC
I received notice today that the chairman is Fouad Tobagi.
Thanks.
∂14-Jan-79 1345 DWW Gene Golub's proposal
To: "@FAC.DIS[1,DPB]" at SU-AI, JED at SU-AI
CC: DWW at SU-AI
Most students here (myself included) would disagree, not on
the grounds that the breadth is bad or useless, but rather on the
grounds that course requirements of any form are pointless at this
stage in our education. I would welcome departmental *encouragement*
of courses outside the department; there is occasionally so thick an
assumption that the students are here to churn out research for the
glory of the department that I suspect many people are put off taking
some of the fine courses available. But course requirements have been
with all our lives and we're all smart enough to know how to beat
them. Requiring a course does not force a student to learn anything
from it, and if I'm not happy about having to take it I simply won't
get anything out of it.
Doing a dissertation is an entirely new activity for most of
us and requires an entirely new form of mental self-discipline. As it
is, the busy-work hurdles of a comp and qual make it hard enough to
get into the real business; I see nothing to be gained by imposing
still more undergraduate-flavored requirements.
-..david
∂14-Jan-79 1511 DCO
To: "@GROUP.DIS[VCG,DCL]" at SU-AI
*** Notice ***
VERIFICATION GROUP SEMINAR
TUESDAY 16th JANUARY
PLACE: ERL 237
TIME: 2:30 p.m.
TITLE: Program Verification based on a Denotational Language Definition
SPEAKER: Wolfgang Polak
ABSTRACT:
We argue that augmenting a program with assertions, generating
verification conditions, and proving these is a practical and useful
paradigm for program verification. However, the use of Floyd-Hoare proof
rules as a basis for a verification system is unsatisfactory. This is not so
much a principal problem with Floyd-Hoare logic; rather it is the often
informal and sloppy use of this formalism.
In this talk a method is presented which allows the generation of
verification conditions for an augmented program based on a denotational
definition of the language. The notion of a reduced language definition is
introduced and it is shown that continuations of a reduced definition are
weakest liberal preconditions. The use of reduced definitions to generate
verification conditions follows immediately.
∂15-Jan-79 0000 JMC*
copy picture, watch tape, make appointment about safe deposit
∂15-Jan-79 0800 JMC*
Filman and Feferman
∂15-Jan-79 0924 Bmoore at SRI-KL (Bob Moore) An index of symbols...
Date: 15 Jan 1979 0926-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: An index of symbols...
To: jmc at SAIL
sounds like a good idea. It may be a day or two before I have a chance to
prepare it. In the meantime, probably the best thing for you to do is to
scan the list of axioms in Appendix A for the first axiom in which the symbol
appears an refer to the associated page number for the explanation of the
symbol.
-------
∂15-Jan-79 1100 JMC*
FILMAN LETTERS
∂15-Jan-79 1300 JMC*
Call Anne Gardner
∂15-Jan-79 1438 TW
To: JED at SU-AI, DWW at SU-AI, "@FAC.DIS[1,DPB]" at SU-AI
I think that our students do in general receive a narrow education
and little encouragement to broaden out. I don't like to do things
by requirements, but the issue is definitely worth raising. --terry
∂15-Jan-79 1522 SAL Computer Forum Brochure
To: "@FAC.DIS[1,DPB]" at SU-AI
The updated Computer Forum brochure has now arrived in my
office. It looks spiffy. Please let me know how many
copies you would like to send to prospective Forum
members or for other purposes.
I am addressed as SAL at SAIL.
Sally Burns
∂16-Jan-79 1140 100 : patte MA
the dsp-66 has come through. Are you going to write a cover letter?
∂16-Jan-79 1404 CLT
Where did "mtcpub.pub[let,jmc]" go? Pub cannot find it!
∂16-Jan-79 1936 Elmasri at SUMEX-AIM Serra House Key
To: S1 at SU-AI
Date: 16 Jan 1979 1826-PST
From: Elmasri at SUMEX-AIM
Subject: Serra House Key
To: s1 at SAIL
The Serra House key that resides in the Serra trailer has been missing, and
none of the everyday users of the trailer have it. This key belongs to Gio, and
was kept in the trailer for our convenience. If anyone has it, please return
it to the trailer.
Ramez El-Masri
-------
∂17-Jan-79 1546 RPG UTIL.>
To: "@USERS.DIS[AID,RPG]" at SU-AI
To speed up compilation (and interpretation), you might try changing your
(require util 1 dsk (aid rpg)) to
(declare (fasload '((dsk (aid rpg)) util fas)))
-rpg-
∂17-Jan-79 1806 Gardner at SUMEX-AIM Legal reasoning paper
Date: 17 Jan 1979 1800-PST
From: Gardner at SUMEX-AIM
Subject: Legal reasoning paper
To: jmc at SAIL, jls at SAIL, suzman at SAIL
The paper's at the lab now, for whoever wants it: LAW.1[1,AVG]
Anne
-------
∂17-Jan-79 2149 MRC
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: 2000
Your files occupy 2377
∂17-Jan-79 2201 MRC
To: S1 at SU-AI
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: 7000
Your files occupy 8676
∂17-Jan-79 2202 MRC modem STILL loses!
To: JMC, LES
I GIVE UP. I went over to VADIC today and brought yet ANOTHER cable
over here. It is the right kind of cable. STILL no luck.
How about getting a hardware hacker (or at least a VADIC person) to
come over HERE to look at it. I simply CANNOT run back and forth
to/from VADIC forever.
Would you take charge of getting Vadic to fix their modem or something?
∂18-Jan-79 0742 RYLAND at RUTGERS Re: comments on Dialnet and Sumex protocols
Date: 18 Jan 1979 1038-EST
From: RYLAND at RUTGERS
To: HEDRICK, jmc at SAIL, les at SAIL, mrc at SAIL, rindFLEISCH at SUMEX-AIM,
yeager at SUMEX-AIM
Cc: LEVY, ROBERTSON
Subject: Re: comments on Dialnet and Sumex protocols
In-reply-to: Your message of 9-Jan-79 1907-EST
I don't have much time right now to leap into the fray, but I did
want to say that I back Chuck's position 100%. We should also
make it clear that we (at least I, and I think Chuck, too) support
the Dialnet 'world view'. What we're asking is that Dialnet simply
be extended a little to handle multiple logical connections over
a single physical connection, as well as user subchannels. This
is not a criticism of Dialnet, as it wasn't conceived in an
envrionment where this would be important, but it doesn't seem
to cost much (let's not be micro-efficient) and lets people use
the Dialnet protocol for high-bandwidth operations. I guess if
Sumex and Dialnet can't somehow come to see eye-to-eye, then I'll
have to go with Dialnet. Sigh. When I get out of this tight
school loop (probably within a week), I'll provide more thoughts
on the subject. I guess the next step is up to all you Sumex
folks. Any responses? Cheers -- Chris
(PS, Mark: I'll be glad to help modify your Twenex code if the
protocol changes slightly; I still intend to put the Dialnet support
into the front end, but this isn't one of the issues (luckily)).
-------
∂18-Jan-79 1242 PAT MA part two (or is it three)
I have mailed off the letters and DSP-66 to both Ma and John Thompson. They
were mailed registered, receipt requested and I took both envelopes to the
U.S. China Relations Office on campus and had the addresses put on in Chinese
as well as English.
∂18-Jan-79 1322 100 : patte proofreading
I have finished proofreading MENTAL for Ringle
∂18-Jan-79 1401 DBL filman
Just talked with him. He's getting out THIS quarter, not Spring or
Summer as I'd assumed, hence there is not enough time for him to
do much more experimenting or even writing. We agreed that I should
read through his first article or two (he's planning one for IJCAI
and one for the AI Journal), and hopefully I can direct them to
the themes I missed in his thesis documnt itself. The journal articles
have such higher readership, anyway, that it's just as good to correct
the deficiencies there as in his thesis.
I tlkaed with AI with him for awhile, and feel he has a pretty good grasp
of the field. This was another thing I was a little neasy about at his oral,
and now feel much better about. All in all, I do not think he will
underrepresent the department in any way.
Doug
∂18-Jan-79 1458 REF
John:
Well, it's that time again (job hunting). Forgoing the use of the
placement office this year puts a slight burden on you. I hope it's not
too much. Would you please send recommendation letters to the following people?
Let me know if it's too much trouble, or if you have any questions or
advice.
Thanks.
bob
Professor Jonathan L. Gross
Department of Mathematical Statistics
618 Mathematics
Columbia University
New York, New York 10027
Prof. Paul W. Purdom, Jr.
Chairman, Computer Science Department
Lindley Hall 101
Indiana University
Bloomington, Indiana 47401
Prof. David Young, Jr.
Chairman, Recruiting Committee
Department of Computer Science
University of Texas
Austin, Texas 78712
Professor Michael A. Arbib
Chairman of the Personnel Committee
Department of Computer and Information Science
Graduate Research Center
University of Massachusetts
Amherst Massachusetts 01003
Prof. Alan Perlis
Chairman, Department of Computer Science
Yale University
New Haven, Conn 06520
Professor Lawrence H. Landweber
Chairman, Computer Sciences Department
University of Wisconsin - Madison
1210 West Dayton Street
Madison, Wisconsin 53706
Professor Jack Minker
Chairman, Department of Computer Science
University of Maryland
College Park, Maryland 20742
Thanks to the magic of electronic computers, ∞ letters is no more trouble
than one. Ask me Monday if I've done it yet.
∂18-Jan-79 1609 LJH Trip to Los Angeles
Mr. Fenaughti and Mr. Fredkin feel it is necessary that you go to L.A. tomorrow.
∂18-Jan-79 1615 CET
To: "@FAC.DIS[1,DPB]" at SU-AI
January 10, 1979
letter received from Graduate Division
Howard Jones
"We need a member of your faculty participate in a conference
held here at the University on March 3.. The conference will invite
women and minority undergraduates from California colleges and
universities, professional scientists from area organizations and
Stanford faculty and graduate students. The member of your faculty
will participate in the CS Workshop and should be able to give information
on the graduate programs in CS, financial aid and generally amplify on
the presentation given by the workshop speaker.
I would appreciate your finding a member of your faculty who will participate
in the conference Computer Science Workshop."
The hours are 10-12 and 1-3. Two volunteers are needed (one for morning and
one for afternoon).
Please call me (72273) or send message. CET@SAIL.
Carolyn Tajnai
∂18-Jan-79 1653 Wiederhold at SUMEX-AIM Last S1 meeting notes
To: S1 at SU-AI
Date: 18 Jan 1979 1648-PST
From: Wiederhold at SUMEX-AIM
Subject: Last S1 meeting notes
To: s1 at SAIL
cc: pickering
S-1 report , Meeting of 10JAN79
PRESENT: ALS, BLH, DRF, FC, GIO, HAD, PLZ, PN, TRG, RANDY BUSCH,
DICK KARPINSKI, Joel McCORMICK
ABSENT: ALL, ARK, HAL, MLB, PEG
Comments and corrections are solicited.
TEST PROGRAMS (PN, FC, DRF)
The FORTRAN benchmarks for Livermore have been run and exhibit
the following behaviour:
SIMPLE: compiled by PCFORT, code generates bad SKIP in SOPA probably
due to peephole optimizer. Found to be false 1/16/79. Source
code was changed to avoid ENTRY statements.
ADI: compiles by PCFORT, code translated by SOPA. Appears to run
correctly, but does not terminate in reasonable time on the
S-1 simulator at SAIL. Had unitialized variables. SECOND
and EXIT were added.
HYDRO: while it used to compile and translate through SOPA it now
shows the same problem as SIMPLE. It previously showed minor
result difference due to initialization failure in Hydro code.
MC: contains bit-wise AND, statement. Was changed, but then failed
in SOPA .
PLANS a) add bit-wise AND, OR, NOT to SOPA, so that FORTRAN can
request these ops on integer and real variables of the
Q, H, S, and D size, in order to comply with FORTRAN
conventions.
b) test long running programs on S-1 itself.
c) PTZ will investigate SOPA bug further and may put in a
'Disable peephole' switch (done 1/11/79). PN will try
to help.
d) Lowell is mailing test results for comparison.
BITADDRESS (HAD)
Hal will soon take a version of SOPA for bit address
implementation.
PCODE, UCODE, etc. (ALS, BUSCH)
Documentation on Ucode to be used by Sites is due to be here
immenently. A global optimizer for testing will follow soon.
To allow use of the optimizer three programs have to be
interfaced: SOPA, PCPASC, PCFORT.
Either the three programs can be changed, or relatively small
conversion routines can be written to accomplish the same,
at higher execution cost.
A number of notes and questions came up. ALS will investigate
the alternatives.
Whereas in SOPA the P-code interface is well localized that is
not true in PCPASC and PCFORT. There is also the question that
since UCODE requires resolution of some forward references
that in addition to the local conversion of a back-patch
pass will be needed. Can that be the same for PCPASC and
PCFORT? How soon will the Sites optimizer be part of the
routine compiling process? Is comparison of FORTRAN programs
on S-1 versus other machines more critical than PASCAL
programs? What code will be generated by Forest and
Hennessy's PASCAL? Will Sites' optimizer include also pre-
or postprocessing passes?
LINKER (ARK)
Is in operation, final documentation due anytime. ARK will
look at advanced File system requirements, and prepare a
proposal. He may also help with PASCAL. Should any more
effort be put into PCPASC given FB/JLH/ARR efforts
SOPA (PEG)
Set change checkout is progressing, changes are both in PCPASC
and SOPA.
MANUAL (BTH, BLH, MLB, LCW, GLS)
Clean copies were produced, but LCW and GLS are embarking on
another pass, to be completed January 20, 1979. No copies
yet distributed beyond narrow circle.
PASMAC (ALL)
Macro processor partially written and checked. Calling of
MACRO's still to be done. Goal is to be done the end of
!
-------
∂18-Jan-79 1729 MINSKY at MIT-AI (Marvin Minsky)
Date: 18 JAN 1979 2028-EST
From: MINSKY at MIT-AI (Marvin Minsky)
To: jmc at SU-AI
Could you bring knuth's paper on photocomposition with you.
As much info as you can get.
Bring yourself, too.
∂18-Jan-79 2338 ME
I have fixed the AP line. It was a bad relay chip. I'm gonna replace
the NYT line's relay too to see if that helps avoid the occasional garbaging.
∂18-Jan-79 2346 ME NYT
Actually the NYT line today looks pretty good, so I'm not going to replace
its relay after all. Let me know any day you think the NYT line is garbaging
a lot.
∂19-Jan-79 1117 DEW computer forum
You asked me to speak at the computer forum and I've put some thought
into it. Today I got messages from SOB and TOB saying they were confused
by conflicting messages from you about whether Juan Bulnes or myself should
speak and that they would assume JB would and I would not unless they
heard differently. I want to get a final say from you about whether or
not I am to speak. Thanks, Dave
I guess I must have gotten confused, but most recently I have been urging
Juan to do it, so unless you have a strong desire to do so, in which case
phone me or see me at the Lab, we'll leave it at Juan.
∂19-Jan-79 1125 DEW
Correction: SOB → SAL (Sally Burns) must have been Freudian slip
∂19-Jan-79 1344 CPP cs 206 grades fall quarter 1978
To: JMC, REP
The fall quarter 206 grades have some mistakes in them. First I talked
to REP about my final and he agreed that I should have gotten about 9
more points on the grade for the final. Second I think that both REP
and JMC believed that CHT and WOL did a final project together and that
I did not do one. Actually CHT and I did the project and WOL did not
do one. What occured was that WOL got credit for my part of the project
and I did not get anything. Therefore my grade should be adjusted to
reflect what actually happened. REP and I agreed that my grade should
be changed from a B- to a B+
Thank you for your attention.
CPP
You are correct about that error, and there will be an adjustment.
∂19-Jan-79 1353 Feldman at SUMEX-AIM My course
Date: 19 Jan 1979 1353-PST
From: Feldman at SUMEX-AIM
Subject: My course
To: jmc at SU-AI
cc: earnest at SU-AI
John,
I am running (with Sue Owicki) a seminar on distributed computing. Only
about six students are doing the work. The problem is that Sail-Plits now
runs only at the lab,and would not be easy to move to LOTS and Tops20-Sail.
May theses few students use the lab machine? Most of them
already work there and Sail-Plits was one thing we told Chern we would
collaborate on.
Jerry F.
-------
The part about our having told Chern we would collaborate on SAIL-Plits
(what is it?) is a decisive consideration, as is the small number of
students. If there is some part of the work not involving it, they
should do the work at LOTS. I don't even teach my Lisp course at SAIL.
So, go ahead but take it as easy if possible.
∂19-Jan-79 1558 RWW
To: "@POTLUK.DIS[S,RWW]" at SU-AI
To: My usual collection of friends and strangers
What: Potluck dinner
When: Tuesday, January 23, at 7:00 PM
Where: 12625 La Cresta Dr., Los Altos Hills
948-2149
As usual please feel compeled bring along lots of friends. I'd like to
meet some new people.
There is an online map in MAP[S,RWW]@SU-AI
RSVP and state what you would like to bring on JAN23[S,RWW]
richard
∂19-Jan-79 1627 SAL FORUM PROGRAM
To: "@FAC.DIS[1,DPB]" at SU-AI
CC: SAL at SU-AI
The program for the llth meeting of the Computer /Forum
has been mailed to each of you today. Please deign
to check your "hard-copy" mail.
With the program is an invitation to attend two
luncheons (Feb 1 & 2) and a banquet on Feb l.
There is a reply form so that I can get a relatively
accurate head count. Please send them back--either
on-line or in hard copy, or both if you want to really
spoil me.
After you've replied, don't forget to come to the
meals you said you would.
I hate to get into a Big Game locker room speech,
but this is Our Forum and it's marvelous if everyone
actively participates. Your industrial colleagues
have joined this program just for the opportunity of
meeting you (and your students) and of exchanging
ideas with you. Don't let them down. Come eat your
head off--at coffee breaks, luncheons, banquet,
wine and cheese party. You'll be a fatter and a
happier person for it.
Sally
∂19-Jan-79 2341 DEW forum
I was sort of looking forward to the Forum, only because I thought it might
open up job possibilities for me. Perhaps that's not true. I'll try to talk
to you on Monday.
∂20-Jan-79 1410 FC How to execute PCFORT
To: S1 at SU-AI
To execute PCFORT, you will have to be in the [for,s1] area (alias).
Have the Fortran source program ready in *.FOR. Say DO TFOR, and
the p-code for the program will be output in *.PCO. The p-code can then
be executed by calling SOPA, just like when executing Pascal.
The other files used in executing PCFORT are:
*.LST The list file.
*.ERR File where error messages only are printed.
∂20-Jan-79 1441 DCO
To: "@GROUP.DIS[VCG,DCL]" at SU-AI
VERIFICATION GROUP SEMINAR
TUESDAY 23rd JANUARY
PLACE: ERL 237
TIME: 2:30 p.m.
TITLE: Formal Proofs as Descriptions of Computation
SPEAKER: Chris Goad
ABSTRACT:
A constructive formal proof of an assertion of the form ∀x∃yA(x,y)
can be used to compute a function f such that ∀xA(x,f(x)) - in this sense
the proof serves as a verified program which satisfies the specification
A. We will discuss methods for the efficient "execution" of proofs, and
the possible advantages of using "proof notation" for the description of
algorithms. Among these advantages is the feasibility of smoothly mixing
deductive processes (theorem proving; program synthesis) and interpretive
processes (execution and optimization of programs) in computations using
proofs.
∂21-Jan-79 1321 Rindfleisch at SUMEX-AIM (Response to message)
Date: 21 Jan 1979 1315-PST
From: Rindfleisch at SUMEX-AIM
Subject: (Response to message)
To: RYLAND at RUTGERS, HEDRICK at RUTGERS, jmc at SAIL,
To: les at SAIL, mrc at SAIL
In response to the message sent 18 Jan 1979 1038-EST from RYLAND at RUTGERS
I've been off doing some other things and have just gotten around to
reading the recent series of msgs that has been exchanged. This is not
intended to be a reaction to all of this as I have not had a chance to
hear Bill Yeager's response yet. I just wanted to acknowledge seeing
the various msgs and we will forward some comments early next week.
Tom R.
-------
∂21-Jan-79 1348 DCL via SU-TIP Modem and phone line
To: LES, HVA
CC: JMC
Hersche, has the phone for my line into the sail computer been ordered.
This is the second time this week i have been unable to dial into
sail. This line was promised me 6 months ago by JMC.
=David
∂21-Jan-79 1353 DCL via SU-TIP remote users
To: LES
CC: JMC
I notice ROZ (MORF) is still using SAIL remotely, thus tying up phone lines
even though he has taken back his modem etc.
Shouldnt he leave his modem?
Also, with all lines taken, DLB is tying one up. Perhaps you could
reword your old policy directive about guest and student users at busy
times and reissue it.
∂21-Jan-79 2216 PEG C1 UDP
To: S1 at SU-AI
To: All users of the C1 User Disk Pack
From: PEG
The C1 UDP is becoming very full (<500 free tracks as of 1/21/79).
Will all users of this pack please go through their directories
on it and delete all useless files (of which there seem to be
many), and also reformat large files (i.e., ET U:FOO.BAR/∞F) to
take up less room. Thank you.
Phil Gerring
∂22-Jan-79 0758 Feigenbaum at SUMEX-AIM need to talk
Date: 22 Jan 1979 0756-PST
From: Feigenbaum at SUMEX-AIM
Subject: need to talk
To: jmc at SAIL
John, I need to talk with you about the musicians. Will you be home
tonight? I could come by about 9:45pm. Otherwise,
will you be on campus today or tomorrow?
Have you been through the new building for a walkthrough? If not, there's
a "private" one you can join up to at 10am this morning (meet in front of
Margaret Jacks Hall)
Ed
-------
∂22-Jan-79 1106 HVA Vera's Tapes
Yesterday evening I completed Side B of Tape #2, leaving 2 tapes to go.
2 (tapes) x 2 (sides)=4 x 6 (aver.hrs.ea.side)= 24 hrs.to go, which could
mean "a month of Sundays", but I hope to go faster than that! Yesterday's
session seemed to go more easily - not so many blank spots - so I am more
encouraged. I have not forgotten the list of questions (peopla and place
names), but I would like to finish yellow drafts first.
∂22-Jan-79 1134 Mcgoveran at SUMEX-AIM Changes in Short-Term Disability Plans (SDI & VDI)
Date: 22 Jan 1979 1134-PST
From: Mcgoveran at SUMEX-AIM
Subject: Changes in Short-Term Disability Plans (SDI & VDI)
To: Faculty distribution:
From: Jeannie K. A. Thomas, Benefits Manager
Date: January 16, 1979
Effective January 1, 1979, the University's short-term disability
benefit plans (State Disability Insurance,SDI, and Stanford University
Voluntary Short-Term Disability Benefit Plan,VDI) have eliminated the
hospital benefit of $12 per day. Action by the State Legislature
removed this benefit from the State plan and the University's VDI
plan has made a parallel change. In all other respects the benefits
under both VDI and SDI remain unchanged.
In dropping the Hospital Benefit from the disability plan, the Legis-
lature noted that health insurance plans normally cover hospitalization
costs, making the disability plan's payment an unnecessary duplication.
Savings to the disability plan from the removal of the hospital benefit
have already been earmarked to fund increased disability payments which
became effective January 1, 1978.
Our faculty and staff are paying for the short-term disability plan, and
they should receive our full support in making use of the plan. When
they become disabled, their departments have the key role in encouraging
them to file an application for short-term disability benefits. The
staff of the Benefits Office provides forms, information, and other
assistance. Please call them with any questions or requests concerning
the use of this disability benefit plan.
-------
∂22-Jan-79 1142 REF
John-
Would you please add the following to your recommendation letter mailing list?
Thanks
Prof. S.D. Conte
Chairman, Department of Computer Science
Mathematical Sciences Building; Room 442
Purdue University
West Lafayette, Indiana 47907
∂22-Jan-79 1350 Bmoore at SRI-KL (Bob Moore)
Date: 22 Jan 1979 1348-PST
From: Bmoore at SRI-KL (Bob Moore)
To: jmc at SAIL
I am planning to go to Boston next Sunday, Jan. 28. If you think there
are any serious problems with my thesis, we should talk about it before
then.
Bob
-------
∂22-Jan-79 1508 REF
And yet another.
Would you please include the following in your mass mailing? Thanks.
Mr. Sidney L. Mantler
Thomas J. Watson Research Center
P.O. Box 218
Yorktown Heights, New York 10598
∂22-Jan-79 1623 ARK via SUMEX-AIM S-1 Meeting Notice
To: S1 at SU-AI
There will be an S-1 meeting in the Serra Conference Room on Wed., 24 Jan
at 2:30 pm.
Arthur
∂23-Jan-79 1104 PAT ringle
what should I do with these proofs? Are you going to call him or what?
Ask me tomorrow about proofs.
∂23-Jan-79 1440 ME
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: 2000
Your files occupy 2394
∂23-Jan-79 1439 CET Change of grade for Chuck Paulson
John, I can put through the change of grade card for Chuck Paulson
if you wish.
He had a B-. It should be ←←←??
Carolyn
Make it B+.
∂23-Jan-79 1552 BOBROW at PARC-MAXC2 A special issue of the AI journal on non-monotonic logics
Date: 23 JAN 1979 1552-PST
From: BOBROW at PARC-MAXC2
Subject: A special issue of the AI journal on non-monotonic logics
To: JMC at SAIL
cc: bobrow, winograd
I have received McDermott's and Weyerauch's papers for inclusion in the
AI journal, and Terry suggested that with his and your paper a
good special issue of the journal might be put together. What do you
think of the idea? Would you send me your paper in any event to be considered for the journal, please. Is there any other paper from the conference (or elsewhere) that you think should also be included?
I like special issues. They get significantly more attention than random ones.
danny
-------
I will send you my paper in its present state, but I would like to revise
it.
∂23-Jan-79 1601 PAT Texas
I will make the hotel and plane reservations if you will tell me when you
want to go (the 28th?) and when you want to return (do you want to stay for
the whole conference?).
Yes the 28th, and I want to stay for whole conference.
∂23-Jan-79 1653 PAT Filman letters
The Filman letters are on my desk for your signature (in folder).
∂23-Jan-79 2228 Jrobinson at SRI-KL (Jane Robinson) YMCA Fitness classes
Date: 23 Jan 1979 2226-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: YMCA Fitness classes
To: jmc at SU-AI
Hello,
Are you still interested in getting your blood pressure down by jogging?
Here's a possibility at the YMCA on Ross Road. There's a fitness class
at 5:00-6:00pm featuring "a variety of exercises [stretching, etc. for about
15-20 minutes], done at each participant's own pace, and a planned progression
from walker to jogger." Heart rates monitored, aerobic conditioning, etc.
Free to adult members of Y. You would be welcome to visit and try it out
--the only requirement, according to the director I talked to, is that if
you are over thirty-five, you should have a statement of some kind from your
physician saying s/he thinks it is safe for you to start jogging. Then, if
after you have tried the Y, you decide to join and participate regularly,
there is a Fitness evaluation ($12.00), perfformed at the Y to assign a
safe-working-heart-rate etc. It's no big deal, but you can see why they
would want it.
Would you care to go with me Monday, January 29 at 5 pm to try it?
I think I'll pass. I balk at the protectiveness (self or otherwise).
Besides my present schedule of running when I get up takes up much
less time than a class.
∂24-Jan-79 0844 REF
How are the recommendation letters coming? Could you get to them today?
(I'm having a harder time lining up interviews without them).
Thanks.
bob
∂24-Jan-79 0946 PN Compiling and running S-1 Fortran programs
To: S1 at SU-AI
See RUNFOR.DOC[DOC,S1]
∂24-Jan-79 1524 ZENON PYLYSHYN at CMU-10A (X320ZP51) Center plans for next yeAR
Date: 24 Jan 1979 1759-EST
From: ZENON PYLYSHYN at CMU-10A (X320ZP51)
Subject: Center plans for next yeAR
To: JMC @ SAIL
Message-ID: <24Jan79 175946 ZP51@CMU-10A>
I spoke to Noam Chomsky today about his visit to Stanford. He said you
had mentioned to him the list of people who would be at the center but he
had forgot most of the names you gave him. It did remind me, though,
that I have been out of touch since coming out here to MIT last summer
(I am still using my CMU mailbox though I use the AI machine here). Fill
me in on what has been transpiring: who is on the final list of visitors
to the Center? Do you know when they are arriving? I am tentatively
thinking about an August 1 arrival. Though I have as yet made no inquiries
about housing, there is some possibility that Ed Feigenbaum's house might
be available for the year, so I am waiting to here from him.
I have been remiss in not sending out a mailing to the group with the
final composition and other information. Anyway here is the list. What
do you think about having a short meeting near the beginning of the study
and perhaps another near the end to which we would invite other people
interested in the field? Do you have ideas about who might come?
McDermott, Chomsky, Dreyfus, Doyle (M.I.T.), Suppes, Creary (Stanford),
Winograd (Stanford), Weyhrauch (Stanford) come immediately to mind, but
there must be many more.
Study group on artificial intelligence and philosophy, Center for Advanced
Study in the Behavioral Sciences - Academic year 1979-80.
Names of participants
Daniel Dennett (Tufts University) Philosophy
John Haugeland (U. Pittsburgh) Philosophy
Patrick Hayes (Essex University) Computer Science
John McCarthy, (Stanford University) Computer Science
Marvin Minsky (M.I.T.) Computer Science and Electrical Engineering
Robert Moore (S.R.I.) Computer Science
Zenon W. Pylyshyn (U. Western Ontario) Psychology
∂24-Jan-79 1720 ME
You have exceeded your disk quota.
The files listed below have been purged to reduce your disk
area to your quota of 2000
Before purging, your files occupied 2400
BACKUP.TMP[W79,JMC]
KNOW.LST[S78,JMC]
DIALNE.LST[DIA,JMC]
CODE.DMP[ 2,JMC]
CRYPT.DMP[ 2,JMC]
LFOL.DMP[F78,JMC]
REV1.DMP[F78,JMC]
ADVEN.DMP[LET,JMC]
ANDREI.XGP[F78,JMC]
FIRST.XGP[W77,JMC]
FINDLE.XGP[F78,JMC]
DREYFU.XGP[F78,JMC]
MRHUG.XGP[S76,JMC]
KYOTO.XGP[W79,JMC]
DIALNE.XGP[DIA,JMC]
FIRST.XGP[F78,JMC]
MENTAL.XGP[F76,JMC]
GOEDEL.XGP[W79,JMC]
DIAL78.XGP[DIA,JMC]
GARDNE.XGP[W79,JMC]
CONCEP.XGP[F78,JMC]
FIRST.XGP[W79,JMC]
∂25-Jan-79 0010 NS
To: JMC
Your following News Service notification
request(s) will expire within a week:
(annapurna)/AP/NYT
∂25-Jan-79 0507 MRC Dialnet blather
To: Ryland at RUTGERS, Hedrick at RUTGERS
CC: LES at SU-AI, JMC at SU-AI
I've thought a little bit about your suggestions. It is important
to me that the packet header remain 32 bits long, so one way I could
allow for the user-number feature is to make the sequence and acknowledgement
fields 4 bits each (like the opcode and channel number are) and put them
into one byte. This would free up an entire header byte for "reserved"
purposes.
This means, of course, that the sequence number bandwidth is decreased
from 255 to 15 (since 0 is reserved for uncontrolled messages). This
makes the maximum practical window size 7 (14 could conceivably be used,
but you would either have spurious NAKs or no NAKs, since you wouldn't
know whether or not to NAK on an out-of-sequence packet. The protocol
says you're just supposed to throw away a packet you've already seen).
Actually 7 is (or will soon be) the maximum Stanford window size anyway,
but perhaps this is unacceptably low on higher-speed communication links?
Certainly at 1200 baud it's no problem. Comment please, since I am
seriously considering making this change, and it IS for you after all.
Another question I have is the problem of acknowledging. Dialnet
uses a scheme whereby acknowledging a packet acknowledges all previous
packets, while PCnet requires explicit acknowledges of all packets.
There are advantages and disadvantages to each way. Dialnet's way is
simpler, since an out-of-sequence packet is discarded; whereas on PCnet
you have to maintain a queue of out-of-sequence packets and run down
this queue every time you get a packet to put it in or to see if you
got it already or to see if you have that packet's next. Retransmission
works differently too; Dialnet requires retransmission of all packets
from the point of the error, meaning that packets that may have been
perfectly okay get discarded, while PCnet does not have that problem.
However, a lost acknowledge would not cause Dialnet to do any
retransmission at all (since the next acknowledge, assuming it goes
through, will take care of it) while on PCnet it does.
It isn't clear which is better, since I can think of cases where
either would lose compared to the other. My own feeling is that the
hair of having a out-of-sequence queue isn't worth it at high speed,
since possibly the medium could retransmit it faster than the CPU
cycles wasted in buzzing down the queue; and at low speed errors
probably don't occur often enough to be worth it (ie, an error is
an infrequent enough even that it is alright for it to take a few
seconds to get back on the right track). I also get the impression
that on a very noisy line PCnet might lose completely.
By the way, as soon as I have a working pair of modems again
I am going to start writing -10 code that will look like a FE TTY
task to DIALNE.MAC on Tops-20. It will be the real FE transfer
vector for TTY lines and go to TTFEDV's one if it isn't the Dialnet
port, otherwise it will kludge things up to look like a Dialnet
task DTE buffer, etc., for DIALNE. It will also have a fake DTEQ
that converts a Dialnet FE buffer into a suitable TTY one. Of
course, doing it this way implies that the level 0 protocol (as
LES calls it) will be in -10 code, impacting both speed and overall
system performance, but at least it will be a running implementation
and will have to do until we have a FE task. With any luck, I might
be able to ship this version to you in a few weeks, and a FTP user
and server program shortly afterwards.
-- Mark
∂25-Jan-79 0930 HVA PATTE'S BIRTHDAY DINNER
CC: JMC, PAT, HVA
John - Patte mentioned that you might return from your trip by Thurs. Feb. 1st,
or Fri. Feb. 2nd, and I wondered if the B-day dinner (Chez Moi), might be
possible for you on the 1st or the 2nd (either night is O.K. for Patte). As
soon as you can let me know your preference (or whether it is possible at all),
I will ask Joan and Les.
∂25-Jan-79 1007 Bmoore at SRI-KL (Bob Moore) suggested participants for AI/Phil meeting
Date: 25 Jan 1979 1007-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: suggested participants for AI/Phil meeting
To: jmc at SAIL
I would strongly recommend Mitch Marcus as a participant in any meetings
we schedule with outsiders, especially if Chomsky or anyone else in his
camp is to be invited.
Other suggestions: David Marr, if we want to get into philosophy of perception
and AI.
Also, I think it would be a tactical mistake not to invite anyone from the
Stanford Philosophy Dept. (I realize that you mentioned Suppes, but he is
not really mainstream philosophy.) Julius Moravczick (probably misspelled)
has had the most contact with us (SRI language group), but I think we should
also invite John Perry. He is much more personable than Julius and I think
his work on personal identity has interesting consequences for self-aware
systems. David Kaplan and possibly even Kieth Donelan (both at UCLA) also
come to mind for similar reasons.
If anyone else occurs to me I will let you know.
By the way, in terms of format, I think that having a paper presented by
the author and then commented on by someone who has taken some time to
study it followed by general discussion is a very good way to organize
these things. For any interdisciplinary meeting such as this, I think
it is very desirable to have the comentator be from a different discipline
than the author.
Bob
-------
∂25-Jan-79 1034 LES Miami reporter
Barry Bearak would still like to talk to you about the Quasar robot, which
he saw demonstrated today. He is at 305 350-2851.
∂25-Jan-79 1117 Mcgoveran at SUMEX-AIM Meeting
Date: 25 Jan 1979 1117-PST
From: Mcgoveran at SUMEX-AIM
Subject: Meeting
To: JMC at SAIL, FXB at SAIL, LES at SAIL
cc: Feigenbaum
Ed would like to meet with you all (and Ralph Gorin, whom I telephoned) on
the 5th of February, late morning, lunch, or early afternoon. I have
reserved Serra conf. rm. for 11 am, though that can be changed.
The meeting is about computers, terminals, and networks. Please let me
know if you can attend at 11 or tell me what other time would be better
for you.
Mary
-------
Feb 5 ok with me. By the way Baskett is changed to FB from FXB.
∂25-Jan-79 1714 100 : patte Texas
reservations have been made in your name for a single at the St. Anthony Hotel,
plane reservations are being made and Jeanette will tell you what they are when
they have been confirmed, tomorrow.
∂25-Jan-79 1918 HENRY at MIT-AI (Henry Lieberman) quasar
Date: 25 JAN 1979 2216-EST
From: HENRY at MIT-AI (Henry Lieberman)
Subject: quasar
To: MINSKY at MIT-AI, EF at MIT-AI, jmc at SU-AI, reid at CMU-10A
I got the following message from Dave Taenzer:
From: dave at BBN-UNIX
Oh, by the way, the Quasar robot thing was on the channel 4 TV news
yesterday with an expose. They showed the guy talking into his sleeve, and
said how you can't trust anyone these days, etc. Then they had the president
of Quasar (whatshisname) on saying that they didn't need that mike guy, so the
TV types closed the door on the mike guy and said how the robot didn't do too
well after that. It was pretty funny. That may be the end of Quasar (I hope).
Pass the word along... Dave
I didn't see the show, does anyone else have more details?
Wonder if our efforts had anything to do with it?
∂25-Jan-79 1927 VENKAT at RUTGERS Request for a copy of your paper
Date: 25 Jan 1979 2221-EST
From: VENKAT at RUTGERS
To: jmc at SU-AI
Cc: VENKAT
Subject: Request for a copy of your paper
Could I get a copy of your paper on "Representation of Recursive
Programs in first order logic"? I would appreciate it , if you could
mail it through the net (VENKAT@RUTGERS), as I understand that the
paper is available on the system at SU-AI. Thank you .
sincerely,
Venkataraman. K.N.
-------
∂25-Jan-79 1927 VENKAT at RUTGERS Request for a copy of your paper
Date: 25 Jan 1979 2216-EST
From: VENKAT at RUTGERS
To: mccarthy at SU-AI
Cc: VENKAT
Subject: Request for a copy of your paper
Could I get a copy of your paper on "Representation of Recursive
Programs in first order logic"? I would appreciate it , if you could
mail it through the net (VENKAT@RUTGERS), as I understand that the
paper is available on the system at SU-AI. Thank you .
sincerely,
Venkataraman. K.N.
-------
An old version of the paper is FIRST.NEW[W77,JMC] and a version currently
being updated is FIRST[W79,JMC]. You are welcome to FTP either version,
but both are PUB source files and a bit strange for direct reading. I could
U.S. mail you a copy of the old version now or a new one in a week or so.
∂26-Jan-79 0152 JBR
To: ME, JMC, LES
CC: WRB, LRM, LLW, LCW
I think SAIL should call Ampex and tell them to give you
a new drive (for DEK) if they can't make it work. This is what we plan
to do for the LLL drive.
∂26-Jan-79 0856 Bmoore at SRI-KL (Bob Moore) another name to consider...
Date: 26 Jan 1979 0856-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: another name to consider...
To: jmc at SAIL
for AI/Phil meeting. I would strongly recommend Barbara Grosz from the
SRI AI Center.
Bob
-------
∂26-Jan-79 1154 HVA Bon Voyage
As today is my Birthday, I'm going to be frivolous and take the afternoon off!
Have a good trip!
Happy birthday!!
∂26-Jan-79 1322 VENKAT at RUTGERS
Date: 26 Jan 1979 1615-EST
From: VENKAT at RUTGERS
To: JMC at SU-AI
In-reply-to: Your message of 25-Jan-79 2342-EST
I FTP'ed both the versions of your paper. But unfortunately we dont
have a "Xerox" printer to get the special characters printed. So I
would appreciate it very much if you could U.S mail the revised
version to the following address:
K.N. Venkataraman.
Department of Computer Science,
Hill Center,
Rutgers University.
New Brunswick. NJ, 08903.
Thank you very much.
-------
∂27-Jan-79 1627 LES
∂27-Jan-79 1609 LES Ampex disks
To: JBR, ME, LLW, LCW
I raised the issue of "fix it or replace it" with Ampex approximately
four weeks ago and they (Bob Trick and Phil McKay) agreed in principle.
I agree that it is now time to demand satisfaction. Unless McKay returns
my phone calls and comes up with a disk rather soon, we shall be forced
to cancel the contract and go after damages.
∂27-Jan-79 1800 BTH SMA3
To: S1 at SU-AI
The S-1 architecture manual is being printed this week. If you want
a copy please be sure your name is in SMA3.DIS[SMA,S1].
∂27-Jan-79 2259 GEOFF at MIT-AI (Geoffrey S. Goodfellow)
Date: 28 JAN 1979 0159-EST
From: GEOFF at MIT-AI (Geoffrey S. Goodfellow)
To: jmc at SU-AI, les at SU-AI
DISTRIB: *CMU, *DM, *MC, *ML, *AI
EXPIRES: 02/27/79 03:33:31
DAVE@BBN-UNIX 01/27/79 03:33:31 Re: Quasar robot
The Quasar robot thing was on the channel 4 TV news on Jan. 23 with
an expose. They showed the guy talking into his sleeve, and said
how you can't trust anyone these days, etc. Then they had the
president of Quasar on saying that they didn't need that mike guy,
so the TV types closed the door on the mike guy and said how the
robot didn't do too well after that.
∨
∂28-Jan-79 1956 REF Dissertation and such
John,
Dissertation draft is in your (physical) mailbox.
Except for fixing up the bibliography, and pretty printing it in the
form the people in the graduate study office like to see, I not planning
to change it.
I'm going away on Wed. morning, first to the Automatic Deduction
Workshop, and then job touring (Texas, Columbia, Maryland, Purdue,
Indiana, Utah and UCSD) and will be back in Palo Alto on Feb. 20.
If you have any suggestions / demands about the dissertation, I hope we
can talk about it then.
Thanks for all your help.
bob
∂29-Jan-79 1303 Bobrow at PARC-MAXC Re:
Date: 29 Jan 1979 12:58 pm (Monday)
From: Bobrow at PARC-MAXC
Subject: Re:
In-reply-to: Your message of 23 Jan 1979 1700-PST.
To: John McCarthy <JMC at SAIL>
cc: bobrow at PARC-MAXC
Delighted to see your paper in its current state. Is it in the mail? How much
work do you want to put in on it in terms of revision?
danny
∂29-Jan-79 1509 DCL
To: AZS, RAK, FWH, WP, SMG, PWM
CC: LES, JMC
∂29-Jan-79 1339 AZS Computer Forum Demonstration
Dave,
As perhaps Sue mentioned to you, we have arranged for a bunch
of computer forum people to go up to the AI Lab at 3:00 PM on Friday,
Februay 2nd. Some will probably want to see a demonstration of the
verifier.
Since I'm a student member of the computer forum committee, I am
charged with the responsiblity of insuring that someone will be available
at that time (for about an hour) to handle the demonstration.
Since you are most knowledgable about the possiblities for
demonstration, I would appreciate it greatly if you could arrange for
someone to be there handle this.
Please advise me if there are any problems or things that I need
to do other than arranging for delivery and return of your audience
Alfred Spector (AZS @ SAIL)
REPLY: Greetings:
The following will coordinate among themselves to organize a demo for you:
Dick Karp (RAK), Freider v. Henke (FWH) , Wolf Polak (WP) and Steve German (SMG)
What would be nice to have is a dedicated machine (i.e., dedicated to
whatever demos you are giving on Friday). Otherwise not much will get shown
at 3.00pm. Please try to arrange this with LES and JMC.
I shall be away at the auto deduction workshop in Texas on Friday.
-David
∂29-Jan-79 1559 SAL FORUM MEALS
To: "@FAC.DIS[1,DPB]" at SU-AI
CC: SAL at SU-AI
This is your VERY LAST CHANCE to sign up for luncheons
and the banquet for the Computer Forum. If you have
not already done so, would you PLEASE send me a message
by tomorrow morning, 30 Jan 79, specifying the meals
you plan to eat during the Forum on Thursday and
Friday.
To recap: luncheons are in TMU 281 at 12:15 both
days and the banquet is at the Faculty Club on
Thursday night (cocktails at 6:00; dinner at 7:00).
To make my head count somewhere within an order of
magnitude of the actual diners, please reply if
you have not already done so.
Thanks.
Sally (SAL at SAIL)
I will come to lunch Thursday if it isn't too late.
∂29-Jan-79 1635 DCL
To: "@GROUP.DIS[VCG,DCL]" at SU-AI
NO VERIFCATION SEMINAR TUESDAY 30TH JANUARY - TOMORROW.
∂29-Jan-79 1815 TOB picture interpretation
I want to know whether Quam has done what I
now propose: to get contour maps of the mountain
(do they exist?) from Arlene Blum and make tighter
estimates of size.
My estimate is that we cannot do much to enhance contrast.
That is limited by Poisson statistics of exposed grains.
You said that there are about 160 grains in that dark area
which gives an estimate to only about 8% error, except that
it appears saturated, which means a good lower bound.
Tom
∂30-Jan-79 1326 PEG PASMAC.TXT[1,ALL]
To: S1 at SU-AI
A copy of the *new* PASMAC.TXT[1,ALL] (description of ALL's
macro preprocessor for PASCAL) has been put in the S1 Library
@ SAIL. A copy will appear in the Serra Trailer branch soon.
∂30-Jan-79 1433 PEG S1 UDP backup tapes
To: S1 at SU-AI
To: Users of the C1 Disk Pack
Subject: Location of the backup tapes
Most of the C1 UDP backup tapes have been moved from the tape rack
in the Gosper/Widdoes/Farmwald/Gilbert office to the top shelf of
the bookshelf in LCW's section of aforementioned office in order to
make room for the tapes for subsequent dumps.
PEG
∂30-Jan-79 1657 LES SUDS licensing
I just had discussions with John Poitras of Tech. Licensing and Ted
Panofshy about the plot to license SUDS distribution by Stanford Computer
Systems (or whatever they're called).
Tentative terms are $1000/system sold with nothing up front for a
limited-time exclusive license (e.g. 2 years). It may be that we will
have to make some demands on Foonly Inc. and III to protect this licensing
arrangement, since they appropriated SUDS without payment or permission.
∂30-Jan-79 1731 Feigenbaum at SUMEX-AIM conversations with Jerry Lieberman
Date: 30 Jan 1979 1731-PST
From: Feigenbaum at SUMEX-AIM
Subject: conversations with Jerry Lieberman
To: jmc at SAIL
Here in brief is it:
1. He wont make a decision about computer music space until John Chowning and I
have talked and have agreed on what to do or have agreed to disagree.
I will call Chowning in Paris. He specifically did not want to go through
Leland Smith on this.
2.He agreed to a 5K raise for Ralph.
Ed
-------
∂31-Jan-79 1343 ECOOPER at BBN-TENEXD Lisp and first order logic
Date: 31 Jan 1979 1642-EST
From: ECOOPER at BBN-TENEXD
Subject: Lisp and first order logic
To: jmc at SU-AI
Professor McCarthy,
I read with interest your article on Lisp in the '78 History of
Programming Languages collection.
I am an undergraduate in mathematics at Harvard, hunting around for
a senior thesis topic. Since one of my interests is mathematical logic,
your reference to the upcoming paper by Cartwright and yourself
intrigued me.
Has it been published yet, and if so, where? In its pre-PUB
form on the Net, it's rather unreadable. Would it be possible to
send me a copy (physically, I mean)?
Thank you very much.
Eric Cooper
c/o Mathematics Department
Harvard University
-------
∂31-Jan-79 1845 MPG Thanks
I wish to thank you for your hospitality during my visit to the AI Lab.
I very much enjoyed our discussions.
Best wishes...Mike.
∂31-Jan-79 2104 AFB
Hello again,
I'm really sorry it has taken me so long to respond to your
last letter-subject-coming to see you about what I did for the
independent study I tookunder you last quarter. I've been
in a sick-try to catch up-relapse cycle for some time now.
I'm trying to get caught up in my classes again now, but I don't
want to push myself anymore so I can lick this bug!! Anyway ,
because of this I've been putting off coming to see you. Now I
think I'd better just tell you in this letter what I did.
I read Winston's Artificial Intelligence (all of it)
quite closely. I also read Nilsson's Problem Solving Methods
in Artificial Intelligence ( the major portion of it).
I also did some low level Lisp experimenting over at LOTS
(low level since I am generally unfamiliar with it).
I also employed Nilsson's pruning
methods on a 2 level lookahead strategy
program (written in Sail).
Still queued up is reading a couple of articles over at the CS
Library as soon as my stamina's built up a bit more. C'est tout.
I hope this is satisfactory. I also hope I haven't been
any trouble or irritation to you. I really appreciate your
having allowed me to study under you even though I never had
to consult you on anything. It was good to
know you were there in case I did. To sum up, Thanks!
Your's truly,
Anne Beetem
OK, that will do for a P. If you want a letter grade, we must talk.
Please get grade change card for Anne Beetem in CS390 from N to P.
∂01-Feb-79 1102 HVA Postponement of Patte's Party
For various reasons, party is postponed until Sat. Feb 10th-hope you can be
present and hope I don't have to postpone again! Oh, and Welcome Back - trust
the trip was O.K, and thank you for the B-day msg.!!
∂01-Feb-79 1849 DCO
I will be away until Feb 25 but would like to talk to you sometime
thereafter about implementing FOPL at Stanford.
Fine. Let's do it.
∂02-Feb-79 0012 NS
To: JMC
Your following News Service notification request(s) have expired:
(annapurna)/AP/NYT
∂02-Feb-79 1331 JLH MEETING
To: S1 at SU-AI
MEETING OF ALL S-1 PERSONAL (ON BOTH PROJECTS).
WEDNESDAY 2:30 PM.
ALL PEOPLE PLEASE ATTEND - OR SEND ME A GOOD EXCUSE WHY YOU CAN NOT.
JOHN.
∂02-Feb-79 1345 LCW
To: S1 at SU-AI
∂02-Feb-79 1331 JLH MEETING
To: S1 at SU-AI
MEETING OF ALL S-1 PERSONAL (ON BOTH PROJECTS).
WEDNESDAY 2:30 PM.
ALL PEOPLE PLEASE ATTEND - OR SEND ME A GOOD EXCUSE WHY YOU CAN NOT.
JOHN.
[I am sure that John means "all Stanford S-1 subcontractors", ie., both
the SAIL and CS S-1 efforts. Curt]
∂02-Feb-79 1341 CET Summer Visitation Program at Hughes Aircraft Co.
To: "@FAC.DIS[1,DPB]" at SU-AI
"The Hughes Aircraft Company's "Professor Visitation Program" (Summer
Employment) for 1979 would like to include Stanford University as a
participant.
Our Ground Systems Group located in suburban Fullerton, CA, is very interested
in having one of your Engineering faculty join their staff during this summer.
They are particularly interested in Professors with primary interests in
one of the following areas:
Phased Array Radar
Advanced Digital Communications
Microwave Systems
Computer Systems
Liquid Crystal Technology
Advanced Signal and Data Processing
The duration of the assignment is flexible depending on your academic calendar,
however, we are plannin on assignments beginning around the middle of June and
finishing around the end of August.
Interested faculty members are invited to submit a resume to me or Mr. Fred
Rodriguez at the addresses below or call (collect) for more details.
Mr. Daniels T. Stewart
Corp. Manager, College Relations
Hughes Aircraft Company
Bldg. 100, M/S 445
P. O. Box 90515
Los Angeles, CA 90009
(213) 670-1515 x6877
Mr. Fred Rodriguez
Head, Employment
Hughes Aircraft Company
Ground Systems Group
P.O. Box 3310
Fullerton, CA 92634
(714) 871-3232 x4080
signed/ Daniels T. Stewart
∂02-Feb-79 1612 TOB
To: JMC at SU-AI, feldman at SUMEX-AIM, JAF at SU-AI,
levinthal at SUMEX-AIM, feigenbaum at SUMEX-AIM
Dr. Chern has requested that we
convene the steering committee to evaluate the research
plans of our NSF research in robotics.
The suggested format is:
1 hr 30 min presentation;
45 min comment, discussion, and recommendation.
There will be a practice session to encourage concise and accurate presentations.
Suggestions are invited for format.
I suggest Wed, Feb 14 at 2:30pm.
Am I on this steering committee? If so Wed Feb 14 at 2:30 is ok.
∂05-Feb-79 0013 LGC via AMES-TIP My IJCAI Paper
Thanks for your comments on my paper; over the weekend I have discovered some
infelicities and one bug in it, all of which are now fixed. I may be able to
have the improved version sent to the IJCAI referees, and I'll also send copies
to Bob Moore, Barbara Partee, and perhaps Nils Nilsson, who has been thinking
about these issues recently. I think Partee is at U. Mass. at Amherst; do you
happen to have her full mailing address handy? If not, I can dig it up elsewhere.
∂05-Feb-79 0854 JLH via SU-TIP Meeting,...
To: S1 at SU-AI
Sorry about the lack of agenda and location.
Meeting: ALL Stanford S-1 people.
Location: ERL 237
AGENDA: Compiler and language efforts.
∂05-Feb-79 1241 TOB
Yes, you are on it.
∂05-Feb-79 1339 100 : patte meeting regarding yao and winograd
Feigenbaum has scheduled a meeting for 2pm tomorrow in Serra house
conference room to discuss promotions for yao and winograd. I have
put it in your calendar.
∂05-Feb-79 1441 PAT expenses
I have the following things on my list of unfinished business:
invoice 9594, Nov 9 to III $476.00 to be paid to travel agent
invoice 10324, jan 19 to III $64.00 to be paid to travel agent
to be invoiced to III
------
$540.00
invoice 10389, Jan 25 to Texas, expense report to be submitted
Patte
∂05-Feb-79 1545 Mcgoveran at SUMEX-AIM Dr. Ramakoti Sadamanda
Date: 5 Feb 1979 1545-PST
From: Mcgoveran at SUMEX-AIM
Subject: Dr. Ramakoti Sadamanda
To: JMC at SAIL, RWW at SAIL
This fellow is visiting from Nehru University, delhi, on Wednesday.
His field is: Predicate Calculus for Pictorial Descriptions of a Computer.
I am looking for people who would be interested in talking with him. Ed gave
me your names. Can you help me?
I found out about this from Dick Umhoefer of the International Center.
If you need more info, you can send me a msg or call him at 71984. His
secretary is Maria Bum (pronounced Boom).
Dr. Sadamanda will be here for Wednesday only.
Mary
-------
∂05-Feb-79 1806 Feigenbaum at SUMEX-AIM re IJCAI travel, for your information
Date: 5 Feb 1979 1759-PST
From: Feigenbaum at SUMEX-AIM
Subject: re IJCAI travel, for your information
To: jmc at SAIL, les at SAIL, reddy at CMUB
Date: 5 Feb 1979 1633-PST
From: Engelmore
Subject: Conversation with Dave Russell
To: Buchanan, Feigenbaum
Russell told me on the phone today that he's very negative on
paying for ARPA contractors' travel to IJCAI this year ("It's a matter of
reciprocity.") In fact, he said he's not going to fund any trips to
the conference. IJCAI75 "nearly drove me up the wall", he said, and
served to make future overseas IJCAI conferences very touchy politically.
He sounded pretty serious about enforcing this policy. Considering
ARPA's key role in funding AI research groups, this could put a big dent
in the attendance this year. If this is a matter of concern for either
of you, it may be a good idea to talk to him and try to change his mind.
Bob
-------
-------
I certainly think we should try to talk him out of it. As for reciprocity,
having the conference in North America every other year more favorable treatment
than most international organizations give the United States. Better
than IFIP, for example. Of course, there should be a U.S. national conference
every other year when IJCAI doesn't meet.
∂05-Feb-79 2345 Feigenbaum at SUMEX-AIM musicians;winograd
Date: 5 Feb 1979 2345-PST
From: Feigenbaum at SUMEX-AIM
Subject: musicians;winograd
To: jmc at SAIL
Hi, John!
two things:
1. Don't forget about the meeting of senior faculty Tuesday afternoon at 2pm
re Winograd tenure decision.
2. Can I meet with you for ten minutes before or after to discuss the
divorce settlement with the musicians. I thnk I have the righ idea for a
divorce settlement (got it from Les), but I need uour agreement,
and I have mental conflicts about the price of the settlement.
Ed
-------
ok, see you at 1:30 or shortly thereafter
∂06-Feb-79 1059 TOB
To: JMC at SU-AI, feldman at SUMEX-AIM, levinthal at SUMEX-AIM,
feigenbaum at SUMEX-AIM
There was a conflict for the first suggested time, Wed 2:30.
Please let me know which are possible and which preferable below.
mon before noon
tues until 11, early pm
Wed before 10, after 2
thur until 11, early pm
fri mostly free; home research
∂06-Feb-79 1356 PAT AIM 1
exists as AIM1[w79,jmc]. The two tables have not been put in because I did
not know exactly what kind of format you wanted (meaning are you going to
want pub and graphics). Also, I'm not exactly sure how some of the features
in memo.pub work.
∂07-Feb-79 1008 CVW Instructor's Statements for Autumn Quarter Courses
To: SJF at SU-AI, DRA at SU-AI, DPB at SU-AI, JGH at SU-AI,
SSO at SU-AI, GIO at SU-AI, FB at SU-AI, DEK at SU-AI,
JMC at SU-AI, TOB at SU-AI, JLH at SU-AI, TW at SU-AI,
gifford at PARC-MAXC
CC: CVW at SU-AI, JSB at SU-AI
On behalf of the Course Evaluation Committee, I invite you
to submit an instructor's statement about the course(s) you taught last
quarter. The statement can include your comments on the goals of the course
and how they were achieved, your approach to the material, your view of the
students, any difficulties (e.g. poor textbook), etc.; it is a potentially
valuable source of information to someone who later teaches the course. If
you wish to provide such a statement, please send it to the Course Evaluation
Committee (CEC), Polya Hall or to CVW@SUAI.
Chris Van Wyk
∂07-Feb-79 1235 HVA Terminal at Home
You skipped off to the volleyball court last evening before I could say
Thank You for asking Lew to bring in your terminal - Lew patiently spent n
hour with me, giving me instruction on the various differences-it was inter-
esting and I would like to learn more about using it. However, I reached the
conclusion that, since I am far from being a "perfect typist", it would be
counter-productive to try to use your portable terminal, thos' I greatly
appreciate your and Lew's generosity. So, tonight, I shall run through Patte's
copies of Tapes #1, and 2 (Patte's copies seem to be more clear than the other
ones), and tomorrow, I shall re-arrange my time in order to start typing on
this terminal, where I can corrct my mistakes more easily! (as you can see,
there are a few "typos" in this message!)
∂07-Feb-79 1524 CLT
Did you make an appointment for your car?
∂08-Feb-79 0809 RAJ.REDDY(A610RR29) at CMU-10B IJCAI TRAVEL SUPPORT FLAP
Date: 8 Feb 1979 1106-EST
From: RAJ.REDDY(A610RR29) at CMU-10B
Subject: IJCAI TRAVEL SUPPORT FLAP
To: balzer at USC-ISI, feigenbaum at USC-ISI, Hart at SRI-KL, mccarthy at SU-AI,
rick at RAND-UNIX, phw at MIT-AI
Some of us got the following message from Bob Englemore.
Russell told me on the phone today that he's very negative on paying
for ARPA contractors' travel to IJCAI this year ("it's a matter of
reciprocity.") In fact, he said he's not going to fund any trips to the
conference. IJCAI75 "Nearly drove me up the wall," he said, and served
to make future overseas IJCAI conferences very touchy politically.
He sounded pretty serious about enforcing this policy, considering
ARPA's key role in funding AI research groups. This could put a big
dent in the attendance this year. If this is a matter of concern for
either of you, it may be a good idea to talk to him and try to change
his mind.
I have just had a talk with Dave Russell regarding the message some of us got
from Bob Englemore about travel restrictions to IJCAI-79. I explained to him
that:
1. Japanese should not be treated like the Russians.
2. There are a large number of U.S. Government-sponsored scientific
meetings already, such as the USA-JAPAN workshops.
3. There is a new joint research program being planned by Fukuda (and I
guess, his successor) and Carter at a billion dollar level (initially
it is to be 200 million).
4. It is unlikely we will be sending hordes of people - given the tight
travel budgets. At most, three or four fully paid and several
partial travel supports per AI group.
He appears to have relented and agreed that he will not take an arbitrary
position and will consider most requests for foreign travel on the normal merit
basis. I promised that together we will propose some general guidelines to
help him make the decisions.
I suggest that we all talk together in the near future and have a common
position. Let us also try not to make too many individual probes to ARPA, but
do it together.
Regards, Raj
∂08-Feb-79 0913 BALZER at USC-ISIB Re: IJCAI TRAVEL SUPPORT FLAP
Date: 8 FEB 1979 0914-PST
From: BALZER at USC-ISIB
Subject: Re: IJCAI TRAVEL SUPPORT FLAP
To: RAJ.REDDY at CMU-10B, balzer at USC-ISI,
To: feigenbaum at USC-ISI, Hart at SRI-KL, mccarthy at SU-AI
cc: BALZER
In response to the message sent 8 Feb 1979 1106-EST from RAJ.REDDY@CMU-10B
I AGREE THAT ARPA SHOULD ADOPT A LESS POLITICAL AND MORE SCIENTIFICALLY
GROUNDED TRAVEL POLICY, AND I'M WILLING TO ASSIST IN THAT EFFORT AS A
UNINVOLVED PARTY.
HOWEVER, I SHOULD MAKE IT CLEAR THAT I IN NO WAY WISH TO SUPPORT IJCAI79. THE
DECISON TO HOLD IT IN JAPAN WAS EXTREMELY DESTRUCTIVE BY DISENFRANCHISING
THE VAST MAJORITY OF POTENTIAL PARTICIPANTS(WHO JUST HAPPEN TO BE AMERICAN).
ONE WOULD HAVE THOUGHT THAT THE ORGANIZING COMMITTEE WOULD HAVE LEARNED FROM
ONE BAD CONFERENCE(IJCAI75), BUT EVIDENTLY IT TAKES AT LEAST TWO. HAVING
AN ATTENDABLE AI CONFERENCE ONCE EVERY FOUR YEARS IS FAR TOO INFREQUENTLY.
ONLY IF AN AMERICAN AI CONFERENCE(POSSIBLY SCHEDULED FOR THE EVEN YEARS) WERE
ORGANIZED WOULD I SUPPORT THE TRAVELING IJCAI ROAD SHOW. UNDER THOSE
CIRCUMSTANCES THE LAUDABLE GOAL OF STRENGTHENING FOREIGN AI PROGRAMS DOES
NOT COME AT THE EXPENSE OF SEVERELY WEAKENING OUR OWN.
FOR THESE REASONS, I HAVE NOT SUBMITTED A PAPER FOR IJCAI79 NOR DO I PLAN
TO ATTEND, EVEN IF TRAVEL FUNDS WERE MADE AVAILABLE.
BOB
-------
∂08-Feb-79 1037 LES FIRST Pub bug
The source file contains two "font b" declarations, one specifying GRFX35 and
the other specifying ZERO30.
∂08-Feb-79 1030 DCL Auto logout
To: LES
CC: JMC
Les, Many months ago, John made me two promises; 1. my own phoneline,
and 2.Exemption from the autologout feature.
Neither has been kept.
I really would appreciate the autologut feature ASAP. And the phone line
too.
-David
∂08-Feb-79 1037 TW
To: feigenbaum at SUMEX-AIM, DEK at SU-AI, JMC at SU-AI, TW at SU-AI
The following note was sent to Ed Feigenbaum in October, and addresses
several of the issues which have come up in the discussion of my
promotion. In addition, I have just given him two papers on KRL, one
to be published in Cognitive Science, and the other an internal
specification document for KRL-1 which has not yet been polished
into a publishable form.
-------------------------
October 17, 1978
As I mentioned in our discussion in August, I am sending you a note
describing some of my plans with regard to my future at Stanford and
explaining the current status of my work. Feel free to distribute this to
other people who will be involved in the tenure decision. It is on line
at the AI lab as TENURE[1,TW]
I have not tried to include a general discussion of my work and teaching
over the past years. Most of the relevant material was put together for
two things: my nomination for the dean's teaching award and my nomination
for the Waterman award. They are in my file and provide a very full
summary. I am also sending you an updated vitae, including current
publications. Here I will just deal with current plans and with
problematic issues which came up in our discussion.
Current research and writing
My original intention was to have my textbook done before now, so that it
could be counted in the decision. About a year and a half ago, I decided
to concentrate on finishing it. Plans shifted, and since the time I
decided to work "only on the book", I have written several papers and
almost completed another book (jointly with Fernando Flores) which is
still in draft state, but for which we already have a firm publisher's
offer. My book is about 2/3 complete, and I hope to finish it during this
academic year. Drafts are available to show what it will be like. I
would be glad to send them to people for review if you feel that would be
useful, but I think my publication record is quite sufficient for tenure,
even if the book didn't exist at all. I have continued to do research on
KRL, along with Danny Bobrow at Xerox, but it too has been somewhat
sidetracked while I did the other papers (which are more theoretical
papers on language and representation). We are currently planning a long
paper describing the KRL-1 design and implementation, to be written some
time during 1979.
Division of time between Stanford and Xerox
One of the questions which is sure to come up regards the amount of time I
have spent at Xerox. For the past few years, my work on Xerox research
has been closely tied to my Stanford research. In addition to being a
consultant at PARC, I have a grant from Xerox which supports two students
and part of my Stanford salary. Work on KRL has been supported by an NSF
grant to Stanford as well. I plan to continue my research connection with
Xerox because it provides an excellent opportunity for both me and my
students to take advantage of both facilities and contacts with top notch
computer scientists. As the hardware developed there becomes more
available, I hope to use whatever influence I have to get more of it here
at Stanford, so the work that students and I do here can be better
integrated with what goes on there.
Over the past few years, my Xerox-related work has occupied about the one
day a week which I am paid for as a consultant there. However, I have
spent much more time than that at PARC in order to make use of the
equipment, particularly for text preparation. I have done most of the
work on my textbook and other papers (all of which are published with a
Stanford affiliation) using the text editors and printers at Xerox.
During my Mellon leave, I was there full time working on my writing.
My original intention was that in the spring of 1978, when both my book
and the new CS dept. building were done, I would switch to spending most
of my time there. Given the delays to both, I have moved my base of
text-preparation to Stanford anyway, partly because I have decided to use
Don Knuth's TEX system for my book instead of the Xerox-based system I had
planned to use. The combination of the much more sophisticated formatting
capabilities and the high quality film printer will make it possible to
produce a much higher quality book than I had anticipated. I am currently
spending all but one day a week in my office at Polya Hall, and expect to
continue that pattern when we move to the new building. The fact that
computing facilities, teaching, and student meetings will all be in one
place (instead of spread between AI lab and campus as they have been
previously) means that I will continue to spend my writing and research
time primarily on campus.
Linguistics and Computer Science
During the past few years, my main focus of activity has been in the
computer science department, even though I have a joint appointment in
linguistics. My courses are cross-listed, and I have served on a number
of linguistics faculty committees. I have one dissertation student
getting a Linguistics Ph.D (in contrast to eight dissertation students,
four beginning Ph.D. students and 5 masters advisees in CS and one
interdepartmental student doing computer-related research). In the
future, I expect to remain more involved with CS than with linguistics,
but would like to increase the linguistics side to some degree. Being in
the main quad will make it much easier to be involved with ongoing
activities in the linguistics department (such as colloquia) and to meet
with students. I am interested in developing an undergraduate course on
computers and language which could be tied in to the undergraduate human
language major. The workshops and visitors associated with the Sloan
foundation grant this year (for which I am one of the recipients) should
serve to increase my contact with linguistics and philosophy.
Teaching
Most of my teaching at Stanford has been in courses connected to my
research specialty. This has been especially useful to me in connection
with writing a textbook, but once that is done I would like to branch out
a bit. The current courses could be offered less frequently, and I would
like to teach other courses in AI and in elementary areas, both in CS and
in the intersection of computers and language as I mentioned above.
Through my research in KRL, and through working with some of my advisees,
I have gotten involved in thinking about programming languages and
systems, and would also be interested in the possibilties of teaching
courses in those areas.
Ph.D. advising
This is an area in which I have learned a lot in the past few years. As a
beginning teacher I made the mistake of becoming heavily overcommitted to
advising (see the numbers given above), and found it difficult at times to
take a strong directive role. I have had a number of discussions both
with students and with other teachers, and have begun to operate in a more
carefully thought out way, using my time to provide strong regular
guidance to the students who need it, while continuing to meet on a less
rigid schedule with the more independent students. I have supervised two
completed Ph.D's (Bill Paxton and Jerry Ginsparg), have three other CS
students (Mitch Model, Alan Borning, and Larry Masinter), one
interdeparmental student (David Shaw), and one linguistics student
(Annette Herskovits) who are finishing this year, and several others
(David Levy, Paul Martin, Dick Gabriel) who are well along in their work,
and will be done in less than two years.
I have also found it quite interesting (and also time consuming) to take a
relatively active part in working with students for whom I am on the
dissertation committee but not as princpal advisor. I would like to be
able to expand the amount of time I spend on this. In the future, I would
like to reduce the number of advisees in order to be able to work more
closely with them. Ideally, I would like to have not more than 3 or 4
actively working on dissertations, and be advisor to a few other beginning
students.
Other issues
This isn't intended to be a complete list -- it's just the things which
have come up, and I would be glad to discuss anything else you or the
others have questions about.
--terry
---------------------------------
∂08-Feb-79 1040 PAT phone call
Prof. Pyster called regarding your talk on Feb. 23 in Santa Barbara. He
would like as soon as possible, a biography (I have one), an abstract of
your talk, your arrival plans (do you want to come early in the day or
just for the evening), do you want a hotel, or do you want to stay at
Martin Davis's house. His number is (805) 961-3236.
∂08-Feb-79 1150 ME
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: 2000
Your files occupy 2383
∂08-Feb-79 1420 PAT
re: Pyster, my mistake, its the 21st.
Remind me to settle travel and call Pyster when I come in.
∂09-Feb-79 0713 Scott at SRI-KL (Scott J. Kramer) LISP MACHINES
Date: 9 Feb 1979 0713-PST
From: Scott at SRI-KL (Scott J. Kramer)
Subject: LISP MACHINES
To: jmc at SU-AI
Hello John,
I am interested in learning more about the LISP machines and would
like to know where to get information on them. Also, are they
accessable via the ARPANET?
Thanks...
Scott Kramer
-------
A manual is available from Greenblatt at mit-ai, and I believe that one
of the machines is accessible via the Arpanet through an M.I.T. local net.
∂09-Feb-79 0957 Scott at SRI-KL (Scott J. Kramer)
Date: 9 Feb 1979 0957-PST
From: Scott at SRI-KL (Scott J. Kramer)
To: JMC at SU-AI
In-reply-to: Your message of 9-Feb-79 0931-PST
Thanks John,
I sent a message to Greenblatt the other night but he hasn't replied.
-------
∂09-Feb-79 1023 CET
∂09-Feb-79 1016 JMC account on department machine
I would like an account with name csd.mccarthy and password 848-0898.
*****
John, you've got it. Carolyn
∂09-Feb-79 1302 Bmoore at SRI-KL (Bob Moore)
Date: 9 Feb 1979 1302-PST
From: Bmoore at SRI-KL (Bob Moore)
To: jmc at SAIL
I turned in my thesis yesterday; everthing has been
a smashing success.
Also, two more names for an AI/Phil conference:
Bill Woods and Brian Smith.
Bob
-------
∂09-Feb-79 1502 RPG ...
To: "@USERS.DIS[AID,RPG]" at SU-AI
I forgot to add: available through (HELP).
-rpg-
∂09-Feb-79 1457 RPG Dislocated open file helper
To: "@USERS.DIS[AID,RPG]" at SU-AI
Sometimes in NEWIO you can get stranded by opening some
file and then losing a pointer to it. This sometimes occurs in
errors; the function CLOSER can help in this case.
When called with 0 arguments, it searches the internal list
of open files and asks about each non-TTY file whether you want it
closed:
(EOPEN '(EDIT /113)) ;loser is asking for trouble!!! No pointer to this one!!!
;unless s/he likes typing this:
#FILE-IN-|DSK:EDIT.113[AID,RPG]|-7774
(EOPEN '(CLOSE 1)) ;there's no end to his/her stupidity
#FILE-IN-|DSK:CLOSE.1[AID,RPG]|-7766
(CLOSER) ;hmmm, maybe not as stupid as all that!
#FILE-IN-|DSK:EDIT.113[AID,RPG]|-7774 - Close this one? Y
#FILE-IN-|DSK:EDIT.113[AID,RPG]|-7774 closed!
#FILE-IN-|DSK:CLOSE.1[AID,RPG]|-7766 - Close this one? N
DONE ;managed to win anyway?!?!?
-rpg-
∂09-Feb-79 1512 Feigenbaum at SUMEX-AIM what Hennessy really feels...
Date: 9 Feb 1979 1512-PST
From: Feigenbaum at SUMEX-AIM
Subject: what Hennessy really feels...
To: dek at SAIL, jmc at SAIL, ejm at SAIL
Mail from SU-AI rcvd at 9-Feb-79 0942-PST
Date: 9 Feb 1979 0941-PST
From: John Hennessy <JLH at SU-AI>
Subject: position
To: feigenbaum at SUMEX-AIM
CC: FB at SU-AI, JLH at SU-AI
Just to confirm our discussion. While I feel it would be great to have Ullman
here, I feel he belongs in the theory or AA group, and is not the right
person for the systems position.
-------
∂09-Feb-79 1528 Bobrow at PARC-MAXC Re:
Date: 9 Feb 1979 3:19 pm (Friday)
From: Bobrow at PARC-MAXC
Subject: Re:
In-reply-to: Your message of 23 Jan 1979 1700-PST.
To: John McCarthy <JMC at SAIL>
cc: bobrow at PARC-MAXC
I received your message about getting your paper in its current state, but not the
paper. Is this a defect in the mails, or is it another kind of bug? Looking
forward
to seeing it.
danny
Please send minima to Dan Bobrow at PARC.
∂09-Feb-79 2351 Kanerva at SUMEX-AIM Today's meeting
Date: 9 Feb 1979 2351-PST
From: Kanerva at SUMEX-AIM
Subject: Today's meeting
To: McCarthy at SU-AI
Thank you for talking to us. Hopefully this time we can take it beyond
mere polite listening.
-------
∂10-Feb-79 1316 ME
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: 2000
Your files occupy 2387
∂12-Feb-79 0922 HVA Some thoughts....
As it's Monday-and you don't teach today-I'm not sure when or if you'll be in-
and I have to leave early. But-I wanted to say I'm glad you came to the party
for Patte and hope you enjoyed it. If I'm not mistaken, today is Vera's Birth-
day-I'm thinking of her and of you....
∂12-Feb-79 1243 Ijcai79 at SUMEX-AIM IJCAI-79 Computers and Thought Award
Date: 12 Feb 1979 1243-PST
From: Ijcai79 at SUMEX-AIM
Subject: IJCAI-79 Computers and Thought Award
To: Reddy at CMU-10B, Newell at CMU-10B, Michie,
To: MCCarthy at SAIL, Feigenbaum, Minsky at MIT-AI,
To: Winston at MIT-AI, Amarel at RUTGERS, Beldsoe at UTEXAS,
To: Waterman at RAND-UNIX, anderson at RAND-UNIX, tob at SU-AI,
To: sacerdoti at SRI-KL, walker at SRI-KL, jed at SU-AI,
To: davis, lenat
cc: Buchanan, ijcai79
We are pleased to announce that Dr. David Marr is the recipient
of the Computers and Thought Award for 1979. He will deliver
the Computers and Thought Lecture at the Sixth International
Joint Conference on Artificial Intelligence, to be held in
Tokyo in August, 1979.
David Marr is Associate Professor of Psychology at M.I.T.
Educated at Trinity College in the University of Cambridge,
he received his B.A. and M.A. in mathematics, and his
Ph.D. in neurophysiology in 1971 for work on the cerebellar
cortex. His education included training in neuroanatomy,
neurophysiology, biochemistry and molecular biology. Dr. Marr
was awarded a research fellowship at Trinity College, and held
a research position in the M.R.C. Laboratory of Molecular
Biology in Cambridge, working with Dr. Sydney Brenner and
Dr. Francis Crick.
Among many publications, two of special note are:
"A Theory of Cerebellar Cortex." J. Physiol. 202
(1969): 437-470.
"Analysis of Occluding Contour." Proc. Roy. Soc. Lond. B
197(1977): 441-475.
Bruce G. Buchanan
Program Chairman, IJCAI-79
-------
∂12-Feb-79 1557 DEW
Using Patterns and Plans to Solve Problems and Control Search
∂12-Feb-79 1723 TOB steering committee meeting
To: JMC at SU-AI, feldman at SUMEX-AIM,
feigenbaum at SUMEX-AIM
The previous suggested time encountered conflicts.
How is Tuesday, Feb 19, at 9 am?
Tom
∂12-Feb-79 1745 ZENON PYLYSHYN at CMU-10A (X320ZP51) Forthcoming visit & housing
Date: 12 Feb 1979 2044-EST
From: ZENON PYLYSHYN at CMU-10A (X320ZP51)
Subject: Forthcoming visit & housing
To: JMC @ SAIL
Message-ID: <12Feb79 204425 ZP51@CMU-10A>
I have let Cutler Know that I am coming next year (now that the sabbatical
is all approved). The biggest problem this year was finding suitable
housing (I had to make several trips to MIT and even then the house I
arranged was withdrawn at the last minute). Could you help out by
posting the announcement I will send in the next mail message. Perhaps
there is a computer announcement bulletin into which you could dump it
for all to see. Any help would be appreciated.
-----------
Zenon
∂12-Feb-79 1753 ZENON PYLYSHYN at CMU-10A (X320ZP51) Housing required for 1980-81 term
Date: 12 Feb 1979 2051-EST
From: ZENON PYLYSHYN at CMU-10A (X320ZP51)
Subject: Housing required for 1980-81 term
To: JMC @ SAIL
Message-ID: <12Feb79 205157 ZP51@CMU-10A>
I will be at the Behavioral Science Center from August 1, 1980 until June
30, 1981 (approx) and need a house with at least 3 bedrooms for that time.
Prefer close to Stanford and walking distance to both primary and high
schools. If you have any ideas please let me know over the TELNET, or
write:
Zenon W. Pylyshyn,
Department of Linguistics and Philosophy,
Massachusetts Institiute of Technology,
Cambridge, Mass 02139.
I have put your notice on the bulletin board and have mailed it to CASBS.
∂13-Feb-79 0937 DEW my oral exam
To: JMC at SU-AI, buchanan at SUMEX-AIM
Hans is coming out for my oral. I plan to get him put on my
committee and to have the oral sometime in May. I'll set the exact date
in late March or April. I'll have a copy of my thesis to you by the end of March.
Let me know if this is not all right for some reason.
Thanks, Dave
Tentatively it is all right, but I might just be travelling. If so,
I propose a rescheduling in good time.
∂13-Feb-79 0948 RWG annihilation
To: MLB
CC: RWG, JMC
Given array A(k,n) k,n = 1,2,..., find C(k) to make the "nth series"
sum(C(k)A(k,n),k≥1) = 0 for all n.
Choose C(1) = 1. Choose C(2) to make the first two terms of the first series
(C(1)A(1,1) + C(2)A(2,1)) sum to 0. (One equation, one unknown.) Choose
the next two terms (C(3) and C(4)) to make the first 4 terms of the first two
series sum to 0. (Two eqns, two unknowns.) Choose the next 4 terms of C(k)
to annihilate the first 8 terms of the first 4 series. (Four eqns, four
unks.) Etc.
After you have done all that, look at the sum for some particular n. We
see that the sum is infinitely often zero in accordance with your construction,
i.e. each sum of 2↑k terms is zero, but for other values of k, the sum can
fluctuate wildly. Is that summing to zero? Alas, convergence questions
cannot be avoided in matters like this.
∂13-Feb-79 1036 RWG
To: JMC, MLB
too true. mlb was in fact concerned with A(k,n) which decayed
with large k,n, so wild fluctuation would have to come from large C(k). A
possible fudge might be to determine more than 2↑(i-1) terms of C(k) on the
ith iteration (which you might have to do anyway in singular cases). The
idea would be to have many more unknowns than equations and then use
"linear programming" to minimize the solutions (C(k)), applying increasingly
stringent conditions with successive iterations. For mlb's harmonic case,
it might even be necessary to consume 2↑n C's for each new n. (If this works
at all).
∂13-Feb-79 1115 Feigenbaum at SUMEX-AIM deal with musicians
Date: 13 Feb 1979 1110-PST
From: Feigenbaum at SUMEX-AIM
Subject: deal with musicians
To: jmc at SAIL
cc: les at SAIL
John,
Still waiting for JMC's confirmation of computer-deal with musicians, and
enough detail so I can contact John Chowning.
Talked with Les a little about it last night at meeting.
Can you give me something definite by this afternoon's faculty meeting?
Ed
-------
∂13-Feb-79 1408 HAYDEN at USC-ISIE LNI Assembly Language Interface
Date: 13 Feb 1979 1315-PST
From: HAYDEN at USC-ISIE
Subject: LNI Assembly Language Interface
To: JMC at MIT-AI
Mike Lyle told me that you have a document that describes how
to use the LNI. I am interested in the assembly level
interface to the hardware. Do you have or know of such a document?
Any help would be apprecieated. Thanks.
-------
Your message was forwarded to me at SU-AI. You must have the wrong name
at MIT-AI, because that is my name there. - John McCarthy
∂13-Feb-79 1528 TW Promotion materials
To: feigenbaum at SUMEX-AIM
CC: TW at SU-AI, JMC at SU-AI, DEK at SU-AI
Ed,
I have a few more thoughts on what to put in the promotion documentation.
I have four papers which are scheduled to appear in refereed journals (one
in CACM, one in AI Journal, and two in cognitive science (one of those two
is joint with Danny)) which do not appear in my vitae as I last sent it
in. I gave Mary a new hardcopy and am appending the relevant part to this
message.
In addition, I today received an advertising blurb for Dan Dennet's new
book on philosophy and AI. They include quotes from several notables
(e.g. Chomsky, Hilary Putnam, Jerry Fodor) one of whom is me. The fact
that they thought my opinion would help sell books says something about
the "fame" issue we talked about earlier. Also, looking at the list for
the La Jolla conference, I note that I am the only one (out of eight) who
is an assistant prof. One other (George Lakoff) is an Associate, and the
rest full professors. The same was true (moreso, I think) for the MIT
panel last year.
If there is anything else I can do to help expedite things, let me know.
--terry
Publications for 1979 and pending
[1979a] "Towards convivial computing" in M. Dertouzos and J.
Moses (eds.), The future impact of computers and
information processing. MIT Press, 1979 (in press).
[1979b] ....... with D. Bobrow, "KRL: Another perspective",
Cognitive Science 3:1 (January, 1979).
[1979c] "Beyond programming languages" accepted for publication
in CACM, to appear in 1979.
[1979d] Extended inference modes in computer reasoning systems,
to appear in Artificial Intelligence, in issue on
non-monotonic logic planned for 1979. Also in J. Cohen
and M. Hesse (eds.), Inductive Logic, Oxford University
Press, 1979 (expected publication date 1979).
In preparation:
[a] ... with Fernando Flores, Understanding cognition as
understanding, Ablex Press, (in preparation: expected
publication date 1979).
[b] "Understanding cognition" text for major address at La
Jolla Cognitive Science Conference, 1979. Papers will be
published in an issue of Cognitive Science and also as a
book.
[c] Language as a Cognitive Process, Addison Wesley, (in
preparation: expected publication date 1980)
-------
∂13-Feb-79 1548 Levinthal at SUMEX-AIM AIL Agreements
Date: 13 Feb 1979 1537-PST
From: Levinthal at SUMEX-AIM
Subject: AIL Agreements
To: tob at SU-AI
cc: feigenbaum, jmc at SU-AI, les at SU-AI, sl at SU-AI,
cc: liebes
Tom,
I am meeting with Ed Feigenbaum tomorrow morning concerning the
resolution of the outstanding issues we discussed.
The question of space can be made minimal, at least for the period
of our present commitments and outstanding proposal ( i.e. to 10/80 ).
If Sid accepts an appointment with CS, there is presumably suitable
office space for him. If we use students as research assistants,
as we plan to, they would have space assigned. Unless arrangements
are made to transfer my appointment to CS, I will retain the space
I have here at SUMC, at least for this period. The only space issue
thus remaining involves the equipment. It is my understanding that
this is unlikely to be a problem. Sid and Les should confirm this
re Stereo Station. You presumably have the space for the display
hardware.
It now seems that we are not proceeding with the previous apreement
re computer access. In view of this, I would like to limit our
understandings to cover just the JPL contract and the MDA proposal.
It is my hope and expectation that our relationship will be more
far-reaching and lasting. However, I believe it is to everyones
interests to deal with the future as it explicitly unfolds rather
than have vague understandings ( which can lead to misunderstandings )
about what I will or will not contribute to the central laboratory
computing facilities or maintenance, or what my needs will be.
Until more durable arrangements are made, I would like to retain
" ownership " of the stereo station, the Varian plotter, and the
display hardware. I would not like to have the $30,000 I will be
contributing spent on equipment other than the display hardware.
I have $ 1800 that can be spent as capital equipment on the plotter
interfacing. As long as this equipment is in the CS laboratories,
it can be used by CS personnel for suitable purposes. Our use will
be limited to carrying out the JPL contract and the MDA proposal.
This proposal has been reduced to a $41,065 effort for the period
4/79 to 10/79 and a $ 85,158 effort for the period 10/79 to 10/80.
It incorporates only three tasks, tasks 1,5, and 8 of the initial
proposal. Task 8 involves no facilitiy use, except perhaps some
text editing. We would, in addition, generate some small amount of
use in connection with developing new opportunities.
I would like your comments on the above. It is my hope that we can
reach a successful conclusion of these discusions in the next few days,
so that we can proceed with the purchase of the display hardware.
Looking foward to a response from you, les and John.
Elliott
-------
∂13-Feb-79 1839 TOB offer to research associate
John
I would like to make an offer to Barry Soroka,
now Assistant Prof at Univ of Kansas, as a
research associate to work in vision. He is
bright, energetic, works hard, and is interested
in research. I think that he would be make a valuable
contribution to our research.
Tom
∂13-Feb-79 1933 TOB space and Levinthal agreement
John
Sid Liebes and I have been talking about Sid
working halftime on stereo vision as a research associate.
He is obligated under a NASA contract to make maps from
Mars Viking stereo on a halftime basis.
Nothing has been agreed on yet about whether he will work here.
I think highly of him.
I mentioned this to you, and would like to talk a moment with
you to get your opinion. He knows how to use the system and
knows some SAIL. He does not know a lot of programming.
His value would be as a scientist. He has excellent analytic
and instrumentation skills.
Tom
∂13-Feb-79 1953 HVA Tapes
I've put most of Side A, Tape 2 in system tonight (there are still 5 pages
of draft left on Side A), and, of course, Tape 1 (both sides were put in
last week) is now behind me...with luck, I hope to have all in system by
end of week, depending on my strength (I am hungry now!), and state of system
at end of day. Should you wish to see the file, it's budget.v; if you log in
as 1/hva, the pass word is hmm.
Good night.
∂13-Feb-79 2208 Feigenbaum at SUMEX-AIM Re: Levinthal memo
Date: 13 Feb 1979 2208-PST
From: Feigenbaum at SUMEX-AIM
Subject: Re: Levinthal memo
To: JMC at SU-AI, LES at SU-AI, TOB at SU-AI
In response to the message sent 13 Feb 1979 1923-PST from JMC at SU-AI
John, I believe you are confusing the discussion of Elliott's
Adjunct-Professor-by-courtesy appointment that got discussed with
Sid Liebes' crossover to CSD work. The former got discussed, but no
action was taken yet. The latter was not discussed with me@(ever)
nor at a faculty meeting. It possibly represents some private initiative
of Tom's. I have to discuss all of this with Elliott on Wednesday morning,
and am totally confused.
In any event, Tom's robotics work (with or without Liebes/Levinthal) is one
of my candidate priority items for fourth floor space when (NOT if)
the musicians go.
Incidentally, speaking of curtailing project activity, I have notified
Djerassi that I am moving the chemists (i.e. the DENDRAL project) out to
live (somewhere; not yet determined) in his space; and I have similarly
exported the medical types to live in Ted Shortliffe's space. So the
claim that JMC reported to have heard, that I am trying to get rid of the
music people for self-space-aggrandisement-reasons is false.
(Also, although the SUMEX machines are just as much a part of CSD research
as the other machines, I have never made a pitch to house them and their
staff in the new building, though by symmetry with the KL-10 and its staff
they deserve to be there). The charge makes me angry. Also, I am beginning to
lose patience with the chairman's role of trying to handle and somehow
resolve the various conflicting interests.
Ed
-------
I hope you will take such comfort as possible from the fact that
the major space allocation fuss will end in finite time.
∂14-Feb-79 0921 TOB steering committee
To: feldman at SUMEX-AIM, JMC at SU-AI, ECL at SU-AI
How is Tuesday, Feb 20, 9am-11am for steering
committee meeting? Please acknowledge.
Tom
Tuesday is ok provided it doesn't run over. My class is at 11.
∂14-Feb-79 1033 PAT letters
Sorry, I will try to be more careful, I had thought they were pretty well covered.
∂14-Feb-79 1103 Bmoore at SRI-KL (Bob Moore)
Date: 14 Feb 1979 1104-PST
From: Bmoore at SRI-KL (Bob Moore)
To: jmc at SAIL
I got back from Boston over the weekend. If you would like to get together
let me know.
Bob
-------
I would like to discuss organizing the meeting associated with the workshop.
Could you phone me Thursday or Friday afternoon.
It would be nice if the command Xreply in E looked at the current page
to find whom the message was from and MAILed the attach buffer or the
page to that person. Fancier yet. XREPLY <rest of line> would find
the sender and transmit to him <rest of line>.
∂15-Feb-79 0831 EF at MIT-AI (Edward Fredkin)
Date: 15 FEB 1979 1131-EST
From: EF at MIT-AI (Edward Fredkin)
To: JMC at MIT-AI
Hi again, The Russians are proposing another
AI conference at Repino, with the following concessions:
The meeting is postponed until after IJCAI (it will start Sept 1 and
run until Sept 5).
As I understand it, the USSR will pay transportation costs from the far east
to Leningrad, and pick up the tab for a one week stop at Lake Biakal. This
comes reasonably close to your idea of what they should do, I think. I'll
be seeing you in LA on Wednesday, regards, Ed.
∂16-Feb-79 1912 HVA Progress...some...
Have now reached Side B of 3, (which has l5 dble.spcd.pages), and 4 has only
6 pages, so tomorrow it should all be in the system, though I have't taken
time to go back and correct typo's yet. Putting it in the system takes a bit
longer than I'd counted on because I have to try to punctuate-paragraphs, etc.-
so it isn't just a running stream. I have also paginated-in a loose sort of
way-and only spooling will tell what changes have to be made, after editing
(when you can fill in some of the blanks), and whether I've allowed too much
space per page-I thought it wise to allow plenty. Wish I'd had a chance to
talk to you about this before you left today, but guess it can wait till
Tuesday. In any case, the immediate goal is to get it in the system - in
moments of despair over not being able to finish it faster, I wonder if
you'd have been happier with a full-time typist! But, it was something I
wanted to do....It's tooo quiet and deserted here now and I am hungry....
Have a nice week-end and, I hope, a holiday....
∂17-Feb-79 0359 ME KA system
To: JMC
CC: LES
∂14-Feb-79 2127 JMC ka system
To: ME
CC: LES
The musicians may get the KA as a separate system with one disk and
controller, Petit channel, tapes and possibly data disk but not the
switch and 256k memory. Is there a system that can readily be
resuscitated to run on such a configuration.
ME - I wouldn't say "readily". The code to run the Petit channel and the
code to make the system work on a KA are no longer around, though they
could be restored from DART tapes and worked into a useable state.
Running the DD without the video switch would also require a bit of
reworking of the display service, essentially back to what it was like
before the switch (with each monitor viewing a fixed DD channel). Out of
curriosity, when might this happen and which 256K of memory are you
thinking of? It certainly wouldn't be impossible to do; it might take a
month. This assumes the hardware still works (the channel).
∂19-Feb-79 1235 TOB staff
John
I woud like your approval for an offer to
Barry Soroka, assistant prof at Univ of Kansas
who got his degree at Penn. Ruzena thinks
very highly of him. He seems bright, energetic,
a hard worker, and serious about research.
Tom
OK on Soroka.
∂19-Feb-79 1405 HAMMER at MIT-XX kudos
Date: 19 Feb 1979 1656-EST
From: HAMMER at MIT-XX
Subject: kudos
To: mccarthy at SU-AI
Cc: jmc at SU-AI
Congratulations on your letter in the Sunday Times.
You are to be complimented on both the content and the
style. You have exposed some of the means by which
unelected bureaucracies and self-appointed "public
interest" groups establish social policy, without
the trouble and embarassment of evaluating its effects;
if all alternatives have been proscribed, there is no
basis for comparison! I was also pleased to note
your pungent and direct style. All too often, those
who go against the currently fashionable pieties
are so apologetic and mealy-mouthed as to eviscerate
their message.
It is especially disturbing to note that Federal funding
often establishes a ratchet effect. It is very easy to
accept Federal funds, very difficult to learn to do without
them; what starts out as a luxury quickly becomes a necessity.
I am also glad that you pointed the finger at the
ideologues with hidden agendas, who seize on a legitimate
issue of public concern and do not seek to solve it, but
to use it as a stick to beat quite another horse. Thus,
many (though not all) of those opposed to nuclear power
are not really concerned with reactor safety; they are
really opposed to modern industrial society. Similarly,
those demanding lower speed limits (and many of those
in favor of "mass transit") are not really concerned
with improved transportation or the energy crisis; they
are seeking to impose their own lifestyle on others.
Again, my congratulations.
M Hammer
-------
∂19-Feb-79 1559 TOB remind
NSF steering committee meeting Tuesday, Feb 20,
9am - 11am in Polya 204
∂20-Feb-79 0215 REM via AMES-TIP New writeup by me on my data-compression
work
To: JMC, RWG
Yesterday morning I finally got the inspiration and energy to write an
article describing all my major results in data-compression since 1975,
up to but not including the internal workings of my invention of last
year. I hope to publish something like it sometime soon. At present
it has no references, no appendices containing detailed results of the
various algorithms, and hasn't been formatted by POX or TEX. Do you
know anyone who has time and ability to read it and offer suggestions
toward getting it published? Note that I have deliberately rushed
through the topics, briefly covering everything before losing energy or
entering typical verbose mode, in order to have something small enough
to be of interest to the reader rather than boring and gigantic. Thus
I have covered in this one paper what I had originally planned to
spend two papers on.
∂20-Feb-79 0856 Feigenbaum at SUMEX-AIM need for sail account for professor kedes
Date: 20 Feb 1979 0856-PST
From: Feigenbaum at SUMEX-AIM
Subject: need for sail account for professor kedes
To: jmc at SAIL
cc: kedes
John,
I would like to request a sail account for Professor Lawrence Kedes,
who is co-principal investigator of our MOLGEN project. He needs the
account to do XGP output.
We have a quid-pro-quo of sorts, in that SUMEX does
printing for SAIL on campus.
Thanks,
Ed
-------
OK for an account for Kedes to be used mainly for XGPing. I assume he can
do his PUB or TEX at SUMEX.
∂20-Feb-79 0858 Nilsson at SRI-KL (Nils Nilsson) Interview
Date: 20 Feb 1979 0901-PST
From: Nilsson at SRI-KL (Nils Nilsson)
Subject: Interview
To: JMC at SU-AI
cc: Nilsson
John, A Mr. Philip Spaulding (data processing engr. from IBM-UK)
has a project to make a film about the use of computers in the home.
They want to interview people concerned with this subject and have
heard about your interests (DIALNET, etc.). Can they talk to you?
Their time preference is the afternoon of Monday, Feb. 26. Second
choice: sometime Friday, March 2. They would come to the AI
lab. --Nils
-------
1:30 feb 26 would be ok for talking with ibmers
∂20-Feb-79 0902 Feigenbaum at SUMEX-AIM musicians/mj hall space
Date: 20 Feb 1979 0859-PST
From: Feigenbaum at SUMEX-AIM
Subject: musicians/mj hall space
To: jmc at SAIL
cc: les at SAIL
John,
I am eager to take the next step. I owe John Chowning a response to
the long shopping list of equipment he said he needed. Our response was
to be given in terms of a KA-10 system configuration, but you and Les were going
to think that one through and send something to me. I would appreciate
closure on that.
Ed
-------
What is Chowning's shopping list?
∂20-Feb-79 1054 DEW
My thesis is back in Patte's office if you want to look at it.
∂20-Feb-79 1134 PAT airline reservations
Your reservations are as follows:
Feb 21 San Jose to Los Angeles
PSA #162 depart 8:30am arrive 9:25am
Feb 21 Los Angeles to Santa Barbara
United #212 depart 4:10pm arrive 4:42pm
Feb 22 Santa Barbara to San Jose
Apollo #311 depart 6:45am arrive 8:00am
I made the reservations directly with the airlines. You should pay for the
tickets when you check in at the airport for each flight.
∂20-Feb-79 1547 RPG New STEP feature
To: "@USERS.DIS[AID,RPG]" at SU-AI
αβM, αM, and βM can be used to step around macro calls gracefully. If you type
either of these at a macro call, the STEPper will assume that it is a macro,
expand it, and present you with the expanded form without any stepping involved
in the expansion. If it isn't a macro, it restates the form.
-rpg-
∂20-Feb-79 2036 JMC
crawford
7:30 859 Cedro
∂20-Feb-79 2209 Feigenbaum at SUMEX-AIM (Response to message)
Date: 20 Feb 1979 2207-PST
From: Feigenbaum at SUMEX-AIM
Subject: (Response to message)
To: JMC at SU-AI, EAF at SU-AI
cc: LES at SU-AI
In response to the message sent 20 Feb 1979 2029-PST from JMC at SU-AI
Chowning's shopping list:
it's a rather lengthy list of terminals, cables of various sorts,
drivers, control devices, etc. I have it at the office (on paper;
John gave it to me over the long-distance phone). I'll type it
in tomorrow from the office.
Ed
-------
∂20-Feb-79 2209 Feigenbaum at SUMEX-AIM winograd papers
Date: 20 Feb 1979 2209-PST
From: Feigenbaum at SUMEX-AIM
Subject: winograd papers
To: jmc at SAIL
Any conclusions yet?
Ed
-------
Results so far are mixed. Need more time.
Richard Gabriel showed signs of having something substantial to say,
so I quizzed him and then asked him to write you.
∂20-Feb-79 2220 Feigenbaum at SUMEX-AIM 2060
Date: 20 Feb 1979 2215-PST
From: Feigenbaum at SUMEX-AIM
Subject: 2060
To: jmc at SAIL
It's up on the net. FTP is very flaky, and telnet is somewhat flaky.
The SCIP cable to SUMEX and our 2060 are not having a very happy marriage.
Ralph continues to chase the bugs. The modems MIGHT be here soon (this week).
32K of memory is not working, so we're running with a very small system at
the moment. Other than that, it's great to have a brand new machine!
I sent you a very interesting report of the MIT committee on the future
of computing at MIT (ten years). Stanford comes out pretty well in
their comparison charts , particularly LOTS vs what-MIT-doesn't-have
(a story that I would subtitle "the beneficial effects of having
McCarthy around").
Ed
-------
∂20-Feb-79 2230 Feigenbaum at SUMEX-AIM (Response to message)
Date: 20 Feb 1979 2220-PST
From: Feigenbaum at SUMEX-AIM
Subject: (Response to message)
To: JMC at SU-AI
In response to your message sent 20 Feb 1979 2214-PST
Gabriel hasn't written to me. I would be interested.
The big fat green form is due at the Dean's office Friday (after a two
week delay/extension). Current vote:
Yes: (N-1) No: 0 Still processing: 1
Ed
-------
∂20-Feb-79 2349 ME Xreply in E
∂20-Feb-79 2017 JMC XREPLY in E
It would be nice if the command Xreply in E looked at the current page
to find whom the message was from and MAILed the attach buffer or the
page to that person. Fancier yet. XREPLY <rest of line> would find
the sender and transmit to him <rest of line>.
ME - Actually, this is sort of on my list of possible things to do
in E, but there is a question about what to do if the original
message was sent to more than one person (as indicated by a TO: line
and/or a CC: line). To reply to all the original receivers requires
E to be able to parse both the network and local MAIL headers.
By the way, if you want to MAIL or SEND only the <rest of line>, you can
use a zero argument in a ⊗0⊗XMAIL or ⊗0⊗XSEND command, as in
⊗0⊗XSEND ME Hey it really works.<cr>
With a zero arg, no text from the page (nor from the attach buffer) is
sent -- only from the command line.
∂21-Feb-79 0838 CET Administrative Vacancy
To: "@FAC.DIS[1,DPB]" at SU-AI
We have received a notice from CA State Univ.,Sacramento
If anyone has interest, please sndmsg to CET
POSITION: Director of Computer Center and Institutional Research
EFFECTIVE DATE: Approx. June 1, 1979
QUALIFICATIONS: Experience of at least 2 years in a responsible administrative
position with a data production and/or computer center.
Univ. or college admin. exp. considered desirable. Teaching
experience in a 4 yr. inst. preferred. Doctorate or equiv.
SALARY $27,708-$33,492 in a five-step range. (12 months service with
24 days-per-year vacation.
JOB DEF. Under the general direction of the acad. vp, charged with
the overall planning, budgeting and operation of the
data processing installations, expected to implement a program
of institutional studies and academic research and complete
other tasks as assigned.
etc.
∂21-Feb-79 0941 Nilsson at SRI-KL (Nils Nilsson) Richard Weyhrauch's Course
Date: 21 Feb 1979 0941-PST
From: Nilsson at SRI-KL (Nils Nilsson)
Subject: Richard Weyhrauch's Course
To: JMC at SU-AI, Feigenbaum at SUMEX-AIM
cc: Nilsson
I've been attending Richard's course on FOL and related topics. I think
he is doing an outstanding job. He's presenting his ideas clearly and
persuasively. It looks to me like he has some of the best stuff going
in AI these days. --Nils
-------
∂21-Feb-79 1004 CVW Numbers and Course Evaluation Committee
To: "@FAC.DIS[CEC,CSD]" at SU-AI
To: CSD Tenure Committee
From: CSD Course Evaluation Committee
Subject: Do we really need all those numbers?
The Course Evaluation Committee collects students' comments on all
courses taught by the department each quarter. We gather both numeric
ratings (e.g., Give the course and overall rating (1 low; 9 high).) and
written comments about instructor and t.a. performance, reasons for taking
the course, and so on.
Most students in the department tell us that the written comment
summaries we provide are much more helpful to them than the numeric summaries,
despite the fact that the latter take up far more than half of each course
evaluation booklet. We also note that for many of the courses, the number
of responses we receive makes the validity of the average ratings questionable.
We believe that we could do a better job by not collecting numbers
at all, instead concentrating on collecting and summarizing written
comments. But since the course evaluation booklets are consulted during
tenure deliberations, we would like to know how this would affect you.
In particular, to what extent do the numeric ratings of courses and instructors
affect any decisions?
Send your comments to either or both of us (BENNETT@SUMEX, CVW@SUAI).
We're also ready to discuss the issues with you if you have the time and are
interested.
∂21-Feb-79 1055 CET Mathematical Sciences Sherry Party
To: "@FAC.DIS[1,DPB]" at SU-AI
The Computer Science Department Faculty is cordially invited to attend
a Sherry Hour, Friday, Feb. 23, 4 p.m., Math Dept. Lounge, 2nd floor
Math Building (Quad).
Sponsored by the Mathematical Sciences Undergraduate Association
∂21-Feb-79 1326 CET
To: "@FAC.DIS[1,DPB]" at SU-AI
∂21-Feb-79 1216 Buchanan at SUMEX-AIM pls distribute to CSD faculty/staff/students
Date: 21 Feb 1979 1216-PST
From: Buchanan at SUMEX-AIM
Subject: pls distribute to CSD faculty/staff/students
To: cet at SU-AI
The CSD library will need to close for about 5 days to move into the
Mathematical Sciences Library. The best time appears to be between
Winter and Spring Quarters. Please plan ahead so you can check out
books you will need between quarters before service is disrupted.
Apologies for the inconvenience.
B.Buchanan
Chm, Library & Publ. Comm.
-------
∂21-Feb-79 1448 PB Noticing your letter on the b-board, I thought you
might be interested to know, if you didn't already, that according to an
article in Science appearing about a year ago, considerable doubt is to be
placed on the premise that the 55 mph speed limit saves lives. The usual
statistic cited is that the death rate on highways went down when the
speed limit was instituted. The authors note that there was also a
corresponding decrease in the statistics in cities, where there was no
change of speed limit.
Thanks. I had heard something about it, and had it occurred to me
I might have tried to add something about the tendency of the government
to reiterate statistics known to be dubious, when they support positions
the government has decided to take.
∂21-Feb-79 1734 HVA Tapes
Tapes are all in system and proofed for "typos" (tho' I may have missed some)
Seeking Patte's assistance, we discovered that text can't be pubbed yet until
all blank spaces are filled (Pub will obliterate them and I hadn't thought
abou that). Also, there are some ... at left hand margins which "confused"
Pub, so they may have to be changed eventually.
∂22-Feb-79 0946 CET SPRING TIME SCHEDULE
To: "@FAC.DIS[1,DPB]" at SU-AI
Spring Time Schedules will be sent to you today. Please check your
class entry. Also, please check out your assigned classroom to make
sure that it is satisfactory. Let me know if you want any changes.
Thanks, Carolyn
∂23-Feb-79 1334 Mcgoveran at SUMEX-AIM Faculty Meeting
Date: 23 Feb 1979 1334-PST
From: Mcgoveran at SUMEX-AIM
Subject: Faculty Meeting
To: faculty.list:
Ed Feigenbaum would like to call a faculty meeting for 4pm on Friday, March 2.
There are a number of items to discuss, and I shall be sending a msg with
the agenda on Monday or so.
Please let me know whether you will be able to make it to this meeting.
The meeting place will be Polya 204.
Mary McGoveran
-------
No problem for me.
∂23-Feb-79 1729 Bmoore at SRI-KL (Bob Moore)
Date: 23 Feb 1979 1732-PST
From: Bmoore at SRI-KL (Bob Moore)
To: jmc at SAIL
I tried to call you a couple of times. Why don't you send me a message
with a convenient time for us to get together. Between 4:00 and 5:00 any
weekday afternnon would be the best time for me. That way I could go from
SRI to SAIL and then go pick up my family without going back to SRI.
Bob
-------
Fine, how about 4pm or a bit later on Monday the 26th. I want to discuss
organizing the auxiliary meeting of the AI and philosophy working group.
∂25-Feb-79 0134 Kanerva at SUMEX-AIM Editing vs. compiling
Date: 25 Feb 1979 0134-PST
From: Kanerva at SUMEX-AIM
Subject: Editing vs. compiling
To: McCarthy at SU-AI
I would like to explain to you briefly what I am trying to do about
the editor and what I am not trying to do. There are a few people of
whom it matters to me that they understand. Many seem unable to.
I am trying to let a technical typist or the author of technical
text compose on the screen--without a formatting program--what he
or she can compose on a typewriter with removable type elements.
(I would like to be able to write the final version of my dissertation
with TVEDIT and without PUB.)
Our editing programs lack underline, overstrike, half line space,
and special symbols. These need to be represented in the file and
indicated on the screen. I have put a lot of thought into the file
representation and am quite satisfied with what I now have.
The Datamedia, and the Hazeltine with the new microcode, is
adequate. Half-line spacing can be indicated by a special symbol
at the margins of the screen. Bright characters can be used for
symbols that the display does not have (lowercase alpha, as well as
underlined "a", will be displayed as bright "a"). What matters is
that horizontal spacing be honest, so that the entries in tables and
expressions line up the way they do in the final paper copy. This
is akin to editing an upper- and lowercase file on an uppercase-only
display, which we do on our TECs.
As for fancy displays, there are enough people worrying about
them. I would rather have an inexpensive display and a good keyboard,
in volume. Ralph can testify to that from our two meetings with Forest
Baskett (they made me feel cheap for not wanting a bit-map display,
much less an "intelligent" terminal). When more characters, underline,
reverse video, and half line space are available, I will use them,
but I would still stay away from proportionally spaced characters
and multiple type sizes, because of what I call the grid postulate:
The difficulty of programming and using an interactive text
editor is directly proportional to the density of the underlying
grid (the number of potential cursor positions per square em).
Independent half lines are a mixed blessing. I chose them for
ease of editing. They eliminate invisible underlying structures and
hence make the visual effects of an editing command predictable. In
this sense independent half lines provide a general solution to the
graphic manipulation of patterns build of characters. But they make
life difficult for compilers that might want to reformat a document.
Here is my excuse: We already bracket difficult format (tables and
displayed formulas) with STOPTEXT and STARTTEXT for PUB, only we
are not always able to express conveniently in the editor what goes
between those brackets.
Being able display and edit formatted material would make it
possible to pass around and work on a technical manuscript without
hard copy. This I take to be very valuable, and the alternative--to
compile the source in order find out what is in it--cumbersome. And
then there is the fact that typewriter style is sufficient for most
of what we write.
We would then need programs that interpret displayed expressions.
I am interested in the conversion into the TEX input language of a
formatted manuscript that contains no information about format beyond
the location of the characters. The question to ask is: What can be
automated and what is best left to be done by hand?
Whether this is gold plated I don't know. I have always thought
of my programs as simple and practical, stainless steel from surface
to the core. In time we will find out.
- Pentti Kanerva
-------
What you are doing may be reasonable for printing your thesis or
as a research project. However, it seems that the world is taking
what you call the compiling option. In particular, better and
better printers are becoming available, and better and better
document compilers are becoming available for using them.
∂26-Feb-79 0240 Feigenbaum at SUMEX-AIM winograd case
Date: 26 Feb 1979 0240-PST
From: Feigenbaum at SUMEX-AIM
Subject: winograd case
To: jmc at SAIL
cc: feigenbaum
John,
The winograd case will have to leave my office by about noon today
(Monday) for inclusion of the Linguistics Department material and arrival
at the Dean's office by 3pm. Do you wish to record your vote, or do you
wish to have it recorded as an abstention?
Ed
-------
I have been reading, and it hasn't led to a definite conclusion. I guess
the vote should be recorded as an abstention for now, but I will continue.
∂26-Feb-79 0243 Feigenbaum at SUMEX-AIM brahms requiem Friday
Date: 26 Feb 1979 0243-PST
From: Feigenbaum at SUMEX-AIM
Subject: brahms requiem Friday
To: jmc at SAIL
John,
I am assumingthat you and Carolyn do want to go to the Brahms performance
after dinner on Friday evening (the whole group of us having dinner will
be going). Is that assumption correct? Let me know since I have to get
tickets for you and Carolyn.
Best,
Ed
-------
Yes, thank you very much. Sorry for being slow. - John
∂26-Feb-79 1050 Bmoore at SRI-KL (Bob Moore) meeting
Date: 26 Feb 1979 0957-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: meeting
To: jmc at SAIL
4:00 PM this afternoon will be fine.
Bob
-------
∂26-Feb-79 1058 EF at MIT-AI (Edward Fredkin)
Date: 26 FEB 1979 1358-EST
From: EF at MIT-AI (Edward Fredkin)
To: JMC at MIT-AI
John, could you give me a pointer towards
someone who know about rotary vane actuators, similar to the type
that was used on that early, powerful hydraulic Stanford arm?
In this case we are interested in pneumatic versions, but
any information you can provide would be helpful.
Regards, Ed
Vic Scheinman would know if anyone. His phone is (415)965-0557,and
his address is Unimation, 154 E. Dana Street, Mountain View, CA. He
built the hydraulic arm as well as our subsequent arms.
∂26-Feb-79 1135 DPB suggestions for instructors needed.
To: "@FAC.DIS[1,DPB]" at SU-AI
Instructors needed for spring quarter in CS101, CS105-1, and CS105-2.
Contact Denny Brown (DPB@SAIL) if you are interested or know someone
who would be a good candidate.
∂26-Feb-79 1256 DPB AI curriculum
To: "@AI.DIS[1,DPB]" at SU-AI
Some of you have indicated that you think that there should be some
modifications made to the AI curriculum. Ed has put me in charge of
getting a meeting together. He will be out of town during the
week of March 5, so it would be best to meet before the fifth.
That means this week.
How does Thursday afternooon, say 2:30-3:30 sound? (Since this
method of determining an appropriate meeting time never works
in one pass, I'll let Carolyn sort out the responses.) If Thursday
afternoon is bad, reply quickly and suggest better times/give
Carolyn your openings. Call her at 7-2273, or msg her as
TAJNAI@SUMEX or CET@SAIL.
-Denny
Thursday 2:30 ok for AI course meeting
∂26-Feb-79 1257 DPB Gray Tuesday Meeting
To: "@FAC.DIS[1,DPB]" at SU-AI
We are due to meet soon to discuss the academic progress of our
PhD students. During the next week or so, I will try to bombard
you with information about each of your advisees in order to
prepare everyone for the meeting.
We have two reasonable dates for the meeting: Tuesday, March 6
and Tuesday, March 20. (March 13 is not available. The admissions
committee has their final PhD admissions meeting that afternoon.)
Please mark your calendars with BOTH dates and send me your
preferred date. I will choose between the two dates on Thursday,
March 1.
Known disadvantages:
March 6 Ed Feigenbaum will be out of town.
There is not much time to generate and digest
advisee information.
March 20 This is during finals week. Faculty membes may be
planning to be away.
I prefer March 20.
-Denny
∂26-Feb-79 1323 TOB soroka
John
Do you have any suggestions about the following?
I am planning to circulate Soroka's vita for
departmental confirmation.
Are there any protocols to be observed?
Tom
No protocols.
∂26-Feb-79 1354 PAT scheinman's address
is in ADDRESS[P,DOC] (corrected).
∂26-Feb-79 1338 TOB video projection unit
To: LES, JMC
I would like to raise the possibility of getting
a video projection unit for group presentation
of computer response. We have in the past used
chained Data Disk terminals, but I think the
video projection might be effective for presentations
and warrants investigation.
Tom
Unless the source of money is one unusable for other purposes, I think
we have more urgent requirements - new xgp, new display system, modern
tapes.
∂27-Feb-79 0308 RWG
who supplied the headline to your 55mph Times letter?
The Times did. Why?
∂27-Feb-79 0932 Levinthal at SUMEX-AIM ai-lab agreements
Date: 27 Feb 1979 0931-PST
From: Levinthal at SUMEX-AIM
Subject: ai-lab agreements
To: jmc at SU-AI
cc: tob at SU-AI, les at SU-AI
John,
I would be glad to met with you to resolve any concerns you have
about the arrangements I am suggesting for my use of the AI-LAb
facilities. In any case I need a response from you. When can I expect
one.
Elliott
-------
We do have to meet, so please phone when convenient.
∂27-Feb-79 1101 Bmoore at SRI-KL (Bob Moore) Aaron Sloman
Date: 27 Feb 1979 1103-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Aaron Sloman
To: jmc at SAIL
There is a paper by Sloman in the TINLAP-1 proceedings titled
"Afterthoughts on Analogical Representations" which appears to
be an updating of the 1971 paper that you referred to. If you
don't have easy access to this paper and are interested, I would
be happy to send you a copy.
Bob
-------
Yes, please send it. The library will send me a copy of his
Computer Revolution book.
∂27-Feb-79 1546 Tajnai at SUMEX-AIM Winter Quarter Final Exams
Date: 27 Feb 1979 1546-PST
From: Tajnai at SUMEX-AIM
Subject: Winter Quarter Final Exams
To: JHL at SAIL, Winograd at PARC, JMC at SAIL, Wiederhold,
To: DEK at SAIL, JGH at SAIL, DPB at SAIL, Lenat
Room scheduling needs the following information.
1. Are you planning NOT to give a final examination winter quarter?
2. Do you require special times or classrooms for your final exam?
Please respond as soon as possible.
Thanks, Carolyn
-------
There will not be a final in CS258.
∂27-Feb-79 1727 Bmoore at SRI-KL (Bob Moore) AI/Phil workshop
Date: 27 Feb 1979 1729-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: AI/Phil workshop
To: jmc at SAIL
When I called Gardner Lindzey this afternoon about the AI/Phil workshop,
he told me that he had spoken to you on another matter earlier
today, and that you had raised the idea at that time. What he told
me (and, I presume, you) was that it was an entirely appropriate activity
and that the Center did have meeting rooms of an adequate size for the
kind of meeting we had thought about. (I estimated that there might be
30-40 participants.) I raised the question of funding, and he said that
the Center would be happy to cooperate in soliciting outside funds, but
that their own funds were "limited". I didn't press him for an explanation
of what that might mean. He is going to send me the current mailing
addresses of the other study group members (I understand that Dennet is
at Oxford this year), so I suppose our next step is to start work on
the letter we are going to send them. I will get to that shortly and
send you a draft for you to go over.
Bob
-------
∂27-Feb-79 1733 Mcgoveran at SUMEX-AIM Agenda
Date: 27 Feb 1979 1733-PST
From: Mcgoveran at SUMEX-AIM
Subject: Agenda
To: faculty.list:
Agenda for March 2 Faculty Meeting
1. Announcements
2. Relations between Computer Science and Computer Music.
3. Visitors. How to allocate our scarce "slots" for visitors.
4. Handouts. How to control costs involved with course handouts.
5. Graduate Study Issues:
a. Students who enter with a Masters degree.
b. Gene Golub's suggestion re courses in other depts.
c. Gray Tuesday announcement.
6. Courses and Degrees / Curriculum
This agenda is long, but we will cover as much of the material as
possible and adjourn by 5:30.
-------
I don't have the time March 2 meeting begins.
∂27-Feb-79 2157 RWG
i just wondered if papers took suggestions on headlines for
letter/editorials.
∂27-Feb-79 2157 REM via AMES-TIP Data-compressed machine language
To: RWG at SU-AI, JMC at SU-AI, WILCOX at SUMEX-AIM
CC: REM at SU-AI
While thinking about an emulated machine-independent language I plan to
maybe implement using bit-aligned data-compressed opcodes and addresses,
I realized something useful: structured (go-to-less) programming, where
all control jumps are nested, is the only type of language that can be
data-compressed using variable-length codes without padding.
With non-nested programming, after computing the offset for a relative
jump, and encoding it using a variable-length code, that code-length
may contradict assumptions about the amount of offset in the jump address,
causing the length of the code to change, etc. resulting in a situation
where a fixed-point isn't guaranteed.
With nested programming, inner loops can be compiled first, thus fixing
once and for all their length in bits (compressed), then non-jumping code
around them can be compiled, then length added to get the offset for jumps
in the next-outer loop, fixing its length in bits, etc. outward until
the entire procedure is compiled.
This may be useful in micro-coded MAINSAIL as well as in my personal
proposed project. Note that in order to compile inside-toward-outer,
it isn't necessary to keep the entire left-context (procedure body,
parsed but not yet compiled) on a stack in-core, nor to thrash with
random-access disk archiving of stack. Buffered I/O of stacks are
much easier to implement than queues or totally-general files, writing
a buffer out to disk only when full, and reading back in only when the
in-core buffer has completely been unstacked. With two buffers, thrashing
at buffer boundaries is eliminated, and thus buffer I/O to/from disk
only need occur when the stack pointer changes by at least one full
buffer size in the "wrong" direction or by two full buffer sizes in
the "right" direction (correction, buffer I/O wait for input or io-done;
after one buffer shift in "right" direction, write can be overlapped
with filling the other buffer).
∂27-Feb-79 2234 Wilcox at SUMEX-AIM data-compressed machine language
Date: 27 Feb 1979 2234-PST
From: Wilcox at SUMEX-AIM
Subject: data-compressed machine language
To: rem at SU-AI
cc: rwg at SU-AI, jmc at SU-AI
One of my representations for MAINSAIL indeed mirrors quite closely the
source code (it is essentially reverse polish), and benefits from the
fact, as you point out, that there are no branches (since MAINSAIL has
no goto), and thus branch resolution proceeds with no problem (this
representation is bit aligned). In fact, there are no branches at all,
but only sizes of "code bodies" which follow. For example, the If
statement and the Iterative statement simply involve specification of
how much code to skip over or iterate, and thus the idea of "branches"
is not used at all. Note how this means the processor knows exactly how
many bits to bring into fast memory in the case of the Iterative
statement, assuming there is a fast memory compared to main slow memory
which normaly holds the program.
Might also point out that this representation uses, for a local variable
reference, only as many bits as are necessary, in the current procedure,
to reference all locals. E.g. if there are 8 locals, then just 3 bits
per local variable reference. Hufman coding might improve on this, but
analysis shows that most procedures have so few locals that 4 or fewer
bits almost always suffice anyway.
With regard to determing branch displacement in "branch-oriented"
representations, you should read the article "Assembling Code for Machines
with Span-Dependent Instructions", by Thomas G. Szymanski, in CACM, Vol. 21,
No. 4, April 1978, pp. 300-308.
I would be interested in hearing of any further thoughts you have in the
area of data-compressed language representations, since that is my current
dissertation research.
-------
∂27-Feb-79 2246 DPB Undergraduate advising
To: "@FAC.DIS[1,DPB]" at SU-AI
Below is a solicitation from the Deans of H&S and Undergraduate Studies.
They want us to provide two CS faculty members to serve as undergraduate
advisors next year. For the past two years, I have been a freshman
advisor. I have found it much less hassle than I anticipated. At times
it is even interesting. I have successfully refused to draft any faculty
members for this task in the past. My position has been that I will
only request volunteers. I prefer being in trouble with the Dean to
saddling some poor freshmen with an unwilling advisor.
If any of you are interested, however, feel free to contact me for
details as to duties and benefits.
-Denny
Date: January 31,1979
To: Edward Feigenbaum, Chairman
Department of Computer Science
From: Halsey Royden, Dean Alexander L. Fetter, Associate Dean
Humanities and Sciences Undergraduate Studies
Subject: GENERAL ADVISING
Discussion in various contexts last year confirmed that under-
graduate advising is a faculty responsibility. In this regard, the
Schools of Earth Sciences, Engineering, and Humanities and Sciences
have agreed to assume eah year the responsibility for providing a
considerable fraction of the 250 general advisors for undecided
majors. You may recall that we pursued this matter in a letter of
January 25, 1978, and we now do so again, seeking to enlist general
advisors for the academic year 1979-80. After considerable analysis,
the Office of the Dean of Undergraduate Studies has requested the
following assignments for 1979-80:
Earth Sciences 6
Engineering 26
Humanities & Sciences 80
Given the distribution of majors and student interest among
freshman, we therefore ask you and your department to designate
2 faculty members as general advisors for 1979-80 and send
the relevant list by March 1, 1979, to the H & S office. We trust
that your recruitment procedure will take account of faculty interest
and enthusiasm for this very important contribution to undergraduate
education at Stanford. These people will typically become freshman
advisors, which ideally entails a two-year commitment until the
students declare a major. We do not, however, want to discourage
willing faculty from serving for a single year, and those with other
plans for 1980-81 are certainly welcome. Either they will be made
advisors for sophomores or transfer students, or they will be fresh-
man advisors with their advisees reassigned for autumn 1980. If
someone prefers to advise transfer students or reassigned sophomores,
please make sure that we are so informed.
It is important to recognize that this assignment is a depart-
mental responsibility. You or your representative should verify
with each nominee his or her willingness to serve for at least one
full year. In addition, if any of your candidates for whatever
reason is unable to fulfill the commitment, the department is
obligated to find a replacement. For your convenience, a list of
your department's current general advisors is attached with those
marked "N" having only advisees retained from 1977-78.
This form of University service is a normal part of faculty
duties, and we urge you to take account of such obligations in
deciding committee assignments. It may even be helpful to list
general advising on your department roster.
We look forward to hearing from you by March 1, 1979. The
specific arrangements will be coordinated by Stephanie Palmer
(7-0968) in my office and by Tania Granoff of the Academic Infor-
mation Center. We are most grateful for your willingness to assist
in arranging for this central function of undergraduate education.
Attachment
∂28-Feb-79 0827 BS Creary
To: JMC
CC: LES, HVA
Ed has approved your salary recommendation for Lew Creary, and it has been for-
warded to the Dean's Office.
Betty
∂28-Feb-79 0853 Mcgoveran at SUMEX-AIM faculty meeting
Date: 28 Feb 1979 0853-PST
From: Mcgoveran at SUMEX-AIM
Subject: faculty meeting
To: JMC at SAIL
The faculty meeting on March 2 will be at 4pm in Polya 204.
-------
∂28-Feb-79 1212 DEK Chinese visitor
This Mrs. Lee I just spoke to showed me a picture of her brother, and
I recognize him as the person who impressed me the most by his
intelligent questions and apparent wide knowledge of the literature.
In my opinion he and Hsu (director of the software part of the Inst.)
were the best scientists I met on that trip.
I also had an excellent impression of Tang. I will draft a letter to
him asking him what are his qualifications and what would he like to
do. It seems to me that we should try to get him as a visitor, but it
may not be possible right away.
∂28-Feb-79 1450 DEW my oral committee
My oral committee will be composed of yourself, Bruce Buchanan, Doug Lenat,
Hans Berliner, and possibly Nils Nilsson. Letting you know so that you
can voice any objection you might have. Thanks, Dave
No objections.
∂28-Feb-79 1646 Bmoore at SRI-KL (Bob Moore) Help us celebrate!
Date: 28 Feb 1979 1625-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Help us celebrate!
To: hart, SACERDOTI, NILSSON, JROBINSON, ROBINSON, GROSZ,
To: hobbs, hendrix, SAGALOWICZ, haas, WALDINGER, konolige,
To: tenenbaum, gaschnig, RAPHAEL, jmc at SAIL, rww at SAIL
cc: bmoore
You are cordially invited to our house on Saturday, March
10, at 8:00 PM to help us celebrate the completion of my
doctoral thesis, which was turned in and accepted on
February 9, 1979. Generous quantities of wine, cheese, and
other goodies will be provided. The address is 737 DeSoto
Dr., Palo Alto, which is just north of Channing, one block
east of Newell. Please send me a message if you can
make it. Join the party!
Bob
-------
Many thanks, but I shall be in Massachusetts on the 10th and 11th.
∂28-Feb-79 1647 PAT
∂28-Feb-79 0058 JMC photos
I chose a photo tuesday. Please call about them, pick up one and airmail
special it.
-----
They called and won't have it (them) ready til about noon tomorrow. They
will call me. Who is the second one for?
Both are for Idaho.
∂28-Feb-79 1733 FB dovers,altos,ethernets,grinnells,networks -- news
To: EAF at SU-AI, JMC at SU-AI, LES at SU-AI, JLH at SU-AI,
admin.gorin at SU-SCORE
I have just returned from a visit to MIT and DEC where Gorin was present.
We picked up information about networks and Grinnells. I am preparing a
short report. My understanding of dovers, altos, and ethernets is that we
are getting the no longer current equipment. For example, the dover is based
on a copier that will not be in the product line. Likewise for altos and
the ethernet. Thus I suspect that most of the surplus that existed has be
been committed and that we can expect to have to pay for additional "dover-
like" Xerox printers. I had understood that a file server and a gateway
were to be part of the package. Is that so?
∂28-Feb-79 2112 REF
John --
The paper
Circumscription Induction - A Way of Jumping to Conclusions
Has it appeared somewhere, or is it going to? Or is it just "unpublished memo"?
Thanks.
bob
reference it as - submitted to Artificial Intelligence
∂01-Mar-79 0949 Mcgoveran at SUMEX-AIM Visitors from Tsing Hwa University
Date: 1 Mar 1979 0949-PST
From: Mcgoveran at SUMEX-AIM
Subject: Visitors from Tsing Hwa University
To: JMC at SAIL, DPB at SAIL
John Linvill called yesterday. He said that a group of visitors from Tsing Hwa
University would like to visit the Computer Science Dept. and the AI Lab on the
12-13 of March. Would you be able to host them?
-------
Yes, depending on the precise time.
∂01-Mar-79 0954 DPB Re: Visitors from Tsing Hwa
To: mcgoveran at SUMEX-AIM
CC: JMC at SU-AI
∂01-Mar-79 0949 Mcgoveran at SUMEX-AIM Visitors from Tsing Hwa University
Date: 1 Mar 1979 0949-PST
From: Mcgoveran at SUMEX-AIM
Subject: Visitors from Tsing Hwa University
To: JMC at SAIL, DPB at SAIL
John Linvill called yesterday. He said that a group of visitors from Tsing Hwa
University would like to visit the Computer Science Dept. and the AI Lab on the
12-13 of March. Would you be able to host them?
-------
[13 March will be full. I will be at the final admissions committee meeting
most of the day. I would be willing to talk to them sometime on the 12th,
but since I teach that afternoon, I wouldn't be in a good position to "host"
them, except for a little while. -Denny]
∂01-Mar-79 1130 PAT
∂28-Feb-79 1658 JMC
please pub and decorate gardne.le1 and include aiphil.pro.
-----
John - for your info, you named it lindze.le1, not gardne.le1., patte
∂01-Mar-79 1135 PAT
∂26-Feb-79 1801 JMC
Patte: Please fill in the references in the last three pages of RENEW.PRO[W79,jmc].
-------
done, yesterday
∂01-Mar-79 1157 PAT expenses
A check to Allways travel should be sent to cover the trips to III
for which you have been reimbursed (Boston and LA). The total
is $540.00. This should bring eveything up to date. Patte
∂01-Mar-79 1225 PAT copies of first
the reference page of FIRST says copies to Albert Meyer and Dana Scott.
I have an address for Scott but I don't know where Meyer is. Also, do
you want copies sent to anyone else? I xeroxed 10 copies and the remainder
are on your reprints shelf. Patte
Albert Meyer is at the Labortory for Computer Science, M.I.T.
∂01-Mar-79 1322 ME
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging,
** ESPECIALLY if there are less than 1000 free disk blocks in the system! **
If you don't delete some of your files, the purger will.
Your disk quota is: 2000
Your files occupy 2435
∂01-Mar-79 1357 Kanerva at SUMEX-AIM Editing and compiling
Date: 1 Mar 1979 1357-PST
From: Kanerva at SUMEX-AIM
Subject: Editing and compiling
To: McCarthy at SU-AI
It is true that better printers and better compilers for them are
becoming available. I really welcome a large set of symbols. At
the same time, display terminals are becoming more and more common,
together with better display editors, and with them the possibility
of working on text with infrequent paper copy.
The world often goes the wrong way and is slow to turn. You know
that about the world of computing better than anybody else I know.
The prevelance of the compiler approach in that world could be due
to inertia. Until recently nothing better than punched cards and
line-oriented editors were generally available, and it is not easy to
maintain a formatted source file with them. So there has developed
a tradition for spelling out format in minute detail, the result of
which is a cluttered source file. That things could now be done
differently--that input format could be used to describe output
format--has not necessarily occurred to people, just as the merits
of display editors don't necessarily occur to people who have grown
up on line editors.
Converting a typewriter-style document into TEX input language or
some other compiled language should indeed be a topic for a research
project. I am approaching it in that spirit.
-------
∂01-Mar-79 1446 HVA DIALNET
CC: JMC, HVA
I tried to reach Weingarten this morning, but he is out of town until
Monday, March 5th, so will pursue next week. Actually, you have money
left in this account--enough to last until 6/30/79 at current rate of
spending--so the wolf is not crying at the door yet.
I am puzzled about having money in the account. It seemed to me we
have been paying Mark Crispin from it, and we started soon enough to
have used it up. Do we get an automatic extension on the spending the
money? Are you taking into account that Mark is half time at CSD now?
∂01-Mar-79 1627 TOB visit
John
I had a visit from Bill Smith, Mostek. He is interested
in product development based on large memories, particularly
in graphics. He indicated that he may want to make some
arrangement, as yet undefined.
Tom
∂01-Mar-79 1702 HVA Ignacio Zabala
CC: JMC, HVA
Betty Scott says he has no local support, only support from Spanish Gov't,
but she will check further and let me know Friday if there is any reason
why he cannot receive some Stanford support.
∂01-Mar-79 1903 FB Elephant
Your language sounds interesting for a number of reasons. If you have
written anything or write anything, please let me know. I will look up
my Lucid references.
∂02-Mar-79 0315 REM
∂25-JUL-76 1440 JMC
Usually the opinion that something is impossible has not been
actually stated, it is simply a boundary condition of people's
thinking. In my opinion, people and societies often die
because they haven't troubled to think about what might save
them. One also needs to understand the known impossibility
results like Godel's theorem, non-constructibility, irrationality,
and the first and second laws of thermodynamics. Trying to make
a real proof that something is impossible often suggests
a way of doing it.
In my Technology Essays there are a number of such things in
partial states of completion.
[REM - Any particular of your essays which at this time you feel
best provide el neato solutions to "impossible" problems or which
establish non-impossibility of survival.... i.e. any particular
essays you are most proud of and would like me to read, or is it
ok to brouse [ESS,JMC] and read all the random files there as
time permits?]
It is ok to browse through my unprotected directories. Not all files
with extension ESS are in ESS,JMC; some are in my quarterly files.
∂02-Mar-79 1010 PRATT at MIT-AI (Vaughan Pratt)
Date: 2 MAR 1979 1303-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: MODAL at MIT-AI
PRATT@MIT-AI 03/01/79 11:26:03
To: MODAL at MIT-AI
Modal logic is receiving attention from computer scientists these days as a
possible source of representational, axiomatic and algorithmic methods for
logics of programs and logics of knowledge. I've set up a mailing list on the
file AI:PRATT;MODAL LIST on MIT-AI, of people who might be interested in
net-mediated discussion of such applications. Right now this message should
reach Leslie Lamport, John McCarthy, Bob Moore, Sue Owicki, and myself. Let me
know if you want out, or if you know of other potential participants. To
address mail to whoever is currently on the list, send it to MODAL at MIT-AI.
Our mailer will then distribute it appropriately.
A possible topic to start the ball rolling is the question, why go to modal
logic when anything you can say in it can be translated into first order logic
in a fairly straightforward way by permitting quantification over situations?
My own answer to this would be, because quantificational calculus is
undecidable while some modal logics are decidable enough to make them useful
for verification (proof checking) purposes, if not for autonomous processing.
The answer about some modal logics being decidable isn't good enough. There
are many decidable theories in predicate calculus languages, and an AI system
that uses predicate calculus at all should be able to use them. The pure
theory of equality, the theory of list structures, the elementary theory
of real numbers, and monadic predicate calculus are useful examples.
The translation of a decidable modal theory into predicate calculus
would be decidable by the same decision procedure provided the translation
was accurate. Decidable theories are entirely usable as subtheories of
larger undecidable ones. My own opinion is that modal logic should normally
be translated into first order logic, i.e. predicate calculus + functions,
conditional expressions, etc. However, all the problems of modality that
are discussed in connection with modal logic will still exist. If one is
interested in studying modal problems in pure form, e.g. the axiomatics
of knowledge, modal logic is a good enough formalism; when one comes to
extensive applications, then a wider system is desirable.
Challenge to modal programming theorists: Express and prove the associativity
of append, and then come back for a harder problem.
∂02-Mar-79 1101 DEW reference
To: JMC at SU-AI, feigenbaum at SUMEX-AIM, berliner at CMU-10A
I am applying for an assistant professorship at UC Santa Cruz, and they have
asked me to solicit letters from my references. Could each of you please
send a letter about me to:
Wilkins letter to:
Chairmain
Dept. of Information Sciences
237 Applied Sciences Building
U.C.S.C.
Santa Cruz, California 95064
If I can help with the paper work in any way, please let me know.
Also, I would appreciate a note of confirmation when the letter is sent
so that I will know things are going smoothly.
Thanks very much.
David
∂02-Mar-79 1140 ARK S-1 Meeting
To: S1 at SU-AI
CC: pickering at SUMEX-AIM
There will be an S-1 meeting on Wednesday, 7 March at 2:30pm in the Serra
House Conference Room.
Arthur
∂02-Mar-79 1501 Susan Owicki <SSO at SU-AI> Why use modal logic
Date: 2 Mar 1979 1500-PST
From: Susan Owicki <SSO at SU-AI>
Subject: Why use modal logic
To: modal at MIT-AI
My reason for wanting to use (some) modal logic is still different from
John's and Vaughan's. Primarily, it provides a convenient notation for
expressing program properties that I wish to verify (or that I need in
verifying other properties). It is an abstraction and abbreviation of
the predicate calculus formulas, the abstraction seems to make it easier
for people to express their reasoning in proving programs correct.
∂02-Mar-79 1503 Susan Owicki <SSO at SU-AI> mailing list
Date: 2 Mar 1979 1502-PST
From: Susan Owicki <SSO at SU-AI>
Subject: mailing list
To: modal at MIT-AI
There are several graduate students here with whom I have been discussing
modal logic. I'm sure they would like to participate in, or at least observe,
these discussions. Having too many people involved may be too confusing,
though. Any opinions?
∂02-Mar-79 1505 PRATT at MIT-AI (Vaughan Pratt)
Date: 2 MAR 1979 1805-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: JMC at SU-AI
Ooops, I hadn't planned on anybody adopting a different strategy from
the one I had suggested, of sending mail to MODAL at MIT-AI. There was
a silent member of the group, namely a secretary file in which all
correspondence accumulates. I think if you put [PRATT;MODAL MAIL]@MIT-AI
as one of your addressees the right thing should happen (assuming your
mailer can parse it - it would depend on whether it treated [] differently
from alphamerics).
∂02-Mar-79 1628 HVA Available Funds in Unrestricted Account 2 FCZ 154
Including the new money-Research Incentive per J. Rosse memo: $56,957.10.
∂02-Mar-79 1849 ME
You have exceeded your disk quota.
The files listed below have been purged to reduce your disk
area to your quota of 2000
Before purging, your files occupied 2441
KNOW.LST[S78,JMC]
DIALNE.LST[DIA,JMC]
MODAL.LST[LET,JMC]
CRYPT.DMP[ 2,JMC]
LFOL.DMP[F78,JMC]
REV1.DMP[F78,JMC]
CODE.DMP[ 2,JMC]
CIRCUM.XGP[W78,JMC]
DIAL78.XGP[DIA,JMC]
PROBLE.XGP[W79,JMC]
DIALNE.XGP[DIA,JMC]
MINIMA.XGP[S77,JMC]
MENTAL.XGP[F76,JMC]
AIM1.XGP[W79,JMC]
CONCEP.XGP[E76,JMC]
GORIN3.XGP[LOT,JMC]
TAKASU.XGP[LET,JMC]
MANNA.XGP[LET,JMC]
PERRY.XGP[LET,JMC]
III.XGP[LET,JMC]
THESIS.XGP[LET,JMC]
TANG.XGP[LET,JMC]
CHANDR.XGP[LET,JMC]
LINDZE.XGP[LET,JMC]
AIPHIL.XGP[W79,JMC]
ELEPHA.XGP[W79,JMC]
FILMAN.XGP[LET,JMC]
FIRST.XGP[W79,JMC]
WILKIN.XGP[LET,JMC]
CRYPT.FAI[ 2,JMC]
VITASK[ 2,JMC]
CHRI.LE1[ 2,JMC]
VIS[ 2,JMC]
VIS1[ 2,JMC]
MIDF76[ 2,JMC]
PARC.2[ 2,JMC]
EC[ 2,JMC]
BLURB[LOT,JMC]
DEC20[LOT,JMC]
TASK[LOT,JMC]
P[LOT,JMC]
TERMIN[LOT,JMC]
BUDGET.2[LOT,JMC]
FRANKL.LE1[LOT,JMC]
LOTS.BLB[LOT,JMC]
MEASUR[LOT,JMC]
TERMIN.2[LOT,JMC]
BOARD[LOT,JMC]
LOTS.REP[LOT,JMC]
SHAW.LE1[LOT,JMC]
GLUGG2[LIT,JMC]
HOTER.BAY[PUB,JMC]
WEIZEN.REP[PUB,JMC]
HOTER.REC[PUB,JMC]
HOTER.OUT[PUB,JMC]
WEIZEN.LED[PUB,JMC]
HOTER.POX[PUB,JMC]
HOTER.WD[PUB,JMC]
WEIZEN.REV[PUB,JMC]
HOTER.BGB[PUB,JMC]
WEIZEN.ANS[PUB,JMC]
SUBSCR[PUB,JMC]
CARGO.LIT[LIT,JMC]
IBM.SNG[LIT,JMC]
BIB.PUB[SEN,JMC]
TECHNO[SEN,JMC]
CALIF[SEN,JMC]
FEB14.NOT[SEN,JMC]
QUOTE.2[SEN,JMC]
STATEM.LET[SEN,JMC]
POSTER[SEN,JMC]
ASHLEY.COV[SEN,JMC]
SIGN[SEN,JMC]
ENDORS[SEN,JMC]
LONG[SEN,JMC]
LEGISL[SEN,JMC]
FEB8.78[SEN,JMC]
ENERGY.ST1[SEN,JMC]
∂03-Mar-79 1026 FB Elephant and Lucid
Both Susan and I are interested in Elephant. I have two more recent Lucid
reports which I will leave at the Lab in your mailbox sometime this weekend.
Please return them to my box (after copying if desired). If you have a
document on Elephant...
∂03-Mar-79 1808 DCO
To: "@GROUP.DIS[VCG,DCL]" at SU-AI
VERIFICATION GROUP SEMINAR
TUESDAY 6th MARCH
PLACE: ERL 237
TIME: 2:30 p.m.
TITLE: Logic Programming
SPEAKER: Sten-Ake Tearnlund, University of Stockholm
ABSTRACT: unavailable
∂03-Mar-79 1929 PRATT at MIT-AI (Vaughan Pratt)
Date: 3 MAR 1979 2229-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: JMC at SU-AI
CC: (FILE [PRATT;MODAL MAIL]) at MIT-AI
From: John McCarthy <JMC at SU-AI>
Our mailer refuses [pratt;modal mail]%ai.
You might pass that problem along to the maintainers of your mailer for
consideration. Anyway, I'll just make a point of remembering to file
manually anything you send on modal logic, unless you feel comfortable
just sending to MODAL%AI, which I know works (that's what Sue Owicki
has been doing).
I don't understand what the different effects of mailing to MODAL and
[pratt ... are intended to be.
∂04-Mar-79 0815 PRATT at MIT-AI (Vaughan Pratt)
Date: 4 MAR 1979 1115-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: MODAL at MIT-AI
(This is mainly for JMC's benefit, though the information should be
of general interest anyway.)
Mailing to MODAL%AI results in distribution to all the people currently
listed on the file AI:PRATT;MODAL LIST. (This is accomplished by putting
a link from MODAL to said file in the pseudo-directory belonging to our
mailer, which then knows to consult that file whenever it receives a
message for MODAL.) Presently there are five humans and one file in the
list. The file acts as a passive recipient by accumulating anything sent
to it. If you send to MODAL%AI you are guaranteed to reach everyone currently
on the list; if you maintain your own list you run the risk of omitting
newcomers (the distributed data base problem). On top of that, it seems you
can't name files as recipients using an SU-local list. Third, you should
find it easier to address mail to MODAL%AI than to MODAL.LST[LET,JMC], though
that isn't a major consideration.
∂04-Mar-79 0845 SSO Net-mediated discussion on modal logic
To: "@MODAL.DIS[1,SSO]" at SU-AI
Vaughan Pratt is initiating a discussion on modal logic and its use in
program semantics -- so far he, John McCarthy, Les Lamport, Bob Moore (I
think) and I are involved. Let me know if you would like to be added to
the mailing list. There is no limit on the number of observers that can
be involved; on the other hand, if too many people want to actively partici-
pate in the discussion, things could get confusing. I think we can start
out by having anyone who wants to say something do so, and then cut back
if there are too many talkers.
Sue Owicki
∂04-Mar-79 2149 BPM More bureaucracy, less $$$
To: EAF at SU-AI, DPB at SU-AI, JMC at SU-AI, DEK at SU-AI
n067 1504 04 Mar 79
BC-AUDIT
By MALCOLM W. BROWNE
c. 1979 N.Y. Times News Service
NEW YORK - The Federal Office of Management and Budget has announced
that a strict new auditing system will go into effect in October to
oversee and regulate all research expenses incurred by universities
receiving federal grants for scientific research.
The new rules disclosed Sunday have been under discussion for the
past year. In their original form as proposed last March, they caused
a storm of protest by such leading research institutions as Stanford
University and the University of California.
Officials of such institutions contended that the rules would, in
effect, drastically reduce the funds available for research, as a
consequence of curbing the use of students as research assistants, of
including such administrative costs as building and library
maintenance in research costs, and other difficulties.
Asked about the new rules, university spokesmen pronounced
themselves dissatisfied, but conceded they had been treated fairly by
the budget office and hoped to have the rules improved on in the
future.
The federal grants affected total $4 billion annually. Although this
money is dispersed among hundreds of academic institutions, the
lion's share traditionally goes to a handful of research leaders,
such as Stanford, California, the University of Rochester and
Massachusetts Institute of Technology.
While federal research grants are made by many government agencies,
they all will be subject to the new guidelines.
John J. Lordan, chief of the financial management branch of the OMB,
said that the new rules represented a compromise to which he hoped
there would be relatively little objection by the universities.
''Because of the greater uniformity of accounting procedures, the
institutions will actually have a smaller burden in accounting for
their research expenditures,'' he asserted in an interview.
Lordan said that although the phrasing of the rules had been changed
to meet some academic objections, the essential principles had been
retained in their original form. In particular, the expenses of
graduate students and faculty members working on government-sponsored
research will be charged against the overall research grants.
Overhead expenses incurred for government research projects,
including building maintenance, electricity and library service, will
not be counted against the grants, however, and will be paid for
separately by the Government.
A contrasting view was expressed by Frank Riddle, assistant
controller of Stanford, which receives $117 million in federal
research funds annually.
''To provide the accounting data the government now requires, we'll
have to revise our whole payroll system and expand our staff,''
Riddle said. ''How do you define what a doctor may be doing in a
special operation - is it research, is it treating a patient or is it
educating the students who are watching? Distinctions like this will
enormously complicate things, and we'll be able to recover even less
of our research expenses than previously.''
Last month, Stanford's president, Richard W. Lyman, sent a letter to
President Carter, acknowledging the ''fairness'' the government had
displayed in formulating the new rules but contending that the rules
would ''not resolve the disputes that have plagued
government-university relations.'' He added that he hoped improvement
of research accounting would be a continuing process.
University officials feel that accounting costs connected with
scientific research have already soared beyond reason, and that the
new rules may make things worse.
''During the past year,'' a Stanford spokesman said, ''we were
compelled to add 33 staff positions to deal with business and finance
at this institution, including 28 who are accountants and lawyers.
The expense of cost accounting has cut deeply into our useful
resources.''
Riddle said that in any case, American universities ''will have to
do what we can to live with the new rules for the time being, to work
with the government and to hope for improvements at some later time.''
∂05-Mar-79 1031 PAT
∂04-Mar-79 0026 JMC
I now have the library copy of Aaron's Sloman's "The Computer Revolution
in Philosophy" - Harvester Press and have concluded that I need to
buy my own copy. Please order it for me.
---
ordered - patte
∂05-Mar-79 1205 CLT
you said you would send a note to Denny Brown about my
MTC qual being postponed until Fall
I spoke to him, and he said that there would be no problem.
∂05-Mar-79 1528 DCO
To: "@GROUP.DIS[1,DCO]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
A Natural Programming Calculus
Sten-Ake Tarnlund, University of Stockholm
2:30 p.m., Tuesday, March 6
ERL 237
ABSTRACT:
We shall take up a natural deduction system for classical logic
and introduce it as a programming calculus. The axioms of the calculus
characterize our data structures, e.g., lists and trees. Our language, L,
is first order predicate logic defined in a standard way. We make use of
thdαcalculus to deduce programs in Horn clauses, which are our programming
language and a subset of L. For reasons of efficiency these Horn clauses
are run in PROLOG, which is a special logic system in our calculus. We
shall show deductions of programs from specifications in L. For example,
alternative programs are deduced from alternative data structures and from
alternative operations on data structures as well as by mappings between
data structures. In particular, we deduce a LISP-program that twists a
binary tree from a logic program defined on labeled binary trees.
Moreover, we solve a problem that Burstall and Darlington could not solve
in their transformation system.
Next speaker: John Reif, University of Rochester, 13 March.
∂05-Mar-79 1834 RPG Book
I have been thinking about your proposal, and I think it would be a
reasonable thing for me to try. The only question is: except for the
feeling on "doing the right thing", what's in it for me?
-rpg-
∂05-Mar-79 2044 JBROWN at MIT-MC (Jordan Brown) ACKERMAN'S FUNCTION
Date: 5 MAR 1979 2338-EST
From: JBROWN at MIT-MC (Jordan Brown)
Subject: ACKERMAN'S FUNCTION
To: jmc at SU-AI
MCLURE@kl said you might be interested in this.
He was going to help me with a lip program to calculate ACK(4,4).
I AM NOT GOING TO WORRY ABOUT ACK(4,4)...A FRIEND OF MINE AND I
JUST FIGURED OUT GENERAL FUNCTIONS FOR ACK(0,N) THROUGH ACK(3,N)
AND AN INCREMENTAL FUNCTION FOR ACK(4,N) [IE GIVEN ACK(4,N), FIND ACK(4,N+1)]
ACK(4,4) IS EQUAL TO:
2↑(2↑(2↑65536))-3
NEEDLESS TO SAY, SUCH A NUMBER IS ASTRONOMICAL. I WAS GOING TO TELL
YOU HOW MANY EXOPARSECS OF PAPER WOULD BE REQUIRED TO PRINT THE
RESULT, BUT THE NUMBER IS PROHIBITIVELY LARGE. AS A MATTER
OF FACT, THE NUMBER OF DIGITS IN THE NUMBER OF EXO PARSECS IS
PROHIBITEVLY LARGE. HOWEVER, THERE ARE APPROXIMATELY 10↑19660
DIGITS IN THE NUMBER OF DIGITS IN THE NMBER OF EXOPARSECS OF PAPER IT
WOULD TAKE TO PRINT.
SIGH. AND ALL FROM SUCH A SIMPLE, MILD-MANNERED FUNCTION...
IN CASE YOU ARE INTERESTED, HERE ARE THE EQUATIONS WE CAME UP WITH:
ACK(0,N)=N+1
ACK(1,N)=N+2
ACK(2,N)=2N+3
ACK(3,N)=2↑(N+3)-3
ACK(4,N)=2↑(ACK(4,N-1)+3)-3
ACK(4,0)=13
∂05-Mar-79 2258 LLW Final ABOX Review
To: "@PLAN[DIS,S1]" at SU-AI, FB at SU-AI, JLH at SU-AI, JMC at SU-AI,
GIO at SU-AI
Note in PLAN.LOG that PMF will be reviewing the ABOX architecture with all
non-MARK2A parties-known-to-be-interested-and-whose-comments-are-welcome,
in the Conference Room of the S-1 Palace tomorrow (Tuesday), at 2 PM. Be
there to carp, or forever (until the Mark III) hold your peace. . . .
Lowell
∂05-Mar-79 2316 LLW ABOX Review Reading
To: "@PLAN[DIS,S1]" at SU-AI, FB at SU-AI, JLH at SU-AI, JMC at SU-AI,
GIO at SU-AI
My last message should have noted that those wishing to indulge in
well-backgrounded carping at Mike's ABOX Review tomorrow PM should have
looked over AMODUL.POX[1,PMF], or one of its hard copy versions floating
about. Blatantly ignorant remarks may be sneered at.
Lowell
∂06-Mar-79 1027 Bmoore at SRI-KL (Bob Moore) Sloman's article...
Date: 6 Mar 1979 1027-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Sloman's article...
To: jmc at SAIL
appeared in the advance papers of a conference titled "Theoretical Issues in
Natural Language Processing," held at MIT in June 1975. The paper seems
fairly reasonable to me except for two rather gross mis-statements (which
nothing much depends on, fortunately). The two statements that I object to
are:
(1) "Roughly, in a complex Fregean symbol the structure of the symbol...
corresponds to the structure of the procedure by which [the] thing [denoted]
is identified, or computed." (p. 165, column 1)
This seems to be just warmed-over verificationism with all its usual problems.
How does the Fregean description "the shortest unprovable true formula of ZF"
give you a procedure for identifying or computing anything, given that ZF is
undecidable?
(2)"Since one can use English to explain how such functions [i.e. functions
whose semantics is claimed to be non-Fregean] work, English also cannot be
restricted to Fregean semantics." (p. 165, column 2)
Statement (1) may at least be contoversial, but this one is just flat-out
wrong. It COULD be the case that semantics of English are completely
Fregean (in Sloman's sense) and that English would still be usable as a
meta-language for describing constructs in some language that did not
have Fregean semantics. A meta-language need not have all of the semantic
constructs of the object language it describes: e.g., axiomatizing the
semantics of modal logics in first-order logic.
-------
Thanks for the Sloman reference. It is clear that he is not a technically
educated logician - or if he is, he doesn't have his logic integrated
with his philosophy. I have gotten half way through his "Computer
Revolution in Philosophy", and I think it is very good. There is an
outside chance of getting him invited to take part in the study.
∂06-Mar-79 1524 Bmoore at SRI-KL (Bob Moore) Modal logic vs. its possible-world translation
Date: 6 Mar 1979 1526-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Modal logic vs. its possible-world translation
To: modal at AI
Here are my thoughts on modal logic vs. its translation into
first-order logic: I believe that, on balance, the current
state of the art of automatic deduction strongly favors
translation into first-order logic. I am not very impressed
with the decision procedures for propositional modal logics
for a number of reasons. First of all, the logics that are
decidable are so weak that I doubt that they have many
practical applications. In particular, no reasonable
axiomatization of the semantics of a real programming
language could possibly be decidable. Consider that the
halting problem is easily expressed in dynamic logic as
[C]F, so to get decidability, either the programming
language must be so weak that its halting problem is
decidable, or the theory must be grossly incomplete so that
the decision procedure doesn't really "try". Are we really
willing to settle for a logic of programs in which we can't
prove that X<-1;WHILE(X>0,X<-X+1) doesn't halt?
A second problem with these decision procedures is that they
are at least exponential in running time. Vaughn has
correctly pointed out that if the theorems to be proved are
relatively short, this is not a problem, and my vauge
understanding is that "relatively short" can be as many as
several hundred symbols on some systems. (I am thinking of
what Richard Wehyrauch once told me about his tautology
checker.) For many problems, however, this is just not
adequate. If progam verification is ever to go beyond toy
programs, we will have to do better. In my own area of
primary interest, common sense reasoning, useful systems
will have to deal with very large data bases of common sense
knowledge. The "theorems" to be proved will be of the form
DB -> P, where DB is the data base and P is the goal to be
deduced. Procedures that are exponential in the size of the
data base are clearly unacceptable. What is needed is not a
uniform decision procedure, but rather specific knowledge
about doing deductions in the problem domain at hand. My
master's thesis was on how to extend PLANNER-like techniques
for encoding this knowledge to the full predicate calculus,
and I feel that much more can be done in this area.
The decision procedures that Vaughn advocates are semantic
in that they actually do construct possible worlds (i.e.
truth assignments to propositional variables) and
accessibility relations. If one attempts to use the axioms
for modal logics directly, there are even more difficulties.
One problem is that in many cases adequate axiomatizations
for modal logics do not exist. In particular, the
axiomatizations for quantified modal logics given by Kripke
("Semantical Considerations") and Hughes and Cresswell are
for languages in which the only terms are variables.
I also believe that the shortest proofs of some theorems may
be exponentially longer using the axioms of standard modal
logics than using semantic methods. Let M be a modality and
let M*N(P) mean N applications of M to P. In T, S4, and S5,
all formulas of the form M*N(P -> Q) -> (M*N(P) -> M*N(Q))
are theorems, but it appears that the length of the shortest
proof is exponential in N. (It appears that to prove an
instance of this schema for N requires proving two different
instances of it for N-1.)
Translating the modal logic into a first-order
axiomatization of its possible world semantics has certain
problems too. For one thing, the resulting formulas tend to
be much less intelligible to people than the original modal
formulas. This can be partially solved by having the
computer do the translation, but if the machine is expected
to be able to explain its reasoning, then we are still in
trouble since the intermediate steps in a deduction in the
possible-world formulation typically do not have modal
equivalents. (This would also be a problem for Vaughn's
decision procedures.) Also, translating into first-order logic
doesn't solve the deduction problem; it merely transfers it
to a framework that we know something about. If we turn a
simple minded (say, resolution) theorem prover loose on the
translation, we are likely to get the same (or even worse)
behavior that we would get from one of Vaughn's decision
procedures. Again, we need domain-specific knowledge to
avoid unacceptable inefficiency. In my doctoral thesis, I
tried to show how this could work for a modal logic of
knowledge and action. The results are only suggestive, as I
did not actually implement anything, but I think the are
very encouraging.
-------
∂06-Mar-79 1559 Mcgoveran at SUMEX-AIM Chinese visitors
Date: 6 Mar 1979 1559-PST
From: Mcgoveran at SUMEX-AIM
Subject: Chinese visitors
To: JMC at SAIL, DPB at SAIL
Dr. Linvill's secretary is making out a schedule for the visitors. Since
Denny's schedule is full for Tuesday, I suggested that he see them on Monday.
Dr. Linvill has invited any interested people to let him know if they want to
have lunch with the group on either day at the Faculty Club. Would it be pos-
sible for Denny to join them at lunch and discuss the CS dept. until 1:30 or
when he has to go to his class?
As for John McCarthy, if you could let me know when you would like to meet with
the group, either day, I could transmit the information to Louise Peterson. She
is awaiting my answers before she puts the rest of the itinerary together since
Denny's schedule seems to be the most crowded.
Mary McGoveran
-------
∂06-Mar-79 1657 LES Ethernet and friends
To: Feigenbaum at SUMEX-AIM
CC: JMC at SU-AI
The Xerox offer of a Dover, Ethernet, etc. looks quite attractive.
I note that nothing has been said about dates, though.
It appears that some new equipment that SAIL plans to acquire might best
be connected to the KL10 and each other through an Ethernet.
This includes the new computer donated by TI and an image display system
that Binford and Levinthal plan to buy. If we are going ahead with the
Xerox deal, it would seem advantageous to start with the Ethernet as soon
as possible.
∂06-Mar-79 1705 PAM Petit channel
Just talked to Rubin about the idea of resurrecting the Petit channel.
He says that not only did it lose its memory bus but it was extensively
modified in order to work with the mappiplexor. He feels that the engineering
effort that would be needed to give it a working memory bus would not be
worth it -- it would be much more sensible to buy a Foonly C1.
∂06-Mar-79 1727 PAM
I think you just got a message from Les logged in as me..Paul
∂07-Mar-79 0905 Feigenbaum at SUMEX-AIM Re: Ethernet and friends
Date: 7 Mar 1979 0905-PST
From: Feigenbaum at SUMEX-AIM
Subject: Re: Ethernet and friends
To: LES at SU-AI
cc: JMC at SU-AI
In response to your message sent 6 Mar 1979 1657-PST
Thanks, Les. I agree with all that. Does anyone from SAIL want to be on
the "liason committee" for the Xerox gift? I intend to begin meeting with
Xerox soon.
Ed
-------
∂07-Mar-79 0917 Feigenbaum at SUMEX-AIM thought you might be interested
Date: 7 Mar 1979 0908-PST
From: Feigenbaum at SUMEX-AIM
Subject: thought you might be interested
To: jmc at SAIL
Date: 7 Mar 1979 0804-PST
From: Hazen
Subject: "Scientific Revolutions" Talk
To: siglunch:
cc: hart at SRI, SACERDOTI at SRI, BOLLES at SRI
This Thursday at 4 p.m. in the Chemistry Department Gazebo,
Doug Lenat will give a talk on "Scientific Revolutions."
Here is a copy of his abstract:
Doug Lenat will discuss Thomas Kuhn's "Structure of
Scientific Revolutions", the criticism it drew, and Kuhn's responses.
The thesis was that science does not progress by gradual, continuous
knowledge accretion, but rather by discrete leaps from one paradigm
to another. Professor Lenat will provide some of the Philosophy of
Science context of this work, especially the work of Lakatos and
Popper, and will focus upon their reactions to Kuhn. A resolution of
the two antithetical views is possible, using the language and
concepts of AI, and this will form the latter part of the talk.
-------
-------
∂07-Mar-79 0917 Feigenbaum at SUMEX-AIM letter to chowning
Date: 7 Mar 1979 0917-PST
From: Feigenbaum at SUMEX-AIM
Subject: letter to chowning
To: jmc at SAIL
John, your letter is excellentand statesmanlike. I will get back to you
with comments just as soon as I get a chance. I am in LA right now, and
will be back at Stanford tonight.
Ed
-------
∂07-Mar-79 1349 SWB Phil Agre
Phil Agre has been admitted to the Ph.D. program for next fall from U. of
Maryland, and he's visiting Stanford March 22-24. He's interested in AI, and
wants to talk to someone at the Lab about where he might fit into the great
cosmic scheme of things. Could he talk to you sometime? - Sam
∂07-Mar-79 1429 ES
A well known but amusing probability paradox
Suppose we have a black box which outputs independent random real
numbers according to a probability density f(x), having mean m and width w,
where the width is measured say from the first to third quartile, which
guarantees that w is finite. Suppose we wish to make an experimental
measurement of m by obtaining a large number, N, of outputs, and then taking the
average. Is this always a good procedure? Surprisingly, the answer is no.
The N-tuple average is itself a random variable satisfying some
probability density f[N](x), with mean m[N] and width w[N]. One's intuition may
claim that
(1) m[N] = m, and
(2) lim w[N] = 0.
However, there exist probability distributions for which condition (2) is
violated.
Let
f[p,w](x) = integrate(exp(-|w*u|↑p + %i*x*u),u,minf,inf)/(2*%pi), 0<p≤2
be the black box output probability density. Special cases are the Gaussian
distribution
f[2,w](x) = exp(-(x/(2*w))↑2)/(2*sqrt(%pi)*w),
and the Cauchy distribution
f[1,w](x) = (w/%pi)*(x↑2 + w↑2)↑(-1).
It is easily seen that f[p,w](x) = f[p,1](x/w)/w, so that w is a measure of
width.
These probability distributions were chosen for their "stability"
property, namely that the N-tuple average has the same shape of distribution as
the individual outputs.
f[p,w,N](x) = f[p,w[N]](x).
Furthermore,
w[N] = N↑(1/p - 1) * w.
For the Gaussian distribution, p=2, we have w[N] = w/sqrt(N), as
expected. For the Cauchy distribution, p=1, we have w[N] = w, independent of N,
so it is futile to average. And for 0<p<1, w[N] actually increases with N, so
in these cases it is better not to average!
Why this unintuitive behavior? In fact, according to the Central Limit
Theorem, shouldn't f[p,w,N](x) approach a Gaussian as N→∞? Well, a sneaky
hypothesis of the Central Limit Theorem is that the variance be finite. But for
p<2, our function has infinite variance. Furthermore, for p≤1, the mean is 0
only by virtue of the density being an even function of x; the first moment
integral is divergent.
∂07-Mar-79 1444 Bmoore at SRI-KL (Bob Moore) letter to study group participants
Date: 7 Mar 1979 1446-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: letter to study group participants
To: jmc at SAIL
Here is a draft of a letter to send to the participants in the AI/Phil
study group concerning our ideas for a workshop this fall. I may have
put in some details that we did not discuss or that you disagree with,
so feel free to add your input. I have added more names to the list
that we discussed, and it occurred to me that any of the people who were
originally suggested for the study group would be obvious candidates for
the workshop, so you might want to review your original list. The letter
is as follows:
Dear XXXXX:
In connection with the forthcoming study of AI and
Philosophy at the Behavioral Sciences Center, we believe
that it would be desirable to have a small conference or
workshop on this topic that would involve a larger group of
scholars than just those in the study group. We would like
to solicit your reactions to this suggestion and ideas on
how it should be organized. A two day meeting (probably a
Friday and Saturday) sometime in October seemed most
appropriate to us. We suggest having invited papers
distributed in sufficient time for the workshop participants
to have read them, but also to select one participant to
serve as official commentator for each paper. The format of
the workshop would then be to have a brief presentation of a
paper, followed by the remarks of the commentator, the
response of the author, and general discussion. This format
would allow for three or four papers presented per day, or a
total of six or eight for the workshop. Since this is to be
an interdisciplinary workshop, it seems desirable that the
authors and commentators be from different disciplines.
We envision officially inviting about twenty people from
outside the study group, with an informal invitation being
extended to the local Philosophy and AI communities. We
have attached a list of prospective participants. We would
like to hear your general reaction to this idea, plus any
ideas you have regarding the date or format of the workshop,
specific topics to be discussed, and potential participants,
speakers, or commentators. At the moment all ideas are
quite tentative, so please do not hesitate to recommend
changes to our suggestions.
Sincerely,
John McCarthy
Robert Moore
SUGGESTED PARTICIPANTS FOR WORKSHOP ON AI AND PHILOSOPHY
ARTIFICIAL INTELLIGENCE
David Marr MIT
Mitch Marcus MIT
Jon Doyle MIT
Brian Smith MIT
Terry Winograd Stanford
Richard Weyhrauch Stanford
Lewis Creary Stanford
Barbara Grosz SRI
Jerry Hobbs SRI
Jane Robinson SRI
Herbert Simon CMU
Bill Woods BBN
Drew McDermott Yale
PHILOSOPHY
Patrick Suppes Stanford
Julius Moravcsik Stanford
John Perry Stanford
Noam Chomsky MIT
Jerry Fodor MIT
Ned Block MIT
Hubert Dreyfus Berkeley
David Kaplan UCLA
Hilary Putnam Harvard
Aaron Sloman Sussex
Barbara Partee U. of Massachusetts
Margaret Boden
∂07-Mar-79 1856 LLW LEE S. PARKS--SU/CS APPLICANT
To: JMC, FB, JLH
CC: LLW
I would like to commend to your attention the application of Lee S. Parks,
presently an MIT senior, for admission to Stanford Computer Science Dept.
for next Fall. (I have never been able to dope out a model of SU/CS
admissions procedures or policies that had predictive value, so I'm taking
the frontal approach of this note of recommendation.)
Parks has been offered and has accepted a full-time, permanent offer to
join the S-1 Project's operating systems group, and will be joining us in
3 months. He is presently finishing up a double major (in physics and
math, peculiarly enough) at MIT, and has extensively sampled both the
computer science course offering and the local research (particularly the
ArchMac Project) there. He appears to be outstandingly bright and
hard-working.
In some (narrow-minded) respect, my supporting his application to SU/CS is
a conflict of interest, since I would like to have his full-time work here
at LLL for the next few years; however, he clearly would benefit from
graduate education at SU, and richly deserves the opportunity--he's one of
the very best I've seen recently.
Lowell
∂07-Mar-79 1904 KENNETH KELLER at CMU-10B school
Date: Wednesday, 7 Mar 1979 2203-EST
From: KENNETH KELLER at CMU-10B
Subject: school
To: jmc @ su-ai
John,
I am a senior at CMU and applied for fall admission to
the Stanford PhD program. I called today and was told that
I was rejected and was not among the top 50 applicants.
I cannot believe the latter. Raj Reddy and Jon Bentley
wrote my letters. Raj told me that I had a very high probability
of acceptance. Bob Sproull and Anita Jones also know my
work. A resume that indicates my performance in the CS
field follows. Could you please reevaluate my application?
Regards,
Ken
KENNETH H. KELLER
school: Box 523, 5115 Margaret Morrison, Pittsburgh, PA 15213, 412-578-4146
home: 901 Regency, Richardson, TX 75080, 214-238-0295
objective Position as a software engineer with a concentration on
design.
experience
1/79 Work on a graphics package with segmented display file
compiler and transformation system for a refresh, vector
display. Department of Computer Science, Carnegie-Mellon
University, Pittsburgh, PA.
5/78-8/78 Work on software tools, education and standards, and
compilers. Advanced Software Technology Department, Texas
Instruments Incorporated, Dallas, TX.
1/78-5/78 Work on display management package used to build user
interfaces. Department of Computer Science, Carnegie-Mellon
! 2
University, Pittsburgh, PA.
5/77-8/77 Work on integrated circuit layout and edit interactive
graphics system, and pattern generator software. Design
Automation Department, Texas Instruments Incorporated, Dallas,
TX.
8/76-5/77,8/75-5/76
Work on data analysis and graphics software. Department of
Physics, Tulane University, New Orleans, LA.
5/76-8/76 Work on data analysis and graphics software. Space Sciences,
University of Texas at Dallas, Richardson, TX.
major In 5/79 expect BS in applied mathematics with concentration in
computer science from Carnegie-Mellon University.
publications Keller, K. H., "Computation Cost Functions for
Divide-and-Conquer Algorithms", Journal of Undergraduate
Mathematics, to appear and presented at the Third Annual
Conference on Undergraduate Mathematics at Wesleyan University.
! 3
courses
computer science
artificial intelligence, structures, comparative languages,
digital systems, combinatorics, discrete mathematics, operating
systems, graphics, compiler design, discrete algorithms,
programming systems.
languages FORTRAN, PASCAL, LISP, SNOBOL, BASIC, COBOL, J73/1, C, SAIL,
assembly languages.
mathematics vector calculus, complex variables, differential equations,
applied probability, linear algebra, numerical methods,
probability, statistics, operations research, data analysis,
modern physics, modern algebra.
! 4
KENNETH H. KELLER
references Available upon request.
personal
born 3/8/57 in New York, NY.
GPA Carnegie-Mellon GPA: 3.7.
pursuits reading, playing organ and synthesizer, wandering outdoors,
member of Sierra Club, Association for Computing Machinery, Tau
Beta Pi, Phi Eta Sigma, and Sigma Pi Sigma national honor
societies, Carnegie Institute, CMU Department of Computer
Science Curriculum Review Committee, and CMU Department of
Mathematics Student Advisory Committee.
I have not been a member of the admissions committee for two years, and I am
not sure how things are done. In particular I don't know how many people
look at each folder in the course of selecting the top 50. If Raj or somone
of equal stature finds the outcome strange, then a message from him to
the current head of admissions might or might not cause re-examination.
∂07-Mar-79 1927 PRATT at MIT-AI (Vaughan Pratt)
Date: 7 MAR 1979 2224-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: jmc at SU-AI
You may be amused at the use to which the new backquote feature in
MACLISP may be put. See AI:PRATT;LSPLAM > and LSPY > for examples
of funarg hackery. (In case you haven't had occasion to try out `
yet, it is a variant of ' that permits evaluation within the
expression. A useful feature is that ` may be nested, raising the
question of how to identify which ` a given , (the eval-invoking
character) refers to. You may enjoy reconstructing how ` and ,
interact from the examples.)
∂07-Mar-79 1928 PRATT at MIT-AI (Vaughan Pratt)
Date: 7 MAR 1979 2226-EST
From: PRATT at MIT-AI (Vaughan Pratt)
To: jmc at SU-AI
Come to think of it, I'll just mail you the code.
(progn (setq base 10. ibase 10.)
(setq s '(lambda (f) `(lambda (g) `(lambda (x) (funcall (,',f x) (,g x))))))
(setq k '(lambda (f) `(lambda (x) ',f)))
(setq i (funcall (funcall s k) k))
(setq b (funcall (funcall s (funcall k s)) k))
(setq c (funcall (funcall s (funcall (funcall b b) s)) (funcall k k)))
(setq w (funcall (funcall c s) i))
(setq succ (funcall s b))
(setq zero (funcall k i) one (funcall succ zero) two (funcall succ one)
three (funcall succ two) four (funcall succ three)
five (funcall succ four))
(setq square (funcall w b))
(funcall (funcall (funcall (funcall square two) (funcall square two)) 'add1) 0))
(prog () (setq
base 10. ibase 10. *nopoint t
y1 ; the least fixpoint operator for one-argument functions
'(lambda (tau)
((lambda (z) (funcall z `',z))
`(lambda (y)
`(lambda (x) (funcall (funcall ',',tau (funcall ,y ',y)) x)))))
ftau ; functional for the factorial function
'(lambda (f)
`(lambda (x)
(cond ((zerop x) 1) (t (times x (,f (sub1 x)))))))
fact ; the factorial function
(funcall y1 ftau)
))
(funcall fact 4)
(funcall fact 20)
(prog () (setq
y2 ; the least fixpoint operator for two-argument functions
'(lambda (tau)
((lambda (z) (funcall z `',z))
`(lambda (y)
`(lambda (x1 x2) (funcall (funcall ',',tau (funcall ,y ',y)) x1 x2)))))
aptau ; functional for appnd
'(lambda (a)
`(lambda (u v) (cond ((null u) v) (t (cons (car u) (,a (cdr u) v))))))
appnd
(funcall y2 aptau)
))
(funcall appnd '(1 2 3) '(4 5 6))
∂07-Mar-79 1937 JMC
To: keller at CMU-10A
CC: reddy at CMU-10A
I have not been a member of the admissions committee for two years, and I am
not sure how things are done. In particular I don't know how many people
look at each folder in the course of selecting the top 50. If Raj or somone
of equal stature finds the outcome strange, then a message from him to
the current head of admissions might or might not cause re-examination.
∂08-Mar-79 0844 DPB
∂07-Mar-79 1950 JMC
Who is head of admissions this year?
[Gene Golub. Committee is also composed of: Herriot, F. Yao, Terry
Winograd, Bill van Cleemput, Sam Bent, and Ann Gardner. They are
now reading about 57 "finalists". Final meeting to determine
20-25 winners is scheduled for Tuesday, March 13. -Denny]
∂08-Mar-79 1016 PHY Vanity plates
If you need new license plates, I suggest that they might read "CDR CAR".
Not only is this historically faithful to the original IMB 704 implementation
of LISP, it also has another neat propert: Most people passing you on the
freeway will think your initials are C.D.R.
- Don Knuth
∂08-Mar-79 1219 Tajnai at SUMEX-AIM MACLISP Manual
Date: 8 Mar 1979 1156-PST
From: Tajnai at SUMEX-AIM
Subject: MACLISP Manual
To: Lenat, JMC at SAIL, Talcott at SAIL
MACLISP MANUAL Part 3 the System
Paul Chase of the Bookstore said they have 50 copies left over
from aut quarter. MIT refuses to take them back. $150 at stake.
Will it be used Spring and or Autumn Quarters?
Please advise.
Carolyn
329-1217 x502
-------
∂08-Mar-79 1234 DCO
To: "@GROUP.DIS[1,DCO]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
Data Flow Analysis of Concurrent Communicating Processes
and
Universal Games of Incomplete Information
John Reif
University of Rochester
2:30 p.m., Tuesday, March 13
ERL 237
John Reif will be giving two talks, each of about a half hour duration.
He will be visiting next Tuesday and Wednesday; anyone interested in talking to
him should contact Derek Oppen.
ABSTRACTS:
Data Flow Analysis of Concurrent Communicating Processes
We consider a set of "distant" concurrent processes each
sequentially executing a distinct program and communicating by the
transmission and reception of messages. (By "distant" we mean that there is
no interference between processes by shared variables, interrupts, or any
other synchronization primitives beyond the message primitives.) We
determine the computational complexity of fundamental problems such as
statement reachability, and also develop an iterative technique for data
flow analysis of communicating processes.
Universal Games of Incomplete Information
In this talk we consider two-person combinatorial games which are
reasonable in the sense that 1. it can be determined if position π' is immediately
reachable from position π in time polynomial in the size of π, and 2. the
number of positions reachable from a given position π is bounded by an exponent
in the size of π. The outcome of game G is the existence or non-existence
of a winning strategy (for a given player at a given position).
G is a two-person game of perfect information if each player is
aware of each other's piece's positions and only the two players determine
the positions; otherwise G is a game of incomplete information.
Relative to a class C of two-person combinatorial games, a game G'
ε C is universal in C if for each G ε C the decision problem for the
outcome of G is log-space-reducible to the outcome of G'. Stockmeyer and
Chandra have shown certain games to be universal for the class of all
two-person reasonable games of perfect information. The outcome problem
for their universal games of perfect information is log-space complete in
exponential time.
This talk presents certain games which are universal for all
reasonable games of incomplete information; the outcome problem for our
universal games are log-space complete in doubly-exponential time.
∂08-Mar-79 1405 LES Foonly channel
Poole's off-hand estimate of availability for a C1 channel is 3-4 months.
He will try to refine that and get back late next week.
He will also cost out another alternative: a combination channel and disk
controller that permits use of higher density disks (300 megabyte) at
about $12,000 each. Unfortunately, this approach would require getting
a whole new set of disks.
∂08-Mar-79 1451 TOB Sid Liebes
John
I would like to talk with you briefly about two
subjects:
1. I have been considering hiring Sid Liebes
half time, with the intent on both sides
that he would make a long term commitment to
robotics.
2. I am making up a prospectus for the consortium
for AL. There are administrative issues for you
and Feigenbaum.
Tom
∂08-Mar-79 1530 Jrobinson at SRI-KL (Jane Robinson) Bob Moore's graduation present
Date: 8 Mar 1979 1532-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: Bob Moore's graduation present
To: hart, SACERDOTI, NILSSON, ROBINSON, hobbs, SAGALOWICZ, haas,
To: WALDINGER, KONOLIGE, TENENBAUM, GASCHNIG, RAPHAEL,
To: jmc at SAIL, rww at SAIL
Soliciting donation - modest. Suggest $1.00 if you are planning to attend
party. Earl will get gift, suitably engraved. IOU's accepted via
sndmsg. Would like to know ASAP.
Reply me or earl.
Jane
-------
∂08-Mar-79 2306 GIO Trip to LIvermore and S-1
To: S1 at SU-AI
CC: pickering at SUMEX-AIM
A visit to Livermore is scheduled for next thursday. Those that wish to come
should before 2:30 pm Friday March 9th send a message to
Ton McWilliams ( aka TM) with a copy to GIO (aka WIEDERHOLD%SUMEX)
Giving Full Name, Citizenship , employer or student status, transport choice.
The transport choices are
Independent - Arrive at Livermore Westgate by 3pm - leave there about 5:30 pm
Wanting Ride- Meet at SERRAhouse/trailer 1:30 pm - eat on way back, arrive Stanford about 7:30 pm
Offering Ride- Either meet at Serra same time or form separate group.
We will let everyone know on Monday if there is Ride overflow or underflow.
Due to need to give advance warning late responders cannot be accomodated.
Se what you have been working on ! Gio
∂09-Mar-79 0933 HVA National Geographic
Thank you for bringing me the copy. I read the article last night and it
evoked so many thoughts, including the mental images I had when doing the
tapes, but they are difficult to express here and now....
∂09-Mar-79 1442 MRC
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging,
** ESPECIALLY if there are less than 1000 free disk blocks in the system! **
If you don't delete some of your files, the purger will.
Your disk quota is: 2000
Your files occupy 2361
∂09-Mar-79 1615 Ijcai79 at SUMEX-AIM IJCAI Reviews
Date: 9 Mar 1979 1615-PST
From: Ijcai79 at SUMEX-AIM
Subject: IJCAI Reviews
To: IJCAI-79 Reviewers:
cc: ijcai79
fDear Reviewer:
Have you received the IJCAI paper(s) that we sent to
you for review We would greatly appreciate your prompt
response. Sndmsg to IJCAI79@sumex-aim.
Thank you, The IJCAI-79 Program Committee
-------
I have received one or more papers.
∂09-Mar-79 1900 JMC*
Sarah grades
∂10-Mar-79 0044 REM via AMES-TIP
To: RWG, JMC, RWW
New version of 79219A.WRU[1,REM] after first edit.
I think I sent it but please check.
∂10-Mar-79 1544 REF
John -
If you haven't already done so, could you please send a current
recommentation letter to:
Aravind K. Joshi
Chairman, Department of Computer & Information Science
Room 268
Moore School / D2
University of Pennsylvania
Philadelphia, Pennsylvania 19104
Thanks.
bob
∂11-Mar-79 1748 HVA
budget.v hmm
∂12-Mar-79 1151 PAT
It has been sent.
∂10-Mar-79 1550 JMC
I think I sent it but please check.
∂10-Mar-79 1544 REF
John -
If you haven't already done so, could you please send a current
recommentation letter to:
Aravind K. Joshi
Chairman, Department of Computer & Information Science
Room 268
Moore School / D2
University of Pennsylvania
Philadelphia, Pennsylvania 19104
Thanks.
bob
∂12-Mar-79 1357 BCM
I tried to POX the "relati" file but no dice. Maybe it needs your area.
∂13-Mar-79 0949 100 : patte
John, I will be working on campus in Polya 202 until about 2pm, patte
the number there is 74776.
∂13-Mar-79 1043 PAT change of plans
They were't ready for me at CS so I am back here at the lab..the mailing
has been rescheduled for Monday.
Sorry for the confusion
∂13-Mar-79 1155 PAT Sarah's tuition
The letter came from benefits today, I talked to Sarah and forwarded it
to her.
∂13-Mar-79 1419 REP Summer Support
Dear John,
I am trying to arrange some support for this comming summer, involving
work in verification or automatic programming. Hopefully this work would
be a spring board into a thesis. Do you have any such support available.
I visited IBM last week and had a brief chat with John Williams (who
works with John Backus) on a functional programming topic. If you are at
all familiar with these people or their work I'd appreciate any brief comments
you could give me. They will, of course, be kept in confidence.
Thanks for your ear, Rich
Please come and see me about possible summer work in verification. I think
that Backus is on a wrong track. Someone with a clear firm point of view
might make something useful from their work, but I think they are unlikely
to produce something useful themselves. Backus's Turing lecture reflects
ignorance of important and relevant work in mathematical logic.
∂13-Mar-79 2120 FML theory of computation quals
I am putting together a CS report containing all qualifying exams given by
the department (since 1965). For theory of computation, I have copies of
written qualifying exams for the following dates:
April 1969, June 1970, October 1971, April 1971, October 1971.
Do you know if there are written versions of any other exams?
In particular, have all the exams since October 1971 been oral?
(If so, do there exist written transcripts of the questions asked?)
I have already done quite a bit of poking around in the CS library and
dept office files, and the above exams were the only ones I could find.
Thanks.
--- Frank Liang
My impression would be that they haven't all been oral, and I will look
for something, but it is unlikely that I will find anything.
∂14-Mar-79 0024 RP
JMC - Scott sent this reply in respose to your request
for a Geomed demonstration for the visitor from England:
∂13-Mar-79 2020 SEK Geomed celebration.
By all means. I'll be there Thursday at 2:00.
∂14-Mar-79 0814 Tajnai at SUMEX-AIM ( Forwarded Mail )
Date: 14 Mar 1979 0812-PST
From: Tajnai at SUMEX-AIM
Subject: ( Forwarded Mail )
To: Lenat, JMC at SAIL, Pat at SAIL, Talcott at SAIL
Date: 8 Mar 1979 1157-PST
From: Tajnai
Subject: MACLISP Manual
To: Lenat, JMC at SAIL, Talcott at SAIL
MACLISP MANUAL Part 3 the System
Paul Chase of the Bookstore said they have 50 copies left over
from aut quarter. MIT refuses to take them back. $150 at stake.
Will it be used Spring and or Autumn Quarters?
Please advise.
Carolyn
329-1217 x502
-------
I have had no response to Bookstore inquiry. Help, please.
-------
∂14-Mar-79 1317 DCL
To: ME
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1236 RAK NOT
To: "@SYSTEM.DIS[DOC,VER]" at SU-AI, PWM at SU-AI
Reminder: we still need to do something about NOT on TOPS20.
Dick
TO ME.
Marty, why is the SAIL reprsentation of ASCII tilda different from
the ASCII standard?
and, could it be changed back?
This deviation from the standard is a real pain.
-DCL
∂14-Mar-79 1333 ME character sets
To: DCL
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1317 DCL
To: ME
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1236 RAK NOT
To: "@SYSTEM.DIS[DOC,VER]" at SU-AI, PWM at SU-AI
Reminder: we still need to do something about NOT on TOPS20.
Dick
TO ME.
Marty, why is the SAIL reprsentation of ASCII tilda different from
the ASCII standard?
and, could it be changed back?
This deviation from the standard is a real pain.
-DCL
ME - Because we have a slightly different character set from ASCII.
And, basically, no, it cannot be changed. We have many times considered
changing our character set, only to conclude each time that the work
involved was far too vast to actually undertake. Essentially, every
program (user or system program) would have to be changed, not to
mention every text file; in addition provision would have to be made for
converting on restoration any files and programs that were saved (e.g.,
with DART) before the conversion. Certain programs, like FTP (in ASCII
mode) convert between the WAITS character set and ASCII. Local programs
that get exported generally have assembly switches in them so that they
know which character set they will actually see when they run.
∂14-Mar-79 1342 DCL
To: ME
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1333 ME character sets
To: DCL
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1317 DCL
To: ME
CC: RAK, DCO, PWM, FWH, LES, JMC
∂14-Mar-79 1236 RAK NOT
To: "@SYSTEM.DIS[DOC,VER]" at SU-AI, PWM at SU-AI
Reminder: we still need to do something about NOT on TOPS20.
Dick
TO ME.
Marty, why is the SAIL reprsentation of ASCII tilda different from
the ASCII standard?
and, could it be changed back?
This deviation from the standard is a real pain.
-DCL
ME - Because we have a slightly different character set from ASCII.
And, basically, no, it cannot be changed. We have many times considered
changing our character set, only to conclude each time that the work
involved was far too vast to actually undertake. Essentially, every
program (user or system program) would have to be changed, not to
mention every text file; in addition provision would have to be made for
converting on restoration any files and programs that were saved (e.g.,
with DART) before the conversion. Certain programs, like FTP (in ASCII
mode) convert between the WAITS character set and ASCII. Local programs
that get exported generally have assembly switches in them so that they
know which character set they will actually see when they run.
TO ME
Perhaps you would help the assemblers of the export version of the verifier
encode either their source or fasl files to get around the tilda glitch.
These people are: RAK,DCO,PWM,FWH
Also, your first comment on slighlty different character sets is unclear;
I know SAIL has different characters; by why does the ASCII subset differ
from the ASCII standard? - DCL
∂14-Mar-79 1743 Amarel at SUMEX-AIM meeting
Date: 14 Mar 1979 1741-PST
From: Amarel at SUMEX-AIM
Subject: meeting
To: jmc at SU-AI
cc: amarel
John, I am visiting Stanford this week, and I wonder whether it would be
possible to get together some time. I would be very interested to find out
more about your research. Is there a chance that you will have some time
on thursday between 4-6 pm, friday after about 3:30 pm (or maybe Saturday
some time). Please let me know (by netmessage or by phone call to Bruce
Buchanan's office). Thanks, Saul
-------
Sure, please phone to arrange time.
∂15-Mar-79 0931 Bmoore at SRI-KL (Bob Moore) Held over by popular(?) demand
Date: 15 Mar 1979 0931-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Held over by popular(?) demand
To: grosz, hendrix, SAGALOWICZ, hobbs, TENENBAUM, WALKER,
To: RAPHAEL, robinson, jmc at SAIL, rww at SAIL, KONOLIGE
cc: bmoore
Since several people expressed disappointment at not being able to attend
our party celebrating the completion of my thesis, we have decided that we
just have to have another one! You are all invited to the Moores' at 8:00
PM, Saturday, March 31. The address, again, is 737 DeSoto Dr., Palo Alto.
Please let me know if you can come.
Bob
-------
I hope I can come.
∂15-Mar-79 1555 HVA Sat. Mar. 17th - 7 p.m.
You slipped out the door when I was on the phone and I won't be in tomorrow
(tho' I'll be in and out at home tomorrow), and I wanted to ask if you will
be able to come to dinner party-hope so. Am off to get badly needed hair cut
and won't be back today. Hersche.
∂15-Mar-79 1739 PEAGRE at MIT-AI (Philip E. Agre) meeting
Date: 15 MAR 1979 2038-EST
From: PEAGRE at MIT-AI (Philip E. Agre)
Subject: meeting
To: jmc at SU-AI
Dr. McCarthy, I am a prospective AI grad student and I'll be
visiting the Stanford AI Lab between next Wednesday (the 21st)
and the succeeding Saturday. SWB says you would agree to speak
to me about research in the lab and what I might do if I came
to the Lab in the Fall. Would you like to set a formal
appointment for this meeting, and if so what times would be
convenient? - Phil Agre at Univ. of Maryland
∂15-Mar-79 2243 RAK Dear staffperson
Let me be the second to confirm that I, too, think you have every right
to express yourself.
Do you suppose that is the kind of letter Dijkstra prefers in place of XGP's?
Dick
∂16-Mar-79 1444 Gifford at PARC-MAXC Meeting to discuss CS258 Problems
Date: 16 Mar 1979 2:34 pm (Friday)
From: Gifford at PARC-MAXC
Subject: Meeting to discuss CS258 Problems
To: John McCarthy <jmc@sail>
cc: Gifford
Unless you have other commitments, I'll assume that our meeting for Tuesday,
March 20, at 11 am at your office at the AI Lab is still on.
/Dave Gifford
∂16-Mar-79 1730 ME
You have exceeded your disk quota.
The files listed below have been purged to reduce your disk
area to your quota of 2000
Before purging, your files occupied 2393
KNOW.LST[S78,JMC]
DIALNE.LST[DIA,JMC]
CRYPT.DMP[ 2,JMC]
LFOL.DMP[F78,JMC]
REV1.DMP[F78,JMC]
CODE.DMP[ 2,JMC]
CONCEP.XGP[E76,JMC]
CHOWN.XGP[LET,JMC]
THESIS.XGP[LET,JMC]
TAKASU.XGP[LET,JMC]
FIRST.XGP[W79,JMC]
CRYPT.FAI[ 2,JMC]
VIS[ 2,JMC]
VIS1[ 2,JMC]
VITASK[ 2,JMC]
CHRI.LE1[ 2,JMC]
MIDF76[ 2,JMC]
PARC.2[ 2,JMC]
EC[ 2,JMC]
IBM.SNG[LIT,JMC]
SABBAT.LIT[LIT,JMC]
CHRIST[LIT,JMC]
BDAY.LIT[LIT,JMC]
NOTES[LOT,JMC]
SPACE[LOT,JMC]
LENG.LE1[LOT,JMC]
BLURB[LOT,JMC]
DEC20[LOT,JMC]
TASK[LOT,JMC]
P[LOT,JMC]
TERMIN[LOT,JMC]
BUDGET.2[LOT,JMC]
FRANKL.LE1[LOT,JMC]
LOTS.BLB[LOT,JMC]
MEASUR[LOT,JMC]
TERMIN.2[LOT,JMC]
BOARD[LOT,JMC]
LOTS.REP[LOT,JMC]
SHAW.LE1[LOT,JMC]
HOTER.BAY[PUB,JMC]
WEIZEN.REP[PUB,JMC]
HOTER.REC[PUB,JMC]
HOTER.OUT[PUB,JMC]
WEIZEN.LED[PUB,JMC]
HOTER.POX[PUB,JMC]
HOTER.WD[PUB,JMC]
WEIZEN.REV[PUB,JMC]
HOTER.BGB[PUB,JMC]
WEIZEN.ANS[PUB,JMC]
SUBSCR[PUB,JMC]
GLUGG2[LIT,JMC]
BIB.PUB[SEN,JMC]
TECHNO[SEN,JMC]
CALIF[SEN,JMC]
FEB14.NOT[SEN,JMC]
QUOTE.2[SEN,JMC]
STATEM.LET[SEN,JMC]
POSTER[SEN,JMC]
ASHLEY.COV[SEN,JMC]
SIGN[SEN,JMC]
ENDORS[SEN,JMC]
LONG[SEN,JMC]
LEGISL[SEN,JMC]
FEB8.78[SEN,JMC]
ENERGY.ST1[SEN,JMC]
MORTEM[SEN,JMC]
SUNDES[SEN,JMC]
MEET.F78[SEN,JMC]
GOSPER[SEN,JMC]
NOON15[SEN,JMC]
CALVO.L1[SEN,JMC]
TECHNO.NOT[SEN,JMC]
CGOLMA[W77,JMC]
CARGO.LIT[LIT,JMC]
CONFIG[LOT,JMC]
VIETNA.NS[S78,JMC]
ALGOL.SNG[LIT,JMC]
NAGEL.LE1[LET,JMC]
F77.OUT[LET,JMC]
∂17-Mar-79 0347 MRC new format packet
To: Dialnet at SU-AI
∂08-Mar-79 1035 Yeager at SUMEX-AIM new packet
==============================
2 bytes SOP framing
==============================
channel # | Opcode
(4 bits) | (4 bits)
------------------------------
MCHANS | packet number
(3 Bits) | (5 Bits)
------------------------------
RESERVED | Acknowledgement
(3 Bits) | (5 Bits)
------------------------------
1 byte Data size
==============================
1 byte Data
additional data bytes
as specified by Data Size
==============================
2 bytes checksum
==============================
2 bytes EOP framing mark
==============================
∂18-Mar-79 1049 CLT
I am at the lab working. Let me know when you are up.
∂19-Mar-79 0944 DPB Help!
To: "@HELP.DIS[1,DPB]" at SU-AI
For the past month or so I have been trying to recruit a CS105
instructor for spring quarter. On Friday I found out that one
of the people I was already planning on will not be available.
So I'm sitting 2 weeks from the start of the quarter needing
2 instructors for CS105. I'm close to exhausting my leads, so
I desperately need help in beating the bushes. If you know anyone
who is:
1. A competent teacher.
2. Modestly familiar with Pascal.
3. Without major spring quarter commitments.
4. Willing.
let me know ASAP.
There seem to be lots of folks with 1 & 2, but few (none so far)
with 3 & 4. Many of our PhD students who would be good candidates
are in critical places in their programs; I don't even want to ask
them. Maybe some of your research associates, visitors, industrial contacts,
etc. might be available.
Help!
-Denny
∂19-Mar-79 1423 DPB Meeting re Jacks space and teaching plans next year
We should meet to discuss details of the Jacks office arrangements.
Your office is probably best, since I need to make a trip to the
XGP soon anyway. Tuesday early afternoon OK? And as long as we're
at it I'd like to get suggestions about how to staff your courses
next year. -Denny
∂19-Mar-79 1458 DCO
To: "@GROUP.DIS[1,DCO]" at SU-AI
PROGRAM VERIFICATION AND ANALYSIS SEMINAR
No seminar this week.
∂19-Mar-79 1830 MRC Dialnet at SAIL
To: IAZ, JMC, LES
Just as strangely as the bug appeared, it disappeared in the new system!
Anyway, it works with Ignazio's FTP and my diagnostic again.
∂19-Mar-79 1941 100 : MRC
You left your briefcase in the front office. Its on the chair near the
scales.
∂19-Mar-79 2210 DPB
To: JMC at SU-AI, TW at SU-AI, DBL at SU-AI, TOB at SU-AI,
DPB at SU-AI, feigenbaum at SUMEX-AIM, buchanan at SUMEX-AIM,
lenat at SUMEX-AIM
The following rather long message can be found as CURRIC.AI[1,DPB]
at SAIL and as <DBROWN>CURRIC.AI at SUMEX with page marks at
appropriate places to allow convenient printing. I meant this
to be a quick summary of what I know about the plans for the
AI curriculum, but it isn't quick. Please look at this and
respond ASAP. We must turn in copy for Courses and Degrees on
Friday (3/23) and Carolyn is out.
I will try to summarize here the bits and pieces I have collected
concerning the AI curriculum. Information contained herein is
distilled from the abortive meeting a couple of weeks ago
attended by Bruce Buchanan, Doug Lenat, Tom Binford, and I and from
subsequent conversations with Lenat, Winograd, Feigenbaum,
and McCarthy.
!It is agreed upon by everyone that there should be one more AI
course. It should be a "course-type" course on intermediate
issues in AI. It would fall between what is currently 224 and
225. In parallel to other parts of the curriculum, it would
make sense to name the new course 225A, and rename the old 225
to be 225B. 225B would still be a "projects-type" course.
Because of leaves next year by Feigenbaum and McCarthy, it was
decided that we should not try to implement 225A during 79-80.
Planning during the next 6-9 months should result in a solid
proposal to include in next year's bulletin, to be implemented
in 80-81.
It looks to be a good idea to split the two versions of 206
into two separately numbered courses. McCarthy thinks that
the McCarthy-Talcott course (Recursive Programming and Proving)
should be numbered 122 to emphasize that "students should be
taught to prove properties of programs at an early stage."
The "AI programming" version of 206 would then be numbered 222.
222 could then assume that students know some LISP, allowing
more time to cover advanced topics.
I agree that students should be taught to prove properties of
programs at an early stage. In fact, I introduce elementary
notions of invariants in my version of CS105/6. In 107, I
introduce simple non-formal Floyd-style proofs of partial
correctness of Pascal programs. My quibble is that labelling
JMC's course at the 100 level will tend to inaccurately imply
that mathematically unsophisticated students are encouraged
to take this course. (From experience, many, many successful
CS107 students have never seen a proof by mathematical
induction until I show them one.) In accordance with the
department's stance opposing the notion of an undergraduate
CS major, I believe that, given a choice between an abstract
mathematics course (Analysis, Algebra, ...) and something like
206, one should do the math course (first).
As a benefit to the course, the instructors could continue to
insist on "real" program proving, and not the hand waving which
is all that some 107 grads are mathematically capable of.
Tom Binford stated that his Robotics course has grown to be
almost worth two courses. He does not want to implement this
change for 79-80, but plans to do so for 80-81. Maybe next
year we'll have 227A,B.
Terry Winograd does not want to stay "locked in" to the
265, 266, 365 sequence forever. The new 225A course would
be one of the courses interesting to him. He also gave me
an outline of a version of CS101 (the CS culture course)
which would be interesting. But due to other current commitments
he wants to continue with 265, 266, 365 for next year.
(I.e. again no change.)
!So for next year the roster of courses continues to be (with some
renaming, according to my current prejudices):
CS101 Culture
CS206 Recursive Programming and Proving
CS222 Programming languages for artificial Intelligence
CS224 As before
CS225 As before
CS226 Epistemological Problems of Artificial Intelligence
CS227 Robotics
CS265 As before
CS266 As before
CS365 As before
The roster of courses in the long term plan is:
CS101 Culture
CS206 Recursive Programming and Proving
CS222 Programming languages for artificial Intelligence
CS224 As before
CS225A Intermediate AI stuff
CS225B AI project course
CS226 Epistemological Problems of Artificial Intelligence
CS227A Robotics I
CS227B Robotics II
CS265 As before
CS266 As before
CS365 As before
I propose the following schedule and staffing for the courses next
year. Some of this was discussed at the meeting; since then I have
had some conflicting feedback. Anybody who doesn't believe this
schedule should contact me soon.
Autumn Winter Spring
-------------------------------------------------------------------
206 Talcott? 222 Lenat 101 Staff
225 Lenat 224 Lenat
227 Binford 226 ?McCarthy gone?
Weyrauch maybe?
265 Winograd 266 Winograd 365 Winograd
The move of 222 and 224 from current slot in spring to next year's
slot in winter is prompted by "making room" for 225A the following
year, by requests by students to do the AI course a little earlier
vis a vis the Comp., and by Doug's prejudices about when he'd like
to have his heavy teaching quarter.
Maybe 226 wants to move from winter to spring for next year, since
JMC won't be doing it anyway.
EAF suggested maybe Bruce would do 225 next fall. (Which would
leave Doug available for 101, if he wants to do that.)
!Long term schedule:
Autumn Winter Spring
-------------------------------------------------------------------
206 McCarthy 222 Staff 101 Staff
225B Staff 224 Staff 225A Staff
227A Binford 227B Binford
226 McCarthy (alt.)
258 McCarthy (alt.)
265 Winograd 266 Winograd 365 Winograd
This plan has 5 Staff slots to be covered by Lenat, Feigenbaum, and
Buchanan. Also remember that Terry will not want to do 265, 266, 365
always. We could, of course, do an alternating years thing.
JMC recently sent the following copy for the Courses and Degrees
bulletin. Doug might want to look at the CS222 description to see
if it accurately describes what he plans to do.
∂23-Feb-79 1337 JMC
Here are some course descriptions. Changing the number of 206 to 122
and introducing 222 was tentatively agreed by the Curriculum Committee
last year, but they or at least its chairman should review the matter.
CS226 will be considered for cross-listing in philosophy.
122. Recursive Programming and Proving - Recursive programming using
the LISP language and techniques for proving the corrrectness of
recursive programs. Computing with symbolic expressions
rather than numbers, e.g. algebraic expressions, logical expressions,
patterns, graphs, and computer programs.
Pattern matching and syntax directed computation. Preparation for
work in artificial intelligence is emphasized.
Prerequisite: 107 or equivalent ability to program.
222. Programming languages for artificial Intelligence - Advanced LISP,
Microplanner, QA4, Scheme. Production systems and other forms of
pattern-directed computation. Applications to problem solving and
language understanding.
Prerequisite: 122, 224 or consent of the instructor.
226. Epistemological Problems of Artificial Intelligence - Formalisms
for representing what a general intelligent program must know about
the world including facts about causality, ability, knowledge and
action. Modes of rigorous and conjectural reasoning for computer
programs. Connections with philosophy, especially philosophical logic.
Related material in logic and elementary set theory.
258. Mathematical Theory of Computation - Abstract syntax and formal
semantics of programming languages. Recursively defined and algolic
programs. Proving assertions about computer programs using formalisms of
Burstall, Cartwright, Floyd, Manna, McCarthy and Scott. The emphasis
is on functional programs rather than sequential. Use of
proof-checking programs. Prerequisite: 156 with 166 recommended.
**** reply to above
I still favor the renaming of 206 to 122. My experience and Carolyn
Talcott's in teaching the course showed that the proving part is
well withing the capabilities of the students who take the course -
even when the proving was introduced without a warning that might
scare of the non-mathematical. The way we do it, it just isn't
that difficult, especially since it is probably somewhat more
natural than the inductive assertions method for sequential
programs. I want to teach 226 in 1980-81, since I will have
spent a year working on just that material. Therefore, I would
prefer that the course not be given with that number or my course
description next year.
∂19-Mar-79 2216 MRC grimlacs
I really can't tell what happened from your description; the network could
have screwed you, you could have typed a funny EMACS command, there could
be a bug in DTN,... I'll have to look at it sometime when you're around.
∂19-Mar-79 2235 TOB your course
John
I wonder whether giving it a 100 series number
devalues your course for graduate students who are
your major audience.
Tom
A point that hadn't occurred to me. I'd better think about it.
∂20-Mar-79 0233 LLW
To: JMC
CC: LLW
∂20-Mar-79 0157 JMC
See CHEMEM[W79,JMC].
[John: CHEMEM seems to be a cogent summary of our discussion earlier this
evening, with some additional helpful detail on exactly what you feel you
need in the way of cellular automata in the memory 'atoms'. I'm most
intrigued by the nearly total lack of required spatial localization of the
memory 'atoms' in your scheme, albeit concerned about some of the
signal-to-noise problems associated with long-term use of such a memory.
I'll keep poking at some possible realizations, and let you know of any
substantial progress. Lowell]
∂20-Mar-79 0955 PAT aimemos
The numbers are as follows:
AIM-324, STAN-CS-79-717
Recursive Programs as Functions in a First Order Theory
Cartwright and McCarthy
ARPA, NSF
AIM-325, STAN-CS-79-724
First Order Theories of Individual Concepts and Propositions
McCarthy
??
AIM-326, STAN-CS-79-725
Ascribing Mental Qualities to Machines
McCarthy
??
I need to know what funding to cite for 325 and 326. Also, in figuring
out how many copies to print, how many of each do you think you need for
you own personal use and are you going to use any for any classes.
Also, for your information, Bob Filman's thesis is
AIM-327, STAN-CS-79-727
The Interaction of Observation and Inference
Robert Elliot Filman
ARPA
Thanks, Patte
CONCEP and MENTAL should cite both ARPA and NSF.
Since I won't be teaching next year and hope to revise the subject matter
then, no copies are needed for classes. I guess I will need about
40 copies each.
∂20-Mar-79 1109 Boyer at SRI-KL (Bob Boyer) A Talk on Automatic Theorem Proving
Date: 20 Mar 1979 1109-PST
From: Boyer at SRI-KL (Bob Boyer)
Subject: A Talk on Automatic Theorem Proving
To: Lenat at SUMEX-AIM, JMC at SU-AI, Pattis at SRI-KL, RWW at SU-AI
Please forward this message on to those who might be interested.
D. Pastre of the University of Paris will deliver a lecture
at 10:30 am on Monday, March 26 in room J1018 of SRI's
Building 30 entitled:
The Importance of Representation, Use of Knowledge, and
Observation of Human Behavior in Automatic Theorem Proving
Pastre will illustrate her ideas with some examples of her
theorem proving program for set theory, which is described
in "Automatic Theorem Proving in Set Theory," AIJ 10, 1,
Feb. 78.
She will be visiting at SRI through Wednesday, March 28.
She is currently visting W.W. Bledsoe at the University of
Texas.
-------
∂20-Mar-79 1116 PAT lib.lst
the first page of LIB.LST[PUB,PAT] contains a list of the papers on your
reading shelf
∂20-Mar-79 1440 DON
Please mail me a one-line description of DIALNET for HELP HELPER.
To MRC: Send him this with a reference to additional documentation.
∂21-Mar-79 0307 MRC Forwarded without comment
To: JMC, LES
∂21-Mar-79 0306 REM at MIT-MC (Robert Elton Maas) PCNet change of direction
Date: 21 MAR 1979 0606-EST
From: REM at MIT-MC (Robert Elton Maas)
Subject: PCNet change of direction
To: MRC at SU-AI
At tonite's meeting, it was the concensus that we have spent too much
time trying to cram a good protocol into 8k PET BASIC and documenting
it, and not enough time working on the higher level of SNDMSG/READMAIL.
We are now going to temporarily suspend efforts in lower-level protocol
and put in a crash effort to adapt PAN (a modem-groking FTP program
written by Doug Guage of the Net-Works in San Diego, inventor of the
IEEE-bus modem used on the PET) for PCNet-style FTP/MAIL but without
all the error-checking and automatic operation. Hopefully after a
large community of users are exchanging electronic mail (with errors
caused by phone-line noise), they will realize how nice the idea is
and also memory costs will be low enough that PET owners can expand
to 16k so that PCNet protocol can be mated to the crude-electronic-mail
program to yield a complete mail system.
(FYI, as principal developer of the "other" dial-telephone protocol)
Personally I'm beginning to think most people don't want electronic
mail, even if they are shown how nice it can be. Maybe there really
isn't a market for ANY dial-phone network. Well, after PAN is ready
for general use, we'll see if any consumer demand occurs.
∂21-Mar-79 1314 TOB soroka appointment
John
Ed has approved making an offer to Prof Barry Soroka
as Research Associate. That falls within our original
agreement and does not involve a request for additional
space.
Tom
∂21-Mar-79 1928 REM
To: JMC at SU-AI, MRC at SU-AI, KLH at MIT-MC,
GREEP at RAND-UNIX
∂19-Mar-79 1016 Fylstra at SRI-KL FTP via direct-dial?
Date: 19 Mar 1979 0955-PST
From: Fylstra at SRI-KL
Subject: FTP via direct-dial?
To: pbaran at USC-ISI, ffm at MIT-MC, cmr at MIT-MC,
To: fjw at MIT-MC, wilber, geoff at SRI-KA, rem at SU-AI,
To: tvr at SU-AI
Via: Rand-Unix
To: Msggroup@mit-mc, Header-People@mit-mc
Subject: Dawning of a New Age?
From: Dcrocker at UDEE
Reply-to: Dcrocker at Rand-Unix
Date: Sun, 18 Mar 79 16:19-EST
Redistributed-To: kunzelman, rom, craighill, fylstra, poggio
Redistributed-By: PICKENS at SRI-KL
Redistributed-Date: 19 Mar 1979
Please to note the From field of this message. As of a couple of
days ago, we are finally able to send messages onto the ArpaNet,
from the University of Delaware. We use dial-up access to a Mail
Relay host (currently Rand-Unix). We should shortly have
automated pickup of mail, also.
The current hack is to have return mail go to our accounts on
Rand. The longer-term solution will remove this and have it go
"directly" back to Delaware.
You might be interested to know that we use 300 baud, which gives
us an effective bandwidth of about a kilobyte per minute. As
atrocious as that sounds, it obviously is somewhat useful. Only
requirement is that the user not be expected to sit and watch it,
so of course we have a background process do the dial-up and
transfer.
Dave.
Folks,
sounds like Dave Crocker is using some kind of FT protocol to
get his mail off of the Rand system via 300 baud direct dial lines.
Does anyone know anything about this?
Dave
-------
[REM - Hmmm, after all that fuss about PCNet is forbidden to
have an automatic gateway into or out-of or through the Arpanet,
because then random people would be able to use Arpanet facilities,
those people at RAND are doing the same thing and getting by wit
it? I don't know what the RANDers are really doing. If they
truly have a gateway, it might set a precedent if it is allowed.
Any information, ideas, skuttlebut, on this rumor?]
∂21-Mar-79 2003 ME
You are exceeding your disk quota.
Files that occupy space beyond your quota are subject to purging,
** ESPECIALLY if there are less than 1000 free disk blocks in the system! **
If you don't delete some of your files, the purger will.
Your disk quota is: 2000
Your files occupy 2367
∂21-Mar-79 2040 REM via AMES-TIP POX and your file
To: JMC, PAT
Your file has two instances of \; at the end of justify mode, which
don't do the expected thing hence are printed on console. Also you
do something that isn't supported, using \& to patch together overlays
which don't have zero kern. It isn't clear what should be done with
such overlays, in regard to \& command, hence this case should be
avoided since it may change suddenly if it is ever decided what should
be done in such case and that is different from what happens now if
you proceed from the two error messages by CTRL-META-P as I did just
now. You should look at your output and see if it is good. If it is,
since POX isn't being developed, your output although not supported
might continue to be good until POX is totally suplanted by TEX sometime
in 1984 or 1985. (It would supplant POX now if the five missing
features were installed in TEX. Perhaps you can hire some student to
go into TEX and finish it?? I really would like to have TEX do what
everyone wants instead of just what Knuth wants for his book. It does
have some nice things that POX and PUB don't have.)
∂22-Mar-79 0858 BPM
To: LLW, JMC
n718 0330 22 Mar 79
BC-Teller 03-22
By BRIAN J. KELLY
(c) 1979 Chicago Sun-Times
Chicago - Government secrecy hinders scientific development
by preventing ''a free exchange of ideas'' and may have
caused the United States to lag behind the Soviet Union in nuclear
development, Edward Teller, the prominent nuclear physicist.
Teller, considered the father of the hydrogen bomb, said he would
like to see the amount of classified material ''drastically''
reduced. His comments come as the federal government is
engaged in an attempt to prevent the Progressive magazine
from printing portions of the H-bomb formula.
Teller is also the author of an Encyclopaedia Americana entry
on the hydrogen bomb which the Progressive contends has much
of the same information they are being barred from publishing.
In an interview this week Teller refused to discuss aspects
of the Progressive case. ''That case is in the courts and I have
no intention of getting involved.''
But the physicist, long an opponent of government secrecy, was
willing to give his views on the secrecy question and included
some veiled references to the Progressive case.
''First, let me say that what is known to a million people
is not a secret. Whether you like it or not, it will leak
out eventually,'' he said.
However Teller, a Hungarian immigrant, said he disagrees with
anyone who tries to circumvent security laws to make a point.
''I have great respect for the U.S. system of laws. If you
are going to do something about secrecy, you should do it
through the Congress.''
The physicist said he has testified on the question before
Congress several times and has tried to influence the various
government projects he's worked on to reduce secrecy requirements.
Teller, who is a fellow at Stanford University's Hoover
Institute, is a consultant to government nuclear laboratories
at Los Alamos, New Mexico and Berkeley, Calif.
''At Livermore (the Berkeley laboratory) I've had the experience
of talented scientists who didn't want to work there because
of the secrecy requirements,'' he said.
Teller said secrecy laws also erect barriers between allies.
''One of the reasons (French President Charles) De Gaulle left
NATO was because we were reluctant to share technology,''
he stated.
As an example of the benefits of sharing ideas, Teller
compares the nuclear program with research in electronics: ''Nuclear
technology is secret and we're not ahead of anyone. Information
about computers and electroncis is not classified and we're far
ahead of the world.''
Teller, who has access to an enormous range of government
secrets, said he sometimes has trouble remembering just what
is classified. He recalled an incident where he let a bit of
classified information slip out during a speech.
''The security people called me on the carpet. I asked them if
the information should be really be classified. They looked into it,
and came back and told me, 'You're right. It's not really
that important,' '' he recalled.
Teller said he'd like to see a system where virtually all
documents could only be classified for one year. ''There may
be a few exceptions and you would have a small panel that would
make those exceptions,'' he said.
He would distinguish between classified and confidential
information, defining confidential as things such as military
troop moments and assignment of CIA agents. But even major
troop movements would fall under the one-year expiration.
''Something like the invasion of Cambodia in 1970 may have
required secrecy when it happened, but all the information about
that operation should have been public after a year,''
Teller said.
∂23-Mar-79 0218 RWG
To: JMC, PAT, REM
instead of struggling with the old macros and poxes, i thought
it would be easier to create RELAT.POX[1,RWG], which uses the new macros and works
in the latest POX. if i hopelessly mangled the content while massaging the form,
note the following typos in RELATI.POX[CUR,JMC]:
univeral
She write to him
whe doesn't love
automation (instead of automaton)
automatia
missing "(" from eq 1)
numerous failures to boldface TIMES, HIST, etc
a prime overstriking the comma in ", but in the trivial case..."
A problem which i didn't fix is that all the primes look like apostrophes.
Beware, however, the genuine(?) apostrophe in "for all j's". Also,
"... depend only on the past of the state ..." sounds slightly garbled maybe.
∂23-Mar-79 0701 PJH via LONDON CASBS
HI JOHN.THANKS (THO THATS RATHER INADEQUATE) FOR GETTING ME OVER THERE.
WHO ELSE WHO ELSE IS COMING, DO U KNOW YET? AM BUSYING GETING THINGS
FINISHED HERE BEFORE COMING. A THOUGHT: ANY CHANCE OF DOING SOME (UNPAID!)
TALKS AT COMPUTER SCIENCE ON MATH. SEMANTICS OF PROGRAMMING LANGUAGES? I
WILL BE CO WRITING A BOOK ON THIS AND WILL NEED FEEDBACK FROM BRIGHT GUYS.
AM CURENTLY LOOKING AT YR MINIMAL MODELS & RELATIONS TO REITERS STUFF AND
MCDERMOTT, WILL WRITE WHEN I CAN . SEE YOU SOON. PAT.
∂23-Mar-79 1019 FRG CS258
John, Just a reminder that the CS258 grade sheets should be returned to Jake
Brown (Polya 253) no later than 5:00 Monday. Thanks. Jake
∂23-Mar-79 1044 FRG CS390 Sec19
John, Another reminder: The grade sheets for CS390 Sec19 should be returned
to Jake Brown (polya 253) no later than 5:00 Monday. Thanks. Jake
∂23-Mar-79 1047 DEW Oral exam
To: JMC at SU-AI, buchanan at SUMEX-AIM, lenat at SUMEX-AIM,
berliner at CMU-10A
CC: DEW at SU-AI
I propose Thursday, May 10, 1979 as the date of my oral exam.
I must get all of you to agree on a date more than a month in advance.
If this date is fine with you, that's great. If the answer is no or maybe,
could you please give me a range of dates near that time for which you
could give me a definite yes.
Thanks, Dave
∂23-Mar-79 1312 PAT Jeff Ullman
Bob Tarjan called and has set up a time for you to talk with Jeff Ullman
on Friday, March 30 at 1:30pm.
∂24-Mar-79 2144 REM via AMES-TIP Second draft of my data-compression paper ready to review
To: JMC at SU-AI, RWG at SU-AI, WD at SU-AI,
WILBER at SRI-KL, LAUREN at UCLA-SECURITY
The file 79219A.WRU[1,REM] contains the second draft of my paper
(to try to publish soon) describing all the major results of my
data-compresson R+D up to but not including my patentable method
of making interval-refinement practical. It still needs bibliographic
references; I hope the NTIS published-search I am ordering will give
me most of the earlier work in this area, directly or indirectly.
I haven't decided what else it needs, but if you find any serious
deficiencies please tell me. I also haven't decided where to
publish, but engineering journals such as IEEE PROCEEDINGS seems to
be where most data-compression is published.
∂25-Mar-79 0302 POURNE at MIT-DMS (Jerry E. Pournelle) Mostly hello
Date: 25 Mar 1979 0601-EST
From: POURNE at MIT-DMS (Jerry E. Pournelle)
To: JMC at MIT-DMS, at SU-AI
Subject: Mostly hello
Message-id: <[MIT-DMS].105198>
Youonce demonstrated the most marvelous information
system, dialing Stanford from my office, and getting summaries
and news and all that sort of good thing.
Is there any way to have access to it? I confess I
find your information utility idea more likely every day--not that
I ever doubted that it would happen, but this damned soon?
We have bought a story from your studen Moravec for
Volume one of THE ENDLESS FRONTIER>
I have open Volume TWO of ENDLESS FRONTIER and a new
anthology called THE SURVIVAL OF FREEDOM, and would love to
have a good essay or other non-fiction thought piece from you
in either or both.
Lots of best regards,
(Is there life after taxes?)
Jerry Pournelle
∂26-Mar-79 0813 BS Telephone Message
800 2236886 toll free
Your number at the Lab didn't answer, and I took the following message for
you:
Please call Walter Glanze, Bantam Books (collect if you wish). Number:
(212) 765-6500, extension 249. Mr. Glanze is a lexicographer and he wants
to include the word "wetware" in the dictionary. He would like your opinion
as to how often the word is used, and whether you think it should be in-
cluded.
Betty
∂26-Mar-79 1011 PAT phone call
from Dean Rosse. He would like to meet with you from 2:45-3:15 on April 2
regarding the department chairmanship. I put it in your calendar. If the
time is not alright then I told her I would call her back otherwise it is
confirmed.
∂26-Mar-79 1533 DEW My oral exam
John,
I earlier suggested May 10 as a date for my oral exam. I would now like
to retract that suggestion until I've had a chance to talk to you about
it. I'm not exactly sure what the oral is for and how far done I should
be before doing it. I'll look you up this week at the lab.
Get well soon, Dave
∂26-Mar-79 1648 DPB
John, In the bulletin copy you sent us re 258, you list 156 as a
prerequisite with 166 recommended. We have no 166, what did you mean?
-Denny
I meant the logic course taught in math or philosophy.
∂27-Mar-79 1039 FRG my whereabouts
I am at Polya until about 2pm, 74776.
∂27-Mar-79 1709 ME
You have exceeded your disk quota.
The files listed below have been purged to reduce your disk
area to your quota of 2000
Before purging, your files occupied 2371
KNOW.LST[S78,JMC]
DIALNE.LST[DIA,JMC]
CRYPT.DMP[ 2,JMC]
LFOL.DMP[F78,JMC]
REV1.DMP[F78,JMC]
CODE.DMP[ 2,JMC]
CHEMEM.XGP[LET,JMC]
CRYPT.FAI[ 2,JMC]
VIS[ 2,JMC]
VIS1[ 2,JMC]
VITASK[ 2,JMC]
CHRI.LE1[ 2,JMC]
MIDF76[ 2,JMC]
PARC.2[ 2,JMC]
EC[ 2,JMC]
CARGO.LIT[LIT,JMC]
IBM.SNG[LIT,JMC]
SABBAT.LIT[LIT,JMC]
CHRIST[LIT,JMC]
BDAY.LIT[LIT,JMC]
NOTES[LOT,JMC]
SPACE[LOT,JMC]
LENG.LE1[LOT,JMC]
BLURB[LOT,JMC]
DEC20[LOT,JMC]
TASK[LOT,JMC]
P[LOT,JMC]
TERMIN[LOT,JMC]
BUDGET.2[LOT,JMC]
FRANKL.LE1[LOT,JMC]
LOTS.BLB[LOT,JMC]
MEASUR[LOT,JMC]
TERMIN.2[LOT,JMC]
BOARD[LOT,JMC]
LOTS.REP[LOT,JMC]
SHAW.LE1[LOT,JMC]
HOTER.BAY[PUB,JMC]
WEIZEN.REP[PUB,JMC]
HOTER.REC[PUB,JMC]
HOTER.OUT[PUB,JMC]
WEIZEN.LED[PUB,JMC]
HOTER.POX[PUB,JMC]
HOTER.WD[PUB,JMC]
WEIZEN.REV[PUB,JMC]
HOTER.BGB[PUB,JMC]
WEIZEN.ANS[PUB,JMC]
SUBSCR[PUB,JMC]
GLUGG2[LIT,JMC]
BIB.PUB[SEN,JMC]
TECHNO[SEN,JMC]
CALIF[SEN,JMC]
FEB14.NOT[SEN,JMC]
QUOTE.2[SEN,JMC]
STATEM.LET[SEN,JMC]
POSTER[SEN,JMC]
ASHLEY.COV[SEN,JMC]
SIGN[SEN,JMC]
ENDORS[SEN,JMC]
LONG[SEN,JMC]
LEGISL[SEN,JMC]
FEB8.78[SEN,JMC]
ENERGY.ST1[SEN,JMC]
MORTEM[SEN,JMC]
SUNDES[SEN,JMC]
MEET.F78[SEN,JMC]
GOSPER[SEN,JMC]
NOON15[SEN,JMC]
CALVO.L1[SEN,JMC]
TECHNO.NOT[SEN,JMC]
CGOLMA[W77,JMC]
CONFIG[LOT,JMC]
VIETNA.NS[S78,JMC]
ALGOL.SNG[LIT,JMC]
MANNA.PRA[LET,JMC]
NAGEL.LE1[LET,JMC]
F77.OUT[LET,JMC]
∂28-Mar-79 0940 Mcgoveran at SUMEX-AIM Faculty Meeting to Vote on Degree Candidates
Date: 28 Mar 1979 0939-PST
From: Mcgoveran at SUMEX-AIM
Subject: Faculty Meeting to Vote on Degree Candidates
To: faculty.list:
This is a notice for a faculty meeting on Tuesday, April 3, to vote on the
degree candidates for next quarter. The meeting will be in Polya 204 at
2:30.
Please let me know whether you will be attending. My extension is 73264.
I am McGoveran at SUMEX.
Mary
-------
∂28-Mar-79 1005 PAT whereabouts day 2
I will be at Polya (again) until About 2pm. If you need me I am reachable
at 74776. Hope you're feeling better. Patte
∂28-Mar-79 1413 REM via AMES-TIP Arpanet protocol violation by SU-AI
To: LES, JMC
Evidence from CCA in the DFTP log is that SU-AI repeatedly violates
Arpanet protocol by getting confused about message and/or bit allocations,
resulting in data overruns sometimes and allocation-deadlocks other times.
Allocation-deadlocks cause DFTP to permanently hang, until the user-DFTP
program is aborted (by zeroing core image, by gunning down job, or by
autologout a half hour or so after the hangage takes effect). The only
way to recover is to start up a new DFTP from scratch. MRC is aware of
the problem, but is probably too busy with SCORE and DIALNET to work on it.
I'm not sure. Does either of you have an opinion to express on whether
this problem is worthy of fixing, and if so then who has time to work
on it? Now that I know what the problem is, I can work around it, getting
a new DFTP each time the old one hangs, so it isn't urgent for me in
terms of DFTP, but the same bug might be causing trouble elsewhere and
the cause (the bug) might not yet have been discovered. Note that it is
possible the problem is really at CCA or in the subnet, but we sort-of
believe it is a SU-AI bug from the evidence so far.
∂28-Mar-79 1439 REM via AMES-TIP Draft version of paper on my data-compression results.
To: "@CRUNCH.DIS[1,REM]" at SU-AI, "@CRUPUB.DIS[1,REM]" at SU-AI
I haven't gotten enough feedback from the two people at SU-AI who have
already been notified of this file, so I am expanding the audience in
an attempt to get more critical feedback. This paper, intended for
publishing in a journal, describes all my major data-compression results
to date, except for the invention for which I am applying for patent.
RWG says the paper is actually understandable, an improvement over my
usual documentation. -- The filename at SU-AI is 79219A.WRU[1,REM].
Feel free to FTP and list it for your private reading only. If you
know someone who is seriously interested in looking at it also, please
send me his/her name and network (or US) address.
∂29-Mar-79 2028 Jrobinson at SRI-KL (Jane Robinson) supper
Date: 29 Mar 1979 2029-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: supper
To: mccarthy at SU-AI
Am surfacing from prolonged contract proposal writing. Haven't seen
you in some time. How are you? And would you care to come over
tomorrow evening (@ 7 pm) for pot-luck supper?
Jane
-------
I'm fine except for intestinal bug that will probably still prevent supper
- unless it consists entirely of Jello and the like. Moreover, I have a
concert Friday also.
How would you like do come here next Wednesday evening. We would also
invite our visiting scholar from Peking University Ma Xiwen, who will be
at the AI Lab for a year and is interested in reasoning in natural
language. I should be better by then, or else we can all have Kung Pao
Jello.
∂30-Mar-79 0710 Feigenbaum at SUMEX-AIM Chowning and your letter
Date: 30 Mar 1979 0652-PST
From: Feigenbaum at SUMEX-AIM
Subject: Chowning and your letter
To: jmc at SAIL
cc: les at SAIL
John,
I am becoming increasingly concerned with John Chowning's
hold-fast policy or attitude with respect to the
music-computer-space problem. Apparently Dean Rosse has been unable to
persuade John to return from paris to sort out this whole problem with us.
Thus there seems to be no forward motion in the face of a request from me in
your letter that the problem be treated with urgency and
that John come back to negotiaate it.
What do you think should be done at this point?
I'll be back from North Carolina tomorrow and we can discuss it.
Ed
-------
∂30-Mar-79 0811 Jrobinson at SRI-KL (Jane Robinson) Understand the problem
Date: 30 Mar 1979 0811-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: Understand the problem
To: mccarthy at SU-AI
Menu for tonight: broiled chicken, baked potato, vegetable that is
not green or leafy, baked custard. Maybe chicken soup. Welcome if
you feel like it. All I need to know is whether to bake one potato
or two at 6 pm.
Accept your invitation for next Wednesday. Do I have
to sing for my supper by reasoning in natural language? Or can I just
bring some jello.
Jane
-------
∂30-Mar-79 0913 Bmoore at SRI-KL (Bob Moore) Reminder
Date: 30 Mar 1979 0913-PST
From: Bmoore at SRI-KL (Bob Moore)
Subject: Reminder
To: grosz, SAGALOWICZ, TENENBAUM, KONOLIGE, hobbs, jmc at SAIL,
To: rww at SAIL
Don't forget the party at our house tomorrow (Saturday) night at 8:00PM.
The address is 737 DeSoto Dr., Palo Alto. Call 328-4253 if you get lost.
See you there.
Bob
-------
∂30-Mar-79 1108 FRG grading procedures
John, Received a note from the Registrar's Office that the Academic Senate
does not allow the assignment of letter grades and 'P' in the same class.
They have changed all P's to symbol '+' on the End-Quarter Report you
submitted. (CS258)
Jake Brown
∂30-Mar-79 1649 RPG LISP reminder
Earlier we discussed the possibility of my helping with
the "applied" parts of your LISP book. If that is still a possibility
with you, it might be nice to further consider these plans before
too many others are made that might preclude them.
-rpg-
∂30-Mar-79 2101 Jrobinson at SRI-KL (Jane Robinson) Tuesday's fine.
Date: 30 Mar 1979 1737-PST
From: Jrobinson at SRI-KL (Jane Robinson)
Subject: Tuesday's fine.
To: Mccarthy at SU-AI
You don't say whether you are coming tonight, so I'll put two potatoes
on to bake anyway.
Jane
-------
∂31-Mar-79 1245 SSO Elephant
To: JMC
CC: FB
I'd like to hear about it (even though I'm biased against introducing time
explicitly). When is a good time?
I don't see how we can get together without intrducing an explicit
time. Monday 1:30 at your place or any other afternoon at the
AI Lab are possible for me.
∂31-Mar-79 1531 DCO corky
Corky will try to phone you Monday and will write you also. He
wants to come this summer and would like some additional support (which will
presumably be the point of his phone call to you!)
∂31-Mar-79 2039 ME disk drives
∂31-Mar-79 1403 JMC 3 user packs
I assume this is just for debugging. As soon as it is safe, there should
be another file pack.
ME - Right. As soon as it is safe. So far one drive is already marginal
and not in use, but it may be fixable since it merely has an intermittent
ID plug detector.
∂01-Apr-79 1149 Mark Crispin <Admin.MRC at SU-SCORE> Dialnet progress
Date: 1 Apr 1979 1145-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Subject: Dialnet progress
To: JMC at SU-AI
Well, I generated a 3A monitor with Dialnet in it and it seems not to crash
the system (but then again it doesn't do useful Dialnet either). I think I
know what's wrong though. I didn't get much else done because GSB is having
hardware problems. I was told that it has been losing this way for a few
weeks. It took me about 15 minutes each shot to get a system loaded; there
were many front-end I/O errors. Bad shape.
I'm going to talk with them Monday and find out when this is going to be
fixed. I won't do anything tonight because it'd be rather futile in the
current state the hardware is there.
-------
Please don't give up easily.
∂01-Apr-79 1150 Mark Crispin <Admin.MRC at SU-SCORE> Sumex ''Dialnet''
Date: 1 Apr 1979 1149-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Subject: Sumex "Dialnet"
To: Yeager at SUMEX-AIM
Cc: JMC at SU-AI
Hi -
How's it going over there with TTYFTP? We now have an installed user and
server FTP program listening on the Dialnet port at SAIL. Do you have any
VADIC 3400 modems to run TTYFTP on? JMC wants to have some kind of demo
of two machines talking Dialnet with each other. By the way, I have our
new protocol running at SAIL as well. GSB is having hardware flakies in
its front-end filesystem (ie, it is a battle to reload it!), so 'thought
I'd see how you were doing.
-- M
-------