perm filename W82.IN[LET,JMC] blob
sn#864698 filedate 1982-04-01 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00473 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00052 00002 ∂05-Jan-82 0900 CL.BOYER at UTEXAS-20 Using your system at Stanford
C00063 00003 ∂05-Jan-82 2116 Oded Anoaf Feingold <OAF at MIT-MC> energy digest #3004
C00099 00004 ∂06-Jan-82 0038 RPG Mail file management
C00101 00005 ∂07-Jan-82 0951 RWW
C00117 00006 ∂07-Jan-82 1601 JJW
C00122 00007 ∂07-Jan-82 1634 YK
C00124 00008 ∂07-Jan-82 2105 KLC
C00127 00009 ∂07-Jan-82 2123 KLC
C00128 00010 ∂07-Jan-82 2359 JMC*
C00129 00011 ∂08-Jan-82 0404 Bill Gosper <RWG at MIT-MC> life block
C00130 00012 ∂08-Jan-82 0910 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Order 2494
C00132 00013 ∂08-Jan-82 0911 JK lispx
C00133 00014 ∂08-Jan-82 0951 JJW SCORE account for Boyer-Moore system
C00134 00015 ∂08-Jan-82 1326 Gene Golub <CSD.GOLUB at SU-SCORE> funds for computing
C00135 00016 ∂08-Jan-82 1543 FFL Final Report on your NSF grant
C00136 00017 ∂08-Jan-82 1545 Susan Hill <CSD.HILL at SU-SCORE> Computer accts
C00137 00018 ∂08-Jan-82 2015 JK ekl bugs and lispx
C00140 00019 ∂09-Jan-82 0124 Bill Gosper <RWG at MIT-MC>
C00141 00020 ∂09-Jan-82 0815 KAHN at USC-ISI Re: ARPA Order 2494
C00143 00021 ∂09-Jan-82 0912 JK
C00146 00022 ∂09-Jan-82 0933 JK
C00147 00023 ∂10-Jan-82 0624 JK
C00149 00024 ∂10-Jan-82 0936 JK
C00150 00025 ∂10-Jan-82 0939 JK flat
C00151 00026 ∂10-Jan-82 1005 JK more on flat
C00152 00027 ∂10-Jan-82 2321 reid@Shasta (SuNet) proposal
C00153 00028 ∂11-Jan-82 1306 CLT
C00157 00029 ∂11-Jan-82 1549 CL.MOORE at UTEXAS-20 Theory of Computation
C00161 00030 ∂11-Jan-82 1550 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Cooperative Research with JPL
C00163 00031 ∂11-Jan-82 2246 JK bug
C00165 00032 ∂12-Jan-82 0809 JK
C00167 00033 ∂12-Jan-82 1033 Jrobinson at SRI-AI Stanford Linguistics Colloquium of interest
C00169 00034 ∂12-Jan-82 2146 RWW
C00175 00035 ∂13-Jan-82 1014 Marilynn Walker <CSD.MWALKER at SU-SCORE> MTC Quals
C00177 00036 ∂13-Jan-82 1135 JJW SCORE account for Boyer-Moore system
C00178 00037 ∂13-Jan-82 1142 JK ekl bug fixed
C00179 00038 ∂13-Jan-82 1436 JK ekl bug
C00185 00039 ∂13-Jan-82 2258 JK ekl bugs
C00186 00040 ∂14-Jan-82 1309 ME system load bottom
C00190 00041 ∂15-Jan-82 0934 JJW
C00204 00042 ∂15-Jan-82 1557 FFL
C00211 00043 ∂15-Jan-82 1617 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Advanced GRE
C00212 00044 ∂15-Jan-82 1633 Buchanan at SUMEX-AIM Re: Advanced GRE
C00213 00045 ∂15-Jan-82 1650 Irmgild Schack <CSD.IRMGILD at SU-SCORE> [Gene Golub <CSD.GOLUB at SU-SCORE>: New Associate Chairman]
C00215 00046 ∂15-Jan-82 1822 Aaai-Office at SUMEX-AIM Robinson Agreement and Accounting Documents
C00254 00047 ∂16-Jan-82 1555 RPG Lisp Conference
C00265 00048 ∂17-Jan-82 1551 CLT
C00267 00049 ∂18-Jan-82 1113 Susan Hill <CSD.HILL at SU-SCORE>
C00268 00050 ∂18-Jan-82 1205 reid@Shasta (SuNet)
C00274 00051 ∂18-Jan-82 1332 FFL
C00277 00052 ∂18-Jan-82 2126 JJW E/EKL interface
C00281 00053 ∂19-Jan-82 0853 FFL
C00285 00054 ∂19-Jan-82 0854 FFL
C00286 00055 ∂19-Jan-82 1616 pratt@Shasta (SuNet) Sun@BBN
C00291 00056 ∂19-Jan-82 1653 JK
C00293 00057 ∂20-Jan-82 1259 JJW Class notes and Lisp textbook
C00303 00058 ∂20-Jan-82 1427 FFL Course in Logic Programming and PROLOG starting soon.
C00305 00059 ∂20-Jan-82 1742 David Warren <WARREN at SRI-AI> Antonio Porto's Visit
C00306 00060 ∂21-Jan-82 0846 LOUNGO at RUTGERS DCS Technical Reports
C00317 00061 ∂21-Jan-82 1250 JJW EEKL single-character macros
C00319 00062 ∂21-Jan-82 2312 Robert Elton Maas <REM at MIT-MC>
C00331 00063 ∂22-Jan-82 0104 Robert Elton Maas <REM at MIT-MC>
C00332 00064 ∂22-Jan-82 1133 JJW EKL syntax
C00333 00065 ∂22-Jan-82 1155 FFL
C00334 00066 ∂22-Jan-82 1338 FFL
C00336 00067 ∂22-Jan-82 1512 FWH Prolog talk at SRI
C00340 00068 ∂22-Jan-82 1611 FFL Visitor from MIT on Feb. 3
C00341 00069 ∂22-Jan-82 2017 Ed Katz <katz.Hp-Labs@UDel> Prolog Course
C00353 00070 ∂23-Jan-82 0226 Arthur Keller <CSD.KELLER at SU-SCORE> Gray Tuesday
C00355 00071 ∂23-Jan-82 1000 JMC*
C00358 00072 ∂23-Jan-82 1038 TOB
C00359 00073 ∂23-Jan-82 1101 TOB
C00363 00074 ∂23-Jan-82 1112 TOB
C00366 00075 ∂23-Jan-82 1952 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Meeting
C00371 00076 ∂23-Jan-82 1957 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Curriculum ctte meeting
C00373 00077 ∂23-Jan-82 2049 Lowell Wood <LLW at S1-A> Rewarding Bad Behavior
C00381 00078 ∂24-Jan-82 1251 CLT
C00385 00079 ∂24-Jan-82 1747 Jim McGrath <CSD.MCGRATH at SU-SCORE> FBI action on our computer breakin?
C00389 00080 ∂24-Jan-82 2138 Feigenbaum at SUMEX-AIM two things
C00400 00081 ∂24-Jan-82 2157 LLW One-Way Exchanges
C00427 00082 ∂24-Jan-82 2246 Kim.fateman at Berkeley NIL/Macsyma
C00431 00083 ∂25-Jan-82 0028 Lowell Wood <LLW at S1-A> Recklessness
C00432 00084 ∂25-Jan-82 0300 ARK Yoram Moses request advisor change
C00433 00085 ∂25-Jan-82 0831 Susan Hill <CSD.HILL at SU-SCORE> Re: Gosper
C00434 00086 ∂25-Jan-82 0940 TOB lunch
C00435 00087 ∂25-Jan-82 1000 JMC*
C00436 00088 ∂25-Jan-82 1000 JMC*
C00437 00089 ∂25-Jan-82 1114 TW Meeting today
C00442 00090 ∂25-Jan-82 1214 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Curriculum mtg.
C00444 00091
C00447 00092 ∂25-Jan-82 2118 ARK Advisors
C00452 00093 ∂26-Jan-82 0823 FFL
C00453 00094 ∂26-Jan-82 0823 FFL
C00454 00095 ∂26-Jan-82 0921 FFL
C00456 00096 ∂26-Jan-82 1138 Woody Bledsoe <ATP.Bledsoe at UTEXAS-20> [CS.DALE: Turing Award]
C00459 00097 ∂26-Jan-82 1311 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Reception for Jerry SALTZER
C00460 00098 ∂26-Jan-82 1343 David Warren <WARREN at SRI-AI> Antonio Porto's Visit
C00461 00099 ∂27-Jan-82 1000 JMC*
C00462 00100 ∂27-Jan-82 1008 FFL
C00463 00101 ∂27-Jan-82 1048 FFL
C00467 00102 ∂27-Jan-82 1136 CLT From SGF
C00468 00103 ∂27-Jan-82 1212 Nancy Dorio <CSD.DORIO at SU-SCORE> Colloquium Notice Week of February 1-5, 1982
C00471 00104 ∂27-Jan-82 1319 JJW Automating induction in EKL
C00474 00105 ∂27-Jan-82 1333 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Nuclear Plants &AI
C00476 00106 ∂27-Jan-82 1719 Oppen at PARC-MAXC Update
C00478 00107 ∂27-Jan-82 1835 Jon L White <JONL at MIT-MC> Two little suggestions for macroexpansion
C00485 00108 ∂28-Jan-82 0959 Gene Golub <CSD.GOLUB at SU-SCORE> Reception
C00493 00109 ∂28-Jan-82 1004 Jrobinson at SRI-AI Tinlunch today
C00494 00110 ∂28-Jan-82 1035 CLT SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
C00495 00111 ∂28-Jan-82 1221 pratt@Shasta (SuNet) 750 <--> 780 timings
C00505 00112 ∂29-Jan-82 0028 BCM prolog
C00519 00113 ∂29-Jan-82 0846 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Meal Reply forms
C00520 00114 ∂29-Jan-82 0946 McCall at PARC-MAXC Re: Course in Programming and PROLOG starting soon
C00525 00115 ∂29-Jan-82 1031 FFL
C00530 00116 ∂29-Jan-82 1058 Morris at PARC-MAXC Re: Course in Programming and PROLOG starting soon
C00533 00117 ∂29-Jan-82 1307 RPG
C00547 00118 ∂29-Jan-82 1321 DBL papers to read
C00548 00119 ∂29-Jan-82 1414 TW via PARC-MAXC Exam trial
C00553 00120 ∂29-Jan-82 1549 JJW PROLOG course
C00554 00121 ∂29-Jan-82 1533 Ichiki at SRI-AI TINLUNCH Meeting 2/4
C00555 00122 ∂29-Jan-82 2148 Lowell Wood <LLW at S1-A> Reporting Games
C00563 00123 ∂29-Jan-82 2149 Kim.fateman at Berkeley Okay, you hackers
C00566 00124 ∂30-Jan-82 1000 JMC*
C00572 00125 ∂30-Jan-82 1000 JMC*
C00573 00126 ∂30-Jan-82 1504 JJW Course in Logic Programming and PROLOG
C00589 00127 ∂31-Jan-82 0939 CLT
C00606 00128 ∂31-Jan-82 1801 JMC*
C00608 00129 ∂31-Jan-82 1836 Feigenbaum at SUMEX-AIM revised statement
C00614 00130 ∂01-Feb-82 0927 TOB
C00617 00131 ∂01-Feb-82 1014 Kim.fateman at Berkeley GETPR and compatibility
C00626 00132 ∂01-Feb-82 1127 Nancy Dorio <CSD.DORIO at SU-SCORE> Correction to Colloquium Notice week of February 1 - 5
C00634 00133 Here are the files that should be put on the disk. OLDPROLOG should be the
C00640 00134 ∂01-Feb-82 2316 Ingalls at PARC-MAXC Re: Course in Programming and PROLOG starting soon
C00642 00135 ∂01-Feb-82 2322 Oppen at PARC-MAXC Hnh???
C00644 00136 ∂01-Feb-82 2316 David Warren <WARREN at SRI-AI> Tops-20 Prolog Tape
C00661 00137 ∂02-Feb-82 0037 ARK Advising
C00669 00138 ∂02-Feb-82 0148 ARK Advice accepted
C00674 00139 ∂02-Feb-82 0803 reid@Shasta (SuNet) The "Masinter report"
C00675 00140 ∂02-Feb-82 1009 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Forum meeting/Guests
C00679 00141 ∂02-Feb-82 1017 Kay at PARC-MAXC Re: Course in Programming and PROLOG starting soon
C00682 00142 ∂02-Feb-82 1018 asprey at PARC-MAXC Re: Course in Programming and PROLOG starting soon
C00683 00143 ∂02-Feb-82 1100 FFL
C00684 00144 ∂02-Feb-82 1109 RINDFLEISCH@SUMEX-AIM (SuNet) Re: The "Masinter report"
C00690 00145 ∂02-Feb-82 1308 FFL
C00693 00146 ∂02-Feb-82 1337 Masinter at PARC-MAXC SUBST vs INLINE, consistent compilation
C00698 00147 ∂02-Feb-82 1504 Konolige at SRI-AI thesis
C00704 00148 ∂02-Feb-82 1524 FFL
C00705 00149 ∂02-Feb-82 1706 Richard Anderson <CSD.ANDERSON at SU-SCORE>
C00710 00150 ∂02-Feb-82 1755 JMC
C00711 00151 ∂02-Feb-82 2050 RPG Thoughts
C00714 00152 ∂02-Feb-82 2051 Lowell Wood <LLW at S1-A> Balance
C00719 00153 ∂02-Feb-82 2109 SGF terminal
C00720 00154 ∂03-Feb-82 0242 Lowell Wood <LLW at S1-A> Follow-Up Or Butter-Up?
C00727 00155 ∂03-Feb-82 1002 FFL
C00728 00156 ∂03-Feb-82 1006 FFL
C00729 00157 ∂03-Feb-82 1036 FFL
C00730 00158 ∂03-Feb-82 1119 FFL
C00731 00159 ∂03-Feb-82 1148 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE Week of FEBRUARY 8 - 12
C00734 00160 ∂03-Feb-82 1405 Richard Treitel <CSL.VER.RJT at SU-SCORE> SCORE Prolog
C00736 00161 ∂03-Feb-82 1534 Patrick H. Winston <PHW at MIT-ML>
C00738 00162 ∂04-Feb-82 0234 Jerry E. Pournelle <POURNE at MIT-MC>
C00746 00163 ∂04-Feb-82 0829 Paul Armer <CSD.ARMER at SU-SCORE> 76-77 Students
C00747 00164 ∂04-Feb-82 0900 JMC*
C00748 00165 ∂04-Feb-82 0946 asprey at PARC-MAXC Re:
C00749 00166 ∂04-Feb-82 0954 Ichiki at SRI-AI TINLUNCH IN BUILDING E
C00750 00167 ∂04-Feb-82 1005 Scott at SUMEX-AIM I wasn't sure if this got to you
C00755 00168 ∂04-Feb-82 1318 FFL
C00756 00169 ∂04-Feb-82 1504 TOB
C00758 00170 ∂04-Feb-82 1618 Betty Scott <CSD.BSCOTT at SU-SCORE> LLL #9628303
C00765 00171 ∂04-Feb-82 1632 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Contract
C00767 00172 ∂04-Feb-82 1704 Gene Golub <CSD.GOLUB at SU-SCORE> Visit by Popek
C00768 00173 ∂04-Feb-82 1905 MMS Match Proof
C00769 00174 ∂04-Feb-82 2356 TOB
C00774 00175 ∂05-Feb-82 0029 TOB
C00779 00176 ∂05-Feb-82 0813 Betty Scott <CSD.BSCOTT at SU-SCORE>
C00780 00177 ∂05-Feb-82 0810 GXG via SRI-KL lisp history
C00782 00178 ∂05-Feb-82 0823 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA contracts
C00784 00179 ∂05-Feb-82 1124 Betty Scott <CSD.BSCOTT at SU-SCORE> Contract N00039-82-C-0250
C00786 00180 ∂05-Feb-82 1313 Raj.Reddy at CMU-10B Conference in Yugoslavia
C00787 00181 ∂05-Feb-82 1350 TOB
C00788 00182 ∂05-Feb-82 1408 JJW Qualifying exam
C00789 00183 ∂05-Feb-82 1451 FFL
C00790 00184 ∂05-Feb-82 1900 JMC*
C00791 00185 ∂06-Feb-82 1722 TW Exams
C00805 00186 ∂06-Feb-82 1709 JJW Comprehensive exam
C00806 00187 ∂06-Feb-82 1832 pratt@Shasta (SuNet) Mass storage
C00808 00188 ∂07-Feb-82 2037 Feigenbaum at SUMEX-AIM mass storage?
C00818 00189 ∂08-Feb-82 0053 Feigenbaum at SUMEX-AIM (Response to message)
C00820 00190 ∂08-Feb-82 0103 REM via SU-TIP Read-Alter mode from MacLISP
C00822 00191 ∂08-Feb-82 0214 pratt@Shasta (SuNet) File system technology update
C00830 00192 ∂08-Feb-82 1000 JMC* via S1-A
C00836 00193 ∂08-Feb-82 1310 Susan Seabrook <CSD.SIS at SU-SCORE> President's Advisory Committee on Computer Science & Computing
C00844 00194 ∂08-Feb-82 1342 Susan Seabrook <CSD.SIS at SU-SCORE>
C00845 00195 ∂08-Feb-82 1425 Betty Scott <CSD.BSCOTT at SU-SCORE> LLL #9628303 Salaries
C00846 00196 ∂09-Feb-82 0319 GXG via SRI-KL account
C00854 00197 ∂09-Feb-82 0909 FFL
C00855 00198 ∂09-Feb-82 1053 nowicki@Shasta (SuNet) ADCOM talk
C00857 00199 ∂09-Feb-82 1105 FFL
C00858 00200 ∂09-Feb-82 1151 ullman@Diablo (SuNet) ADCOM
C00859 00201 ∂09-Feb-82 1331 Paula Bauman <PJB at S1-A> Consulting Bill
C00862 00202 ∂09-Feb-82 1331 MAS Visit of Cdr Ohlander from DARPA
C00863 00203 ∂09-Feb-82 1434 Ichiki at SRI-AI TINLUNCH 2/11
C00864 00204 ∂10-Feb-82 0325 HST via SRI-KL account
C00865 00205 ∂11-Feb-82 0023 RWG
C00866 00206 ∂11-Feb-82 0937 FFL
C00867 00207 ∂11-Feb-82 1151 RPG Papers
C00868 00208 ∂11-Feb-82 1157 RPG
C00875 00209 ∂12-Feb-82 1008 FFL
C00876 00210 ∂12-Feb-82 1305 CLT
C00877 00211 ∂12-Feb-82 1434 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
C00878 00212 ∂12-Feb-82 1434 Gail Stein <OR.STEIN at SU-SCORE> Re: NSF Funding
C00879 00213 ∂12-Feb-82 1439 Gail Stein <OR.STEIN at SU-SCORE> Re: NSF Funding
C00880 00214 ∂12-Feb-82 1439 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
C00881 00215 ∂12-Feb-82 1633 Konolige at SRI-AI boolos
C00882 00216 ∂12-Feb-82 1657 Gene Golub <CSD.GOLUB at SU-SCORE> Salaries for 82-83
C00884 00217 ∂12-Feb-82 1705 Gene Golub <CSD.GOLUB at SU-SCORE> Salaries for 82-83
C00886 00218 ∂12-Feb-82 1757 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
C00887 00219 ∂12-Feb-82 2145 ME LOGIN[1,2]
C00888 00220 ∂13-Feb-82 1613 RPG
C00889 00221 ∂14-Feb-82 0517 JPM via SU-SCORE My account on SAIL
C00891 00222 ∂14-Feb-82 1045 JK new EKL re-writing system
C00892 00223 ∂14-Feb-82 1239 REM via SU-TIP MACLISP RANDOM ACCESS USES
C00894 00224 ∂14-Feb-82 1352 CLT
C00895 00225 ∂15-Feb-82 0007 REM
C00896 00226 ∂15-Feb-82 0845 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Our Previous Conversation
C00899 00227 ∂15-Feb-82 1731 CLT
C00900 00228 ∂15-Feb-82 1915 ME spooling files from E
C00901 00229 ∂16-Feb-82 0848 FFL
C00902 00230 ∂16-Feb-82 1050 Jake Brown <CSD.JAKE at SU-SCORE> Special Faculty Luncheon TODAY
C00903 00231 ∂16-Feb-82 1115 JMM NRO grades
C00904 00232 ∂16-Feb-82 1221 SAGALOWICZ at SRI-AI (Daniel Sagalowicz) Tinlunch on Thursday February 18
C00906 00233 ∂16-Feb-82 1224 JMC*
C00907 00234 ∂16-Feb-82 1355 FFL
C00908 00235 ∂16-Feb-82 1418 FFL
C00909 00236 ∂16-Feb-82 1645 Jake Brown <CSD.JAKE at SU-SCORE> Lost raincoat
C00910 00237 ∂16-Feb-82 1759 CLT
C00911 00238 ∂16-Feb-82 2248 CLT SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
C00912 00239 ∂16-Feb-82 2358 Jerry E. Pournelle <POURNE at MIT-MC> SHACKLETON
C00914 00240 ∂17-Feb-82 0045 RPG At some point
C00915 00241 ∂17-Feb-82 0136 Jerry E. Pournelle <POURNE at MIT-MC> text of advertisement
C00917 00242 ∂17-Feb-82 0140 Jerry E. Pournelle <POURNE at MIT-MC>
C00918 00243 ∂17-Feb-82 0142 ME
C00919 00244 ∂17-Feb-82 0152 Jerry E. Pournelle <POURNE at MIT-MC>
C00921 00245 ∂17-Feb-82 0159 Jerry E. Pournelle <POURNE at MIT-MC>
C00923 00246 ∂17-Feb-82 0423 JPM reply to message
C00924 00247 ∂17-Feb-82 0901 JMC*
C00925 00248 ∂17-Feb-82 1141 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
C00927 00249 ∂17-Feb-82 1315 MDD via NYU SU account
C00928 00250 ∂17-Feb-82 1320 MDD via NYU ATP Prize
C00929 00251 ∂17-Feb-82 1337 LGC Castaneda's Stuff
C00930 00252 ∂17-Feb-82 1535 FFL
C00931 00253 ∂17-Feb-82 2040 ullman@Shasta (SuNet) meeting
C00932 00254 ∂17-Feb-82 2051 Paula Bauman <PJB at S1-A> Commute Mileage Reimbursement
C00934 00255 ∂17-Feb-82 2116 Nancy Dorio <CSD.DORIO at SU-SCORE> ADDENDUM TO COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
C00936 00256 ∂18-Feb-82 0005 Jerry E. Pournelle <POURNE at MIT-MC> film in space
C00937 00257 ∂18-Feb-82 0208 Lowell Wood <LLW at S1-A> Shackleton
C00940 00258 ∂18-Feb-82 0419 Bill Gosper <RWG at MIT-MC> Stop me if you've heard this one before...
C00942 00259 ∂18-Feb-82 0923 Ichiki at SRI-AI TINLUNCH Meeting in Building A Today
C00943 00260 ∂18-Feb-82 1046 RPG@Sail (SuNet) meeting
C00945 00261 ∂18-Feb-82 2012 Ingalls at PARC-MAXC Programming exercise
C00950 00262 ∂18-Feb-82 2013 reid@Shasta (SuNet) REM
C00952 00263 ∂18-Feb-82 2014 DWALTZ at BBND AAAI Program Committee
C00955 00264 ∂18-Feb-82 2155 Nilsson at SRI-AI Re: AAAI Program Committee
C00958 00265 ∂19-Feb-82 0141 Jerry E. Pournelle <POURNE at MIT-MC> Shackleton
C00962 00266 ∂19-Feb-82 0830 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Re: more on simulation
C00964 00267 ∂19-Feb-82 0857 WEBBER at BBND Re: AAAI Program Committee
C00966 00268 ∂19-Feb-82 1548 WOODS at BBND Re: AAAI Program Committee
C00968 00269 ∂19-Feb-82 1652 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Commencement
C00969 00270 ∂20-Feb-82 2152 CLT
C00972 00271 ∂20-Feb-82 2207 CLT
C00983 00272 ∂21-Feb-82 0647 Andrew M. Odlyzko <AMO at MIT-MC>
C00986 00273 ∂21-Feb-82 1111 CLT Kriesels Godel obituary
C00987 00274 ∂21-Feb-82 1235 Aaai-Office at SUMEX-AIM AAAI-82 conference ...
C00995 00275 ∂21-Feb-82 1557 RPG ∂21-Feb-82 1409 JMC
C00996 00276 ∂21-Feb-82 2357 MOON at SCRC-TENEX Fahlman's new new sequence proposal, and an issue of policy
C01003 00277 ∂22-Feb-82 0831 Paul Armer <CSD.ARMER at SU-SCORE> IBM Fellowship
C01009 00278 ∂22-Feb-82 0928 Bob Moore <BMOORE at SRI-AI> Re: disjunctive knowledge
C01011 00279 ∂22-Feb-82 0947 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Participation in Educational Technology Study Group
C01013 00280 ∂22-Feb-82 1002 TOB dept equipment buy
C01014 00281 ∂22-Feb-82 1105 PLATEK at USC-ECLB Visiting Stanford
C01015 00282 ∂22-Feb-82 1143 Irmgild Schack <CSD.IRMGILD at SU-SCORE> General Advisors
C01016 00283 ∂22-Feb-82 1206 Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE> equip
C01017 00284 ∂22-Feb-82 1408 HHB chair
C01018 00285 ∂22-Feb-82 1657 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting reminder
C01019 00286 ∂22-Feb-82 1659 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting reminder
C01020 00287 ∂22-Feb-82 1722 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Gene Golub's Birthday (leap-year)
C01021 00288 ∂23-Feb-82 1006 Ichiki at SRI-AI TINLUNCH Meeting 2/25
C01022 00289 ∂23-Feb-82 1216 Paul Armer <CSD.ARMER at SU-SCORE> Compton Fellowships
C01023 00290 ∂23-Feb-82 1523 RPG Woes
C01026 00291 ∂24-Feb-82 0108 Jerry E. Pournelle <POURNE at MIT-MC> Council Report
C01029 00292 ∂24-Feb-82 0130 Bill Gosper <RWG at MIT-MC> Shop at Bernie's for an 83% rebate
C01031 00293 ∂24-Feb-82 0157 Jerry E. Pournelle <POURNE at MIT-MC>
C01035 00294 ∂24-Feb-82 1038 Feigenbaum at SUMEX-AIM (Response to message)
C01036 00295 ∂24-Feb-82 1223 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
C01040 00296 ∂24-Feb-82 1303 FFL
C01041 00297 ∂24-Feb-82 1351 Nancy Dorio <CSD.DORIO at SU-SCORE> CHANGE IN COLLLOQUIUM NOTICE OF MARCH 1 - 5, 1982
C01042 00298 ∂24-Feb-82 1441 David Warren <WARREN at SRI-AI> Visit by Ehud Shapiro
C01044 00299 ∂24-Feb-82 1513 Nancy Dorio <CSD.DORIO at SU-SCORE> ADDENDUM TO COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
C01045 00300 ∂25-Feb-82 0902 Betty Scott <CSD.BSCOTT at SU-SCORE> Report Processing
C01047 00301 ∂25-Feb-82 0939 Ichiki at SRI-AI TINLUNCH Reminder
C01048 00302 ∂25-Feb-82 1000 JMC*
C01049 00303 ∂25-Feb-82 1042 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
C01052 00304 ∂25-Feb-82 1143 RWW
C01054 00305 ∂25-Feb-82 1541 Gene Golub <CSD.GOLUB at SU-SCORE> recruiting
C01056 00306 ∂25-Feb-82 1551 Gene Golub <CSD.GOLUB at SU-SCORE> Visit
C01057 00307 ∂25-Feb-82 1652 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
C01061 00308 ∂25-Feb-82 1933 Perlis.EE at UDel-Relay
C01063 00309 ∂25-Feb-82 2009 RWW
C01065 00310 ∂25-Feb-82 2011 RWW
C01066 00311 ∂25-Feb-82 2353 HST via SRI-KL lisp history
C01068 00312 ∂25-Feb-82 2358 Bill Gosper <RWG at MIT-MC>
C01069 00313 ∂26-Feb-82 0040 Feigenbaum at SUMEX-AIM (Response to message)
C01074 00314 ∂26-Feb-82 0934 Gene Golub <CSD.GOLUB at SU-SCORE> Re: recruiting
C01075 00315 ∂26-Feb-82 1156 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Recruitment
C01076 00316 ∂26-Feb-82 1332 CLT
C01077 00317 ∂26-Feb-82 1938 RPG Selker
C01078 00318 ∂27-Feb-82 1746 YM shapiro visit
C01080 00319 ∂28-Feb-82 0046 TOB
C01082 00320 ∂28-Feb-82 1417 Gene Golub <CSD.GOLUB at SU-SCORE> [ Ehud Shapiro <Shapiro at YALE>: Re: Visit]
C01101 00321 ∂28-Feb-82 1700 Kim.fateman at Berkeley smoking things out of macsyma
C01109 00322 ∂28-Feb-82 2048 Brian Reid <reid@Shasta at Sumex-Aim> systems candidate Paul Hilfinger
C01115 00323 ∂01-Mar-82 0639 Oded Anoaf Feingold <OAF at MIT-MC> another delayed mailing
C01123 00324 ∂01-Mar-82 0946 DWALTZ at BBND AAAI Program Committee once again
C01126 00325 ∂01-Mar-82 1000 JMC*
C01127 00326 ∂01-Mar-82 1000 JMC*
C01128 00327 ∂01-Mar-82 1124 pratt@Shasta (SuNet) Proposed logic course for this Fall
C01130 00328 ∂01-Mar-82 1149 pratt@Shasta (SuNet) revised description
C01132 00329 ∂01-Mar-82 1209 David Warren <WARREN at SRI-AI> Veronica Dahl
C01134 00330 ∂01-Mar-82 1403 Gene Golub <CSD.GOLUB at SU-SCORE> Next meeting
C01135 00331 ∂01-Mar-82 1404 Gene Golub <CSD.GOLUB at SU-SCORE> Next Meeting
C01136 00332 ∂01-Mar-82 1449 David Warren <WARREN at SRI-AI>
C01141 00333 ∂01-Mar-82 1527 Ichiki at SRI-AI TINLUNCH meeting 3/4
C01143 00334 ∂01-Mar-82 2000 JMC*
C01144 00335 ∂02-Mar-82 0834 Doug Lenat <CSD.LENAT at SU-SCORE> Re: meta-cognition: reasoning about knowledge
C01155 00336 ∂03-Mar-82 0924 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF FEBRUARY 8 - 12, 1982
C01172 00337 ∂03-Mar-82 1227 Nancy Dorio <CSD.DORIO at SU-SCORE> correction to Colloquium notice
C01173 00338 ∂03-Mar-82 1613 Nancy Dorio <CSD.DORIO at SU-SCORE> conference rooms, spring quarter
C01177 00339 ∂03-Mar-82 1803 RPG
C01180 00340 ∂04-Mar-82 2152 Purger exceeding your disk quota
C01186 00341 ∂05-Mar-82 1132 Howard Trickey <CSD.TRICKEY at SU-SCORE> Course Evaluations
C01188 00342 ∂05-Mar-82 1252 pratt@Shasta (SuNet) dinner with Ehud Shapiro
C01193 00343 ∂05-Mar-82 1432 David E. Smith <CSD.SMITH at SU-SCORE> Re: Hayes quote
C01195 00344 ∂05-Mar-82 1607 Gene Golub <CSD.GOLUB at SU-SCORE> [Mark Crispin <Admin.MRC at SU-SCORE>: SCORE reliability]
C01203 00345 ∂05-Mar-82 1621 Mullen at SUMEX-AIM AI quals
C01204 00346 ∂05-Mar-82 2009 YM
C01205 00347 ∂06-Mar-82 1134 CLT
C01206 00348 ∂07-Mar-82 1256 Feigenbaum at SUMEX-AIM Re: AI quals
C01209 00349 ∂08-Mar-82 0126 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
C01221 00350 ∂08-Mar-82 0838 FFL
C01222 00351 ∂08-Mar-82 1000 JMC*
C01223 00352 ∂08-Mar-82 1010 Kanerva at SUMEX-AIM Advise
C01225 00353 ∂09-Mar-82 0806 Betty Scott <CSD.BSCOTT at SU-SCORE>
C01226 00354 ∂09-Mar-82 1645 FFL
C01227 00355 ∂10-Mar-82 0142 RAH via S1-GATEWAY l5 convention
C01228 00356 ∂10-Mar-82 0847 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLLOQUIUM NOTICE WEEK OF MARCH 15 - 19, 1982
C01230 00357 ∂10-Mar-82 1047 Ichiki at SRI-AI TINLUNCH tomorrow
C01236 00358 ∂10-Mar-82 1534 Gene Golub <CSD.GOLUB at SU-SCORE>
C01239 00359 ∂10-Mar-82 1704 Gene Golub <CSD.GOLUB at SU-SCORE> [Dennis Allison <CSL.LAB.DRA at SU-SCORE>: Gary Lindstrom]
C01241 00360 ∂10-Mar-82 1844 Gene Golub <CSD.GOLUB at SU-SCORE> Faculty appointments
C01242 00361 ∂10-Mar-82 1846 Gene Golub <CSD.GOLUB at SU-SCORE> Files
C01243 00362 ∂10-Mar-82 2043 Gene Golub <CSD.GOLUB at SU-SCORE> Shapiro
C01244 00363
C01251 00364 ∂11-Mar-82 0759 Kanerva at SUMEX-AIM Thanks for the talk
C01252 00365 ∂11-Mar-82 1124 Ehud Shapiro <Shapiro at YALE> Thanks and parallel Prolog
C01261 00366 ∂11-Mar-82 1632 Gene Golub <CSD.GOLUB at SU-SCORE> [Buchanan at SUMEX-AIM: statement on Adjunct Professors in the CSD -- from EAF & me]
C01266 00367 ∂11-Mar-82 1655 Gene Golub <CSD.GOLUB at SU-SCORE> next Faculty meeting
C01267 00368 ∂12-Mar-82 0852 FFL
C01268 00369 ∂12-Mar-82 1320 Gene Golub <CSD.GOLUB at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
C01271 00370 ∂12-Mar-82 1341 Gene Golub <CSD.GOLUB at SU-SCORE> Yao appointment
C01273 00371 ∂12-Mar-82 1526 csl.jlh at SU-SCORE (John Hennessy) Re: [John McCarthy <JMC at SU-AI>:]
C01274 00372 ∂12-Mar-82 2123 RAH via S1-GATEWAY saturday visit
C01275 00373 ∂13-Mar-82 1147 TOB
C01279 00374 ∂13-Mar-82 1935 Oded Anoaf Feingold <OAF at MIT-MC> energy digest #3005.5
C01285 00375 ∂14-Mar-82 1423 Tom Dietterich <CSD.DIETTERICH at SU-SCORE> [Tom Dietterich <CSD.DIETTERICH at SU-SCORE>: Re: [John McCarthy <JMC at SU-AI>:]]
C01291 00376 ∂14-Mar-82 1925 Solomon at MIT-MULTICS (Richard Jay Solomon) account on SAIL
C01293 00377 ∂14-Mar-82 2009 Lowell Wood <LLW at S1-A> Missed Connections
C01295 00378 ∂14-Mar-82 2058 CLT
C01296 00379 ∂14-Mar-82 2323 Lowell Wood <LLW at S1-A> Aluminum *Energy* Costs
C01298 00380 ∂14-Mar-82 2355 Feigenbaum at SUMEX-AIM confidential
C01301 00381 ∂15-Mar-82 1440 AMAREL at RUTGERS Japanese AIers
C01303 00382 ∂15-Mar-82 1442 Don Walker <WALKER at SRI-AI> Westinghouse Science Talent Search Winner
C01305 00383 ∂15-Mar-82 1501 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Dennis Heimbigner
C01306 00384 ∂15-Mar-82 1538 Ichiki at SRI-AI 3/18 TINLUNCH meeting
C01308 00385 ∂15-Mar-82 1629 FFL
C01309 00386 ∂15-Mar-82 1647 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Faculty Lunch
C01310 00387 ∂15-Mar-82 2227 RPG
C01311 00388 ∂15-Mar-82 2241 Arthur Keller <CSD.KELLER at SU-SCORE> Comprehensive Reading List
C01313 00389 ∂16-Mar-82 1139 LES
C01314 00390 ∂16-Mar-82 1159 Paula Bauman <PJB at S1-A> PSQ
C01316 00391 ∂16-Mar-82 1215 FFL
C01317 00392 ∂16-Mar-82 1301 CLT Seminar at SRI
C01319 00393 ∂16-Mar-82 2230 LGC Advice-Taker Discussion
C01320 00394 ∂16-Mar-82 2235 TOB
C01323 00395 ∂17-Mar-82 0853 FFL
C01324 00396 ∂17-Mar-82 0902 CL.BOYER at UTEXAS-20 String Search and Syllabus
C01326 00397 ∂17-Mar-82 0903 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 22 - 26, 1982
C01328 00398 ∂17-Mar-82 1005 TOB
C01329 00399 ∂17-Mar-82 1119 RPG
C01330 00400 ∂17-Mar-82 1143 CLT truth
C01331 00401 ∂17-Mar-82 1205 CLT truth again
C01332 00402 ∂17-Mar-82 1213 RPG via S1-A
C01333 00403 ∂17-Mar-82 1213 RPG via S1-A Stupid mailer
C01334 00404 ∂17-Mar-82 1332 CLT truth again
C01336 00405 ∂18-Mar-82 0843 REG
C01337 00406 ∂18-Mar-82 1237 Colmerauer.TextMgt at MIT-MULTICS
C01338 00407 ∂18-Mar-82 1707 David Warren <WARREN at SRI-AI> Re: Colmerauer is on ARPAnet.
C01340 00408 ∂18-Mar-82 1736 David Warren <WARREN at SRI-AI>
C01341 00409 ∂18-Mar-82 1835 CLT
C01342 00410 ∂19-Mar-82 0201 Peter G. Neumann <NEUMANN at SRI-AI> VERkshop III?
C01344 00411 ∂22-Mar-82 0657 PJH via ROCHESTER golux
C01350 00412 ∂22-Mar-82 0810 FFL
C01351 00413 ∂22-Mar-82 0939 MILTON at SRI-AI Database Seminar, Friday, 26 March, by Umesh Dayal
C01354 00414 ∂22-Mar-82 1000 JMC*
C01355 00415 ∂22-Mar-82 1135 Jeff Rubin <JBR at S1-A>
C01356 00416 ∂22-Mar-82 1345 Sharon A. Aglito <CSD.AGLITO at SU-SCORE> Interviews with Dennis Heimbigner on April 2.
C01358 00417 ∂23-Mar-82 0847 RPG Griss
C01359 00418 ∂23-Mar-82 0903 FFL
C01360 00419 ∂23-Mar-82 1220 Don Walker <WALKER at SRI-AI> David McDonald, Thursday, 3-25, at 10:30 and for TINLUNCH
C01364 00420 ∂23-Mar-82 1441 Perlis.EE at UDel-Relay faculty position
C01366 00421 ∂23-Mar-82 2105 CLT question
C01367 00422 ∂23-Mar-82 2143 CLT
C01368 00423 ∂24-Mar-82 1113 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 29 - APRIL 2, 1982
C01370 00424 ∂24-Mar-82 1219 LGC On Sortal Predicates in Quantifiers
C01371 00425 ∂24-Mar-82 1357 TOB
C01372 00426 ∂25-Mar-82 1000 JMC*
C01373 00427 ∂25-Mar-82 1100 JMC*
C01374 00428 ∂25-Mar-82 1101 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Forum Update
C01375 00429 ∂25-Mar-82 1634 FFL
C01376 00430 ∂26-Mar-82 0119 CSD.ULLMAN@SU-SCORE (SuNet) EQUIPMENT
C01377 00431 ∂26-Mar-82 0839 RPG
C01378 00432 ∂26-Mar-82 0846 RPG EQUIPMENT
C01380 00433 ∂26-Mar-82 1000 JMC*
C01381 00434 ∂26-Mar-82 1022 FFL
C01383 00435 ∂26-Mar-82 1445 ullman@Shasta (SuNet) meeting
C01384 00436 ∂26-Mar-82 1602 CSD.ULLMAN@SU-SCORE (SuNet) [KAHN at USC-ISI: Re: RE Compiler Project]
C01388 00437 ∂26-Mar-82 1637 RINDFLEISCH@SUMEX-AIM (SuNet) Re: meeting
C01389 00438 ∂26-Mar-82 1740 Hobbs at SRI-AI Stanford Course Announcement
C01397 00439 ∂27-Mar-82 1742 CLT
C01398 00440 ∂27-Mar-82 2333 RWG Cheshire binders
C01400 00441 ∂28-Mar-82 0217 REM EMACLS.RPG[UP,DOC] last written by JMC
C01402 00442 ∂28-Mar-82 1828 Purger exceeding your disk quota
C01403 00443 ∂28-Mar-82 1844 GOSPER at PARC-MAXC [RWG: Life.]
C01412 00444 ∂29-Mar-82 0219 Bill Gosper <RWG at MIT-MC> hic on h's ifu design, etc.
C01414 00445 ∂29-Mar-82 0820 FFL
C01415 00446 ∂29-Mar-82 1029 David Warren <WARREN at SRI-AI>
C01416 00447 ∂29-Mar-82 1434 David Warren <WARREN at SRI-AI>
C01417 00448 ∂29-Mar-82 1440 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Faculty Lunch and Meeting
C01418 00449 ∂29-Mar-82 1451 TOB
C01420 00450 ∂29-Mar-82 1628 David Warren <WARREN at SRI-AI> Prolog Architectures / DEC
C01422 00451 ∂29-Mar-82 1629 David Warren <WARREN at SRI-AI> Draft DEC Proposal
C01444 00452 ∂29-Mar-82 1757 Chandra-at-OhioState <Chandrasekaran at RUTGERS> visit to SU-AI
C01445 00453 ∂29-Mar-82 1756 Chandra-at-OhioState <Chandrasekaran at RUTGERS> help on funding
C01447 00454 ∂29-Mar-82 2110 TOB
C01461 00455 ∂30-Mar-82 1007 RPG@SU-AI (SuNet) Vaxes
C01462 00456 ∂30-Mar-82 1327 RPG@Sail (SuNet) Clarification
C01465 00457 ∂30-Mar-82 1429 Gene Golub <CSD.GOLUB at SU-SCORE>
C01466 00458 ∂30-Mar-82 1704 Ichiki at SRI-AI 4/1 TINLUNCH MEETING
C01467 00459 ∂31-Mar-82 0743 Gene Golub <CSD.GOLUB at SU-SCORE> [Irmgild Schack <CSD.IRMGILD at SU-SCORE>: Dennis Heimbigner]
C01469 00460 ∂31-Mar-82 0918 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Dennis Heimbigner
C01470 00461 ∂31-Mar-82 1135 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF APRIL 5 - 9, 1982
C01473 00462 ∂31-Mar-82 1307 C.S./Math Library <ADMIN.LIBRARY at SU-SCORE> Journal of Information Processing
C01474 00463 ∂31-Mar-82 1325 JK nsf proposal
C01475 00464 ∂31-Mar-82 1942 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting Reminder
C01476 00465 ∂01-Apr-82 0832 MDD via NYU reference
C01477 00466 ∂01-Apr-82 0849 Ichiki at SRI-AI Remember TINLUNCH
C01478 00467 ∂01-Apr-82 0858 Paul Armer <CSD.ARMER at SU-SCORE> Award Nominations
C01480 00468 ∂01-Apr-82 0858 FFL
C01481 00469 ∂01-Apr-82 0859 FFL
C01482 00470 ∂01-Apr-82 0909 FFL
C01483 00471 ∂01-Apr-82 0914 FFL
C01484 00472 ∂01-Apr-82 0959 ullman@Shasta (SuNet) SUVAXES
C01485 00473 ∂01-Apr-82 1010 Ted Anderson <OTA at S1-A> On the road again!
C01487 ENDMK
C⊗;
∂05-Jan-82 0900 CL.BOYER at UTEXAS-20 Using your system at Stanford
Date: Tuesday, 5 January 1982 10:56-CST
From: CL.BOYER at UTEXAS-20
To: John McCarthy <JMC at SU-AI>
Cc: moore at UTEXAS-20
Subject: Using your system at Stanford
You are most welcome to use our theorem-prover in your
class. If SCORE is a 2060 running TOPS-20, there would be
no technical problems in moving our system to SCORE
provided there are several thousand pages of disk available.
The work required to bring the theorem-prover up there might
be as simple as FTPing some files from [UTEXAS-20]<CL.TP1>.
There would be no proprietary problems. The system is
entirely in the public domain.
Our first guess is that it is unlikely that SRI would
allow the free use of their F2s, but they might be
willing to sell time. Let us know if you would like
us to enquire.
Bob & J
∂05-Jan-82 2116 Oded Anoaf Feingold <OAF at MIT-MC> energy digest #3004
Date: 5 January 1982 23:39-EST
From: Oded Anoaf Feingold <OAF at MIT-MC>
Subject: energy digest #3004
To: energy
Administrivia
North Pole
North Pole
Transporting energy with space technology
Transporting energy with space technology
Re: Transporting energy with space technology
Cables to an SPS
----------------------------------------------------------------------
Date: 5 January 1982 22:50-EST
From: Oded Anoaf Feingold <OAF at MIT-MC>
Subject: Administrivia
To: Energy
Happy new year - this is the first energy mailing in about three weeks.
(Vacation am responsible).
Several of the contributions below are duplicated in other mailing lists,
specifically SPACE. This aint terrible per se, but if it becomes the norm
then the reason for maintaining ENERGY as a separate mailing list will
erode. Please be judicious in your broadband mailings. Note that
virtually everybody on the energy mailing list is also on SPACE, so ....
Anyway, enjoy.
Oded
------------------------------
Date: 29 December 1981 02:54-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: North Pole
To: CS.WEBB at UTEXAS-20
cc: ENERGY at MIT-MC
Date: Sunday, 27 December 1981 20:32-CST
From: Jon A. Webb <CS.WEBB at UTEXAS-20>
I think it's clear. You can't make a skyhook that's attached to the
North Pole. You can only make them if they are attached to the equator.
Objects can orbit in geosynchronous orbit only in the plane of the
equator.
Running a cable directly up from the North (or South) Pole is simple.
You don't have to worry about orbits, you just have to have something
at the top to hold it up and enough strength to hold it together
(or have boosters along the way to alleviate the load on the top part
of the cable). An SPS will have a lot of surplus power. I originally
suggested (several months ago) stationing an SPS above a pole,
using its surplus energy to levitate (for example by throwing
ionized moon dust at relativistic velocity). But if the inhabitants
of the North (or South) pole are terribly worried about the warmth
from microwaves causing health hazards, perhaps a hanging cable can
be used instead. With the cable, you want the SPS close to the Earth
to minimize the length of the cable, but with the original idea the
SPS should be as far away as possible considering antenna focusing
because the farther away it is the less energy it takes to levitate.
(For example, the SPS could be put several times as far away as the Moon,
rather than having to be much closer than the Moon like the equatorial
synchronous stations are; LEO=100 miles, Synch=20k miles, Moon=200k miles,
roughly).
------------------------------
Date: 29 December 1981 03:07-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: North Pole
To: JPM at SU-AI
cc: ENERGY at MIT-MC
Date: 27 Dec 1981 1936-PST
From: Jim McGrath <JPM at SU-AI>
First, you CAN have a skyhook at the North (or South) Pole. ...
However, all this speculation is really meaningless because a skyhook,
in the sense we have been talking about, cannot be placed at the
poles. Why? Well, our skyhooks rely upon the tension created
throughout the structure by the rotation of the earth. At the poles,
this tension cannot be generated (since the structure does not
rotate). Thus the weight of the structure has to be supported by the
base, just like any other building.
You're confused (an euphemism for wrong).
A skyhook isn't a point mass, it's a long thing with different things
happening at different places. At orbital point it's subject to zero
force in the rotating frame, but below that point it's hanging from
above (like a chandelier) and above that point it's hanging from below/inside
(like a slingshot; the kind you swing around, not the rubber-band kind).
At the north pole you just have the inner (bottom) part, the part that
hangs from above (the outside). See my other messages about the SPS
using its surplus energy to levitate above the North pole.
Second, I am puzzled that no one has mentioned the obvious solution to
the SPS power beaming problem. Naturally the power transmission from
space to ground is done via superconducting cables along a skyhook.
So why not extend this logic further? Instead of beaming the power
from the SPS to the skyhook (incurring transmission losses, conversion
losses, and still some safety problems), simply string a
superconducting cable from the SPS to the skyhook! That is, do not
use beam transmission at all - always use cables.
That was my original proposal. But somebody misunderstood me and
created a "straw man" of beaming the energy along part of the path
from SPS to ground. That's an interesting idea, and might work with
some geometries (see my earlier message about tangential beaming
geometry), but not the original idea. Your proposed modification
is the original idea. -- Oh well, minor communication difficulty,
back to the main question/discussion...
------------------------------
Date: 29 Dec 1981 21:17:12-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
To: energy at mit-mc, space at mit-mc
Subject: Transporting energy with space technology
Although oil is not, in my opinion, the way to go, we still have to face
up to the fact that large quantities of money are going to be spent on digging
oil out of the ground in the next few decades. The Canadian government alone
expects 300 GIGABUCKS will be spent in the next decade or so on matters such
as the tar sands and arctic oil.
This is a lot of money, enough to buy a whole passle of Shuttle flights or
to put a solar power unit in orbit. Can we get some of it spent on space?
People are looking anxiously for a way to get oil out of the north down to the
consumers in Canada and the US. It's all in ice, so tankers can't reach it
unless they are submarines, pipelines are hard to build, and are very
vulnerable to very costly sabatoge by natives who don't want them.
What can space technology do to ship the power.
It can either be shipped as oil or in another form. Is it possible to build
one of the catapults talked about in the Space Digest to send oil to a touchdown
off the coast from some refinery?
What about in other forms? If we can build such a plant there, could the
oil be burned, and the power sent up to reflectors or collectors in orbit
to be beamed back down to the surface again?
This may all sound like it will perpetuate oil, but it puts lots of those
nice petrodollars into space. Or am I dreaming?
Brad Templeton (p-btempl@cca-unix or decvax!watmath!bstempleton@Berkeley)
------------------------------
Date: 30 December 1981 11:10-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: Transporting energy with space technology
To: p-btempl at CCA-UNIX
cc: ENERGY at MIT-MC, SPACE at MIT-MC
Gigabucks spent just trying to get oil out of ground and send it to
where it's needed is a good reason for developing space for energy.
If a teensy fraction of that money were invested in space, we could
get bootstrapped. There's a political problem, do we want some big
petrolium company owning the whole space-industry project? But if
nobody else will invest the money to develop the technology,
maybe we have to go that way. We carefully watch their progress:
If they develop it, fine; If they suppress it to protect their interests,
we get on their backs.
------------------------------
Date: 30 Dec 1981 23:58:39-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
To: REM at MIT-MC
Subject: Re: Transporting energy with space technology
Cc: ENERGY at MIT-MC, SPACE at MIT-MC
In response to your message of Wed Dec 30 11:13:04 1981:
If we're going to develop space and viable energy sources (I think those
two go hand in hand), then the only way to see it in our lifetimes may be
with petrodollars. Great leaps often only occur with paranoia, be it
from WWII, the Cold War or the space race. Right now one of the big
sources of fear is the problem of energy - people are willing to spend
gigabucks at the drop of a hat if you mention oil to them. Moving
heavy industry into space and powering it with a Solar Power Satellite
is one of the best answers I know of for a lot of our problems today, and
it paves the way for who knows what else better.
(for those of you in space@mc, sorry for the preaching to the converted)
------------------------------
Date: 31 December 1981 18:59-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: Re: Cables to an SPS
To: p-btempl at CCA-UNIX
cc: ENERGY at MIT-MC
Date: 30 Dec 1981 23:52:03-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
well, microwaves for power are nice, but will we be able to convince the
world of it? We have to find a way around complaints that the microwaves
are sapping our precious bodily fluids.
What idiot would seriously propose that low-level microwaves are sapping
our bodily fluids? (I'm of course not speaking of somebody dumb enough to
set camp on top of the rectenna and then complain that it's too warm there
and he's dehydrating from the excessive warmth.)
P.s. there's an article in yesterday's newspaper about frying and other
high-temperature cooking methods possibly generating enough mutations to
generate dangerous amounts of carcinogens. Listed as most dangerous
are frying, grilling, broiling. Intermediate are baking and roasting.
Least danger of all are from simmering, stewing, boiling and MICROWAVE
COOKING. (Capitals by REM for emphasis.) If <microwave> cooking food
enough to kill all life in it by general heating doesn't make mutations
or carcinogens (compared to other cooking methods) I rather doubt that
tiny amounts of microwaves from side lobes of SPS will. (Besides, any
body at a temperature more than a few degrees above absolute zero, i.e.
everything around us, gives off microwaves profusely. Anybody who really
thinks microwaves are inherently bad ought to be put in a deep freeze
for protection from hirself!)
P.s. the newspaper info is from Leonard Bjeidanes, assoc prof of food tox,
Marvin Morris, et al in Dept of Nutr Sciences, at UC Berkeley.
------------------------------
End of energy digest
********************
Date: 29 December 1981 02:54-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: North Pole
To: CS.WEBB at UTEXAS-20
cc: ENERGY at MIT-MC
Date: Sunday, 27 December 1981 20:32-CST
From: Jon A. Webb <CS.WEBB at UTEXAS-20>
I think it's clear. You can't make a skyhook that's attached to the
North Pole. You can only make them if they are attached to the equator.
Objects can orbit in geosynchronous orbit only in the plane of the
equator.
Running a cable directly up from the North (or South) Pole is simple.
You don't have to worry about orbits, you just have to have something
at the top to hold it up and enough strength to hold it together
(or have boosters along the way to alleviate the load on the top part
of the cable). An SPS will have a lot of surplus power. I originally
suggested (several months ago) stationing an SPS above a pole,
using its surplus energy to levitate (for example by throwing
ionized moon dust at relativistic velocity). But if the inhabitants
of the North (or South) pole are terribly worried about the warmth
from microwaves causing health hazards, perhaps a hanging cable can
be used instead. With the cable, you want the SPS close to the Earth
to minimize the length of the cable, but with the original idea the
SPS should be as far away as possible considering antenna focusing
because the farther away it is the less energy it takes to levitate.
(For example, the SPS could be put several times as far away as the Moon,
rather than having to be much closer than the Moon like the equatorial
synchronous stations are; LEO=100 miles, Synch=20k miles, Moon=200k miles,
roughly).
------------------------------
Date: 29 December 1981 03:07-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: North Pole
To: JPM at SU-AI
cc: ENERGY at MIT-MC
Date: 27 Dec 1981 1936-PST
From: Jim McGrath <JPM at SU-AI>
First, you CAN have a skyhook at the North (or South) Pole. ...
However, all this speculation is really meaningless because a skyhook,
in the sense we have been talking about, cannot be placed at the
poles. Why? Well, our skyhooks rely upon the tension created
throughout the structure by the rotation of the earth. At the poles,
this tension cannot be generated (since the structure does not
rotate). Thus the weight of the structure has to be supported by the
base, just like any other building.
You're confused (an euphemism for wrong).
A skyhook isn't a point mass, it's a long thing with different things
happening at different places. At orbital point it's subject to zero
force in the rotating frame, but below that point it's hanging from
above (like a chandelier) and above that point it's hanging from below/inside
(like a slingshot; the kind you swing around, not the rubber-band kind).
At the north pole you just have the inner (bottom) part, the part that
hangs from above (the outside). See my other messages about the SPS
using its surplus energy to levitate above the North pole.
Second, I am puzzled that no one has mentioned the obvious solution to
the SPS power beaming problem. Naturally the power transmission from
space to ground is done via superconducting cables along a skyhook.
So why not extend this logic further? Instead of beaming the power
from the SPS to the skyhook (incurring transmission losses, conversion
losses, and still some safety problems), simply string a
superconducting cable from the SPS to the skyhook! That is, do not
use beam transmission at all - always use cables.
That was my original proposal. But somebody misunderstood me and
created a "straw man" of beaming the energy along part of the path
from SPS to ground. That's an interesting idea, and might work with
some geometries (see my earlier message about tangential beaming
geometry), but not the original idea. Your proposed modification
is the original idea. -- Oh well, minor communication difficulty,
back to the main question/discussion...
------------------------------
Date: 29 Dec 1981 21:17:12-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
To: energy at mit-mc, space at mit-mc
Subject: Transporting energy with space technology
Although oil is not, in my opinion, the way to go, we still have to face
up to the fact that large quantities of money are going to be spent on digging
oil out of the ground in the next few decades. The Canadian government alone
expects 300 GIGABUCKS will be spent in the next decade or so on matters such
as the tar sands and arctic oil.
This is a lot of money, enough to buy a whole passle of Shuttle flights or
to put a solar power unit in orbit. Can we get some of it spent on space?
People are looking anxiously for a way to get oil out of the north down to the
consumers in Canada and the US. It's all in ice, so tankers can't reach it
unless they are submarines, pipelines are hard to build, and are very
vulnerable to very costly sabatoge by natives who don't want them.
What can space technology do to ship the power.
It can either be shipped as oil or in another form. Is it possible to build
one of the catapults talked about in the Space Digest to send oil to a touchdown
off the coast from some refinery?
What about in other forms? If we can build such a plant there, could the
oil be burned, and the power sent up to reflectors or collectors in orbit
to be beamed back down to the surface again?
This may all sound like it will perpetuate oil, but it puts lots of those
nice petrodollars into space. Or am I dreaming?
Brad Templeton (p-btempl@cca-unix or decvax!watmath!bstempleton@Berkeley)
------------------------------
Date: 30 December 1981 11:10-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: Transporting energy with space technology
To: p-btempl at CCA-UNIX
cc: ENERGY at MIT-MC, SPACE at MIT-MC
Gigabucks spent just trying to get oil out of ground and send it to
where it's needed is a good reason for developing space for energy.
If a teensy fraction of that money were invested in space, we could
get bootstrapped. There's a political problem, do we want some big
petrolium company owning the whole space-industry project? But if
nobody else will invest the money to develop the technology,
maybe we have to go that way. We carefully watch their progress:
If they develop it, fine; If they suppress it to protect their interests,
we get on their backs.
------------------------------
Date: 30 Dec 1981 23:58:39-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
To: REM at MIT-MC
Subject: Re: Transporting energy with space technology
Cc: ENERGY at MIT-MC, SPACE at MIT-MC
In response to your message of Wed Dec 30 11:13:04 1981:
If we're going to develop space and viable energy sources (I think those
two go hand in hand), then the only way to see it in our lifetimes may be
with petrodollars. Great leaps often only occur with paranoia, be it
from WWII, the Cold War or the space race. Right now one of the big
sources of fear is the problem of energy - people are willing to spend
gigabucks at the drop of a hat if you mention oil to them. Moving
heavy industry into space and powering it with a Solar Power Satellite
is one of the best answers I know of for a lot of our problems today, and
it paves the way for who knows what else better.
(for those of you in space@mc, sorry for the preaching to the converted)
------------------------------
Date: 31 December 1981 18:59-EST
From: Robert Elton Maas <REM at MIT-MC>
Subject: Re: Cables to an SPS
To: p-btempl at CCA-UNIX
cc: ENERGY at MIT-MC
Date: 30 Dec 1981 23:52:03-EST
From: p-btempl at CCA-UNIX (Brad Templeton)
well, microwaves for power are nice, but will we be able to convince the
world of it? We have to find a way around complaints that the microwaves
are sapping our precious bodily fluids.
What idiot would seriously propose that low-level microwaves are sapping
our bodily fluids? (I'm of course not speaking of somebody dumb enough to
set camp on top of the rectenna and then complain that it's too warm there
and he's dehydrating from the excessive warmth.)
P.s. there's an article in yesterday's newspaper about frying and other
high-temperature cooking methods possibly generating enough mutations to
generate dangerous amounts of carcinogens. Listed as most dangerous
are frying, grilling, broiling. Intermediate are baking and roasting.
Least danger of all are from simmering, stewing, boiling and MICROWAVE
COOKING. (Capitals by REM for emphasis.) If <microwave> cooking food
enough to kill all life in it by general heating doesn't make mutations
or carcinogens (compared to other cooking methods) I rather doubt that
tiny amounts of microwaves from side lobes of SPS will. (Besides, any
body at a temperature more than a few degrees above absolute zero, i.e.
everything around us, gives off microwaves profusely. Anybody who really
thinks microwaves are inherently bad ought to be put in a deep freeze
for protection from hirself!)
P.s. the newspaper info is from Leonard Bjeidanes, assoc prof of food tox,
Marvin Morris, et al in Dept of Nutr Sciences, at UC Berkeley.
------------------------------
End of energy digest
********************
∂06-Jan-82 0038 RPG Mail file management
I wrote some e/lisp hacks the other day that Lew thought
you might find useful. When I want to clean up my mail file I just
put a `↓' at the beginning of the first line of the page to
signify that I want that page flushed. If I want to put that
page in another file I put `→<filename>' on the first line.
Then I start up the program EDIT under E and invoke the function
E:DISPOSE-FILE. Lisp then proceses the file for me while I do something else.
Here are some E macros I have to make this process easier to remember:
αXdefine e⊗↔αXSL EDIT⊗↔αXSAY Running EDIT⊗↔αβ⊗↓
αXdefine k⊗↔αβ-αXsl⊗↔αXSAY Killing LISP⊗↔αβ⊗↓
αXdefine cleanup⊗↔α0α=(e:dispose-file)⊗↔αXSAY Cleaning Up File⊗↔αβ⊗↓
which I have in my EINIT.CMD file. Then αZe starts EDIT, αZk
will kill the Lisp job, and αZcleanup will perform the function.
-rpg-
Thanks for mail file info.
∂07-Jan-82 0951 RWW
∂07-Jan-82 0039 JMC
What would be involve in reviving LCF for class experimentation?
I believe that the core image (LCF.DMP) could be revived and it would
work perfectly. Maybe we should pumpkin it and see.
∂07-Jan-82 1601 JJW
Here's a message that I just got from Boyer. I'm going to ask MRC
to give me sufficient space at SCORE, but you will probably have to
arrange for some account to pay for it. My "Interlisp question"
that he refers to was a statement that I know next to nothing about
Interlisp and would appreciate help if it is needed.
∂07-Jan-82 1522 CL.BOYER at UTEXAS-20 Theorem-prover
Date: Thursday, 7 January 1982 17:19-CST
From: CL.BOYER at UTEXAS-20
To: Joe Weening <JJW at SU-AI>
Subject: Theorem-prover
Cc: Moore at UTEXAS-20
Here's some information to get you going. It is probably
sufficient. Moore and I would like the use of our
theorem-prover in McCarthy's course to be successful, so bug
us whenever you want.
1. Create a directory with at least 3200 pages. If you can,
name the directory <CL.TP1>. FTP all of the files from
[UTEXAS-20]<CL.TP1>. You can use the Multiple Get command.
Login as ANONYMOUS, password FOO. There are 79 files, but
you'll need very few of them. It's easier to start with
everything and delete what you don't want. For example, any
file of the form *.proofs is for reading only, and can be
deleted to save space.
2. Make sure that LISP.EXE.133 is on <LISP>. You probably
already have it. If not, ask your local Interlisp guru to
put it there; we have resisted going to higher versions of
Interlisp because of bugs in them.
3. Read the files CODE.TXT and CODE.DOC. The second is a
users manual somewhat out of date; the first is a less than
perfect update. Then follow the instructions therein to run
the theorem-prover. As your first step, duplicate the very
simple examples in CODE.DOC.
Your Interlisp question is too general to answer. After
you've run the simple examples, ask me again.
If you can, please describe how the theorem-prover is to be
used. We have it on a back burner to redo the users-manual;
if the theorem-prover might be seriously used by some
students, we'd have an incentive (vanity) to polish up the
manual.
Whoever is going to use the theorem-prover might want to
know about the existence of "Metafunctions: proving them
correct and using them efficiently as new proof procedures."
It appears in The Correctness Problem in Computer Science,
Academic Press, 1981. It presents precisely the slight
changes in the theory presented in "A Computational Logic"
that are now implemented in the version of the
theorem-prover you are getting. The main changes concern
what signs can appear in literal atoms and variables. One
noteworthy change is that the CAR and CDR of things not
produced by CONS is 0.
∂07-Jan-82 1634 YK
COME TO OUR HOUSE
to help us celebrate our wedding
We were married on December 22, 1981 and we would like to invite you to
share in our good fortune. Supper and snacks will be served throughout
the evening so bring an appetite, good cheer and some friends.
Place: 12625 La Cresta Drive, Los Altos Hills
Time: Sunday January 24 open house from 4pm to ...
Richard and Yasuko Weyhrauch
R.S.V.P. (415) 948-2149 or RWW@SU-AI or YK@SU-AI
∂07-Jan-82 2105 KLC
Hi...
I'm trying to get an understanding of the relationship between
non-monotonic logic and inexact or probabilistic reasoning. I think I
understand the situations in which one is clearly preferable to the other,
but it's not clear to me how (or whether) to mix the two properly. Could
you give me a few pointers to the literature or comments about such
issues?
Also, I'm wondering if the following idea is interesting. One of Polya's
heuristics could be stated as: If A(x)⊃B(x), and B(c) is true, then A(c)
is more likely. This amounts to the assumption that B-A (abusing
notation) is small, which would seem to be a circumscription of the
property B. This would suggest that circumscription could be used prior
to probabilistic methods, perhaps in choosing a sample space.
I would be grateful for any comments you might have.
Ken Clarkson
Also, I'd like to apologize for some remarks I made on Wednesday which,
heard out of context, would be pretty offensive. If you don't know what
I'm referring to, the point is moot, however....
There is no literature that I know of that mentions both except a sentence
or two in my paper. I agree that circumcscription might be used to establish
a sample space for probabilistic reasoning. I would be glad to discuss the
matter with you at a mutually convenient time - tomorrow afternoon any time
after 2:30 would be convenient for me. Since I don't remember any Wednesday
remarks, any apology is indeed moot.
∂07-Jan-82 2123 KLC
I'll be by sometime tomorrow afternoon after 2:30...Ken Clarkson
∂07-Jan-82 2359 JMC*
Edozien
∂08-Jan-82 0404 Bill Gosper <RWG at MIT-MC> life block
Date: 8 January 1982 07:00-EST
From: Bill Gosper <RWG at MIT-MC>
Subject: life block
To: JMC at SU-AI
My SEND was just observing that a block colored
W B
B W
would "blink". Similarly for tubs, barges, etc.
W B
B W
won't blink according to my rules, because flipping requires a majority
including self. Of course, the rules can be as we choose, but for
the competitive game, it seems to me that my rules are more likely
to lead to a situation that can be called a victory. Experiment will
tell, however.
∂08-Jan-82 0910 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Order 2494
Date: 8 Jan 1982 0908-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: ARPA Order 2494
To: Kahn at USC-ISI
cc: JMC at SU-AI, TOB at SU-AI, CSD.BScott at SU-SCORE
The Sponsored Projects Office at Stanford tells me that there is some
confusion among George Daniels, Jane Hensley and Gene Stubbs concerning
the additional $102K which is due on AO 2494 to finish the Formal
Reasoning/Image Understanding work on Contract MDA903-80-C-0102. We
were told in October that the funds would be coming, but have heard
nothing further. We will much appreciate your looking into the matter
and letting us know when we can expect receipt of a modification adding
the funds.
Thanks very much,
Betty Scott
-------
∂08-Jan-82 0911 JK lispx
I completed the proof (more or less) - internal form is in
lispx.prf[ekl,jk] - there is a faster way of doing it by
applying [1]*der*nil at the end of rewriting mode of line 43
but I encountered a bug doing that; I will fix it. This would
save the last two lines.
I'll look.
∂08-Jan-82 0951 JJW SCORE account for Boyer-Moore system
To get 3200 pages, as Boyer says we need, requires about 60 aliquots
on SCORE, which comes to over $790 per month. Shall I ask Lynn Gotelli
to open an account of that size, and if so, who gets charged for it?
I'll be back to you. It will be some combination of the department
and overhead, but I'll have to talk to Golub.
∂08-Jan-82 1326 Gene Golub <CSD.GOLUB at SU-SCORE> funds for computing
Date: 8 Jan 1982 1321-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: funds for computing
To: jmc at SU-AI
cc: bscott at SU-SCORE
We shall try to cover the computer costs from the residual IBM funds.
GENE
-------
∂08-Jan-82 1543 FFL Final Report on your NSF grant
To: JMC, FFL
Final report on Basic Research in AI will be due on March 3l, l982
∂08-Jan-82 1545 Susan Hill <CSD.HILL at SU-SCORE> Computer accts
Date: 8 Jan 1982 1539-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
Subject: Computer accts
To: jmc at SU-AI
cc: csd.gotelli at SU-SCORE
Betty said something about your asking for permission for students in your
class to have computer accounts. How do you plan to handle this?
-------
∂08-Jan-82 2015 JK ekl bugs and lispx
Have been fixed, I think.
There is a new proof in lispx.prf[ekl,jk] with pretty-proofed version
in lispx.ppr[ekl,jk] that combines the last three steps of the proof.
∂09-Jan-82 0124 Bill Gosper <RWG at MIT-MC>
Date: 9 January 1982 04:20-EST
From: Bill Gosper <RWG at MIT-MC>
To: JMC at SU-AI
Well, W
B B
W W
B etc should "ko", at any rate.
Indeed!
∂09-Jan-82 0815 KAHN at USC-ISI Re: ARPA Order 2494
Date: 9 Jan 1982 0809-PST
Sender: KAHN at USC-ISI
Subject: Re: ARPA Order 2494
From: KAHN at USC-ISI
To: CSD.BSCOTT at SU-SCORE
Cc: Kahn at USC-ISI, JMC at SU-AI, TOB at SU-AI,
Cc: mckenney at USC-ISI
Message-ID: <[USC-ISI] 9-Jan-82 08:09:03.KAHN>
In-Reply-To: Your message of 8 Jan 1982 0908-PST
betty,
the 102k was put out by our office as part of the
fy82 incremental funding actions last summer. i'm
on travel now and wont be back in the office until the 18th
of january. In the meantime, ill ask dona mckenney to
see if she can determine what happened to the funds
after they reached dss-w.
bob
[Dona - Can you sort this one out for Betty Scott at Stanford?]
p.s. actually, if i recall correctly the 102k was really more
like 100k (i seem to remember $100,638) which was all that
remained on the contract t for johns AI tasks. I
believe additional $ were added to the contract by dss-w
on other arpa orders however on a separate arpa order
that covered what was originally johns portion of
the net graphics work and perhaps even a bit of luckhams
work a year or two ago. However, without the documentation in hand,
its hard to be sure.
∂09-Jan-82 0912 JK
∂09-Jan-82 0019 JMC lispx inelegances
lispx.lsp[w82,jmc] is a source file, and lispx.ppr and lispx.prf are the others.
I wanted to get the sort information and the facts about append entirely
into a symbolic form. In order to do this I followed each line containing
such information by a LINENAMES that updated the line in question. It worked,
but there are the following inelegances.
1. I had to answer the question of whether I really meant it after
each use of LINENAMES for updating purposes, so there is a line containing
y after each such line.
----------------------
That checking is controlled by the global CAREFUL - if you
setq it to nil, this will not occur. Since CAREFUL also controls checking
for extra lines that need to be deleted, it may be best to define a new
function like LNAME:
(defun lname fexpr (x) (let ((careful nil)) (apply 'linename x)))
and use that, including it into your ekl.ini file.
2. SORTINFO has extra lines.
3. In the .ppr file, the command still comes out with some of the
numerical line number information instead of the purely symbolic information
that went in.
---------------------
I don't understand what you mean by 2 and 3 - I went thru the linename code
again and fixed a few other bugs so this may have cleared up now.
∂09-Jan-82 0933 JK
∂09-Jan-82 0150 JMC more lispx
I have added a definition of
flat(x,u) = if atom x then x.u else flat(car x,flat(cdr x, u))
but don't understand rewriting well enough yet to know how far I
can get toward
∀x u.listp flat(x,u)
in one ∀e step.
------------
Assuming that ∀x u.sexp flat(x,u) is clear, one could just instantiate
a case of car induction for ∀u.listp flat(x,u)
and that should go in one step.
------------
∀x u.sexp flat(x,u) isn't clear, since that would also amount to the
termination of the program. Even so I don't see how it would be
used.
∂10-Jan-82 0624 JK
∂10-Jan-82 0044 JMC
Is the following a bug? If not how do I get use an axiom ¬p to get
if p then a else b = b?
the proof BUG:
(DECL (P) |TRUTHVAL| CONSTANT)
(DECL (A B) |GROUND| CONSTANT)
(ASSUME |¬P|)
3. ¬P
ctxt: (1) deps: (3)
(TRW |IF P THEN A ELSE B| |*3*NIL|)
4. IF P THEN A ELSE B=IF P THEN A ELSE B
ctxt: (1 2) deps: (3)
(ASSUME |P|)
5. P
ctxt: (1) deps: (5)
(TRW |IF P THEN A ELSE B| |*5*NIL|)
6. IF P THEN A ELSE B=A
ctxt: (1 2) deps: (5)
----------------
This has been fixed.
∂10-Jan-82 0936 JK
∂09-Jan-82 0150 JMC more lispx
I have added a definition of
flat(x,u) = if atom x then x.u else flat(car x,flat(cdr x, u))
but don't understand rewriting well enough yet to know how far I
can get toward
∀x u.listp flat(x,u)
in one ∀e step.
-----------
The file flat.ppr[ekl,jk] contains my approximation of this.
∂10-Jan-82 0939 JK flat
In fact, the whole thing can be done in one line with the new
decision procedure (the old one is not always very intelligent
about sorts)
∂10-Jan-82 1005 JK more on flat
The file flat.ppr[ekl,jk] contains now a one-line proof of
∀x u.listp flat(x,u). I under estimated the old decision
procedure.
∂10-Jan-82 2321 reid@Shasta (SuNet) proposal
Date: 10 Jan 1982 23:21:05-PST
From: reid at Shasta
To: equip
Subject: proposal
As soon as I get vitae for JMC, EAF, and JLH, and a paragraph from Bob E.
explaining why we are buying instead of leasing this equipment (beats me),
then we are ready to roll. The Dover is down again, of course.
Press file for the existing proposal, incorporating comments by JDU, JLH, and
a few by RPG, is /mnt/reid/proposals/eqp81/proposal.press. Keep those packets
and streams rolling in.
Brian
BIOJMC[1,JMC] is a vita in PUB.
∂11-Jan-82 1306 CLT
I'm playing with annette tonight, won't be around for supper
∂11-Jan-82 1549 CL.MOORE at UTEXAS-20 Theory of Computation
Date: Friday, 8 January 1982 12:56-CST
From: CL.MOORE at UTEXAS-20
To: John McCarthy <JMC at SU-AI>
Subject: Theory of Computation
Dear Professor McCarthy,
We will be teaching Theory of Computation this Spring and
we'd like to know what you teach in your course. Would you
please send us a syllabus?
Thanks,
Bob and J
I forgot to answer your request for a syllabus of my course in
mathematical theory of computation. Anyway there is, alas, no
syllabus. However, covered the following:
1. representation of recursive programs by sentences in first order
logic
2. A Scott axiomatization of typed lambda calculus. (1969 CUCH, ISWIM and
OWHY paper - not the models of lambda calculus that came later.
3. Use of Ketonen's EKL interactive theorem prover.
4. The Boyer-Moore system. Because of problems in getting enough
disk space and some laziness, both on my part and on the students'
part, they never got to run any problems on the computer.
5. My Elephant formalism for representing sequential programs as
sentences in first order logic.
6. Manna's method (his PhD thesis) for proving total correctness.
7. Axiomatization of flow charts by sentences of first order logic
(multiple entries and exits).
8. Formalization of dirty LISP, including rplaca s and imbedded
setq s.
This doesn't amount to a syllabus, because I don't assert that this
is what should be covered, and it may not represent what I'll do
next month in Marseille or what I'll do when I teach the course again
in two years.
∂11-Jan-82 1550 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Cooperative Research with JPL
Date: 11 Jan 1982 1541-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Cooperative Research with JPL
To: CSD-Faculty: ;
cc: ADMIN: ;
The California Institute of Technology and NASA plan to make funds available
during calendar year 1982 to support cooperative research on topics and in
areas related to the interests and capabilities of the Jet Propulsion
Laboratory. Such research is to be conducted by university and college
personnel working in close cooperation with people at JPL. From these funds,
a number of grants will be awarded.
Proposals for research to be initiated within calendar year 1982 will be
accepted for consideration not later than February 1, 1982.
The documentation for preparing the proposals is in my office.
GENE
-------
∂11-Jan-82 2246 JK bug
has been fixed, it seems. It was in the old unification routine.
∂12-Jan-82 0809 JK
∂12-Jan-82 0135 JMC
New bug
Step 47 in lispx.lsp[w82,jmc] or lispx.prf[w82,jmc] is produced by the
command below it. Step 51 above (not in lispx.prf) was produced by
the same command with $ in place of the &. The & was an attempt to
get it to expand flatten without again expanding flat. How is that
to be done? I want a mode that will use all applicable occurrences
of the line range referred to but won't expand the results of expansions.
Then I could expand all definitions in part of the formula with a
single reference to DEFINFO without danger of a loop.
The bug is that the term (x~y)~nil*u should be
(flatten(x)*flatten(y))*u.
--------------
I don't quite understand the problem - maybe we can talk about it
later today.
Would using + help?
∂12-Jan-82 1033 Jrobinson at SRI-AI Stanford Linguistics Colloquium of interest
Date: 12 Jan 1982 1032-PST
From: Jrobinson at SRI-AI
Subject: Stanford Linguistics Colloquium of interest
To: tlgrp:
Notice of two linguistic colloquia at Stanford.
1. Manfred Bierwisch. Today, Jan. 12, at 3:15 School of Education
Room 334. "Some More Speech Errors".
2. Emmon Bach and Barbara Partee (UMASS and Amherst)
Thursday, January 21, 4:15, CERAS Room 112.
"Events and Processes, Things and Stuff."
-------
∂12-Jan-82 2146 RWW
do you know how to contact steve ness in the city?
I know of no way to relieve your nesslessness.
∂13-Jan-82 1014 Marilynn Walker <CSD.MWALKER at SU-SCORE> MTC Quals
Date: 13 Jan 1982 0903-PST
From: Marilynn Walker <CSD.MWALKER at SU-SCORE>
Subject: MTC Quals
To: jmc at SU-AI
cc: csd.mwalker at SU-SCORE
Please send message re results of students who took MTC Quals.
Thank you.
Marilynn Walker
-------
Yonatan Malachi was a high pass. Frank Yellin was a pass. Joe Weening
didn't pass but will take it again in late February.
Weening retaking the mtc qual has been until Zohar and I are next
both in town - probably in September.
∂13-Jan-82 1135 JJW SCORE account for Boyer-Moore system
I asked Lynn Gotelli for 60 aliquots, which is 3180 blocks
(Boyer said we need 3200), but she has told me that there are only
555 blocks left to be sold on SCORE. So I don't know how we will
be able to bring up the theorem-prover unless something is done to
free up disk space at SCORE.
I'll start politicking.
∂13-Jan-82 1142 JK ekl bug fixed
TRW and RW were not reading in all the sort information.
It should work now.
∂13-Jan-82 1436 JK ekl bug
I looked at it - it does not seem to be a bug at all:
The reason $40 fails that one wants to apply it to the conditional
(if false then .. else flatten(x)*flatten(y))*u (or something like that)
If you stick a nil before $40 everything is o.k.
Thanks; you can expect at least a few to be my bugs.
∂13-Jan-82 2258 JK ekl bugs
The pretty-proof and show bug has been fixed.
Save-proofs worked correctly for me - I could not
reproduce that. Was there any indication of
an infinite loop going on? This could be a bug
in the Emaclisp interface - there were problems with open files
before.
∂14-Jan-82 1309 ME system load bottom
∂13-Jan-82 1327 JMC system
I have noticed that in the last few months, the load levels have been
somewhat lower, especially the residual load level when no-one is
doing anything. Is this a real change or just a change in the way
things are measured?
Formerly the load was (as I remember it) never beow .10, and now it
goes down to .04.
ME - As far as I know, the measuring method is totally unchanged. I
suppose this might mean that some phantom is either doing less or working
more efficiently, since with no real users the phantoms generate all of
the load. I don't think we've done away with any phantoms (which might
have caused this).
∂15-Jan-82 0934 JJW
∂14-Jan-82 1950 JMC improved lname
Could lname generate a (comment) line of the form (COMMENT LNAME foo baz)
so that .ppr files would have the line names attached to the axioms?
----------
The LNAME in LNAME.LSP[EKL,JJW] now generates a COMMENTL line after the line
to which it is applied. Note this introduces two blemishes:
(1) The COMMENTL line gets a number (though EKL doesn't show it), so there
appear to be "missing" line numbers.
(2) If a linerange has only one member, then in the .PPR file, EKL inserts a
(LINENAME FOO *) line after that line, which appears together with the
comment.
∂15-Jan-82 1557 FFL
To: JMC, FFL
Telegram from Porto says his departure date is 1/25. I will get the ticket
for him on Monday and wire him to pick it up. Anything else we need to tell
him?
∂15-Jan-82 1617 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Advanced GRE
Date: 15 Jan 1982 1610-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Advanced GRE
To: csd.buchanan at SU-SCORE
cc: JMC at SU-AI, csd.golub at SU-SCORE
In our current "Dear Applicant" letter we discourage applicants from taking the
advanced GRE in Engineering. Perhaps you could examine whether we should
allow this advanced test. I think our letter should list the possible test
with an ordering.
GENE
-------
∂15-Jan-82 1633 Buchanan at SUMEX-AIM Re: Advanced GRE
Date: 15 Jan 1982 1629-PST
From: Buchanan at SUMEX-AIM
Subject: Re: Advanced GRE
To: CSD.IRMGILD at SU-SCORE, csd.buchanan at SU-SCORE
cc: JMC at SU-AI, csd.golub at SU-SCORE
In response to the message sent 15 Jan 1982 1610-PST from CSD.IRMGILD at SU-SCORE
I'll have the committee look at the letter and we'll revise it.
thanks,
bgb
-------
∂15-Jan-82 1650 Irmgild Schack <CSD.IRMGILD at SU-SCORE> [Gene Golub <CSD.GOLUB at SU-SCORE>: New Associate Chairman]
Date: 15 Jan 1982 1643-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: [Gene Golub <CSD.GOLUB at SU-SCORE>: New Associate Chairman]
To: CSD-Faculty: ;
Mail-From: CSD.GOLUB created at 15-Jan-82 16:29:43
Date: 15 Jan 1982 1629-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: New Associate Chairman
To: bboard at SU-SCORE
I'm pleased to announce that PAUL ARMER has accepted the position of Associate
Chairman, effective January 26. Paul comes to the department with a rich
background in computer science and computing. He has been associated with
Rand, Stanford, and the Babbage Institute. He is a very thoughtful and
considerate person and I'm sure the department will benefit by his presence.
GENE GOLUB
-------
-------
∂15-Jan-82 1822 Aaai-Office at SUMEX-AIM Robinson Agreement and Accounting Documents
Date: 15 Jan 1982 1703-PST
From: Aaai-Office at SUMEX-AIM
Subject: Robinson Agreement and Accounting Documents
To: Bledsoe at UTEXAS-20, Bobrow at PARC-MAXC, Grosz at SRI-AI,
Feldman at SUMEX-AIM, JMC at SU-AI, Simon at CMU-10A, GJS at MIT-AI,
DWaltz at BBND, Woods at BBND, Tenenbaum at SRI-KL, Reddy at CMU-10A
cc: Minsky at MIT-AI, Nilsson at SRI-AI, Feigenbaum at SUMEX-AIM,
AAAI-OFFICE at SUMEX-AIM
To:Drew McDermott via U.S. Mail
To:Charles Rieger via U.S. Mail
To: AAAI Executive Council
From: Lou Robinson
The following agreement statement has been hammered out between myself
and the AAAI Finance Committee. It has already been presented to
Marvin Minsky, Nils Nilsson and Ed Feigenbaum. They have approved the
statement. Please review and send along any comments.
Mail-from: ARPANET host SRI-AI rcvd at 7-Jan-82 1115-PST
Date: 7 Jan 1982 1059-PST
From: Don Walker <WALKER at SRI-AI>
Subject: Robinson agreement documents
To: Minsky at MIT-AI, Nilsson at SRI-AI, Feigenbaum at SUMEX-AIM, Raj.Reddy at CMU-10B
cc: aaai-office at SUMEX-AIM
AGREEMENT BETWEEN AAAI AND L. G. ROBINSON
The Finance Committee of AAAI, consisting of Reddy, Tenenbaum and Walker,
negotiated the 1980 AAAI contract with Lou Robinson and the new 1981 and 1982
contract. This contract is attached to this memo as Appendix I.
With the establishment of a AAAI office and the centralization of many of the
functions, such as publications and membership, there is a need to realign and
increase the responsibilities of the Executive Director, Lou Robinson. At the
last executive council meeting, the Finance Committee was directed to negotiate
a new contract. The following draft of a revised contract statement will
become effective after the President approves it.
From the current contract given in Appendix I, you will note that Lou is
guaranteed a range of $35K for performing certain duties in connection with
AAAI-82. The revised contract assumes that the routine aspects of the AAAI-82
conference coordination and exhibits programs will be handled by the new
Administrative Assistant at the AAAI office, Rose Marie Berry. This will
reduce the work load for Lou. But the centralization of the publication
functions of the AI magazine and the Proceedings will significantly increase
the responsibilities.
PROPOSED CONTRACT
Lou Robinson will act as the Executive Director of AAAI during 1982. He
agrees to spend at least 1/2 time performing the following functions with a
guaranteed minimum of $30,000 total income (fee plus commissions) and a maximum
of $60,000.
- Take complete responsibility for smooth operation of publications and
membership functions (Fee: $15,000).
- AAAI-82 Conference supervision (Fee: $5,000). R. Berry will do most
of the routine work.
- Organize the publicity and arrangements for tutorials (10% of gross
receipts - Minimum $5,000).
- Supervise AAAI office including accounting, quotation and bids, AAAI-
82 finances and reporting to the Treasurer of AAAI (Fee: $5,000).
Optionally Robinson may engage in other revenue generating activities for
AAAI during the remaining 1/2 time, earning commissions on the income. The
intention is that each such activity will result in net positive cash flow to
AAAI. The activities may include:
- Arrange exhibit program with the help of R. Berry (Commission: 10% to
Lou Robinson, 25% to Rosemarie Berry of Gross).
- Act as Advertising Agent for AAAI publications (Commission: 25% to
50% of revenues based on the industry norm).
- Tutorial video tape publicity and distribution (Commission: 25% of
the net income after deducting video taping, duplication and
advertising expenses).
- Fund-raising help for Minsky (Commmission: 5%).
! Secretarial assistance, mailing, telephone and advertising expenses will be
handled as part of the AAAI office. All other expenses incurred by Robinson as
part of his AAAI duties will be reimbursed fully. However, all expenditures
not already budgeted and approved by the AAAI Treasurer require specific
authorization. AAAI will pay for 2 trips to conference sites: one
approximately a year in advance and one at the time of the conference.
All items not covered by this contract shall be first negotiated with the
Finance Committee and approved by the Executive Council.
RATIONALE
Lou received $18,500 for services in coordinating AAAI-80 and $16,500 for
helping with IJCAI-81. As a result of starting the AAAI office and hiring a
full-time Administrative Assistant and increasing the minimum for Lou to
$30,000, our annual expenses will probably be close to income rather than the
substantial surplus we enjoyed in 1980. For your information, we are providing
the financial accounting for AAAI 1980 and 1981 in Appendix II and III.
Appendix IV contains the budget for 1982. Appendix V contains the budget for
the AAAI-82 conference.
Note that this is a one year contract. This will give us an opportunity to
renegotiate the contract in 1982 in case of financial stringency.
Appendix I: Current Contract with Robinson
Appendix II: AAAI - 1980 Financial Statement
Appendix III: AAAI - 1981 Financial Statement
Appendix IV: AAAI - 1982 Budget
! APPENDIX I: CURRENT CONTRACT WITH ROBINSON
Agreement between AAAI and Garcia-Robinson, Inc. for services to be performed
by Garcia-Robinson, Inc. for IJCAI-81 and the AAAI 1982 Conference:
For IJCAI-81, Garcia-Robinson, Inc. will:
- Produce the following publications: A poster/flyer, a conference
brochure, a conference program, an R&D exhibit guide, and any other
publications as needed, not including proceedings.
- Serve in an advisory capacity to Pat Hayes, Conference Chairman, and
Richard Rosenberg, local arrangements chairman, on all aspects of
conference planning, as needed.
- Organize and promote a AAAI tutorial program and a Research and
Development Expostion.
Garcia-Robinson, Inc. will receive $10,000 for services rendered to IJCAI-81,
to be paid quarterly.
For the tutorials, Garcia-Robinson will receive 5% (five percent) of the
gross tutorial receipts.
For the Research and Development Exposition, Garcia-Robinson will receive a
commission of 50% (fifty percent) of the net receipts of the Exposition.
AAAI guarantees Garcia-Robinson a minimum of $15,000 total income (fee plus
commissions) and a maximum of $35,000 for 1981.
For the 1982 AAAI Conference, to be held at Carnegie-Mellon University,
Garcia-Robinson, Inc. will act as Conference Coordinator for the entire
conference. This will include organizing and administering non-technical parts
of AAAI-82 at CMU, such as publicity, publications, local arrangements (with
CMU assistance), and final accounting.
Garcia-Robinson will receive a fee of $15,000 for acting as Conference
Coordinator, to be paid quarterly.
Garcia-Robinson will also organize and promote a AAAI tutorial program and a
Research and Development Exposition.
For the tutorials, Garcia-Robinson will receive 5% (five percent) of the
gross tutorial receipts.
For the Research and Development Exposition, Garcia-Robinson will receive a
commission of 50% (fifty percent) of the net receipts of the Exposition.
AAAI guarantees Garcia-Robinson a minimum of $20,000 total income (fee plus
commissions) and a maximum of $35,000 for 1982.
Expenses incurred by Garcia-Robinson will be billed directly to AAAI. These
include, but are not limited to, secretarial assistance, mailing, telephone,
advertising, and travel expenses. AAAI will pay for at least two trips to each
conference site: one approximately a year in advance, and one at the time of
the conference.
! All expenses over $1000 are subject to approval by the Conference Chairman,
Jay M. Tenenbaum.
Participants in the R & D Exposition are subject to approval by the
Conference Chairman, Jay M. Tenenbaum.
Garcia-Robinson, Inc. is willing to be responsible for the sale and
distribution of the audio and videotapes of the AAAI 1980 Conference and future
conferences on a commission basis. The details of such an arrangement will be
handled in a separate agreement.
! APPENDIX II: AAAI - 1980 FINANCIAL STATEMENT
1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1980 FINANCIAL STATEMENT
INCOME EXPENSES
Dues 8,342.73
Membership Committee 933.82
AI Magazine 4,007.13
AAAI-80 Conference 71,487.00 47,198.43
Conference Coordination 6,500.00
AAAI-80 Tutorials 35,885.00 16,997.15
AAAI-80 Exhibits 12,050.00 3,137.83
Exhibits Coordination 6,575.00
Coordinator Bonus 5,500.00
AAAI-80 Proceedings (post conference) 726.76
AAAI-81 Tutorials 492.31
Legal fees (incorporation) 577.50
Interest 336.00 143.45
---------- ---------
TOTALS 128,827.49 92,062.62
NET INCOME 36,764.87
ASSETS
Corporate Income Fund 25,000.00
InterCapital Liquid Asset Fund 3,537.22
Membership Account (Stanford University) 2,905.16
Conference Account 2,605.74
Association Checking 2,750.77
---------
TOTAL 36,798.89
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1980 FINANCIAL STATEMENT
Membership Account
INCOME EXPENSES
Dues 8,342.73
Salaries and benefits 229.61
Postage 643.77
Supplies 60.44
---------- ---------
TOTALS 8,342.73 933.82
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1980 FINANCIAL STATEMENT
AI Magazine Account
Volume 1, No. 1 Summer 1980 (31pp.)
Printing 2,823.29
Layout 495.00
Art Supplies 27.87
Typesetting 493.80
Postage 138.30
--------
Total (1980, 1:1) 3,978.26
Advertising Rate Sheet 28.87
--------
TOTAL 4,007.13
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
AAAI-80 FINANCIAL STATEMENT
31 December 1980
CONFERENCE INCOME EXPENSES
Registration (818 attendees) 57,365.00
611 regular
207 student
Proceedings page charges 5,300.00
Proceedings sales at conference 1,180.00
Banquet 5,883.00 5,660.83
Stanford housing 1,759.00 1,130.50
Facilities 4,946.45
Receptions and refreshments 9,935.95
Secretarial services 1,313.60
Communications 949.43
Advertising 1,780.55
Printing 5,511.20
Proceedings (2000 copies) 9,535.70
Program Committee 3,869.62
Fredkin Prize expenses 1,019.20
Miscellaneous 823.80
Bad checks and adjustments 721.60
--------- ---------
CONFERENCE SUBTOTALS 71,487.00 47,198.43
Conference Coordinator 6,500.00
--------- ---------
CONFERENCE TOTALS 71,487.00 53,698.43
CONFERENCE NET INCOME 17,788.57
TUTORIALS INCOME EXPENSES
KBS Tutorial (169) 17,420.00
NL Tutorial (81) 8,140.00
R&D Managers Tutorial (109) 10,325.00
Lecturer fees 8,971.26
Facilities 903.97
Refreshments 314.14
Secretarial services 1,040.18
Communications 300.00
Advertising 1,780.55
Course materials 494.06
Videotaping 3,192.99
--------- ---------
TUTORIALS TOTALS 35,885.00 16,997.15
TUTORIALS NET INCOME 18,887.85
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
AAAI-80 FINANCIAL STATEMENT (concl.)
31 December 1980
EXHIBITS INCOME EXPENSES
Space rental 12,050.00
Facilities 1,346.75
Booth dividers 210.00
Secretarial services 200.00
Communications 557.07
Advertising 400.00
Printing 424.01
--------- ---------
EXHIBITS SUBTOTALS 12,050.00 3,137.83
Coordinator 6,575.00
--------- ---------
EXHIBITS TOTALS 12,050.00 9,712.83
EXHIBITS NET INCOME 2,337.17
---------- ---------
AAAI-80 SUBTOTALS 119,422.00 80,408.41
Coordinator Bonus 5,500.00
---------- ---------
AAAI-80 TOTALS 119,422.00 85,908.41
AAAI-80 NET INCOME 33,513.59
! APPENDIX III: AAAI - 1981 FINANCIAL STATEMENT
[Please note that the accounting for 1981 is not quite complete; in particular
I want to allocate some of the office expenses to IJCAI-81; that certainly
reflects 50% of Lou's time during the year. However, the statement is close
enough to provide a basis for the current decision. Don Walker]
1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1981 FINANCIAL STATEMENT
INCOME EXPENSES
Membership 9,153.57 5,101.94
AI Magazine 125.00 10,727.26
AAAI-80 Conference 1,160.00 304.26
AAAI-80 Tutorials 174.00
AAAI-80 Proceedings (G-R,Inc. 93) 898.06
AAAI-80 Proceedings (AAAI 50) 650.00 175.39
Sales Tax 16.43
AAAI-81 Tutorials 48,778.10 18,121.05
Coordinator Fees 2,429.00
AAAI-81 Exhibits 7,400.00 1,508.85
Coordinator Fees 3,700.00
Office
Personnel 15,783.13
Facilities 1,000.00 5,314.00
Materials 11,714.25
Legal fees 495.00
California state fees 7.50
Interest (through 12-2-81) 775.48
---------- ---------
TOTALS 69,940.21 75,572.05
NET INCOME (5,631.84)
ASSETS
InterCapital Liquid Asset Fund 25,199.03 (not final)
Membership Account (Stanford University) 1,319.83
Conference Account (Lou Robinson) 4,459.76
Association Account 666.00
---------
TOTAL 31,644.62
1980 ASSETS 36,798.89
DIFFERENCE (5,154.27)
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1981 FINANCIAL STATEMENT
Office Account
INCOME EXPENSES
Personnel
Coordinator 10,000.00
Office Staff 5,685.63
Services 97.50
---------- ---------
Personnel Subtotal 15,783.13
Facilities
Rental Deposit 900.00
Rent 1,000.00 4,200.00
Insurance 214.00
---------- ---------
Facilities Subtotal 1,000.00 5,314.00
Materials
Supplies 1,250.01
Furniture 1,251.68
Equipment 4,355.00
Telephone 1,553.65
MCI 185.83
Answering Service 99.16
Postal 793.01
PO Box Rental 13.00
Design/Typography 641.72
Printing 1,450.01
Copying 20.05
Miscellaneous 101.13
---------- ---------
Materials Subtotal 11,714.25
---------- ---------
TOTALS 1,000.00 32,811.38
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1981 FINANCIAL STATEMENT
Membership Account
INCOME EXPENSES
Dues 9,153.57
Salaries and benefits 2,885.45
Postage 863.82
Supplies 335.33
Printing 88.40
Copying 110.30
Computer 818.64
---------- ---------
TOTALS 9,153.57 5,101.94
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1981 FINANCIAL STATEMENT
AI Magazine Account
Volume 1, No. 1 Summer 1980 (31pp.)
Envelopes 52.08
Labeling 150.00
--------
Total (1980, 1:1) 202.08
Volume 2, No. 1 Winter 1980-81 (41 pages, 3000 copies)
Printing 3,726.30
Design and Layout 577.92
Art Supplies 35.00
Typesetting 508.00
Postage 156.15
Labeling 94.00
--------
Total (1980-81, 2:1) 5,097.37
Volume 2, No. 2 Summer 1981 (48 pages, 3000 copies)
Printing 3,776.70
Design and Layout 686.86
Art Supplies 26.84
Typesetting 322.67
Postage 142.50
Labeling 122.24
--------
Total (1981, 2:2) 5,077.81
Miscellaneous
Postal fees 100.00
Overhead 250.00
--------
Total Miscellaneous 350.00
TOTAL 10,727.26
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
AAAI-81 FINANCIAL STATEMENT
31 December 1981
TUTORIALS INCOME EXPENSES
Vision (78) 6,430.00 1,607.50
Robotics (114) 9,075.00 2,268.75
Computing Environments (189) 15,350.00 3,837.50
Expert Systems (217) 17,625.00 4,406.25
--------- ---------
Registration 48,480.00
Speaker fees (25% of 48,480) 12,120.00
Other 77.51
Syllabus sales 220.59
Refunds 664.62
Bad checks 719.15
Syllabus Production 2,516.44
Videotaping 2,100.84
--------- ---------
TUTORIALS SUBTOTALS 48,778.10 18,121.05
Professional Fees (5% of 48,580) 2,429.00
--------- ---------
TUTORIALS TOTALS 48,778.10 20,550.05
TUTORIALS NET INCOME 28,228.05
EXHIBITS INCOME EXPENSES
Space rental 7,400.00
Telephone 406.34
Mailing 196.56
Copying 11.73
R&D Guide 894.22
--------- ---------
EXHIBITS SUBTOTALS 7,400.00 1,508.85
Professional Fees (50% of 7,400) 3,700.00
--------- ---------
EXHIBITS TOTALS 7,400.00 5,208.85
EXHIBITS NET INCOME 2,191.15
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
AAAI-80 FINANCIAL STATEMENT
Revised 31 July 1981
(figures in parentheses identify changes from 31 December 1980 accounting)
CONFERENCE INCOME EXPENSES
Registration (818 attendees) 58,325.00 (+960)
611 regular
207 student
Proceedings page charges 5,500.00 (+200)
Proceedings sales at conference 1,180.00
Banquet 5,883.00 5,660.83
Stanford housing 1,759.00 1,130.50
Facilities 5,052.75 (+106.30)
Receptions and refreshments 9,935.95
Secretarial services 1,313.60
Communications 949.43
Advertising 1,780.55
Printing 5,511.20
Proceedings (2000 copies) 9,535.70
Program Committee 3,869.62
Fredkin Prize expenses 1,217.16 (+197.96)
Miscellaneous 823.80
Bad checks and adjustments 721.60
--------- ---------
CONFERENCE SUBTOTALS 72,647.00 47,502.69
Conference Coordinator 6,500.00
--------- ---------
CONFERENCE TOTALS 72,647.00 54,002.69
CONFERENCE NET INCOME 18,644.31
TUTORIALS INCOME EXPENSES
KBS Tutorial (169) 17,420.00
NL Tutorial (81) 8,140.00
R&D Managers Tutorial (109) 10,325.00
Lecturer fees 8,971.26
Facilities 903.97
Refreshments 488.14 (+174)
Secretarial services 1,040.18
Communications 300.00
Advertising 1,780.55
Course materials 494.06
Videotaping 3,192.99
--------- ---------
TUTORIALS TOTALS 35,885.00 16,997.15
TUTORIALS NET INCOME 18,713.85
! 1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
AAAI-80 FINANCIAL STATEMENT (concl.)
Revised 31 July 1981
EXHIBITS INCOME EXPENSES
Space rental 12,050.00
Facilities 1,346.75
Booth dividers 210.00
Secretarial services 200.00
Communications 557.07
Advertising 400.00
Printing 424.01
--------- ---------
EXHIBITS SUBTOTALS 12,050.00 3,137.83
Coordinator 6,575.00
--------- ---------
EXHIBITS TOTALS 12,050.00 9,712.83
EXHIBITS NET INCOME 2,337.17
---------- ---------
AAAI-80 SUBTOTALS 120,582.00 80,886.67
Coordinator Bonus 5,500.00
---------- ---------
AAAI-80 TOTALS 120,582.00 86,386.67
AAAI-80 NET INCOME 34,195.33
! APPENDIX IV: AAAI - 1982 BUDGET
1-4-82
AMERICAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE
1982 BUDGET
INCOME EXPENSES
Membership 35,000.00 7,500.00
AI Magazine 5,000.00 15,000.00
AAAI-82 CONFERENCE 75,000.00 55,000.00
AAAI-82 TUTORIALS 75,000.00 35,000.00
AAAI-82 EXHIBITS 20,000.00 12,000.00
Videotape Distribution 5,000.00 2,750.00
Proceedings 2,500.00 800.00
Office
Executive Director 30,000.00
Office Staff 20,000.00
Services 1,500.00
Rent 2,400.00 8,400.00
Insurance 214.00
Telephone 1,000.00
Mailing 200.00
Supplies 500.00
Equipment 2,000.00
Copying 250.00
Legal fees 500.00
Interest 1,000.00
---------- ----------
TOTALS 220,900.00 192,614.00
PROJECTED NET INCOME 28,286.00
!
-------
∂16-Jan-82 1555 RPG Lisp Conference
For this trip to Indiana to review papers, is Stanford supposed to
pay for me to go? Should I try to get LLNL to pay 1/2?
-rpg-
Try to get LLNL to pay half.
∂17-Jan-82 1551 CLT
I think I won't come back for supper tonight.
Artis Wodehouse is doing `Great violinists of the Past'
(from the record archives) tonight at 8.
Do you want to go? I think I will.
∂18-Jan-82 1113 Susan Hill <CSD.HILL at SU-SCORE>
Date: 18 Jan 1982 1104-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
To: JMC at SU-AI
In-Reply-To: Your message of 18-Jan-82 1056-PST
Right, he uses only SAIL; but he has 3 aliquots or 150 blocks. A minimum
or guest account is 25 and there is no charge for that.
-------
I'll find out something of what he's doing. There are other possibilities
for his disk. I'll be back to you.
∂18-Jan-82 1205 reid@Shasta (SuNet)
Date: Monday, 18 Jan 1982 12:04-PST
To: John McCarthy <JMC at SU-AI>
Cc: csd.schreiber at SU-SCORE, csd.herriott at SU-SCORE
Reply-to: BKR at SU-AI
In-reply-to: Your message of 18 Jan 1982 1110-PST.
From: reid at Shasta
Course 108 is for sophomores. 156 is usually considered challenging by
graduate students. Perhaps the course description should mention this.
I have taught this course to sophomores at CMU twice, once as a teaching
assistant and once for real. I find that it is extremely valuable to get
them started early at thinking that there are formalisms for describing and
grounding programs. The existence of Turing machines and the notion of
computability forms a single lecture in the course; formal languages form
another lecture. The reason for including them in the course description
even though they form such a small part of the directly presented material
is that I wanted to make sure students understood that this was not just
another hacking course.
It sounds like I should get on the stick and make up the week-by-week
curriculum for this course in order that people who haven't been attending
the meetings of the committee in which it was planned and discussed can get
a better idea of what will be taught in it. In my opinion the course
material, as I intend to present it, should be part of the education of
every professional programmer. If there were an undergraduate-level
course like 156 (or 142, for another domain) then I would argue that a
person who intends to become a Ph.D.-level researcher in computer science
should learn this material from those courses, but since there isn't, I will
argue that the would-be computer scientist should take this course sequence
also.
Brian
Perhaps I was hasty. I don't think a week-by-week outline should be asked
of you, any more than from anyone else. I'll react further later.
∂18-Jan-82 1332 FFL
To: JMC, FFL
Sarah called. She asks that you call her tonight at 10 if you can.
∂18-Jan-82 2126 JJW E/EKL interface
To: JMC, JK
I have written a set of E macros and Maclisp functions whose purpose is to
make the E/Maclisp interface more palatable with an EKL subjob. They are
described on EEKL.DOC[EKL,JJW]. I would appreciate any suggestions for
additional functions or ways in which the current functions can be improved.
∂19-Jan-82 0853 FFL
To: JMC, FFL
Edward Teller's office called. Would like you to call re a possible
meeting with him on Friday, the 22nd. 7-0601.
∂19-Jan-82 0854 FFL
To: JMC, FFL
Edw. Feigenbaum's secretary called to remind you of a meeting with him
at 10 a.m. and a meeting with George Lindemude at ll a.m.
∂19-Jan-82 1616 pratt@Shasta (SuNet) Sun@BBN
Date: 19 Jan 1982 16:16:07-PST
From: pratt at Shasta
To: equip, sun
Subject: Sun@BBN
For an outside perception of Sun workstations in the role of "General
Computing Element" (a term that should trigger visions of PDP-11/03's), see
<CSD.PRATT>GCE on Score or /mnt/pratt/gce on Shasta. (Suns don't get
mentioned until about page 8, be patient.)
The bottom line is that Suns are ideal for what BBN wants in a GCE.
Apparently BBN is gearing up for a massive Sun software effort, which would
make yet another source of Sun software in addition to Stanford and
Lucasfilm (and MIT at Sun/Nu-independent level).
I have heard the argument that the reason for paying twice as much for a
Star as for a Sun was software. That's a position that is getting harder
to maintain these days. One could reasonably take the position that Star
software has already fallen behind Sun software now that Lucasfilm Unix runs
on Suns, with all that that implies in access to Unix software the world
over. Moreover about 300 Sun processors have been delivered to date by the
various manufacturers, a situation very conducive to further Sun software
development. (I don't have the corresponding figure for Stars.)
Vaughan
∂19-Jan-82 1653 JK
∂19-Jan-82 1444 JMC commutativity and associativity
To: JK
CC: JJW
I have declared *, i.e. append, to be associative, and this seems to
work fine; it makes one of the proofs go without special appeal to
associativity. I have advocated to Yoram that he declare + and times
to be associative also. However, it occurs to me that declaring them
associative may cause problems in using their commutativity. Suppose,
for example, that we want to prove a+b+c+a = a+a+b+c. If EKL doesn't
allow us to group the terms freely, then there will be difficulty in
using associativity. There will also be a problem in matching
subexpressions. It seems to me that allowing operators to be
declared associative is good and even necessary, but it imposes further
requirements on interactive theorem provers such as EKL. These seem to
include
1. Parts designators should be able to pick out segments
-------------------
You should be able to prove a+a+b+c=a+b+c+a using ∀x.a+x=x+a; i.e.
the rewriter knows some about associative operators and attempts
to match segments.
∂20-Jan-82 1259 JJW Class notes and Lisp textbook
To: "@CLASS.DIS[258,JJW]"
To cover the cost of reproducing class handouts, please bring $5.00 to
class as soon as possible. This is for an estimated 100 pages at 5 cents
per page. Also, if you want a copy of McCarthy and Talcott, "LISP:
Programming and Proving", you can get one from me; the cost is $6.00.
Checks may be made payable to Stanford University.
Joe
∂20-Jan-82 1427 FFL Course in Logic Programming and PROLOG starting soon.
To: SU-Bboard at SU-AI, Bboard at SRI-AI, Bboard at PARC-MAXC,
FFL at SU-AI, JMC at SU-AI
Starting in the week of February 2 and continuing through the quarter
there will be a course in logic programming and the PROLOG language, taught
by Antonio Porto of the University of Lisbon. Logic programming in general
and PROLOG in particular have attracted wide interest in Europe but are only
getting started in the United States. There will be an opportunity to use
PROLOG.
There will be an organization meeting to fix hours on Thursday,
January 28, at 3 p.m., in 252 Jacks Hall (Computer Science Dept.). Anyone
interested who can't make the meeting or who has questions should contact
John McCarthy by telephone or net message.
∂20-Jan-82 1742 David Warren <WARREN at SRI-AI> Antonio Porto's Visit
Date: 20 Jan 1982 1741-PST
From: David Warren <WARREN at SRI-AI>
Subject: Antonio Porto's Visit
To: JMC at SU-AI
cc: Warren at SRI-AI
I hear Antonio Port is visiting next month and will
give a course on Prolog. Please send me news of further
developments, and I'd be glad to assist in any way I can.
David Warren.
-------
He'll arrive on the 26th.
∂21-Jan-82 0846 LOUNGO at RUTGERS DCS Technical Reports
Date: 21 Jan 1982 1138-EST
From: LOUNGO at RUTGERS
Subject: DCS Technical Reports
To: Arpanet-mailing: ;
cc: loungo at RUTGERS, petty at RUTGERS
By the way, below is the publication order form list. You can access
the list of abstracts for these reports via FTP with user account <anonymous> with any password. The file name is:
<library>online-tecrpts.doc
PUBLICATION ORDER FORM
RUTGERS UNIVERSITY
Computer Science Department
New Brunswick, New Jersey 08903
January 1982
The Department of Computer Science at Rutgers University, New Brunswick, is
pleased to announce the availability of new Technical Reports. An abstract of
each is attached.
Use this form to order the announced publications. Due to limited quantities,
we will fill orders on a first-come, first-served basis.
To order, please check the box next to the title of the publication(s) that you
are interested in. Then detach this sheet, refold so that our address is
facing out, attach a first-class stamp, and mail.
[ ] DCS-TR-103 NUMERICAL UPSTREAM BOUNDARY CONDITIONS THAT REDUCE SPURIOUS
REFLECTION, R. Vichnevetsky, E. Sciubba, Y. Pak, August 1981.
[ ] DCS-TR-105 MEDIT USER MANUAL, Robert N. Goldberg, March l981.
[ ] DCS-TR-106 KNOWLEDGE REPRESENTATION AND PROBLEM SOLVING IN MDS, Chitoor
V. Srinivasan, November 1981.
[ ] DCS-TR-107 NOTE ON LEARNING IN MDS BASED ON PREDICATE SIGNATURES, Chitoor
V. Srinivasan, November 1981.
[ ] DCS-TR-108 SINGULAR INTEGRAL EQUATIONS - THE CONVERGENCE OF THE NYSTROM
INTERPOLANT OF THE GAUSS-CHEBYSHEV METHOD, Apostolos
Gerasoulis, October 1981.
[ ] DCS-TR-109 IMPROVED CONSTRAINT SATISFACTION ALGORITHMS USING INTER-
VARIABLE COMPATIBILITIES, Bernard Nudel, December l981.
[ ] LCSR-TR-20 SEMANTIC INTEGRITY TRANSACTIONS IN DESIGN DATABASES, Charles
M. Eastman, Gilles M. E. Lafue, October 1981.
[ ] LCSR-TR-22 A FILE ORGANIZATION BASED ON EXTENSIBLE HASHING, Richard King,
December 1981.
[ ] LCSR-TR-23 A TRANSFORMATIONAL FRAMEWORK FOR AUTOMATIC DERIVED DATA CONTROL
AND ITS APPLICATIONS IN AN ENTITY-RELATIONSHIP DATA MODEL,
Shaye Koenig, December 1981.
[ ] LCSR-TR-24 THE CASE FOR A SETL BASED QUERY LANGUAGE, Ravinderpal
S. Sandhu, December 1981.
[ ] CBM-TM-92 KNOWLEDGE BASED ACQUISITION OF RULES FOR MEDICAL DIAGNOSIS,
George A. Drastal, Casimir A. Kulikowski, October 1981.
[ ] CBM-TR-124 COMMON VIRTUAL ARRAYS IN PDP-11 FORTRAN: AN EXERCISE IN
SOFTWARE ENGINEERING, Bob Schwanke, November 1981.
[ ] CBM-TR-125 REPRESENTING KNOWLEDGE IN AIMDS: INTRODUCTION USING TAXMAN
EXAMPLES, N. S. Sridharan, December 1981.
-------
∂21-Jan-82 1250 JJW EEKL single-character macros
Having EEKL[EKL,JJW](2) in your OPTION.TXT file no longer defines
single-character macros. To get the single-character macros as
described in the documentation, you now need EEKL[EKL,JJW](2:3).
∂21-Jan-82 2312 Robert Elton Maas <REM at MIT-MC>
Date: 22 January 1982 02:02-EST
From: Robert Elton Maas <REM at MIT-MC>
To: JMC at SU-AI
Please send me instructions as to how to find a wife.
I have no special reason to suppose that I can be much help, but if you
would like to have dinner some evening and discuss your problems, I would
be glad to do it - perhaps early next week.
∂22-Jan-82 0104 Robert Elton Maas <REM at MIT-MC>
Date: 22 January 1982 04:02-EST
From: Robert Elton Maas <REM at MIT-MC>
To: JMC at SU-AI
My problem is that I'm 36.7 years old already and although I've found
a few women that I would like to be married to none of them have
been interested in starting a serious relationship with me (like going
out once in a while, like getting together to do things together),
not to mention wanting to marry me.
How do I find somebody I want who wants me?
Would discussion help any?
∂22-Jan-82 1133 JJW EKL syntax
To: JK, JMC, YOM
I've created a file called EKL.DEF[EKL,JJW], whose purpose is to collect
comments and suggestions for a more formal definition of the syntax and
semantics of EKL's input language. Please feel free to write on this file.
∂22-Jan-82 1155 FFL
To: JJW, FFL, JMC
The latest is a call from our Travel Agent saying that Porto leaves Lisbon
on TWA Flight 903 which arrives in NYC at 4 p.m. He leaves NYC at 6:45 p.m
on TWA Flight 901 which arrives SFO at 9:51 p.m., January 29, Friday.
Agent says to check with TWA to see if Flight 903 arrived on time. Flight
901 may be on time, but if he missed it, he wonit arrive as planned.
∂22-Jan-82 1338 FFL
To: JMC, FFL
Eric Martinez of the Office of Tech. Licensing would like you to call him
at 7-0651.
∂22-Jan-82 1512 FWH Prolog talk at SRI
To: "@TEMP.DIS[1,FWH]" at SU-AI
Date: 20 Jan 1982 1701-PST
From: Waldinger
Subject: prolog talk by komorowski, friday, jan 29
To: AIC-Associates:, CSL:
time: 10:30 am, friday, jan 29
place: aic conference room, ek242
title and abstract follow:
AN ABSTRACT PROLOG MACHINE
ABSTRACT
An abstract Prolog machine is specified in a dialect of Meta-IV. The
purpose of this specification is to provide a formal AND
implementation-oriented definition of the Prolog prtogramming language. The
specification is written in a purely APPLICATIVE style. Due to a good
modularity it is possible to identify main problems in implementing Prolog.
The specification has been used elsewhere for a rigorous implementation of
an interpreter of the language, for a formal description and implementation
of partial evaluation transformations, and a a specification of interactive
programmer's tools (eg break, trace, spy). It can be also further refined
toward an improved efficiency.
-------
jan komorowski will be visiting sri from linkoping, sweden. his talk
will be in building e, on ravenswood avenue across from the pine street
intersection, in menlo park. visitors who arrive in the lobby will be
escorted to the seminar room.
-------
∂22-Jan-82 1611 FFL Visitor from MIT on Feb. 3
To: JMC, FFL
Jeff Ullman asks if you would like to speak with Tom Layton sometime on
Feb. 3 in the a.m., from CSL, MIT.
∂22-Jan-82 2017 Ed Katz <katz.Hp-Labs@UDel> Prolog Course
Date: 22 Jan 82 15:48:29-PST (Fri)
From: Ed Katz <katz.Hp-Labs@UDel>
To: JMC at Su-Ai
cc: bs.Hp-Labs at UDel
Subject: Prolog Course
Via: HP-Labs; 22 Jan 82 23:15-EST
I would like to attend the class, but will be unable to make
the organizational meeting. Will the class meet at the same time
as the organizational meeting?
Ed Katz
The point of the organizational meeting is to set a time for the class.
As it happens, Porto won't arrive till the day after the organizational
meeting which will be held as announced. Therefore, all it can do is
set a time. It would be best for you to try to be represented.
∂23-Jan-82 0226 Arthur Keller <CSD.KELLER at SU-SCORE> Gray Tuesday
Date: 23 Jan 1982 0225-PST
From: Arthur Keller <CSD.KELLER at SU-SCORE>
Subject: Gray Tuesday
To: CSD-Faculty: ;
cc: CSD.MWalker at SU-SCORE, CSD.Trickey at SU-SCORE,
CSD.Dietterich at SU-SCORE
Reply-to: ARK@SU-AI
Gray Tuesday is tentatively scheduled for Tuesday, 23 February at 2:30pm.
If we are all prepared it should take beyond 4pm. If this day or time is
a problem for any of you, please send me a message quickly.
Gray Tuesday is part of the annual evaluation of all students in the
department. At this meeting, each student is reviewed to see if he/she
is making satisfactory progress. Warning letters are sent out to those
who are not making satisfactory progress. During late spring, the Black
Friday meeting takes place. This is a serious evaluation of students
who were warned at Gray Tuesday. Actions to take are decided at Black
Friday.
In about a week, each of you will receive a list of all of the students
that you are listed as advising, as well as information about each
student. Before Gray Tuesday, you should meet with each student to discuss
his/her progress.
Please contact me if you have any questions about this.
-------
∂23-Jan-82 1000 JMC*
cannon,tob,scheinman,bloom
∂23-Jan-82 1038 TOB
John
The file is protected.
Tom
∂23-Jan-82 0125 JMC
See also DAILY.7[LET,JMC] for a draft.
∂23-Jan-82 1101 TOB
1) I think the statement by Bernie Roth that
we learned about robotics from the Russians is a total exaggeration, and the
*** I agree. I think that we have learned next to nothing from the Russians.
They publish very little outside. Much of their work in vision is in secret
institutions. I might tell you of a visit there. Perhaps Bernie has learned
something on his visits, but I have gotten nothing.
2) statement that it is impractical to put restrictions on Umnov if he visits
is essentially false.
*** I agree that it is false. Restrictions may involve personal supervision.
It may be enough just to warn some of us about the visit and ask us not to
accept visists or disclose what we are doing, and to make a statement to the
visitor that he is not to visit robotics.
I think overall that it is reasonable to make efforts to favor US exploitation
of our research. I disagree with the efforts concerning VLSI and robotics
to avoid any restrictions on technology transfer. I do not think that there
is free information flow from the Japanese in VLSI, which is regarded as
very important and is done in secret in industry labs (so I was told by
Japanese in high positions). I have no evidence of secrecy in robotics in
public institutions, but there is secrecy in industry, as we would expect.
I personally accept most Japanese visitors, and would accept Russian visitors,
probably. I strongly differentiate between visitors from US companies and
foreign visitors in what I tell them. I don't want you to tell anyone that
because I would like it not to be known.
∂23-Jan-82 0055 JMC Umnov and Roth
Please look at UMNOV.NS[W82,JMC]. I think the statement by Bernie Roth that
we learned about robotics from the Russians is a total exaggeration, and the
statement that it is impractical to put restrictions on Umnov if he visits
is essentially false. I am thinking about a letter saying so either to the
Stanford Daily or the New York Times. This is is independent of the question
of whether such restrictions are desirable. What is your opinion?
Would you be willing to take part in a joint letter or even a press conference
dissenting from the official Stanford view? We might ask a few other people,
e.g. Cannon and deBra, but we'd do something even if they weren't inclined
to participate. I guess a letter is appropriate in any case and a press
conference if there are enough, say 4 or 5.
∂23-Jan-82 1112 TOB
Potential military applications are
''very small. ... I don't think anyone will win a war on the basis of
those machines. It's certainly not of critical national interest.''
**** I disagree with the statement. Does the cruise missile have an
military importance? I think that military applications are part of the
reason that much Soviet work is in secret institutions. I don't like
to stress military applications of robotics, but it is a reality. I
think Roth may refer to manipulators and locomotion.
He added that nothing is being done in the field that the Soviets
''either don't already know or can easily figure out for themselves.''
*** Could be true, but I would be surprised if they can't learn from
us, even from a different approach and different problems addressed.
Since they don't make available what they do, we don't have the
opportunity to look at their contributions.
One of us has visited the Soviet Union more than ten times,
and our experience is that access to Soviet work is best achieved by
hard bargaining.
*** I am not sure about this, not sure that there is an effective access.
We are so open about most things that anyone with talent and imagination
could find out what he wants to know from open sources.
∂23-Jan-82 0125 JMC
See also DAILY.7[LET,JMC] for a draft.
∂23-Jan-82 1952 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Meeting
Date: 23 Jan 1982 1947-PST
From: Robert Schreiber <CSD.SCHREIBER at SU-SCORE>
Subject: Meeting
To: csd.schreiber at SU-SCORE, csd.rwf at SU-SCORE, csd.genesereth at SU-SCORE,
csd.pratt at SU-SCORE, csd.dbrown at SU-SCORE, csd.mwalker at SU-SCORE,
csd.reges at SU-SCORE, csl.jlh.noah at SU-SCORE, csd.gass at SU-SCORE,
csd.jock at SU-SCORE, csd.sis at SU-SCORE, csd.golub at SU-SCORE,
csl.lantz at SU-SCORE, csl.bkr at SU-SCORE
cc: csd.mccarthy at SU-SCORE
IMPORTANT: Since Brian will not be here next week, the meeting
is postponed. We'll try for the following Tuesday. I'll keep you posted.
ROB
-------
∂23-Jan-82 1957 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Curriculum ctte meeting
Date: 23 Jan 1982 1955-PST
From: Robert Schreiber <CSD.SCHREIBER at SU-SCORE>
Subject: Curriculum ctte meeting
To: csd.mccarthy at SU-SCORE
John,
We hope to hold a meeting of the curriculum committee soon to
discuss the content of cs108 in greater detail. I hope you will come
and present your views on the suitability of automata theory for undergrads,
and anything else you wish. I very strongly hope that 108 becomes the course
for math sci majors -- I hope we can, therefore, reach an agreement on
its contents.
ROB SCHREIBER
-------
∂23-Jan-82 2049 Lowell Wood <LLW at S1-A> Rewarding Bad Behavior
Date: 23 Jan 1982 2044-PST
From: Lowell Wood <LLW at S1-A>
Subject: Rewarding Bad Behavior
To: jmc at SU-AI
CC: LLW at S1-A
∂22-Jan-82 2319 John McCarthy <JMC at SU-AI>
Date: 22 Jan 1982 2322-PST
From: John McCarthy <JMC at SU-AI>
To: llw at S1-A
Congratulations on Lawrence award.
[John: Thanks. It has been widely believed that I've opposed The Will Of
Heaven so often and so annoyingly that I couldn't ever get the Lawrence
Prize. However, it appears that my friends, feeling frisky after the last
national elections, abruptly ganged up on The System and staged a
micro-coup-d'etat (for how else could this be regarded?). Lowell]
∂24-Jan-82 1251 CLT
did you want to go to richards party this evening?
Sure. For a while anyway. You?
∂24-Jan-82 1747 Jim McGrath <CSD.MCGRATH at SU-SCORE> FBI action on our computer breakin?
Date: 24 Jan 1982 1745-PST
From: Jim McGrath <CSD.MCGRATH at SU-SCORE>
Subject: FBI action on our computer breakin?
To: su-bboard at SU-SCORE
cc: reg at SU-AI, jmc at SU-AI
The following message was sent to the LOTS BBoard yesterday, and I
thought it would be of general interest to the community. Does anyone
know which machine was broken into? And whether this story is in
anyway accurate? If it is, I am rather surprised that no one has sent
out some sort of announcement on this topic.
The following item appeared in this morning's San Jose Mercury-News:
COMPUTER `TAPPED' AT STANFORD
The FBI has begun a preliminary investigation into the unauthorized use of
a Stanford University research computer connected to a Defense Department
agency, an FBI spokesman said Friday.
University police said that "an unidentified person gained access and used
a nationwide computer system without authorization" sometime before Jan.20.
Ralph Gorin, who is in charge of the $3.5 million computer system in
Stanford's Computer Science Department, said the possible damage to
information contained in the computer was minor, but the fact that someone
broke through computer codes to gain access "is alarming to us."
Stanford computer-science professor John McCarthy said someone apparently
called into the computer from an outside telephone.
McCarthy said the computers involved are tied into the Defense Advanced
Research Project's Agency.
-------
What I know came from Ralph. Perhaps I should not have spoken to the reporter
at all given that fact, but I felt obliged to convince her (Alice Cuneo in case
you should have the misfortune to meet her) that the computer couldn't be used
to fire missiles or write checks. If someone offers $3.5 million for the
system in Margaret Jacks, the Department should accept.
∂24-Jan-82 2138 Feigenbaum at SUMEX-AIM two things
Date: 24 Jan 1982 2135-PST
From: Feigenbaum at SUMEX-AIM
Subject: two things
To: jmc at SU-AI
cc: feigenbaum at SUMEX-AIM
Hi, John! I tried calling you this evening, but no answer so I decided
to send a message.
First thing: I am helping a group of students (about 10-15) with a paper-
studying seminar that meets on Monday at 3:30PM in Rm. 301. This monday
(tomorrow, or today depending on when you read this) they are discussing
Doyle's AI Journal paper on Truth Maintenance System. You're on top
of this, I think. Would you care to drop in for an hour or so to discuss
this?
Second, we're coming down to the wire on the Lenat reappointment. Lenat has
given me some papers, representing writeups of what he has done since last
year. The faculty will consider the issue a week from Tuesday (Feb. 2).
As you recall, when Lenat's 3-yr renewal as an Assistant Professor came up
last year, we renewed him for only one year, and reserved judgment on the
other two years until now, to spur him on to make some scientific
accomplishments. I will have the material xeroxed and sent to you so you
can read the material and form the necessary judgments. I have not yet
read the material myself, so I dont have a judgment yet.
Best regards,
Ed
-------
∂24-Jan-82 2157 LLW One-Way Exchanges
To: JMC at SU-AI
CC: llw at S1-A, rah at S1-A, HPM at SU-AI
∂23-Jan-82 HPM 24-Jan-82 JMC Secure robots
a021 0047 23 Jan 82
PM-Scientist, Bjt,580
Universities Rebuff State Department Restrictions on Soviet Visitor
STANFORD, Calif. (AP) - A widening dispute over the federal
government's attempt to extend anti-Soviet sanctions to academia has
lopped two universities from the itinerary of a Soviet scientist whose
specialty is robots.
Stanford University has been declared off-limits to Nikolay Umnov
because the school refused to honor State Department restrictions on
his visit. The university said it could not successfully police a
visit, and that to make the attempt would disrupt the free environment
needed for creative work to take place.
The University of Wisconsin followed Stanford's lead Friday, saying
it was an ''open institution.''
Wisconsin Asssociate Dean Camden A. Coberly and professor Ali A.
Seireg said they plan to withdraw their acceptance of Umnov's visit.
That move will reduce the scientist's itinerary to Auburn University
and Ohio State University.
But at Ohio State University, where Umnov was to spend six weeks,
professor Robert B. McGhee said Umnov would be ''welcome for two or
three days. I'd accept restrictions for a visit of that length, but
not any longer.''
Umnov has been caught in a dispute over what visiting Soviet
scientists can see and do during a three-month scientific exchange
program that is sponsored by the National Academy of Sciences and
funded by the U.S. government.
As part of the Reagan administration's attempt to restrict Soviet
access to U.S. technology, the State Department said Umnov couldn't
see anything not already published in open literature.
It also said he should be restricted to mechanical theory of robotic
locomotion, with no industrial visits and no access to control units
or computer programming techniques that have given robots new
flexibility and made them more useful in the workplace.
The measures are justified because the government is paying for the
visit, said Sue Pittman, a State Department spokeswoman. She also
cited ''possible technology transfer risks,'' an administration term
for thefts or leaks of U.S. scientific secrets.
The restrictions reached universities hosting Soviet scholars via
letters from the national academy. As criticism of the restrictions
mounted, the NAS on Tuesday halted distribution of the letters.
The restrictions are ''absurd,'' according to Bernard Roth, a
professor in Stanford's Department of Mechanical Engineeering.
''Guys in the U.S. who know something about it (robotics) learned
from the Russians,'' he said. Potential military applications are
''very small. ... I don't think anyone will win a war on the basis of
those machines. It's certainly not of critical national interest.''
He added that nothing is being done in the field that the Soviets
''either don't already know or can easily figure out for themselves.''
In refusing to honor the restrictions, Stanford Vice-Provost Gerald
J. Liberman said, ''We believe the best interests of American science
and technology are served by open exchanges of university research
activities.''
He said in a letter to the NAS, ''The campus is completely open; and
the thousands of visiting scholars to campus each year have free
access to any of the university's programs. Even if we had the means
to monitor or police the activities of visitors, such actions would
drastically disrupt the academic environment which is essential in
fostering creative research endeavors.''
Attempts to restrict visiting scientists are not new, and last
February Stanford President Donald Kennedy expressed ''grave concern''
over federal attempts to apply export controls to academic teaching
and research. He was joined in writing protests to the secretaries of
state, defense and commerce by presidents Marvin L. Goldberger of the
California Institute of Technology, Paul E. Gray of MIT, Frank H.T.
Rhodes of Cornell and David S. Saxon of the University of California.
Kennedy told the faculty senate earlier this month the universities
had done ''sporadic negotiating'' with some agencies and had had
''skirmishes'' with others.
ap-ny-01-23 0346EST
***************
kjk - Question for the robotics people: isn't Umnov in robot locomotion?
Aren't the Russians significantly ahead of us in locomotion?
What are the expected gains from Russian→US technology transfer vs.
US→USSR?
In other words, do we stand to gain more than Russia from Umnov's
visit? If so, a letter to the State Department informing them that they
are applying an import restriction rather than an export restriction
may be in order.
A brief review of the status of known Russian work in robotics might
be useful for clarifying the tradeoffs. Is anyone here willing to do that?
Lest people get the wrong impression--I'm opposed to applying
export restrictions to academic work. I think the State Dept is wedged.
However, a specific instance of the loss we incur from these restrictions
may do more than a thousand pages of moral, ethical, and philosphical
argument.
TAW - If the US Gov't doesn't want Umnov to know anything about
state-of-the-art robotics, why are they letting him in?? That seems
to be the way to handle these things, as opposed to making every
US technologist in a sensitive field take Diplomacy courses.
JMC - I think that both Roth and Lieberman have made misstatements in
order to make their case stronger. In my knowledge and Tom Binford's,
the Russians are mainly behind us in robotics. Only in walking machines
have they done anything substantial. Of course, they are very secretive
about what they do, and before they receive any visitor at an institute,
they determine what he may and may not be shown and what may be discussed.
Many institutes are secret and known only by Post Office Box numbers within
the country. When people from one such institute (the Institute of Applied
Mathematics in Moscow) wanted to talk with me in 1965, they pretended to
be from a different institute. Only later, when that institute wanted to
buy an American computer, did its existence emerge from secrecy; I never
got to visit it.
[LLW - IAM is well-known for its central role in Soviet nuclear weapons
development, among other military activities; their mathematics is highly
`applied.']
My attempts to collaborate with Ershov on MTC broke down mainly because
it was illegal for him to put scientific information in a letter except
by sending published documents. It is illegal for a Russian to give a
foreigner written scientific information except in the form of published
books or papers. In the 1960s, these rules were often bent, but exceptions
have been few since then.
[Very few Westerners are aware of the `published documents only'
restriction on Soviet scientists, but it is quite rigidly obeyed (and
enforced, I'm told). I can't recall an instance where I learned technical
information from Soviet scientists that which I couldn't have read--in
unclassified but somewhat obscure publications--in the US beforehand.]
Roth is also wrong in saying that robotics has no
military applications. Lieberman's statement
''We believe the best interests of American science
and technology are served by open exchanges of university research
activities.'' contains the assumption that putting no restrictions
on Umnov would achieve that result.
[Roth is so completely wrong that he is clearly either an ideologue or a
fool. Robotics is one of the technological fields most replete with military
applications, both present and (near) future; his assertion is absolutely
incredible. Lieberman's remarks are those of a parasite who is totally
uncaring for the health of his host; the world does *not*, contrary to his
evident belief, begin and end with the wonderfulness of `university
research activities.' The security of his fellow citizens and their
government is of some moment, which he completely fails to acknowledge:
American science-and-technology exists only in the context of the American
political system, which supports its *play* generously and asks for little
in return.]
Finally, the statement that restrictions are impractical and would
disrupt the University is phoney. A visitor's schedule is made by
the University's "Office of Foreign Visitors", and nothing is easier
than leaving something off the schedule and notifying the Principal
Investigators of a few DoD sponsored projects that this has been done.
It doesn't prevent someone who wants to from telling all in the coffee
shop, but it would probably satisfy the State Department and have a
high probability of accomplishing what they want.
[Again, most perceptive remarks.]
It is often argued that the Soviets can get all the information they
want by monitoring the literature. It isn't easy to tell to what
extent this is true. Certainly they are often behind in many areas,
and in many areas of computer science, their work is derivative - e.g.
putting on a Soviet computer systems done earlier in the U.S. Of course,
there is the possibility that derivative work is all we hear about.
Overall, it seems that restrictions have important but random effects.
Of course, the restrictions the Soviets place on travel by their own
people are far more effective than our restrictions, but the effects
are additive.
[The mere fact of existence of several very large (thousands of
professionals each) KGB and GRU Directorates charged exclusively with the
identification and covert transfer of technology from the West gives the
lie to this `argument,' which is really just a declaration of faith, not
an argument at all. The Soviets are manifestly behind the West in
essentially *every* area of technology, except for a few which are pivotal
to current (and previous) military technology. I challenge anyone to name
three notable counterexamples to this assertion! Really effective
restrictions on technology transfer imposed today would eliminate the
Soviet Union as a world power by the end of this century, without any
doubt, with little cost and inconvenience. It is mindlessly greedy
businessmen and ego-blinded academics in the West--primarily the US--who
prop up the Soviet system day by day with myriad tiny streams of high
technology which this `genetically defective' system can't effectively
generate on its own and absolutely must have to remain a major power.]
Stanford is imposing restrictions on access to information, because the
Center for Integrated Systems will give special access to the companies
that are paying for it. Even the Computer Forum gives special access to
its members. The CIS has been agonizing for the better part of a year
about how best to give privileges to the 12 companies.
All the above constitutes criticism of the misstatements by Roth and
Lieberman. There still remains the question of whether the particular
action of the State Department is reasonable. If I had to make a
decision myself, I'd want more information about what the Russians do,
how open they have been, and what the military applications are.
On the basis of my experience and the information contained in the
article, their position seems a reasonable compromise.
[The State position is indeed reasonable in the abstract, but its
implementation was botched (as usual), primarily through untimely motion.
The synaptic delays in the State bureaucracy would have been a disgrace in
a Brontosaurus; in the aggregate, they are a (continuing) National scandal.]
In contrast to this, Stanford seems to be taking an absolutist position,
and its spokesmen are inventing facts to fit the desired conclusions.
Perhaps the position is a leftover from the 1960s.
[The basic problem is that it became fashionable in Academe to reflexively
oppose to whatever the Government might be doing, 1-1.5 decades ago, and
such fashions die hard (the more so as some work actively to keep them
alive). Inventing facts is one of the *more* savory features of such
opposition.]
Apart from cost-benefit considerations, there is a political issue.
The AP story will be summarized in the Soviet press
as follows: "In spite of the efforts of
the mad dog Reagan to whip up anti-Soviet hysteria over the false
issue of Poland, American scientists realize the leading role of
the Soviet Union in robotics as in other fields. They know the
Polish issue is false". For more restricted circles,
the assurance may be offered that Comrade Umnov is politically
mature and will make certain that he gets more than he gives.
Should anyone wish to be actually constructive about exchanges
with the Soviet Union, I would suggest the following:
[`Exchanges' with the Soviets are `tutorials'--let's be honest about it.
I've *never* been told something really worthwhile by a Soviet scientist
(e.g., something striking which I didn't already know from the
literature--usually ours). Who has a notable counterexample to offer?]
1. Form an estimate of what the Russians are doing in certain fields,
e.g. robotics or computer science generally. Make a guess as to where
they are being open and where they are being secretive. The CIA, which
works for us you know, could be asked to help; it interviews many
emigrants from the Soviet Union and has other sources of information.
[The CIA is a surprisingly high quality source of information, virtually
across the technical spectrum. Unfortunately, they're so sensitive about
revealing (by omission) what they *don't* know (and about compromising
their sources, both East and West) that they hopelessly over-classify
their assessments of Soviet technological positions. As a result, these
assessments are available only to (almost invariably) technically
incompetent political types. . .]
2. If we decide there is something we want to know or institutions we
think we would like to visit, the State Department or the National
Academy of Sciences can bargain on our behalf. Their bargaining will
be more effective if they get good co-operation. Our problem will be
that very few scientists will find the Soviet Union interesting enough
for long visits.
[We will *never* get into Soviet institutes of real interest, i.e., where
they are doing technology development comparable to the best in the West,
simply for the reason I noted above: all such institutes are committed
to military work. Indeed, visiting the Soviet Union is like entering
into a vast prison--some of the inmates are quite interesting, but the
prison atmosphere (and lack of amenities) is utterly pervasive.]
Maybe it isn't worth the trouble, however. Perhaps the effort
would be better spent on deciding what open Japanese publications to
translate.
[Hear, hear!!]
Comments on the comments: KJK wants to be sure people won't get him
wrong. Let it be noted that his views are orthodox.
[They are academically orthodox, indeed, at least on this subject:
strongly opinionated, founded squarely on nearly perfect ignorance (which,
to be sure, deviate from academic political norms by being candidly
declared), and utterly false-to-fact!]
TAW wants to know why the Government let's Umnov in at all if they want to
restrict what he learns. Usually the Defense Department wants to restrict
Soviet visitors and argues against letting them in many cases. The
present case has the earmarks of one where the Defense Department opposed
the visit, and the State Department got a compromise. The State Department is
in the business of making agreements and it prefers letting them in.
However, the State Department is often more aware than others of
bargaining considerations. For example, Americans were able to visit
Novosibirsk before any other foreigners including Soviet satellites in the
1960s while it was still a closed city, because the State Department had
mapped it on Pittsburgh and wouldn't let Soviets visit Pittsburgh until
Americans visited Novosibirsk. The National Academy of Sciences likes
exchanges. Individual American scientists like visits and don't much care
if the exchange is very uneven. In fact, I have found it good for the ego
to be on the informative side of such conversations. Only later do I
notice that I have learned almost nothing.
[Brontosauri probably walked awkwardly, too--it's very difficult to keep
the left hind foot informed in a timely fashion of where the right front
foot is stepping, if the animal is sufficiently large. The NAS
corporately worships at the shrine of the universal brotherhood of seekers
after knowledge, and ever more frantically scrounges for those
sufficiently amoral or ignorant to engage in the tutoring of those trusted
by a government dedicated to destruction of the political system which
sustains the NAS and its little `exchange' follies.]
[Lowell]
∂24-Jan-82 2246 Kim.fateman at Berkeley NIL/Macsyma
Date: 24 Jan 1982 22:40:50-PST
From: Kim.fateman at Berkeley
To: gjc@mit-mc
Subject: NIL/Macsyma
Cc: common-lisp@SU-AI
Since it has been possible to run Macsyma on VMS sites (under Eunice or
its precursor) since April, 1980, (when we dropped off a copy at LCS),
it is not clear to me what GJC's ballyhoo is about. If the physics
sites are only now getting a partly working Macsyma for VMS, it only
brings to mind the question of whether LCS ever sent out copies of the VMS-
Macsyma we gave them, to other MIT sites.
But getting Maclisp programs up under NIL should not be the benchmark,
nor is it clear what the relationship to common lisp is.
Having macsyma run under common lisp (whatever that will be)
would be very nice, of course,
whether having macsyma run under NIL is a step in that direction or
not. It might also be nice to see, for example, one of the big interlisp
systems.
∂25-Jan-82 0028 Lowell Wood <LLW at S1-A> Recklessness
Date: 25 Jan 1982 0023-PST
From: Lowell Wood <LLW at S1-A>
Subject: Recklessness
To: jmc at SU-AI
CC: LLW at S1-A
John, as the hour gets later, I get more reckless. Go ahead and post my
comments on your Exchanges note on BBOARD (removing, if it seems appropriate
to you, the ad hominen remarks about kjk), or not, as you see fit. Lowell
∂25-Jan-82 0300 ARK Yoram Moses request advisor change
∂25-Jan-82 0246 YOM Advisor
Hi Arthur!
Please change the listing of my advisor from Floyd to McCarthy.
Thanks, Yoram.
ARK - Is this OK with you?
JMC - Yes.
∂25-Jan-82 0831 Susan Hill <CSD.HILL at SU-SCORE> Re: Gosper
Date: 25 Jan 1982 0829-PST
From: Susan Hill <CSD.HILL at SU-SCORE>
Subject: Re: Gosper
To: JMC at SU-AI
In-Reply-To: Your message of 23-Jan-82 1801-PST
Thanks.
-------
∂25-Jan-82 0940 TOB lunch
Is it on for today?
Tom
∂25-Jan-82 1000 JMC*
Porto meeting, Karl Cohen, Chandrasekharan.
∂25-Jan-82 1000 JMC*
tob, susie, deBra, Cannon, Lieberman
∂25-Jan-82 1114 TW Meeting today
To: "@COMP.DIS[1,TW]" at SU-AI
This is a reminder that we will meet at 1:30 today in 252 to complete
the version of the exam that will be trial-taken. Everyone should
have a completed version of their questions and also a set of worked
out answers.
-t
∂25-Jan-82 1214 Robert Schreiber <CSD.SCHREIBER at SU-SCORE> Curriculum mtg.
Date: 25 Jan 1982 1059-PST
From: Robert Schreiber <CSD.SCHREIBER at SU-SCORE>
Subject: Curriculum mtg.
To: csd.schreiber at SU-SCORE, csd.rwf at SU-SCORE, csd.genesereth at SU-SCORE,
csd.pratt at SU-SCORE, csd.dbrown at SU-SCORE, csd.mwalker at SU-SCORE,
csd.reges at SU-SCORE, csl.jlh.noah at SU-SCORE, csd.gass at SU-SCORE,
csd.jock at SU-SCORE, csd.sis at SU-SCORE, csd.golub at SU-SCORE,
csl.lantz at SU-SCORE, csl.bkr at SU-SCORE
cc: csd.mccarthy at SU-SCORE
I have rescheduled the meeting to discuss cs108's content for
TUESDAY, FEB 9, 2:30-4:00 PM
Rob
-------
∂25-Jan-82 1323 Betty Scott <CSD.BSCOTT at SU-SCORE> LLLContract 9628303
Date: 25 Jan 1982 1316-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: LLLContract 9628303
To: JMC at SU-AI, CSL.FB at SU-SCORE, CSL.JLH at SU-SCORE
cc: CSD.BScott at SU-SCORE
Proposal Entitled: "An Operating System and Memory Switch for the S-1
Computer," John McCarthy, P.I.
A final report on this contract was apparently due at the end of
December, 1979. It is difficult for me to determine whether John
McCarthy was actively engaged in the research, or whether Forest Baskett
and John Hennessy performed most of the research. Anyway, I am sending
each of you a copy of the proposal. The performance period was January 1,
1978, through December 31, 1979, and the funding totaled $304,472.
A final payment of approximately $3,500 is due Stanford from LLL, and they
won't pay until the report is submitted--15 copies of it. If the report
is not submitted LLL could conceivably disallow the entire contract,
leaving us holding a $304K "bag."
Would you please let me have a final report just as soon as possible. I
will have it duplicated and sent.
Thanks very much,
Betty
P.S. The LLL delay in notifying us of the above is apparently due to their
inability to locate their paperwork until recently--after hiring some
consultants to get their files in order. -- Thanks again.
-------
Lowell:
I believe the people supported under that contract now mainly work
for LLL. What would be the possibility of either waiving that report or
getting the Livermore contract people to accept a one page report? Otherwise,
you could lend us Jeff for a few weeks.
John
∂25-Jan-82 2118 ARK Advisors
∂25-Jan-82 0925 JJW Advisors
Although it is true that Manna was assigned to me as "academic advisor"
at the beginning of last year, I consider McCarthy to be my advisor since
I do research work for him.
ARK - Here's another one for you. OK?
ok
∂26-Jan-82 0823 FFL
To: JMC, FFL
A call from Math-Sci remining you of the Tversky lecture tonight in Ceras
7:15.
∂26-Jan-82 0823 FFL
To: JMC, FFL
A Stanford student, Chinese, called and asked to see you. I suggested that
he come in this afternoon about 3 p.m.
∂26-Jan-82 0921 FFL
To: JMC, FFL
A John Markoff has called twice from INFORMATION WORLD. I suggested he
speak with R. Gorin and he said that he had. I told him that I doubted you
would have anything to add as Ralph was handling questions about the problem.
He said that you had spoken to the DAILY and therefore were obviously speaking
with someone about it and to ask you to call him.
Woops, the Markoff number is 328-4602.
Thanks Fran. Professor McCarthy has no further comment.
∂26-Jan-82 1138 Woody Bledsoe <ATP.Bledsoe at UTEXAS-20> [CS.DALE: Turing Award]
Date: 26 Jan 1982 1335-CST
From: Woody Bledsoe <ATP.Bledsoe at UTEXAS-20>
Subject: [CS.DALE: Turing Award]
To: jmc at SU-AI
cc: ATP.Bledsoe at UTEXAS-20
John,
Here is the message from Al Dale regarding his talk with Bachman.
I propose to write a letter to Bachman right away and trust you will
do the same. However, I will wait a day or two, in case you
have a different idea on how we should handle this (like a combined
letter -- which is fine with me if you think it best).
Woody
---------------
Date: 26 Jan 1982 1327-CST
From: CS.DALE
Subject: Turing Award
To: atp.bledsoe
I spoke with Charlie Bachman (Chairman, Nominating Committee) this morning.
He said that he is ready to receive nominations for the award, and that one
or two letters of nomination are sufficient. He feels that any extensive mail
campaign is non-productive, and that the nominations should be fairly concise.
I believe that the most appropriate course would be for you and McCarthy
to make the nominations since you can speak to the credentials and contributionsof Bob and J with most authority.
His address is Charles Bachman
Cullinane Data Base Systems
400 Bluehill Drive
Westwood, MA 02090.
-------
-------
I think we should write separate letters and will write mine.
∂26-Jan-82 1311 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Reception for Jerry SALTZER
Date: 26 Jan 1982 1306-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Reception for Jerry SALTZER
To: faculty at SU-SCORE
cc: su-bboard at SU-SCORE
There will be a reception for Jerry Saltzer on Thursday, January 28
immediately following his last lecture, about 5:30, in the Margaret
Jacks Hall Lounge.
Incidentally, Jerry's first talk on Monday was quite interesting and
lively and I'm sure you will find the succeeding ones of interest
GENE
-------
∂26-Jan-82 1343 David Warren <WARREN at SRI-AI> Antonio Porto's Visit
Date: 26 Jan 1982 1337-PST
From: David Warren <WARREN at SRI-AI>
Subject: Antonio Porto's Visit
To: JMC at SU-AI
Please ask Antonio to get in touch with me when he arrives.
Work: 859-2128 Home: 322-3842
Thanks, David.
-------
∂27-Jan-82 1000 JMC*
Call deBra
∂27-Jan-82 1008 FFL
To: JMC, FFL
David Cheriton stopped by to ask if you would join the group at the faculty
club at 12 noon today, to eat with Jerry Salster, the Forsythe lecturer.
∂27-Jan-82 1048 FFL
To: JMC, FFL
I am not able to include Prof. DeBra in your meeting - his times are not
the hours available to others.
∂27-Jan-82 1136 CLT From SGF
To: JMC at SU-AI, konolige at DARCOM-KA
The philosophy colloquium on Feb3 may be of interest to you, John
and Kurt. George Bealer will talk about ideas he has about self-
application in situations involving thought, belief, knowledge.
Bealer is visiting here from Reed C.
∂27-Jan-82 1212 Nancy Dorio <CSD.DORIO at SU-SCORE> Colloquium Notice Week of February 1-5, 1982
Date: 27 Jan 1982 1144-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: Colloquium Notice Week of February 1-5, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
2/1/82 Math 380C Ole Hald
Monday Numerical Analysis U.C. Berkeley
4:15p.m. Seminar ``Reconstruction of the Density in the Mantle''
2/2/82 MJ 301 Dr. Jim Brinkley
Tuesday Medical Computing Stanford University
1:30p.m. Journal Club ``Recent Journal Articles of Interest''
2/2/82 MJ 301 Dr. Lanny Forgy
Tuesday Knowledge Carnegie Mellon University
2:30p.m. Representation ``Representation of Data and Control in OPS-5''
Group
2/2/82 Jordan 041 Professor Don Knuth
Tuesday Computer Science Stanford University
4:15 p.m. Colloquium ``The WEB System of Structured Documentation''
2/3/82 MJ 301 Tom Leighton
Wednesday Computer Systems MIT
3:15p.m. Laboratory Seminar ``New Lower Bound Techniques for VLSI''
2/5/82 MJ 352 Pierre Wolper
Friday Ph.D Oral Stanford University
9:00A.M. ``Synthesis of Communicating Processes From
Temporal Logic Specifications''
2/5/82 MJ 301 Lorna Shinkle
Friday Database Research U.C. Berkeley
3:30p.m. Seminar ``A Database Design Tool for INGRES''
-------
∂27-Jan-82 1319 JJW Automating induction in EKL
I'm having some difficulties with what you gave me in MACRO.LSP[W82,JMC].
First, I don't think the use of CASES is going to work as intended. It
wants its first argument to be the number of a line in the proof (not a
term) which is of the form P∨Q∨R. The idea is that you have, say,
... (some amount of proof)
5. P∨Q∨R
deps: (FOO)
(assume p)
6. P
deps: (6)
(assume q)
7. Q
deps: (7)
(assume r)
8. R
deps: (8)
...
12. RESULT
deps: (6)
...
15. RESULT
deps: (7)
...
19. RESULT
deps: (8)
(cases 5 12 15 19)
20. RESULT
deps: (FOO)
In the case of induction, you have assumptions such as
(assume |atom(x)|)
(assume |phi(car x)|)
(assume |phi(cdr x)|)
but there is no line of the form ATOM(X)∨PHI(CAR X)∨PHI(CDR X), so we can't
use CASES.
Another question that I have is whether the assumptions for the inductive case
shouldn't actually be
¬ATOM(X)∧PHI(CAR X)∧PHI(CDR X)
I've combined them into one line, since the rewriter will use this line in
rewriting by trying each of the conjuncts separately. We could also use the
"upwards" form of the induction axiom in LISPAX by assuming PHI(X)∧PHI(Y)
and proving PHI(X~Y).
Also, a few steps in the proof could be saved by using DEVAL. That is, when
finishing up, instead of doing ∀I's followed by a ∀E in the induction axiom,
the whole thing can be done in DEVAL mode and the result will be a single
proof line with the proper dependencies. The disadvantage of this is that
when you (or EKL) later looks at that proof line, there is no record of what
command created it.
∂27-Jan-82 1333 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Nuclear Plants &AI
Date: 27 Jan 1982 1632-EST
From: Chandra-at-OhioState <Chandrasekaran at RUTGERS>
Subject: Nuclear Plants &AI
To: JMC at SU-AI
cc: CHANDRASEKARAN at RUTGERS
I just got back from a trip abroad, and got in touch with
Dr. Teller's secretary Julie at Hoover. She told me
that you and Dr. Teller had a recent meeting at which my
letter to him was discussed, and that I could contact
you for some advice on how to proceed with some
possible sources of support. I'd also be very happy to
have further technical discussions with you on
this matter.
Please let me know how we can best get in touch.
-------
∂27-Jan-82 1719 Oppen at PARC-MAXC Update
Date: 27 Jan 1982 17:03 PST
From: Oppen at PARC-MAXC
Subject: Update
To: ME@SAIL
Cc: JMC@SAIL
John asked me to update my directory entry but I dont know who is in charge of
that anymore. Maybe you could pass along the following correct entry:
Dr. Derek C. Oppen SAIL Guest [DCO] --- or whatever
495 Arbor Road, Menlo Park, Ca. 94025
Phone: 325-8927
Thanks, Derek
Well, I see you're still keeping your PARC location secret.
∂27-Jan-82 1835 Jon L White <JONL at MIT-MC> Two little suggestions for macroexpansion
Date: 27 January 1982 21:34-EST
From: Jon L White <JONL at MIT-MC>
Subject: Two little suggestions for macroexpansion
To: jmc at SU-AI
I didn't want people to be getting two copies of this; if both
mailing lists were on the same host, then that host's mailer should
be able to prevent multiple copies, but COMMON-LISP is at SU-AI,
and LISP-FORUM is at MIT-AI.
So here follows the text of my msg to LISP-FORUM:
Several times in the COMMON LISP discussions, individuals have
proffered a "functional" format to alleviate having lots of
keywords for simple operations: E.g. GLS's suggestion on page 137
of "Decisions on the First Draft Common Lisp Manual", which would
allow one to write
((fposition #'equal x) s 0 7) for (position x s 0 7)
((fposition #'eq x) s 0 7) for (posq x s 0 7)
This format looks similar to something I've wanted for a long time
when macroexpanding, namely, for a form
foo = ((<something> . . .) a1 a2)
then, provided that <something> isn't one of the special words for this
context [like LAMBDA or (shudder!) LABEL] why not first expand
(<something> . . .), yielding say <more>, and then try again on the form
(<more> a1 a1). Of course, (<something> . . .) may not indicate any
macros, and <more> will just be eq to it. The MacLISP function MACROEXPAND
does do this, but EVAL doesn't call it in this circumstance (rather EVAL does
a recursive sub-evaluation)
FIRST SUGGESTION:
In the context of ((<something> . . .) a1 a2), have EVAL macroexpand
the part (<something> . . .) before recursively evaluating it.
This will have the incompatible effect that
(defmacro foo () 'LIST)
((foo) 1 2)
no longer causes an error (unbound variable for LIST), but will rather
first expand into (list 1 2), which then evaluates to (1 2).
Similarly, the sequence
(defun foo () 'LIST)
((foo) 1 2)
would now, incompatibly, result in an error.
[Yes, I'd like to see COMMON LISP flush the aforesaid recursive evaluation,
but that's another kettle of worms we don't need to worry about now.]
SECOND SUGGESTION
Let FMACRO have special significance for macroexpansion in the context
((FMACRO . <fun>) . . .), such that this form is a macro call which is
expanded by calling <fun> on the whole form.
As a result of these two changes, many of the "functional programming
style" examples could easily be implemented by macros. E.g.
(defmacro FPOSITION (predfun arg)
`(FMACRO . (LAMBDA (FORM)
`(SI:POS-HACKER ,',arg
,@(cdr form)
':PREDICATE
,',predfun))))
where SI:POS-HACKER is a version of POSITION which accepts keyword arguments
to direct the actions, at the right end of the argument list.
Notice how
((fposition #'equal x) a1 a2)
==>
((fmacro . (lambda (form)
`(SI:POS-HACKER X ,@(cdr form) ':PREDICATE #'EQUAL)))
a1
s2)
==>
(SI:POS-HACKER X A1 A2 ':PREDICATE #'EQUAL)
If any macroexpansion "cache'ing" is going on, then the original form
((fposition #'equal x) a1 a2) will be paired with the final
result (SI:POS-HACKER X A1 A2 ':PREDICATE PREDFUN) -- e.g., either
by DISPLACEing, or by hashtable'ing such as MACROMEMO in PDP10 MacLISP.
Now unfortunately, this suggestion doesn't completely subsume the
functional programming style, for it doesn't directly help with the
case mentioned by GLS:
((fposition (fnot #'numberp)) s) for (pos-if-not #'numberp s)
Nor does it provide an easy way to use MAPCAR etc, since
(MAPCAR (fposition #'equal x) ...)
doesn't have (fposition #'equal x) in the proper context.
[Foo, why not use DOLIST or LOOP anyway?] Nevertheless, I've had many
ocasions where I wanted such a facility, especially when worrying about
speed of compiled code.
Any coments?
∂28-Jan-82 0959 Gene Golub <CSD.GOLUB at SU-SCORE> Reception
Date: 28 Jan 1982 0954-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Reception
To: SU-bboard at SU-SCORE
cc: faculty at SU-SCORE
Just a reminder that there will be a reception for Jerry Saltzer
in the MJH Lounge immediately after his talk--about 5:30.
GENE GOLUB
-------
∂28-Jan-82 1004 Jrobinson at SRI-AI Tinlunch today
Date: 28 Jan 1982 1004-PST
From: Jrobinson at SRI-AI
Subject: Tinlunch today
To: tlgrp:
Administration Bldg. Conference Room A. Ellen Prince article on
inferencing of indefinite-this NPs. 12 noon.
-------
∂28-Jan-82 1035 CLT SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
To: "@LOGIC.DIS[1,CLT]" at SU-AI
SPEAKER: Ian Mason
TITLE: "Stationary logic and models of set theory"
TIME: Tues. February 2, 4:15-5:30
PLACE: 381-T Math. Dept.
Note change of time and place. Meetings will now be alternate Tuesdays.
∂28-Jan-82 1221 pratt@Shasta (SuNet) 750 <--> 780 timings
Date: 28 Jan 1982 11:57:46-PST
From: pratt at Shasta
To: equip, sun
Subject: 750 <--> 780 timings
I'd forgotten all about the following item until I stumbled across it just
now. Of interest now that we're expecting 750's.
7-May-81 10:41:55-PDT,6040;000000000001
Mail-from: ARPANET site SUMEX-AIM rcvd at 7-May-81 1041-PDT
Date: 7 May 1981 1029-PDT
From: Rindfleisch@SUMEX-AIM
Subject: FYI RE VAX 750/780 BENCHMARKS (KASHTAN)
To: SUMEX STAFF:
cc: ADMIN.GORIN@SU-SCORE, CSL.FB@SU-SCORE, PRATT@SUMEX-AIM,
cc: CSL.LANTZ@SU-SCORE, CSL.BKR@SU-SCORE, MOGUL@SCORE,
cc: NOWICKI@SCORE
Mail-from: ARPANET host SRI-AI rcvd at 7-May-81 0009-PDT
Date: 7 May 1981 0001-PDT
From: Wilcox at SRI-AI (Clark Wilcox)
Subject: [KASHTAN: VAX-11/750 <--> VAX-11/780 benchmarks]
To: rindfleisch at SUMEX-AIM
cc: Wilcox at SRI-AI
Thought you might be interested.
---------------
Date: 6 May 1981 2331-PDT
From: KASHTAN
Subject: VAX-11/750 <--> VAX-11/780 benchmarks
To: quam, witkin, hanson, jirak, wilcox, meyers, larson, kennard, sad,
heathman at SRI-AI, ryland at SRI-AI, burback at SRI-AI, mcghie,
sword
Here are the complete results of the 11/750 - 11/780 benchmarks. Looks
like the 11/750 gets to memory faster (and is optimized w.r.t. getting
to memory faster) than the 11/780. It loses VERY badly when it comes to
actually executing instructions, as the execution unit is very much slower
in the 750 than the 780. This is particulary born out by the execution
benchmarks for the convolution program in various languages. The languages
vary from BLISS (which keeps the whole world in registers) to LISP (which
keeps the whole world in memory). Even though the 750 gets to memory faster,
it doesn't do you much good when it takes so long to process what you got
from memory (even a simple move).
The 750 does a good job of operand processing (especially given its relative
CPU speed) but this doesn't seem to help too much in actual program execution,
as on the 750 the execution time seems to be dominated by the instruction
execution time rather than the on the operand fetch time (as is the case on
the 780).
A note on the Richard Fateman's 750 benchmarks. Seems that all they did was
run a Liszt (Franz Lisp compiler) compile on one of Bell Labs UNIX systems.
A compiled Franz Lisp program (as Liszt is) tends to be very heavy on CALLS
and on moving things around in memory (i.e. to and from the stack). No
intermediate results are kept in registers at all. What this does is skew
the results somewhat towards a faster looking 750 (since the 750 will benefit
from any benchmarks that are heavily involved in memory referencing). What
he reported was that the 750 was indeed about 60% of the 780 in this case.
PLEASE NOTE that large IU and VLSI programs, while we might consider them
memory intensive, are really virtual memory intensive (i.e. have very large
working sets). This is not the same as the above benchmark. Most IU and
VLSI programs when compiled with good compilers will tend to do a small amount
of computation (even just an add or multiply) with each datum fetched from
memory. You can expect the performance of the 750 relative to the 780 to
drop quite a bit from the above mentioned 60%. It should become very much
like the following convolution benchmarks (a very good example of a virtual
memory intensive program that does a small amount of computation with each
datum fetched). An interesting side note: CARs and CDRs in compiled lisp
tend to come out as "movl x(r),dst" (which executes at about 60% of 780
speed).
My feeling from playing with the two systems is that the 750 is best used
as an entry level system for those sites which need to acquire the smallest
possible VAX configuration (i.e. the lowest possible price). An entry level
750 goes for about $90K while an entry level 780 system with approximately
the same configuration would go for about $140K. Clearly there is a big
difference here (almost all of it in the price of the CPU). As the systems
get larger the price advantage goes away (as the price will note be dominated
by the CPU price, which is the case in the smaller systems, but by memory /
peripheral prices). Here you will save about $50K on a $250K system and get
less than 1/2 the machine.
I am somewhat confused by the divide instruction timings. There are a couple
of possibilities here - 1) a stupidity in the 780 was fixed in the 750
2) I muffed the 780 test (don't thing so, as I
triple checked it)
3) I muffed the 750 test.
I find it incredible that MULL is 4x as fast on the 780 while DIVL is a bit
slower on the 780. I did not do any floating point tests, as there is no
floating point accelerator on the 750.
David
-------------------------------------------------------------------------------
VAX-11/750 vs VAX-11/780
------------------------
Simple 2D convolution program:
11/750 11/750 (% of 11/780) 11/780
------ -------------------- ------
BLISS-32 5.45 sec 45% 2.5 sec
VMS PASCAL 12.9 sec 38% 4.9 sec
UNIX C 11.3 sec 44% 5.0 sec
UNIX F77 39.9 sec 29% 11.4 sec
Compiled
Franz Lisp 76.5 sec 53% 41.0 sec
Instruction timings:
movl r,r 1000nSec 40% 400nSec
movl x(PC),r 1760nSec 45% 800nSec
movl r,x(PC) 2300nSec 52% 1300nSec
movl (r),r 1330nSec 60% 800nSec
Addressing modes:
r 0nSec -- 0nSec
# (short) 0nSec -- 0nSec
# (long) 700nSec 57% 400nSec
(r) 330nSec 120% 400nSec
(r)+ 330nSec 120% 400nSec
-(r) 330nSec 120% 400nSec
@(r)+ 900nSec 111% 1000nSec
x(r) 500nSec 80% 400nSec
@x(r) 1150nSec 86% 1000nSec
[r] 1000nSec 60% 600nSec
Instructions:
MOVL 1000nSec 40% 400nSec
ADDL
SUBL
etc
MULL 8000nSec 25% 2000nSec
DIVL 8000nSec 112% 9000nSec
CALLx/RET 20000nSec+1800nSec/register 15000nSec+
100% 2000nSec/Reg
JSB/RSB 6000nSec 50% 3000nSec
SOBGxx 2000nSec 50% 1000nSec
ACBL 5600nSec 71% 4000nSec
MOVC3 350nSec/byte 107% 375nSec/byte
3 operand +500nSec 40% +200nSec
instructions
-------
---------------
-------
-------
∂29-Jan-82 0028 BCM prolog
How can Prolog be accessed on Score or Sail? Ben
mrc:<prolog>prolog at SCORE is the command.
It has been removed from the disk; watch for its return.
∂29-Jan-82 0846 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Meal Reply forms
Date: 29 Jan 1982 0845-PST
From: Carolyn Tajnai <CSD.TAJNAI at SU-SCORE>
Subject: Meal Reply forms
To: CSD-Faculty: ;
cc: CSL faculty: ;
I must give a guaranteed number for meals on Monday, Feb. 1.
Please let me know if you plan to attend the luncheons and the banquet.
electronic mail is fine.
Thanks,
Carolyn
-------
∂29-Jan-82 0946 McCall at PARC-MAXC Re: Course in Programming and PROLOG starting soon
Date: 29 Jan 1982 09:45 PST
From: McCall at PARC-MAXC
Subject: Re: Course in Programming and PROLOG starting soon
In-reply-to: Bobrow's message of 29 Jan. 1982 9:33 am PST (Friday)
To: JMC@SU-AI
cc: McCall
I am a Stanford student and would like to know when and where the course will
be meeting.
Kim McCall
Monday and Wed at 11 in 301 mjh.
∂29-Jan-82 1031 FFL
To: JMC, FFL
Dave Gifford of Xerox called re book you are writing with Caroline. Please
call at 494-4478.
∂29-Jan-82 1058 Morris at PARC-MAXC Re: Course in Programming and PROLOG starting soon
Date: 29 Jan 1982 10:58 PST
From: Morris at PARC-MAXC
Subject: Re: Course in Programming and PROLOG starting soon
In-reply-to: Bobrow's message of 29 Jan. 1982 9:33 am PST (Friday)
To: JMC@SU-AI
cc: Morris
John,
I would like to attend the course, at least for part of the time. What should I do?
Jim Morris
Just come to the course which will be Mon and Wed at 11 in Room 301 Margaret Jacks
Hall.
∂29-Jan-82 1307 RPG
John:
Do you think that this is worth pursuing? For example,
do you know this Arthur Norman?
∂28-jan-82 1711 100 on tty162 1711: mark j denne via london
Dear Dick,
Thanks very much for taking the trouble to come back
at me about this one. i'm mark denne from cambridge university
england and i'm talking to arpa via the university college
london gateway . i'm currently applying to do a phd at stanford
and i'm interested in lisp. dr arthur norman here at cambridge
has been telling us about lisp's obvious connections with
stanford and i wondered if i could get permission to use
sail to become more aquaited with the lisp system and its
ai applications. i understand you also have an algol-like
language for ai on sail and would love to find out more about it.
since i don't have my own arpa mailbox and can't read
mail anyway unless i'm logged in, it's probably going to be
difficult to get back to me. if the london gateway is up
and running tomorrow night i'll try and get back to
you at approx 18.00 pm pacific time. if all else fails
try leaving a message for me in dr arthur norman's mailbox
many thanks for your help.
mark denne
selwyn college
university of cambridge
england.
I don't know Arthur Norman, and I'm not inclined to give Denne an
account to be come acquainted with LISP and SAIL. If it were something
that exists here uniquely like EKL, I would do it, but there
are LISPs and Algol-like languages in England and presumably at
Cambridge, and the differences aren't worth fussing about.
∂29-Jan-82 1321 DBL papers to read
John,
I dropped off the latest versions of some of my recent papers to you
this morning. The latest VLSI and MetaCognition papers can be
distinguished because they bear January 28 dates.
I am looking forward to your readings of the papers, especially
the Nature of Heuristics and the Eurisko ones. If you have
some comments or suggestions, either at the textual level or
at the level of planning for future years' research, I'd like
to get together sometime to discuss them.
Doug
∂29-Jan-82 1414 TW via PARC-MAXC Exam trial
To: "@COMP.DIS[1,TW]" at SU-AI
CC: MMS at SU-AI
Mike SPreitzer has the exam and will take it on Saturday.
He will be in touch with Joe Weening, who will have the
papers. It would be good if people could get hold of them
on Monday morning and look them over before the meeting.
Mike will be there to give us feedback. Unfoutunately
I won't be there today to coordinate, so if you could
work things out with Joe (JJW) that would be helpful.
Thanks --t
∂29-Jan-82 1549 JJW PROLOG course
My list of people for the PROLOG course is on PROLOG.DIS[1,JJW].
If you want to add people to this, please do so. I haven't yet sent
out a message giving the room number.
∂29-Jan-82 1533 Ichiki at SRI-AI TINLUNCH Meeting 2/4
Date: 29 Jan 1982 1533-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH Meeting 2/4
To: tlgrp:
The TINLUNCH meeting Thursday, February 4th will be held in Building
E, Conference Room EK242 at lunch time. The paper to be discussed is:
Of Tense and Aspect: One Analysis
By Michael Bennett
(pp. 13-29 in "SYNTAX AND SEMANTICS--Tense and Aspect",
Volume 14, edited by Philip Tedeschi and
Annie Zaenen)
Michael Bennett will not attend the meeting.
Copies of the above paper are on top of Barbara's file cabinet.
-------
∂29-Jan-82 2148 Lowell Wood <LLW at S1-A> Reporting Games
Date: 29 Jan 1982 2147-PST
From: Lowell Wood <LLW at S1-A>
Subject: Reporting Games
To: jmc at SU-AI
CC: LLW at S1-A
[John: I can't waive the report requirement, as I am only the `technical
adviser' of the actual contract monitor (on paper, at least). I suggest
the 1-2 report page route, consisting of a very high-level summary (which
I'm sure Jeff would be willing to help you with), and containing much
reference to oral reports made and software delivered. The basic problem
is that the Lab can't award grants (which could entail *no* obligations of
any kind), and that *some* `deliverable' is required under the contract
format. Sorry for the hassle! Lowell]
∂27-Jan-82 1520 John McCarthy <JMC at SU-AI>
Date: 27 Jan 1982 1526-PST
From: John McCarthy <JMC at SU-AI>
To: llw at S1-A
Lowell:
I believe the people supported under that contract now mainly work
for LLL. What would be the possibility of either waiving that report or
getting the Livermore contract people to accept a one page report? Otherwise,
you could lend us Jeff for a few weeks.
John
∂25-Jan-82 1323 Betty Scott <CSD.BSCOTT at SU-SCORE> LLLContract 9628303
Date: 25 Jan 1982 1316-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: LLLContract 9628303
To: JMC at SU-AI, CSL.FB at SU-SCORE, CSL.JLH at SU-SCORE
cc: CSD.BScott at SU-SCORE
Proposal Entitled: "An Operating System and Memory Switch for the S-1
Computer," John McCarthy, P.I.
A final report on this contract was apparently due at the end of
December, 1979. It is difficult for me to determine whether John
McCarthy was actively engaged in the research, or whether Forest Baskett
and John Hennessy performed most of the research. Anyway, I am sending
each of you a copy of the proposal. The performance period was January 1,
1978, through December 31, 1979, and the funding totaled $304,472.
A final payment of approximately $3,500 is due Stanford from LLL, and they
won't pay until the report is submitted--15 copies of it. If the report
is not submitted LLL could conceivably disallow the entire contract,
leaving us holding a $304K "bag."
Would you please let me have a final report just as soon as possible. I
will have it duplicated and sent.
Thanks very much,
Betty
P.S. The LLL delay in notifying us of the above is apparently due to their
inability to locate their paperwork until recently--after hiring some
consultants to get their files in order. -- Thanks again.
-------
∂29-Jan-82 2149 Kim.fateman at Berkeley Okay, you hackers
Date: 29 Jan 1982 20:31:23-PST
From: Kim.fateman at Berkeley
To: guy.steele@cmu-10a
Subject: Okay, you hackers
Cc: common-lisp@SU-AI
I think that when GJC says that NIL/Macsyma runs the "X" demo, it
is kind of like the dog that plays checkers. It is
remarkable, not for how well it plays, but for the fact that it plays at all.
(And I believe it is creditable [if] NIL runs Macsyma at all... I
know how hard it is, so don't get me wrong..)
Anyway, the stardard timings we have had in the past, updated somewhat:
MC-Macsyma, Vaxima and Lisp Machine timings for DEMO files
(fg genral, fg rats, gen demo, begin demo)
(garbage collection times excluded.) An earlier version of this
table was prepared and distributed in April, 1980. The only
column I have changed is the 2nd one.
MC Time VAXIMA 128K lispm 192K lispm 256K lispm
4.119 11.8 sec. 43.333 sec. 19.183 sec. 16.483 sec.
2.639 8.55 sec. 55.916 sec. 16.416 sec. 13.950 sec.
3.141 14.3 sec. 231.516 sec. 94.933 sec. 58.166 sec.
4.251 13.1 sec. 306.350 sec. 125.666 sec. 90.716 sec.
(Berkeley VAX 11/780 UNIX (Kim) Jan 29, 1982, KL-10 MIT-MC ITS April 9, 1980.)
Kim has no FPA, and 2.5meg of memory. Actually, 2 of these times are
slower than in 1980, 2 are faster.
Of course, GJC could run these at MIT on his Franz/Vaxima/Unix system, and
then bring up his NIL/VMS system and time them again.
∂30-Jan-82 1000 JMC*
Karl
∂30-Jan-82 1000 JMC*
porto, tell elliott to put Oliver on SE2 list.
∂30-Jan-82 1504 JJW Course in Logic Programming and PROLOG
To: "@PROLOG.DIS[1,JJW]" at SU-AI
Starting on February 1 and continuing through the quarter there will be a
course in logic programming and the PROLOG language, taught by Antonio
Porto of the University of Lisbon. Logic programming in general and
PROLOG in particular have attracted wide interest in Europe but are only
getting started in the United States. There will be an opportunity to use
PROLOG.
The class will meet twice a week, on Mondays and Wednesdays at 11:00 a.m.,
in 301 Margaret Jacks Hall.
∂31-Jan-82 0939 CLT
i'm sorry
∂31-Jan-82 1801 JMC*
Loebner and Suppes and Lipset.
∂31-Jan-82 1836 Feigenbaum at SUMEX-AIM revised statement
Date: 31 Jan 1982 1832-PST
From: Feigenbaum at SUMEX-AIM
Subject: revised statement
To: jmc at SU-AI
FACULTY STATEMENT ON SOVIET VISITORS
If the enclosed Associated Press story accurately reflects Stanford
policy, we would like to urge upon the university administration a
more moderate and balanced policy with respect to visitors from
the Soviet Union.
The story gives the impression that Stanford takes an
attitude of total opposition towards the Government's efforts to restrict
technology transfer to the Soviet Union. Moreover, this attitude seems to
be supported by a number of untrue statements, namely (1) The Soviets are
ahead of the U.S. in robotics. (2) Robotics has no military applications.
(3) There are no possible restrictions on Soviet visitors that won't
disrupt the functioning of the University.
Consider the following not-so-hypothetical implication of the story:
that when the Soviet Consulate in San Francisco reported on the situation
to Moscow, this report probably reinforced the idea in Moscow that
American protests about Poland, Afghanistan, Sakharov,
Scharansky, etc. were a governmental sham
without any support in the institutions of the country such as
universities.
We believe that Stanford should adopt a policy toward Soviet
visitors that balances the following considerations:
1. Preserving the orderly functioning of the academic processes of
research, publication and teaching and avoiding the creation of a security
atmosphere.
2. Showing the visitors a society more open than their own.
3. Minimizing transfer of militarily useful technology so as to
minimize our own defense costs; perhaps we can help obviate the need for a
return to the draft. Ways of implementing this without security measures
need study.
4. Increasing reciprocity so that access to Soviet research
becomes available to Americans to the same extent that our research
activities are accessible to them. This involves some degree of support
to the bargaining efforts on our behalf conducted by the State Department
and the National Academy of Sciences.
5. Maintaining good relations with the State Department, the
Defense Department and other Government agencies.
Since these goals are not entirely compatible, some degree of
compromise is needed.
As scholars who have interchanges with Soviet scientists
and who occasionally visit Soviet universities and institutes, we are
particularly affected by item 4. Our access to ideas and laboratories
is always severely limited - for example, by the Soviet law that foreigners
can only be given published information without special permission.
A certain reciprocity that demonstrates that
we understand what is happening to us and will not tolerate it is
indicated.
We recognize that Stanford is an open institution. But with care
and planning, we can affect what Soviet visitors see. Stanford, of
course, as an institution, cannot control what people say, but people
who are concerned can be careful about what they say. We realize that
many in the Stanford community are not concerned as we are, but the
university policy should be evenhanded and neutral as regards this concern.
We suggest that a committee be appointed to develop a policy. If
an interim statement clarifying the impression given by the Stanford press
release could be made, it might help mitigate the unfortunate signals that
may have been sent to Moscow.
∂01-Feb-82 0927 TOB
John
I think that we should point out that the quotes of Roth were
cut in such a way as to lead to two of the mistaken impressions
you cite.
Tom
∂31-Jan-82 1925 JMC
The current version of the statement is page 3 of STATE[W82,JMC].
∂01-Feb-82 1014 Kim.fateman at Berkeley GETPR and compatibility
Date: 1 Feb 1982 10:11:13-PST
From: Kim.fateman at Berkeley
To: common-lisp@su-ai
Subject: GETPR and compatibility
There are (at least) two kinds of compatibility worth comparing.
1. One, which I believe is very hard to do,
probably not worthwhile, and probably not
in the line of CL, is the kind which
would allow one to take an arbitrary maclisp (say) file, read it into
a CL implementation, and run it, without ever even telling the CL
system, hey, this file is maclisp. And when you prettyprint or debug one of
those functions, it looks pretty much like what you read in, and did
not suffer "macro←replacement←itis".
2. The second type is to put in the file, or establish somehow,
#.(enter maclisp←mode) ;; or whatever, followed by
<random maclisp stuff>
#.(enter common←lisp←mode) ;; etc.
The reader/evaluator would know about maclisp. There
are (at least) two ways of handling this
a: any maclisp construct (e.g. get) would be macro-replaced by
the corresponding CL thing (e.g. getprop or whatever); arguments would
be reordered as necessary. I think transor does this, thought generally
in the direction non-interlisp ==> interlisp. The original maclisp
would be hard to examine from within CL, since it was destroyed on read-in
(by read, eval or whatever made the changes). (Examination by looking
at the file or some verbatim copy would be possible). This makes
debugging in native maclisp, hard.
b: wrap around each uniquely maclisp construction (perhaps invisibly)
(evaluate←as←maclisp <whatever>). This would preserve prettyprinting,
and other things. Functions which behave identically would presumably
not need such a wrapper, though interactions would be hard to manage.
I think 2a is what makes most sense, and is how Franz lisp
handles some things which are, for example, in interlisp, but not in Franz.
The presumption is that you would take an interlisp (or maclisp)
file and translate it into CL, and at that point abandon the original
dialect. In view of this, re-using the names seems quite possible,
once the conversion is done.
In point of fact, what some people may do is handle CL this way.
That is, translate it into another dialect, which, for whatever
reason, seems more appropriate. Thus, an Xlisp chauvinist
might simply write an Xlispifier for CL. The Xlispifier for CL
would be written in Xlisp, and consist of the translation package
and (probably) a support package of CL functions. Depending on
whether you are in CL-reading-mode or XL-reading-mode, you would
get one or the other "getprop".
Are such "implementations of CL" "correct"? Come to think of
it, how would one determine if one is looking at an implementation
of CL?
∂01-Feb-82 1127 Nancy Dorio <CSD.DORIO at SU-SCORE> Correction to Colloquium Notice week of February 1 - 5
Date: 1 Feb 1982 1112-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: Correction to Colloquium Notice week of February 1 - 5
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Correction:
Seminar entitled Computer Systems Laboratory Seminar should have been:
Special Seminar ← Wednesday, February 3, 1982, at 3:15p.m. in MJ 301.
Tom Leighton, from MIT will speak on ``New Lower bound Techniques for VLSI''
Computer Systems Laboratory Seminar - Wednesday, February 3, 1982, at 4:15p.m.
John Limb, From Bell Labs, Murray Hill, N.J. will speak on ``High Speed
Operation of Local Computer Networks''.
-------
Here are the files that should be put on the disk. OLDPROLOG should be the
file accessed as prolog. There are two copies of each file on the tape
just to be sure.
∂01-Feb-82 2316 David Warren <WARREN at SRI-AI> Tape Listing Itself
Date: 1 Feb 1982 1724-PST
From: David Warren <WARREN at SRI-AI>
Subject: Tape Listing Itself
To: JMC at SU-AI
DUMPER tape # 1, documentation, Monday, 1-Feb-82 1649
file last write size (pages) checksum
<PROLOG>READ.ME.2 12-Jul-81 2045 1
<PROLOG>PROLOG.DOC.1 12-Jul-81 2037 58
<PROLOG>GUIDE3.MEM.1 12-Jul-81 2036 36
<PROLOG>DEBUG.MEM.1 12-Jul-81 2036 22
<PROLOG>PROLOG.HLP.1 12-Jul-81 2039 1
<PROLOG>TUTORI.LPT.1 12-Jul-81 2045 8
<PROLOG>TUTORI.PL.1 12-Jul-81 2045 3
DUMPER tape # 1, oldprolog, Monday, 1-Feb-82 1650
file last write size (pages) checksum
<WARREN>PROLOG.EXE.26 31-Oct-81 1050 118
DUMPER tape # 1, newprolog, Monday, 1-Feb-82 1651
file last write size (pages) checksum
<PROLOG>PROLOG.EXE.1 12-Jul-81 2037 115
<PROLOG>PROLOG.EXE.26 16-Dec-81 0908 116
<PROLOG>PROLOG.EXE.28 21-Dec-81 1615 116
DUMPER tape # 1, documentation1, Monday, 1-Feb-82 1652
file last write size (pages) checksum
<PROLOG>READ.ME.2 12-Jul-81 2045 1
<PROLOG>PROLOG.DOC.1 12-Jul-81 2037 58
<PROLOG>GUIDE3.MEM.1 12-Jul-81 2036 36
<PROLOG>DEBUG.MEM.1 12-Jul-81 2036 22
<PROLOG>PROLOG.HLP.1 12-Jul-81 2039 1
<PROLOG>TUTORI.LPT.1 12-Jul-81 2045 8
<PROLOG>TUTORI.PL.1 12-Jul-81 2045 3
DUMPER tape # 1, oldprolog1, Monday, 1-Feb-82 1652
file last write size (pages) checksum
<WARREN>PROLOG.EXE.26 31-Oct-81 1050 118
DUMPER tape # 1, newprolog1, Monday, 1-Feb-82 1653
file last write size (pages) checksum
<PROLOG>PROLOG.EXE.1 12-Jul-81 2037 115
<PROLOG>PROLOG.EXE.26 16-Dec-81 0908 116
<PROLOG>PROLOG.EXE.28 21-Dec-81 1615 116
End of tape.
!
-------
∂01-Feb-82 2316 Ingalls at PARC-MAXC Re: Course in Programming and PROLOG starting soon
Date: 1 Feb 1982 12:40 PST
From: Ingalls at PARC-MAXC
Subject: Re: Course in Programming and PROLOG starting soon
In-reply-to: Bobrow's message of 29 Jan. 1982 9:33 am PST (Friday)
To: JMC@SU-AI
cc: Ingalls
I am interested in attending the Prolog course. Having missed the first session,
I'm interested in what I missed, and if there is a place where I could pick up
any materials which were distributed. I don't know how you feel about people
free-loading - is there some sort of auditing arrangement, or should I just come.
I've just gotten a tape with Tops-20 version of Prolog from David Warren at SRI
which I am going to attempt to install at Xerox for people to try out.
Thanks for your attention
- Dan
Nothing was distributed. I don't know how the department feels about
free loaders. The instructor, Antonio Porto, is AP@SU-AI.
∂01-Feb-82 2322 Oppen at PARC-MAXC Hnh???
Date: 1 Feb 1982 15:45 PST
From: Oppen at PARC-MAXC
Subject: Hnh???
To: JMC@SAIL
John! I don't know what you mean that I am keeping my being at PARC
secret! It is no secret! However, if you mean --- why don't I include PARC in
my PLAN file --- I am here only a few times a week as a consultant and then
in either of two buildings, so it is usually impossible to find me here. My home
phone is still the best way to find me, or else by netmail.
Derek
Sorry. What set it off was that my secretary was asked by someone how to
find you, and I sent you a message without thinking about looking at
a PLAN file.
∂01-Feb-82 2316 David Warren <WARREN at SRI-AI> Tops-20 Prolog Tape
Date: 1 Feb 1982 1722-PST
From: David Warren <WARREN at SRI-AI>
Subject: Tops-20 Prolog Tape
To: JMC at SU-AI
cc: Warren at SRI-AI
John, The listing of the tape will be in the following message.
Basically, you need the PROLOG.EXE from the save set "oldprolog"
for a reliable version. This version, however, only allows
Tops-10 style filenames. The version of PROLOG.EXE.28 in the
save set "newprolog" is the very latest version which permits
full Tops-20 filenames but which has a number of (minor) bugs
and snags. The first save set, "documentation", contains all
the current documentation. The tape contains a duplicate
copy of each save set for good measure. A catalog of the
various files follows. -- David.
; Files needed to use DEC-20 Prolog (see READ.ME):
<WARREN>PROLOG.EXE ; Prolog interpreter + compiler
<PROLOG>READ.ME ; Fuller description of these files
PROLOG.DOC ; User's Guide
GUIDE3.MEM ; Guide to Version 3
DEBUG.MEM ; Guide to debugging facilities
PROLOG.HLP ; Prolog help file
TUTORI.LPT ; Short tutorial on Prolog
TUTORI.PL ; Examples for the tutorial
-------
∂02-Feb-82 0037 ARK Advising
∂02-Feb-82 0036 JMC
Don't you have Kurt Konolige listed as my thesis advisee?
ARK - Nope, according to my records, his advisor is Lenat. Is that wrong?
He is my thesis advisee.
∂02-Feb-82 0148 ARK Advice accepted
∂02-Feb-82 0144 JMC
He is my thesis advisee.
ARK - I'll correct the file and get you an abstract of his record. Thanks
for the correction.
∂02-Feb-82 0803 reid@Shasta (SuNet) The "Masinter report"
Date: 2 Feb 1982 07:50:43-PST
From: reid at Shasta
To: Masinter@Parc, equip
Subject: The "Masinter report"
Is there a copy of the "Masinter report" available online anywhere at stanford?
Offline?
∂02-Feb-82 1009 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Forum meeting/Guests
Date: 2 Feb 1982 1004-PST
From: Carolyn Tajnai <CSD.TAJNAI at SU-SCORE>
Subject: Forum meeting/Guests
To: CSD-Faculty: ;
cc: CSL faculty: ;
This is a list of the expected affiliate attendees to the Forum.
Thursday and Friday we will serve a continental breakfast starting at
8 a.m. This will give everyone an opportunity to register, get name
tags, programs, viewgraph booklets, and get acquainted, and start the
meeting at 9 a.m. It would be great if you all could attend, but it
is particularly desirable for the faculty liaison (finders and maintainers)
to attend.
Bell Labs:
Al Aho
Elliot Pinson
Bill Coughran
BNR:
Robert Gaskins
Glenn Stewart
Data General:
Jack Crawford
Dick Bushroe
DEC:
Doug Clark
Jud Leonard
Sam Fuller
LM Ericsson:
none
Fairchild:
Robert Burnett
Robert Clough
?
Fujitsu:
Fumihiro Maruyama
General Electric:
Richard Shuey
Francis Wang
Ron Taylor
Harris Corporation:
Sam Mathan
Carmen J. Palermo
Hewlett-Packard:
Dr. James Bell
Bill Gimple
Bert Raphael
Hughes:
Hitachi America:
Kuzufumi Yoshida
Fumitoshi Monizo
IBM:
Abraham Peled
Lee Hoevel
George Fan
Intel:
Eric McLeod
Roger Swanson
Dan Hammerstrom
Eastman Kodak Co.:
Eugene A. Trabka
Peter E. Castro
Martin-Marietta:
Jeffrey Van Baalen
Mitsubishi:
no one
Nippon Electric Co:
Masato Kawai
PG&E:
Paul Gribik
Matthew Halter
Ho-Nien Liu
Philips Research Laboratories:
J. Kostelec
Schlumberger:
Dave Vanderschel
Peter Hart
Harry Barrow
Siemens:
Dr. Hans-Jorg Pfleiderer
Dr. Christian Mueller-Schloer
Sohio:
Lou Malito
David Beck
Hans Loffeld
STC Computer Research:
David Anderson
Lin Wu
Tektronix:
Rick LeFaivre
Kit Bradley
Rick Potter
Texas Instruments:
Dr. John Fish
Herman Dierks
Tom Kehler
TRW:
Joshua Hurvitz
Dr. Barry Boehm
Welex-Halliburton:
Dr. Suresh Thadani
-------
∂02-Feb-82 1017 Kay at PARC-MAXC Re: Course in Programming and PROLOG starting soon
Date: 2 Feb 1982 10:15 PST
From: Kay at PARC-MAXC
Subject: Re: Course in Programming and PROLOG starting soon
In-reply-to: Bobrow's message of 29 Jan. 1982 9:33 am PST (Friday)
To: JMC@SU-AI
cc: Kay
Is it to late to get in on the PROLOG course. I would like to if it is still possible.
--Martin.
No problem. Monday and Wednesday at 11, 301 Margaret Jacks.
∂02-Feb-82 1018 asprey at PARC-MAXC Re: Course in Programming and PROLOG starting soon
Date: 2 Feb 1982 10:17 PST
From: asprey at PARC-MAXC
Subject: Re: Course in Programming and PROLOG starting soon
In-reply-to: Bobrow's message of 29 Jan. 1982 9:33 am PST (Friday)
To: JMC@SU-AI
Yes, I am interested in the PROLOG class, but at what time will it meet?
Thankyou,
Peggy Asprey
It meets at 11am on Mondays and Wednesdays. First meeting has occurred.
∂02-Feb-82 1100 FFL
To: JMC, FFL
Nils Nilsson called to say he would like to speak with you briefly.
859-2311
∂02-Feb-82 1109 RINDFLEISCH@SUMEX-AIM (SuNet) Re: The "Masinter report"
Date: 2 Feb 1982 0834-PST
From: T. C. Rindfleisch <Rindfleisch at SUMEX-AIM>
Subject: Re: The "Masinter report"
To: reid at SU-SHASTA, Masinter at PARC-MAXC at SUMEX-AIM, equip at SU-SHASTA
cc: Rindfleisch at SUMEX-AIM, CMiller at SUMEX-AIM
In-Reply-To: Your message of 2-Feb-82 0750-PST
Online copies of Masinter's report are available on SUMEX and may be
FTP'ed anonymously. The file names are:
<LISP>INTERLISP-VAX-RPT.TXT
or <LISP>INTERLISP-VAX-RPT.PRESS
Hardcopy versions of the report are just now coming out of the CSD
technical report machinery. Below is a recent postscript from the ISI
development group...
**********************************
Mail-from: ARPANET host PARC-MAXC rcvd at 29-Jan-82 1658-PST
From: Masinter at PARC-MAXC
Subject: Addendum to Interlisp-VAX report
The folks at ISI have asked that the following postscript be included
when my Interlisp-VAX report is circulated. I have no objection.
- - - - - - - - - - - - - - - - - - - - - - - - - -
Addendum to Interlisp-VAX: A report
Dave Dyer
20 Jan. 1982
Since Larry Masinter's "Interlisp-VAX: A Report" is being circulated
widely, it is important that it be as accurate as possible. This note
represents the viewpoint of the implementors of Interlisp-VAX, as of
January 1982.
The review of the project and the discussions with other LISP
implementors that provided the basis for "Interlisp-VAX: A report", took
place in June 1981.
We believed at the time, and still believe now, that Masinter's report
is largely a fair and accurate presentation of Interlisp-VAX, and of the
long term efforts necesary to support it. We now have the advantage of
an additional 6 month's development effort. There are some areas where
progress and performance has been better than anticipated in the
Interlisp-VAX report, and we would like to report on our current status.
AVAILABILITY AND FUNCTIONALITY
-----------------------------
Interlisp-VAX has been in use for testing purposes both here at ISI and
at several sites around the ArpaNet, since since November of 1981.
We are planning the first general release for February 1982 - ahead of
the schedule that was in effect in June, 1981.
The current implementation incudes all of the features of Interlisp-10
with very few exceptions. There is no noticable gap in functionality
among Interlisp-10, Interlisp-D and Interlisp-VAX, except for features
that are inherently peculiar to some implementation (eg. windows on the
Dolphin, JSYS and TENEX on the PDP-10)
Among the Interlisp systems we are running here are KLONE, AP3, HEARSAY,
and AFFIRM.
PERFORMANCE
-----------
Masinter's analysis of the problems of maximizing performance, both for
Interlisp generally and for the VAX particularly was excellent. It is
now reasonable to quantify the performance based on experiance with real
systems. The analysis of the performance of Lisp programs is quite
complex, and single numbers are often more misleading than
representative. It is hard to give a complete analysis, so we will only
give general performance numbers.
CPU speed (on a single-user vax/780) is, for many of the programs we
have measured, currently in the range of 1/4 the speed of Interlisp-10
(on a single-user DEC 2060). We believe that a factor of two overall
performance improvement is acheivable.
Currently, it seems reasonable to allow 1 mb. real memory per active
user.
-------
∂02-Feb-82 1308 FFL
To: JMC, FFL
Prof. John Holst who is a visiting professor here has a colleague coming
from Sweden with 12 students, John Schmittger, of the Royal Inst. of
Technology. Prof. Holst says that you have met Schmittger - was here on
a sabbatical some time ago. He wishes to speak with you on Monday, Feb. 8,
between 10 and 11 a.m. Prof. Holst would like you to call him to confirm -
7-0960.
∂02-Feb-82 1337 Masinter at PARC-MAXC SUBST vs INLINE, consistent compilation
Date: 2 Feb 1982 13:34 PST
From: Masinter at PARC-MAXC
Subject: SUBST vs INLINE, consistent compilation
To: Common-Lisp@SU-AI
cc: Masinter
I think there is some rationale both for SUBST-type macros and for INLINE.
SUBST macros are quite important for cases where the semantics of
lambda-binding is not wanted, e.g., where (use your favorite syntax):
(DEFSUBST SWAP (X Y)
(SETQ Y (PROG1 X (SETQ X Y]
This isn't a real example, but the idea is that sometimes a simple substitution
expresses what you want to do more elegantly than the equivalent
(DEFMACRO SWAP X
\(SETQ ,(CADDR X) (PROG1 ,(CADR X) (SETQ ,(CADR X) ,(CADDR X]
These are definitely not doable with inlines. (I am not entirely sure they can be
correctly implemented with SUBST-macros either.)
-----------------
There is a more important issue which is being skirted in these various
discussions, and that is the one of consistent compilation: when is it
necessary to recompile a function in order to preserve the equivalence of
semantics of compiled and interpreted code. There are some simple situations
where it is clear:
The source for the function changed
The source for some macros used by the function changed
There are other situations where it is not at all clear:
The function used a macro which accessed a data structure which
has changed.
Tracing the actual data structures used by a macro is quite difficult. It is not
at all difficult for subst and inline macros, though, because the expansion of
the macro depends only on the macro-body and the body of the macro
invocation.
I think the important issue for Common Lisp is: what is the policy on consistent
compilation?
Larry
∂02-Feb-82 1504 Konolige at SRI-AI thesis
Date: 2 Feb 1982 1501-PST
From: Konolige at SRI-AI
Subject: thesis
To: jmc at SAIL
John, work proceeds apace. At the beginning of January I
started reading Boolos seriously, and refining and writing up the
modal and syntactic logics for belief. By the end of the week I will
have something written to give to you and Fefferman for criticism
and comment. --kk
-------
OK. It's "Feferman" by the way. I need to see you before Gray Tuesday,
so I can say something other than, "He seems to still exist" when they
ask, but I don't need to press you right now.
∂02-Feb-82 1524 FFL
To: JMC, FFL
James Ransom of the Aerospace Corp. in LA called to invite you to attend
a conference on space development there, April 2-4. Will have a one-hour
session on teleoperators and robotics and interested in having your chair
that session. Will call you back about this on Wednesday afternoon.
∂02-Feb-82 1706 Richard Anderson <CSD.ANDERSON at SU-SCORE>
Date: 2 Feb 1982 1608-PST
From: Richard Anderson <CSD.ANDERSON at SU-SCORE>
To: jmc at SU-AI
I have joined a project with the HPP. I thought you knew that I was nolonger
with the Formal Reasoning Group.
Richard Anderson
-------
∂02-Feb-82 1755 JMC
To: "@PROLOG.DIS[1,AP]" at SU-AI
<prolog>prolog enters prolog on SCORE.
∂02-Feb-82 2050 RPG Thoughts
If you think that there is a reasonable chance of success,
I will apply for the faculty position. Should I gather references?
-rpg-
I don't know how great the chances of success are. For example, I
don't know Terry's opinion. Only the names of references need be
gathered.
∂02-Feb-82 2051 Lowell Wood <LLW at S1-A> Balance
Date: 02 Feb 1982 2045-PST
From: Lowell Wood <LLW at S1-A>
Subject: Balance
To: jmc at SU-AI
CC: LLW at S1-A
∂31-Jan-82 1723 John McCarthy <JMC at SU-AI>
Date: 31 Jan 1982 1724-PST
From: John McCarthy <JMC at SU-AI>
To: llw at S1-A
You may be interested that some faculty are meeting with Lieberman on
the subject (he doesn't want us to get the wrong impression). The following
is my attempt to be diplomatic.
FACULTY STATEMENT ON SOVIET VISITORS
We are distressed by the situation described in the attached
Associated Press story. It gives the impression that Stanford takes an
attitude of total opposition towards the Government's efforts to restrict
technology transfer to the Soviet Union. Moreover, this attitude seems to
be supported by a number of untrue statements, namely (1) The Soviets are
ahead of the U.S. in robotics. (2) Robotics has no military applications.
(3) There are no possible restrictions on Soviet visitors that won't
disrupt the functioning of the University. We are further distressed that
that when the Soviet Consulate in San Francisco reported on the situation
to Moscow, this report probably re-inforced the idea in Moscow that
American protests about Poland and Afghanistan were a governmental sham
without any support in the institutions of the country such as
universities.
We believe that Stanford should adopt a policy toward Soviet
visitors that balances the following considerations:
1. Preserving the orderly functioning of the academic processes of
research, publication and teaching and avoiding the creation of a security
atmosphere.
2. Showing the visitors a society more open than their own.
3. Minimizing transfer of militarily useful technology so as to
minimize our own defense costs; perhaps we can help obviate the need for a
return to the draft. Ways of implementing this without security measures
need study.
4. Increasing reciprocity so that access to Soviet research
becomes available to Americans to the same extent that our research
activities are accessible to them. This involves some degree of support
to the bargaining efforts on our behalf conducted by the State Department
and the National Academy of Sciences.
5. Maintaining good relations with the State Department, the
Defense Department and other Government agencies.
Since these goals are not entirely co-incident, some degree of
compromise is needed.
We suggest that a committee be appointed to develop a policy. If
an interim statement clarifying the impression given by the Stanford press
release could be made, it might help mitigate the unfortunate signal that
has been sent to Moscow.
[John: Hear, hear!! Lowell]
Well, we didn't get to first base with the Vice-provost, who took an
extremely defensive attitude. We'll try the President next.
∂02-Feb-82 2109 SGF terminal
iI don't have any funds on my grant to be able to get amy own terminal,
i I appreciate very much having the use of this, and hope it is
possible to continue borrowing it. Thanks, Sol.
∂03-Feb-82 0242 Lowell Wood <LLW at S1-A> Follow-Up Or Butter-Up?
Date: 03 Feb 1982 0236-PST
From: Lowell Wood <LLW at S1-A>
Subject: Follow-Up Or Butter-Up?
To: jmc at SU-AI
CC: LLW at S1-A
∂02-Feb-82 2049 John McCarthy <JMC at SU-AI>
Date: 02 Feb 1982 2054-PST
From: John McCarthy <JMC at SU-AI>
To: llw at S1-A
Well, we didn't get to first base with the Vice-provost, who took an
extremely defensive attitude. We'll try the President next.
[John: From the little I know of Stanford politics, you might aptly try to
first present the matter to a like-minded Stanford Trustee, and then have
*him* call the President and ask him to hear you on this matter, as said
Trustee is `quite concerned about the policy aspects of Stanford
administration handling of this and possible future incidents.' This is
likely to get you seriously listened to by your President, *and* to get
some follow-up action (rather than on-the-spot butter-up action, to be
followed by very little). Good luck! Lowell]
∂03-Feb-82 1002 FFL
To: JMC, FFL
Lisa Hall of the Philosophy Dept. called re the colloquia. The date had been
given tentatively to Sol Feferman. She is sorry and suggests you may perhaps
want a date in spring quarter.
∂03-Feb-82 1006 FFL
To: JMC, FFL
Yes, Meng Lee has applied for admission.
∂03-Feb-82 1036 FFL
To: JMC, FFL
∂02-Feb-82 1717 JMC
Check that we are no longer paying Richard Anderson who is with HPP.
Thank you for letting me know. I stumbled upon it when the computer charges
were assigned and caught it then. The salary which you paid him for the
first half of January will be transferred next month to HPP.
∂03-Feb-82 1119 FFL
To: JMC, FFL
George Morales, who says he knows you, asks that you call him. He would like
to have an appointment with you. He is free at almost any time. 408 733 6590.
∂03-Feb-82 1148 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE Week of FEBRUARY 8 - 12
Date: 3 Feb 1982 1128-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE Week of FEBRUARY 8 - 12
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
2/8/82 Math380C Todd Dupont
Monday Numerical Analysis University Of Chicago
4:15 p.m. Seminar ``Mesh modification and evolution equations''
2/9/82 MJ301 Mr. John Kunz
Tuesday Medical Computing Stanford University
1:30 p.m. Journal Club ``Recent Medical Computing Articles of Interest''
2/9/82 MJ301 Dr. Milt Grinberg
Tuesday Knowledge Stanford University
2:30 p.m. Representation ``CSA: A Causal Representation Language''
Group Meeting
2/9/82 Jordan041 John Ousterhout
Tuesday Computer Science U.C. Berkeley
4:15 p.m. Colloquium ``Caesar: An Interactive Editor for VLSI
Layouts''
2/10/82 Skilling Auditorium Jerry Popek
Wednesday Computer Systems UCLA
4:15 p.m. Laboratory Seminar To be Announced
2/12/82 MJ301 Alberto O. Mendelzon
Friday Database Research University of Toronto
3:30 p.m. Seminar ``In Praise of the Weak Universal Relation
Model''
-------
∂03-Feb-82 1405 Richard Treitel <CSL.VER.RJT at SU-SCORE> SCORE Prolog
Date: 3 Feb 1982 1347-PST
From: Richard Treitel <CSL.VER.RJT at SU-SCORE>
Subject: SCORE Prolog
To: jmc at SU-AI
cc: CSL.VER.RJT at SU-SCORE
Flipping back and forth between editor and Prolog seems easy enough, using
the "exit to Monitor" command described in sec. 3.7 of the manual, and the
"reconsult" built-in function allows you to patch up your programs in some
way or other, assuming you know enough Emacs (or any other editor) to create
a file of changes separate from your program file!!
However, there seems to be an undocumented feature: try starting up the system
and immediately giving a "save" command with some random filename.
- RJT
-------
∂03-Feb-82 1534 Patrick H. Winston <PHW at MIT-ML>
Date: 3 February 1982 18:36-EST
From: Patrick H. Winston <PHW at MIT-ML>
To: jmc at SU-AI, tw at SU-AI, reddy at CMU-10A, simon at CMU-10A
cc: PHW at MIT-ML, KRD at MIT-ML, BKPH at MIT-ML, GJS at MIT-ML,
KAREN at MIT-ML
Today some brochures arrived implying that P. Winston, B. K. P. Horn,
R. Davis, and G. Sussman are all thinking of attending a conference
in Dubrovnik, Yugoslavia. Of these, only Winston actually considers
the question open (having foolishly getting drawn in some time ago).
The rest have declined and a letter was sent to the Yugoslavs saying
so well in advance of the brochure mailing. We intend to express
irritation and to insist on a second mailing rectifying the
misimpression given.
PHW, KRD, BKPH
∂04-Feb-82 0234 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 4 February 1982 05:32-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: SPACE at MIT-MC
The whole space community, with, I thnk, particular
credit to L-5 Society, deserves a couplee attaboys. I'll take a
bit of the plaudits because of the Citizens Council activity
(and Danny Graham's efforts, plus Newt Gingrich's were somewhat
influenced and aided by the Council.) Anyway--it is not what we
wanted, but it is less than we feared.
We could get into next year's state of the union if we
worked it right; it means more coordinated work...
Date: 03 Feb 1982 2335-PST
From: John McCarthy <JMC at SU-AI>
... remainder deleted
∂04-Feb-82 0829 Paul Armer <CSD.ARMER at SU-SCORE> 76-77 Students
Date: 4 Feb 1982 0825-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: 76-77 Students
To: csd.McCarthy at SU-SCORE
Stanford-Phone: (415) 497-2274
The provost's office is inquiring as to where our 76-77 students got
positions after getting their Ph.D. degrees from Stanford. Can you
help me with Robert Bolles, Robert Cartwright and Todd Wagner?
Thanks, Paul.
-------
Bolles is at SRI, Cartwright is at Rice and Wagner is at Intel.
Put me on your mailing list as JMC@SAIL, although mail addressed
as you did is automatically forwarded.
∂04-Feb-82 0900 JMC*
gilliam
∂04-Feb-82 0946 asprey at PARC-MAXC Re:
Date: 4 Feb 1982 09:43 PST
From: asprey at PARC-MAXC
Subject: Re:
In-reply-to: Your message of 02 Feb 1982 1048-PST
To: John McCarthy <JMC at SU-AI>
cc: asprey
And now I've missed two PROLOGUE classes. Is it worth coming now and, if so,
where is the class held? Sorry to be such a bother.
Peggy Asprey
If you know something or will do some reading, it may be
worthwhile to come. The class is in 301 Margaret Jacks Hall.
∂04-Feb-82 0954 Ichiki at SRI-AI TINLUNCH IN BUILDING E
Date: 4 Feb 1982 0953-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH IN BUILDING E
To: tlgrp:
Just a quick reminder that the TINLUNCH meeting today at lunch time
will be back in the old conference room EK242 in our Building E.
-------
∂04-Feb-82 1005 Scott at SUMEX-AIM I wasn't sure if this got to you
Date: 4 Feb 1982 1005-PST
From: Scott at SUMEX-AIM
Subject: I wasn't sure if this got to you
To: jmc at SU-AI
Mail-from: ARPANET host SU-SCORE rcvd at 4-Feb-82 0813-PST
Date: 4 Feb 1982 0810-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: Re: for you
To: Scott at SUMEX-AIM
cc: CSD.BSCOTT at SU-SCORE
In-Reply-To: Your message of 31-Jan-82 1604-PST
Sorry for the delay, John. I've been away for a few days. Will try to get
the information to you today.
Betty
-------
-------
∂04-Feb-82 1318 FFL
To: JMC, FFL
Oliver called and said that he intended to have a technical meeting next
Thursday or Friday. Please call him at home tonight. 617 862 5438.
∂04-Feb-82 1504 TOB
What do you think about the worth of chairing a session in L5?
Well, I sicked the guy on you. Pournelle, who sponsored the meeting,
is a prolific and well known science fiction writer. He has also
written on defense policy, etc. in collaboration with a now retired
Hoover professor. He is also a Republican politician. His Citizen's
Advisory Council on Space Policy, in which I have taken part, together
with the L-5 society, deserves a good part of the credit for Reagan's
decision, reported last night, to save the planetary program of NASA
against Stockman's advice.. In general, I think Pournelle's enterprises,
which involve both scientists and space fans, are surprisingly effective.
It would be worthwhile for you to do some thinking about robotics in
space. Except that I will be in France, I would have accepted the
invitation myself.
∂04-Feb-82 1618 Betty Scott <CSD.BSCOTT at SU-SCORE> LLL #9628303
Date: 4 Feb 1982 1615-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: LLL #9628303
To: JMC at SU-AI
cc: CSD.BScott at SU-SCORE
John, here is the list of those who were paid on this contract--
in alphabetical order and excluding secretarial:
Baskett, Forest
Earnest, Les
Eichenberger, P.
Elmqvist, Hilding
Frost, Martin
Gilbert, Erik
Hennessy, John
Lam, Gary
Lebrun, Marc
McCarthy, John
Panofsky, Edward
Rodriguez, Armando
Rubin, Jeffrey
Samuel, Arthur
Let me know if you need further information.
Betty
-------
I guess I need to know the amounts paid to the different people. I didn't
realize that I was paid at all on the contract.
∂04-Feb-82 1632 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA Contract
Date: 4 Feb 1982 1621-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: ARPA Contract
To: JMC at SU-AI, TOB at SU-AI, DCL at SU-AI, ZM at SU-AI,
Wiederhold at SUMEX-AIM
cc: FFL at SU-AI, MAS at SU-AI, CSD.BScott at SU-SCORE
I have just learned today that the ARPA contract papers were mailed
by Navelex almost two weeks ago. They have not been received at
the Stanford SPO office. Navelex promised to Federal Express a copy
to Stanford in tonight's mail. Just as soon as I have more information
I'll send you a message--and just as soon as the papers are here
I will try and get an advance budget account number for each of you.
Betty
-------
∂04-Feb-82 1704 Gene Golub <CSD.GOLUB at SU-SCORE> Visit by Popek
Date: 4 Feb 1982 1701-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Visit by Popek
To: faculty at SU-SCORE
cc: csd.dietterich at SU-SCORE
Popek of UCLA will be visiting on Wednesday, February 10. He is being
considered for a CIS/CSD appointment. He'll be speaking at the EE 380
seminar at 4:15. Please let me know if you wish to see him.
GENE
-------
∂04-Feb-82 1905 MMS Match Proof
To: JJW, JMC
Are Alists made as lists of 2-element lists, or
lists of dotted pairs, or maybe even
lists of lists?
the |alistp ⊃ | axiom in LISPAX looks like its
for lists of 2-element lists. The statement
of MATCH given out seems to construct lists of
dotted pairs.
I would think that for the purposes of this
proof, lists of dotted pairs would be slightly
easier, because it takes one less operation
to take them apart.
Lists of dotted pairs are what was intended.
∂04-Feb-82 2356 TOB
John
I will consider it. I have served on two NASA committees and
think that I have done enough public service in that direction.
I agree that robotics in space has interest.
I have thought about robotics in space in those two committees
and could stand to think about it more. I would be more inclined
to involve myself if I saw a place to get some support. I have
been reluctant to make a strong effort to get support from NASA given
their economic situation. I have in mind advanced telemanipulator work
for constructing and servicing communication systems, large antennas.
Tom
∂04-Feb-82 1511 JMC
Well, I sicked the guy on you. Pournelle, who sponsored the meeting,
is a prolific and well known science fiction writer. He has also
written on defense policy, etc. in collaboration with a now retired
Hoover professor. He is also a Republican politician. His Citizen's
Advisory Council on Space Policy, in which I have taken part, together
with the L-5 society, deserves a good part of the credit for Reagan's
decision, reported last night, to save the planetary program of NASA
against Stockman's advice.. In general, I think Pournelle's enterprises,
which involve both scientists and space fans, are surprisingly effective.
It would be worthwhile for you to do some thinking about robotics in
space. Except that I will be in France, I would have accepted the
invitation myself.
∂05-Feb-82 0029 TOB
To: csd.golub at SU-SCORE, JMC at SU-AI
∂30-Jan-82 2004 TOB
I think that the adjunct professor system should be greatly expanded to
permit the University to respond to needs and opportunities in important
areas with more flexibility in faculty slots than Stanford now has.
Adjunct faculty apparently serve varied roles. In language departments, it is
my impression that they function as lecturers, e.g. native speakers
teaching language performance. In science areas, adjunct faculty perform
primarily research and scholarship functions. This note refers to the
adjunct faculty role of research and scholarship.
There should be a title which reflects this role: I propose Research Faculty.
Research Assistant Professor, etc.
A key issue is maintaining quality control. Research faculty should be
able to serve as principal investigators. However, to maintain quality
control, appointments should be made for fixed terms as for regular faculty.
Reappointment would be necessary to continue, so that merely obtaining
funding does not make Research Faculty independent of the selection system.
Research and education in VLSI, molecular genetics, computer science,
Artificial Intelligence and robotics are important areas for Stanford,
areas in which Stanford has had a leading position with large efforts.
There are opportunities for substantial growth in some of these areas.
As long as it is possible to maintain top quality of research and education,
there are benefits to be had which are only possible with a critical mass
program. Indeed, to be sub-critical mass is to be second rate in several
areas.
Tenure slots are too few to allow much flexibility
in recruiting staff to respond to opportunities and needs such as these.
The university has no commitment of tenure to adjunct faculty. A major
reason for inflexibility in assigning faculty slots is the financial
commitment of tenure. Adjunct faculty do not have tenure and
depend on research funds.
I think there is a need for "critical mass" only in a few areas. Also
the adjunct faculty in language will be justly offended by your misinformation
that they are mainly native speakers - they aren't. At least the
adjunct professor in charge of teaching French has a non-French name
and is the author of a series of textbooks.
∂05-Feb-82 0813 Betty Scott <CSD.BSCOTT at SU-SCORE>
Date: 5 Feb 1982 0811-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
To: JMC at SU-AI
cc: CSD.BSCOTT at SU-SCORE
In-Reply-To: Your message of 4-Feb-82 1624-PST
I'll get the salary amounts today.
Betty
-------
∂05-Feb-82 0810 GXG via SRI-KL lisp history
hello ,prof.mccarthy.we have access to your computer via net.therefore it is
possible to receive comment on my pages without asking you to write paper
letters.did you read anything?what,when?did you gave to david luckham something?
it's this time the same as with your letters.nobody will would answer.
you understand:nobody did answer,even you yourself not.obviously the
project as whole is in question now.the springer verlag is intyλerested to
take the book but it is not yet settled.it would be encouraging if you
could send me comments and remarks.or is it too hard to read the stuff?
bye,herbert stoyan
∂05-Feb-82 0823 Betty Scott <CSD.BSCOTT at SU-SCORE> ARPA contracts
Date: 5 Feb 1982 0819-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: ARPA contracts
To: JMC at SU-AI, TOB at SU-AI, Feigenbaum at SUMEX-AIM, CSL.Lantz at SU-SCORE,
CSD.Ullman at SU-SCORE
cc: MAS at SU-AI, FFL at SU-AI, CSD.BScott at SU-SCORE
I have just received word that the McCarthy and Feigenbaum contracts
(the current ones, not the new five P.I. contract) have been fully
funded. This message came from SPO. As soon as I receive the
modifications, I will allocate the funds to the appropriate accounts.
Will let you know when this is done.
Betty
-------
∂05-Feb-82 1124 Betty Scott <CSD.BSCOTT at SU-SCORE> Contract N00039-82-C-0250
Date: 5 Feb 1982 1119-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: Contract N00039-82-C-0250
To: JMC at SU-AI, TOB at SU-AI, DCL at SU-AI, ZM at SU-AI,
Wiederhold at SUMEX-AIM
cc: FFL at SU-AI, MAS at SU-AI, CSD.BScott at SU-SCORE
The ARPA contract papers are here. Note that since it was processed
through Navelex, it has a Navy number.
The contract has to be approved by both CS and EE. I will try to get
this done this afternoon, and on Monday will have budget account
numbers for you. Pending receipt of a finally approved contract,
we have to guarantee expenditures against a department unrestricted
account. I will do that for CS, and will have SEL take care of it
for David.
Betty
-------
∂05-Feb-82 1313 Raj.Reddy at CMU-10B Conference in Yugoslavia
Date: 5 February 1982 1607-EST (Friday)
From: Raj.Reddy at CMU-10B
To: PHW at MIT-ML, JMC at SU-AI, TW at SU-AI, Simon at CMU-10A,
Raj.Reddy at CMU-10B, KRD at MIT-ML, BKPH at MIT-ML, GJS at MIT-ML,
KAREN at MIT-ML
Subject: Conference in Yugoslavia
Message-Id: <05Feb82 160742 RR29@CMU-10B>
I have the same problem as the rest of you. I do not plan to be in Yugoslavia
for this conference. I am also somewhat annoyed at their using our names
without our approval.
Raj
∂05-Feb-82 1350 TOB
John
Thanks for the comments.
tom
∂05-Feb-82 0036 JMC
To: TOB at SU-AI
CC: csd.golub at SU-SCORE
I think there is a need for "critical mass" only in a few areas. Also
the adjunct faculty in language will be justly offended by your misinformation
that they are mainly native speakers - they aren't. At least the
adjunct professor in charge of teaching French has a non-French name
and is the author of a series of textbooks.
∂05-Feb-82 1408 JJW Qualifying exam
Zohar has told me that he plans to be back again in April, and thinks
that would be a better time to give me the exam. I hope this is all
right with you.
Joe
I will in Europe in April and May, but I don't have to be present.
Alternatively, the exam could be postponed till Zohar returns again,
which I believe is in September.
∂05-Feb-82 1451 FFL
To: JMC, FFL
John Cate asks that you call him. 321 1225.
∂05-Feb-82 1900 JMC*
Selfridge
∂06-Feb-82 1722 TW Exams
To: "@COMP.DIS[1,TW]" at SU-AI
The exam went fine on Saturday due to help from Tom, Joe, Gio and
Ginger. The papers have all been distributed to the appropriate
graders. Rosemary is expecting to have final scores (i.e. after
the reconciliation between the two graders on any one section)
by 10 on Tuesday, in order to prepare a summary sheet that we
can work from in the meeting at 12. The easiest would abe if
you could give her a sheet correlating student numbers and grades
(rather than have her try to decipher your markings on the blue
books). Thanks --t
∂06-Feb-82 1709 JJW Comprehensive exam
I have the exam books for the MTC section, and will take them home
tonight (Saturday) to start grading. I'll bring them back in
tomorrow.
Joe
∂06-Feb-82 1832 pratt@Shasta (SuNet) Mass storage
Date: 6 Feb 1982 18:17:02-PST
From: pratt at Shasta
To: equip
Subject: Mass storage
One product that came on the market last month is the M860 MSS (Mass Storage
System), made by Masstor Systems Corp, Sunnyvale. One 55-gigabyte storage
module and control unit is $450,000. This is about half the price of the
comparable IBM mass storage product, the 3850. Both are cartridge-based
systems. The M860 has 316 3.5" data cartridges, each holding 175 megabytes.
It packs its data 18 times as densely as the 3850, giving a substantially more
compact unit.
55 gigabytes equals the capacity of 1000 reels of 2400' tape recorded at
1600 bits per inch (the density used in our present archiving). This is one
promising direction to look for tertiary storage at Stanford.
Vaughan
∂07-Feb-82 2037 Feigenbaum at SUMEX-AIM mass storage?
Date: 7 Feb 1982 1954-PST
From: Feigenbaum at SUMEX-AIM
Subject: mass storage?
To: jmc at SU-AI
Mail-from: ARPANET host SU-AI rcvd at 6-Feb-82 1818-PST
Date: 6 Feb 1982 18:18:05-PST
From: pratt at Shasta
Reply-to: csd.pratt at SCORE
To: equip
Subject: Mass storage
One product that came on the market last month is the M860 MSS (Mass Storage
System), made by Masstor Systems Corp, Sunnyvale. One 55-gigabyte storage
module and control unit is $450,000. This is about half the price of the
comparable IBM mass storage product, the 3850. Both are cartridge-based
systems. The M860 has 316 3.5" data cartridges, each holding 175 megabytes.
It packs its data 18 times as densely as the 3850, giving a substantially more
compact unit.
55 gigabytes equals the capacity of 1000 reels of 2400' tape recorded at
1600 bits per inch (the density used in our present archiving). This is one
promising direction to look for tertiary storage at Stanford.
Vaughan
John, is this concept good for us? Is the device any good (do you know?)
Ed
-------
In reply to Ed's question about the M8l60.
It doesn't find data as quickly as a disk, since it has to spin tape,
but of course it can provide very large file backup. I fear it's too
expensive for us, but it would certainly be nice to have in the long
run. In any case, it's best to postpone serious discussion till we have
our money.
∂08-Feb-82 0053 Feigenbaum at SUMEX-AIM (Response to message)
Date: 8 Feb 1982 0048-PST
From: Feigenbaum at SUMEX-AIM
Subject: (Response to message)
To: JMC at SU-AI
In response to your message sent 07 Feb 1982 2302-PST
Thanks, John (re answer on tape device)
I havn't yet set up a party since I havn't decided yet on small vs large one.
And I want to check with Herb on dates. He just got in to town.
Ed
-------
∂08-Feb-82 0103 REM via SU-TIP Read-Alter mode from MacLISP
To: REG, JMC
I finally got around to figuring out how to allocate pages of memory
in MacLISP that are exempt from the GC, figuring out how to do page
mapping between memory and file on ITS, and doing the grungy work to
interface Read-Alter mode on SU-AI. Now I have a MacLISP package to
allow random read/write access on disk files on both ITS and SU-AI.
After I finish cleaning it up, I may be able to get back to some
volunteer projects that were suspended due to lack of Read-Write I/O
in MacLISP.... ask me if you're interested in using this facility for
your own purposes. File is REAWRI.LSP[1,REM] together with LAP822.LSP[1,REM]
for the machine-language interface.
I suggest you send this message to RPG who is the MACLISP guru here.
∂08-Feb-82 0214 pratt@Shasta (SuNet) File system technology update
Date: 8 Feb 1982 02:06:42-PST
From: pratt at Shasta
To: equip
From: pratt
Subject: File system technology update
Date: 8 February 1982 02:06:29-PST (Monday)
The following is intended as a file system technology update for the
Computer Facilities committee. My goal is not to recommend either a
particular file system architecture or components for such an architecture,
but rather just to give people an idea of what things cost today for one
reasonable file system architecture.
That the particular architecture assumed below is reasonable depends mainly
on the following considerations.
1. We are rapidly moving towards a processor-per-user world.
2. Performance demands that secondary storage be efficiently accessible from
such single-user processors.
3. The continued high cost of disks makes it too expensive to allocate one
per user. Furthermore disks are unwelcome office mates, being both noisy and
bulky. Thus we should be thinking in terms of each disk being shared by
several users, connected by something with the performance of an Ethernet.
4. To maintain reasonable response a disk arm should not be called on to
service more than about 10 processors (plus or minus 5).
5. More than 2 or 3 disks on one processor or Ethernet can turn that
processor or Ethernet into the main bottleneck, inappropriate when the disks
are by far the most expensive component.
6. A delay of several seconds in retrieving data older than a month
should be quite acceptable.
7. The following products are among the more attractive file system
components available today. The Fujitsu 2884, a 160-megabyte 1 megabyte/sec
Winchester, costs $8800 in quantity one, and $5500 in OEM quantities. The
Fujitsu Eagle, a 480-megabyte 1.8 megabyte/sec 10" Winchester, costs $18,000
in quantity one. The Masstor M860, a 55-gigabyte cartridge system, with a
latency on the order of a few seconds, costs $450,000 in quantity one.
These considerations suggest the following file system architecture.
* Up to 10 single-user computers net-connected to one or possibly two
Winchester disks under the control of a computer. The disks supply secondary
storage and contain files accessed within the past month. Call this a
cluster.
* Clusters are net-connected to each other and to some source of tertiary
storage. No specific topology is specified for cluster interconnection.
The tertiary storage could be provided by something having the specifications
of the M860, although I'd prefer something more decentralized for the sake of
redundancy. The capacity and access time of the M860 are both more than
adequate for our needs for the next several years, by which time video disks
or better should be well established.
Performance
Cluster performance should be essentially indistinguishable from the
performance of a conventional timeshared computer having 10 users and the
same number of disks, except that compute-bound response will be improved due
to the per-user processors. This follows from the observation that between
disks, computers, and Ethernets, Ethernets constitute the least bottleneck.
Cost
The costs of such an approach are as follows. A 160 Mb cluster disk with its
controller, processor, and net interface (i.e. a complete file server for a
cluster) should cost around $16,000. 20 clusters supporting 200 users would
then have a secondary memory cost of around $300,000. This is in the same
ballpark as the tertiary storage cost if an M860 is used. A 480 Mb disk could
be used if it turns out that 160 Mb is too small for this application,
which however I doubt.
Let me say again that I am not proposing this architecture or its components
as something to spend the current supply of ARPA money on, but to bring the
committee up to date on what's available for one plausible architecture. I
would hope however that the above figures could serve as a rough guide in
evaluating other architectures to get a lower bound on how far from optimal
such architectures might be, both in terms of cost and performance.
Performance is particularly important when considering secondary storage for
a community of the order of hundreds of users. The size and number of disks
must be chosen with due regard for latency and throughput as well as raw
capacity. I am under the impression that file system performance has
received little attention from the committee to date compared to capacity.
The users served by the committee could well be disappointed in the system
performance if inappropriate choices are made.
Vaughan
∂08-Feb-82 1000 JMC* via S1-A
gilliam, LLNL bill
∂08-Feb-82 1310 Susan Seabrook <CSD.SIS at SU-SCORE> President's Advisory Committee on Computer Science & Computing
Date: 8 Feb 1982 1304-PST
From: Susan Seabrook <CSD.SIS at SU-SCORE>
Subject: President's Advisory Committee on Computer Science & Computing
To: CSD-Faculty: ;
Stanford-Phone: (415) 497-4365
TO: Computer Science Department Faculty ---
The President's Advisory Committee on Computer Science and
Computing will meet February 10th and 11th. I am forwarding a
copy of the agenda to each of you. Please feel free to drop in
if you wish. Note that at times they will split into two groups,
as Margaret Jacks 252 may get crowded. You are also invited to a
reception on Wednesday Feb. 10th at 5:30 p.m. in the Piazza Room
of the Holiday Inn in Palo Alto.
-- Paul Armer --
President's Advisory Committee
on Computer Science and Computing
1981 - 82 Meeting
February 10 - 11, 1982
*** AGENDA ***
Wed. February 10th
Margaret Jacks 146
8:30 - 9:00 I. Introduction
Stephen G. Lukasik
Edward E. Shaw
9:00 - 9:30 II. General Review of Information Technology at
Stanford
Edward E. Shaw
9:30 - 10:00 III. General Review of the Computer Science
Department
Gene Golub
10:00 - 10:15 BREAK
10:15 - 11:30 IV. Instructional and Research Computing
Michael Carter, Ralph Gorin, Jerry Miller
11:30 - 12:15 V. Networking at Stanford:
Part I: Voice, Data & Video Strategies
William Yundt
12:15 - 1:30 CATERED LUNCH
1:30 - 2:15 VI. Networking at Stanford:
Part II: Digital Networking Plan
William Yundt
2:15 - 3:00 VII. Center for Integrated Systems
John Linvill
3:00 - 3:15 BREAK
!
UNIVERSITY - WIDE CSD
PERSPECTIVE PERSPECTIVE
MJ 252 MJ 146
3:15 - 4:00 VIII-A. Text Networking VIII-B Curriculum Changes
Jon Sandelin Rob Schreidber
4:00 - 4:45 IX-A. Computer Systems & Services IX-B CSD's Computer
Edgar Williams Facilities
Ralph Gorin
5:30 - 6:30 RECEPTION - Holiday Inn, Piazza Room
Thursday, February 11th
UNIVERSITY - WIDE CSD
PERSPECTIVE PERSPECTIVE
MJ 220 MJ 252
8:30 - 9:15 X-A. Data Base Management X-B. Student Concerns
John Sack Tom Dietterich
9:15 - 10:00 XI-A. Administrative XI-B. Proposal for New Equipment
Information Systems Jeff Ullman
David Ernst
10:00 - 10:15 BREAK
Margaret Jacks 252
10:15 - 11:00 XII-A. Summary
Edward E. Shaw
11:00 - NOON XIII. Executive Session
-------
∂08-Feb-82 1342 Susan Seabrook <CSD.SIS at SU-SCORE>
Date: 8 Feb 1982 1335-PST
From: Susan Seabrook <CSD.SIS at SU-SCORE>
To: CSD-Faculty: ;
Stanford-Phone: (415) 497-4365
Re the possible crowding in MJ252 mentioned in the previous note on the
meeting of the Advisory Committee, the intent was to point out that
sessions in MJ252 may be crowded. That potential crowding is independent
of whether or not the group is split at the time.
-------
∂08-Feb-82 1425 Betty Scott <CSD.BSCOTT at SU-SCORE> LLL #9628303 Salaries
Date: 8 Feb 1982 1425-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: LLL #9628303 Salaries
To: JMC at SU-AI
cc: CSd.BScott at SU-SCORE
Here are the salaries charged to this contract, John:
Baskett $15,345
Earnest 8,538
Eichenberger 2,400
Elmqvist 3,450
Frost 14,208
Gilbert 737
Hennessy 4,148
Lam l,463
Lebrun 13,133
McCarthy 722
Panofsky 12,772
Rodriguez 16,879
Rubin 18,123
Samuel 16,530
Betty
-------
∂09-Feb-82 0319 GXG via SRI-KL account
hello,dear prof.mccarthy.could you be so kind and reestablish an
account for me at sail machine?i prpose user name hstoy and
password xxxxxx.thank you.herbert stoyan
An account for Herbert Stoyan has been created. The account name
is HST (account names cannot have more than three letters), and
the password is the one requested.
∂09-Feb-82 0909 FFL
To: JMC, FFL
I need to know if Stoyan is to have a "guest" account, 25 pages, or 1 aliquot,
about 50 pages, please?
A guest account.
∂09-Feb-82 1053 nowicki@Shasta (SuNet) ADCOM talk
Date: 9 Feb 1982 10:52:20-PST
From: nowicki at Shasta
To: equip
From: nowicki
Subject: ADCOM talk
Date: 9 February 1982 10:52:08-PST (Tuesday)
This message was sent to me, even though I have been kicked off this
committee. Evidently the "equip" list on Diablo had been corrupted.
I changed equip@Diablo to forward to equip@Shasta instead.
----------------
From ullman@Diablo Tue Feb 9 10:26:37 1982
Date: 9 Feb 1982 10:26:32-PST
From: ullman at Diablo
To: equip
Subject: ADCOM talk
I desperately need someone to fill in for me and speak to the advisory
committee on the equipment proposal. The talk is scheduled for 9:15-10:00AM
Wednesday 2/10. I have slides prepared (source in /usr/ullman/adcomslides
@diablo). In case of ties, first postmark gets the honor of explaining
to a surly and uninterested bunch of distinguished people why it is we
can't behave more sensibly.
I'll do it if you like, but ...
∂09-Feb-82 1105 FFL
To: JMC, FFL
Please call Alice Cuneo of the San Jose Mercury. 941-4813.
∂09-Feb-82 1151 ullman@Diablo (SuNet) ADCOM
Date: 9 Feb 1982 11:49:26-PST
From: ullman at Diablo
To: equip at Diablo
Subject: ADCOM
Lucky me; it turns out that my piece was scheduled for Thursday, so
I'll be able to give the talk after all. Thanks to all of you who
volunteered or who will volunteer after reading my first message but before
you read this one.
∂09-Feb-82 1331 Paula Bauman <PJB at S1-A> Consulting Bill
Date: 09 Feb 1982 1327-PST
From: Paula Bauman <PJB at S1-A>
Subject: Consulting Bill
To: jmc at SU-AI
I found out that you have two choices of mileage from Stanford to
LLNL, depending on which route you take. If you drive over the
San Mateo Bridge, the mileage is 52 miles each way. If you drive
over Dumbarton Bridge, the mileage is 48 miles each way. You can
also claim the bridge toll of 75 cents for each trip. The mileage
rate for reimbursement is 22 1/2 cents per mile.
You can enter your mileage claim on the next "Claim For Consulting
Services" form you submit, under the "For Local Travel Only" section.
I am still trying to submit a request for your mileage for those 12 days
you worked. Could you give me the dates, please and indicate which
route you took, and I will get this taken care of (I hope). Thanks.
I actually took a third route by highway 237 which I suppose is 50
miles. As to the dates of my summer travel, I think it would be best
to make it the 12 days for which they paid me. As you may recall, I
told you that they over paid me, so I worked off the extra days during
the Fall, but I don't want to overload the accounting system. It seems
to me that my first day was September 21 and got paid for the next 12
working days. Thanks in advance for your efforts to make it all come
out even.
∂09-Feb-82 1331 MAS Visit of Cdr Ohlander from DARPA
Commander Ohlander from DARPA will be here Thursday, February 18 and would
like to talk to you. Is 1:30 pm a good time for you?
Marianne
∂09-Feb-82 1434 Ichiki at SRI-AI TINLUNCH 2/11
Date: 9 Feb 1982 1433-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH 2/11
To: tlgrp:
Before the TINLUNCH Meeting February 11th, there will be a formal
seminar by Martin Kay in Building E, Conference Room EK242 from 10:30
to 11:30. The TINLUNCH meeting will then be held in EK242 at lunch
time. The paper to be discussed is:
An Algorithm for Compiling Parsing Tables from a Grammar
By: Martin Kay
Martin Kay will also attend the meeting.
Copies of the above paper are on top of Barbara's file cabinet.
-------
∂10-Feb-82 0325 HST via SRI-KL account
hello,dear prof.mccarthy.thanks for the account.did you show to dave
luckhmam the pages on early lisp history?bye.herbert srtoyan.
∂11-Feb-82 0023 RWG
no noodles til 1530 today.
∂11-Feb-82 0937 FFL
To: JMC, FFL
Chauncey Starr's offic called to remind you of lunch date, Wed., Feb. l7.
12:30 in his office at EPRI. Brn bag or sandwiches and drinks available there.
Is on calendar.
∂11-Feb-82 1151 RPG Papers
I grabbed them off your shelf.
-rpg-
∂11-Feb-82 1157 RPG
John:
Here's a list of possible references. ! means a good recommendation,
? means I'm not sure, ?! etc means could be good, depending on moods.
Terry will give a good recommendation for Lisp/systems type position.
David Waltz will give the best by far, but he might be unhappy to see me
not going exactly into AI. I talked a little to Bruce Buchannon, who gave some
support. I'm also not sure how the systems faculty will receive this, especially
Baskett (is he still faculty). Lantz won't like it either. Reid will be amenable,
but he may not count.
Cordell Green (Kestrel)!
William L. Scherlis (CMU)!
Martin Griss (Utah)!
Barry Soroka (USC)!
Gerry Sussman (MIT)?!
Pat Winston (MIT)?
Jon L. White (MIT)!
Scott Fahlman (CMU)?!
Rod Brooks (MIT)!
Jeff Rubin (LLL)!
Tom McWilliams (LLL)?
Richard Weyhrauch (Stanford)!
Ed Davidson (EE at Univ of Illinois)!
Janak Patel (CSL at Univ of Illinois)!?
I suggest you offer the following as references, suitably combining people
who will give good references and those who will be asked anyway.
David Waltz (Univ of Illinois)!
Terry Winograd (Stanford)?
John McCarthy (Stanford)?
Guy L. Steele Jr (CMU)!
Lowell Wood (LLL)!
Nils J. Nilsson (SRI)!
∂12-Feb-82 1008 FFL
To: JMC, FFL
Dr. Teller's office called. He has seen the news that computers can be
used to find out whether a number is a prime number. He wants to know
who he can talk to about it. He also wants to talk with you and would
like to see you on Thursday, Feb. 18, late in the afternoon, if possible.
Pls. reply to Liz at 7-0601 or 7-0425.
∂12-Feb-82 1305 CLT
I have ordered (3) tickets for Tartuffe for tomorrow.
The theatre is at Market and SanCarlos in San Jose.
Box office opens at 5pm
We could go pickup the tickets and have supper some where nearby,
or get there around 7:40 to pickup tickets (performance at 8pm)
We should decide, so we can tell Antonio what the plan is.
∂12-Feb-82 1434 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
Date: 11 Feb 1982 1031-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Al Thaler"s talk
To: bboard at SU-SCORE, faculty at SU-SCORE
cc: bscott at SU-SCORE
Al Thaler will talk today at 11:00 am in room 383M, Math Building on
"Everything you always wanted to know about NSF and were afraid to
ask"
-------
∂12-Feb-82 1434 Gail Stein <OR.STEIN at SU-SCORE> Re: NSF Funding
Date: 11 Feb 1982 1427-PST
From: Gail Stein <OR.STEIN at SU-SCORE>
Subject: Re: NSF Funding
To: CSD.GOLUB at SU-SCORE, faculty at SU-SCORE
cc: su-bboard at SU-SCORE, OR.STEIN at SU-SCORE
In-Reply-To: Your message of 11-Feb-82 0830-PST
Thanks for the message re Al Thaler.
_
-------
∂12-Feb-82 1439 Gail Stein <OR.STEIN at SU-SCORE> Re: NSF Funding
Date: 11 Feb 1982 1427-PST
From: Gail Stein <OR.STEIN at SU-SCORE>
Subject: Re: NSF Funding
To: CSD.GOLUB at SU-SCORE, faculty at SU-SCORE
cc: su-bboard at SU-SCORE, OR.STEIN at SU-SCORE
In-Reply-To: Your message of 11-Feb-82 0830-PST
Thanks for the message re Al Thaler.
_
-------
∂12-Feb-82 1439 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
Date: 11 Feb 1982 1031-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Al Thaler"s talk
To: bboard at SU-SCORE, faculty at SU-SCORE
cc: bscott at SU-SCORE
Al Thaler will talk today at 11:00 am in room 383M, Math Building on
"Everything you always wanted to know about NSF and were afraid to
ask"
-------
∂12-Feb-82 1633 Konolige at SRI-AI boolos
Date: 12 Feb 1982 1528-PST
From: Konolige at SRI-AI
Subject: boolos
To: jmc at SAIL
John, although it's not in the draft I gave you, I think I
understand the relationship between SB and the provability logic that
Boolos uses. I'll try writing it up. --kk
-------
∂12-Feb-82 1657 Gene Golub <CSD.GOLUB at SU-SCORE> Salaries for 82-83
Mail-from: ARPANET site SU-SCORE rcvd at 12-Feb-82 1652-PST
Date: 12 Feb 1982 1651-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Salaries for 82-83
To: faculty at SU-SCORE
I have now received notice from the Dean that the average salary
increment for the next academic year is 10 per cent. It is his
intention to reward service rendered and I am in agreement with
this. In general I have very little information on which to base
my recommendation. Therefore I am asking each of you to submit
some evidence to me as to your accomplishments by Friday, Feb 19.
Letters of commendation, papers published and other bits of
puffery will be of interest. Thanks for your help on this.
GENE
-------
∂12-Feb-82 1705 Gene Golub <CSD.GOLUB at SU-SCORE> Salaries for 82-83
Mail-from: ARPANET site SU-SCORE rcvd at 12-Feb-82 1652-PST
Date: 12 Feb 1982 1651-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Salaries for 82-83
To: faculty at SU-SCORE
I have now received notice from the Dean that the average salary
increment for the next academic year is 10 per cent. It is his
intention to reward service rendered and I am in agreement with
this. In general I have very little information on which to base
my recommendation. Therefore I am asking each of you to submit
some evidence to me as to your accomplishments by Friday, Feb 19.
Letters of commendation, papers published and other bits of
puffery will be of interest. Thanks for your help on this.
GENE
-------
∂12-Feb-82 1757 Gene Golub <CSD.GOLUB at SU-SCORE> Al Thaler"s talk
Date: 11 Feb 1982 1031-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Al Thaler"s talk
To: bboard at SU-SCORE, faculty at SU-SCORE
cc: bscott at SU-SCORE
Al Thaler will talk today at 11:00 am in room 383M, Math Building on
"Everything you always wanted to know about NSF and were afraid to
ask"
-------
∂12-Feb-82 2145 ME LOGIN[1,2]
∂12-Feb-82 2001 JMC message files
Is it normal for someone coming from utexas-20 to be logged into
the message file area?
ME - That means they've said "L<cr>" but haven't given a programmer
name (or password) yet. The program running should be LOGIN. Such a
LOGIN runs as 1,2 until it actually has logged the user in, and 2 is
the message file "programmer" name (as in [2,2]).
∂13-Feb-82 1613 RPG
∂11-Feb-82 1704 JMC
I suggest you offer the following as references, suitably combining people
who will give good references and those who will be asked anyway.
David Waltz (Univ of Illinois)!
Terry Winograd (Stanford)?
John McCarthy (Stanford)?
Guy L. Steele Jr (CMU)!
Lowell Wood (LLL)!
Nils J. Nilsson (SRI)!
Gerald Sussman (MIT)!
I think Sussman will give a good recommendation now, better than Nils.
-rpg-
∂14-Feb-82 0517 JPM via SU-SCORE My account on SAIL
CSD has finally caught up with me and has taken away authorization for
my SAIL account as a MS student (MS students get either a SAIL or a
SCORE account - I use to have both under a grandfather clause, but
that is no more). I was wondering if the Lab itself or any group
you know of would authorize my account with a min of storage space
(I use approximately 10 blocks plus whatever is in my mail file)
in return for any sort of work on my part. My past hacking
activities have been primarily in SAIL and PASCAL, so applications
programming seems like a good bet.
Thanks.
Jim
∂14-Feb-82 1045 JK new EKL re-writing system
To: JJW, JMC
Has been installed. It calls the new decision procedure which
can handle equalities and the like better. The standard
re-writing part has also been considerably strengthened.
∂14-Feb-82 1239 REM via SU-TIP MACLISP RANDOM ACCESS USES
Now that I have MacLISP fitted with Read-Alter mode, I have several
things I might do with it:
New SPINDLE program that people have been asking for, would combine
the best of the two existing programs, namely would allow binary
as well as text files but would support data-compression qua text
(binary files wouldn't get compressed well but at least the program
could handle them without crashing).
Continuation on program to convert between XGP and PRESS files which was
suspended when I found MacLISP didn't support read/alter mode.
... (I thought I had a third idea but it slipped my mind) ...
Any preferences or other ideas?
Both would be good, but I supposet the .xgp to .pre is more urgent.
∂14-Feb-82 1352 CLT
steve mentioned that sony, but seemed to think the NAD was better for
if you don't care about fancy gadgets
∂15-Feb-82 0007 REM
∂14-Feb-82 1326 JMC
Both would be good, but I supposet the .xgp to .pre is more urgent.
Oh, I thought in the meantime (while my work was stopped waiting for
random access in MacLISP) somebody else already did xgp/pre conversion??
Not that I know of.
∂15-Feb-82 0845 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Our Previous Conversation
Date: 15 Feb 1982 1123-EST
From: Chandra-at-OhioState <Chandrasekaran at RUTGERS>
Subject: Our Previous Conversation
To: JMC at SU-AI
cc: Chandrasekaran at RUTGERS
You might remember that you had indicated that you'd discuss
with prof. Teller possible support sources for the nuclear
safety and AI proposal. A possibility was his writing
a letter to Mr Palladino of NRC, another was to explore
some DOE sources. Have you had an opportunity to
discuss this matter with him yet?
Thank you for your attention.
-------
I have discussed the prospects once with Dr. Teller, and we have to talk
again. Incidentally, I have been looking at the EPRI Journal and noted
Simulation Methods for Nuclear Power Systems.
WS-81-212, $32.00
"This report constitutes the proceedings of a conference on nuclear power
plant simulation that was sponsored by NRC and EPRI in Tucson, Arizona in
January 1981. Papers from government, industry, natuional laboratories,
and universities are presented; the topics covered include simulation
needs, simulator design and performance, engineering simulations,
model development methods and verification. EPRI Project Manager
P.G. Bailey
Electric Power Research Institute, P. O. Box 10412, Palo Alto,
CA 94303.
It seems to me that your efforts to get funding would meet greater
success if your proposals mentioned such efforts and explained how
your proposals supplemented what was being done. Second, it occurs
to me that EPRI is a possible source of funding.
∂15-Feb-82 1731 CLT
you might want to prepare to talk to you class tomorrow
if the dover and the xgp continue to be non functioning
i won't be able get any notes ready
and i'm not keen on making that sort of presentation
`acappela'
∂15-Feb-82 1915 ME spooling files from E
∂15-Feb-82 1351 JMC
Would it be difficult to modify Xpo so that one option would be to dover
or xgp the file whose name was found?
ME - Sounds like a kludge, and doesn't sound very useful. Do you think
you would really use it very much?
∂16-Feb-82 0848 FFL
To: JMC, FFL
Davis is shown at all the California campuses and then there is an address
in Kensington, California. Is that the correct one to use?
∂16-Feb-82 1050 Jake Brown <CSD.JAKE at SU-SCORE> Special Faculty Luncheon TODAY
Date: 16 Feb 1982 1046-PST
From: Jake Brown <CSD.JAKE at SU-SCORE>
Subject: Special Faculty Luncheon TODAY
To: faculty at SU-SCORE
Stanford-Phone: (415) 497-4879
There will be a special faculty luncheon in the Boy's Town Conference Room
at noon today in honor or Dr. Herbert Simon.
-------
∂16-Feb-82 1115 JMM NRO grades
Got your message late,as I was not around over the weekend. How about sometime
tomorrow-I will try and finish grading my share of the questions of the final
by then.
Jitendra
∂16-Feb-82 1221 SAGALOWICZ at SRI-AI (Daniel Sagalowicz) Tinlunch on Thursday February 18
Date: 16 Feb 1982 1209-PST
From: SAGALOWICZ at SRI-AI (Daniel Sagalowicz)
Subject: Tinlunch on Thursday February 18
To: tlgrp:
cc: simon at CMU-10A, feigenbaum at SUMEX-AIM
Our guest for TINLUNCH this Thursday, February 18, will be
Professor Herb Simon of CMU. We will be discussing his paper
entitled, "Why Should Machines Learn?" Copies are on the top
of Barbara Grosz's file cabinet.
TINLUNCH will be in conference room A, building 1. Professor
Simon will be available around 10:00 for discussions. These
will take place in conference room EK242, interested people may
join us there.
-------
∂16-Feb-82 1224 JMC*
stoyan
∂16-Feb-82 1355 FFL
To: JMC, FFL
Jerry Pournelle called. He would like to speak to you about Project Shackleford.
213 762 2256.
∂16-Feb-82 1418 FFL
To: JMC, FFL
Please call Donald Loveland at Duke Universiy. 919 493 2216. This is his hom
number.
New York, June 7-10, Automatic Deduction Conference, June 8, courant insitute
non-monotonic reasoning or program correctness
title: 2 weeks
March 10 deadline for manuscript
Martin Davis host
∂16-Feb-82 1645 Jake Brown <CSD.JAKE at SU-SCORE> Lost raincoat
Date: 16 Feb 1982 1438-PST
From: Jake Brown <CSD.JAKE at SU-SCORE>
Subject: Lost raincoat
To: faculty at SU-SCORE
Stanford-Phone: (415) 497-4879
A gray raincoat was left in the Boy's Town Conference Room and can be picked
up at the receptionist's desk in Margaret Jacks Hall on the 2nd floor.
Jake
-------
∂16-Feb-82 1759 CLT
i've gone back to the house to practice
∂16-Feb-82 2248 CLT SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
To: "@LOGIC.DIS[1,CLT]" at SU-AI
SPEAKER: Shuzuo Takahashi
TITLE: "Friedman's independence proofs in graph theory
(Kruskal-type theorems)"
TIME: Friday February 19, 4:15-5:30
PLACE: 381-T Math. Dept.
NOTE change of day.
∂16-Feb-82 2358 Jerry E. Pournelle <POURNE at MIT-MC> SHACKLETON
Date: 17 February 1982 02:57-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: SHACKLETON
To: JMC at SU-AI, LLW at SU-AI
Took liberty of discussing Shackleton with James Vaughn
with view to see if money operation is fesible. Jimmy has
arranged for me to be in conference call with him and a friend
of his from First Boston Bank. Friend is in business of raising
sums in ranges of 10↑7 and 10 ↑8 dollars.
Conference happens Friday. Jimmy's friend discrete (in that
business you better be) and will confine discussin to his
opinion of whether sums in 500 x 10 ↑6 are ridiculous or
possible. In initial reaction with Vaughn his opinion was
"difficult but possible; have you ANY government support? Will
Dyson, Teller, other famous names in science support? Will
possibly President Reagan say 'attaboy' for project. Etc."
Also wanted to know profit possibilities if any.
I have my guesses as to answers to those questions, but
should I know more? I will be up tonight for a while, and my
regular phone line is OK, I now have separate phone for login on
net.
Should we discuss this before Friday?
JERRY
∂17-Feb-82 0045 RPG At some point
I should tell you some of the things I'm doing relevant to my case.
-rpg-
Late tomorrow afternoon would be a good time.
∂17-Feb-82 0136 Jerry E. Pournelle <POURNE at MIT-MC> text of advertisement
Date: 17 February 1982 04:35-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: text of advertisement
To: POURNELLE at MIT-MC, JMC at SU-AI, LLW at SU-AI
Following is quoted in "STANDING OVATION OR POLITE APPLAUSE?" a
book on (obviously) public speaking fro St George Press.
Classified advertisement appearing in London papers in 1900:
"MEN WANTED FOR HAZARDOUS Journey. Small wages, bitter cold,
long months of complete darkness, constant danger, safe return
doubtful. Honor and recognition in case of success.
-- Ernest Shackleton"
Shackleton commented later, "It seemed as though all the men of
Great Britain were determined to accompany me, the response was
so overwhelming."
Thanks for the quote.
∂17-Feb-82 0140 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 17 February 1982 04:40-EST
From: Jerry E. *ournelle <POURNE at MIT-MC>
To: JMC at SU-AI
Did you get earlier message regarding conference call ths
friday? Is there anything we need to discuss?
Somehow I missed your previous message. A conference call is ok with
me if it's ok with Lowell. Somehow I think we haven't done enough work
yet, but if your friends won't be put off by a preliminary discussion,
that's fine with me. Have you reached Lowell or should I try.
After noon or evening is fine with me for call.
∂17-Feb-82 0142 ME
∂17-Feb-82 0103 JMC
What is a .spi file?
ME - Got me. I dunno. Where do you see one?
There were three in 1,jmc. I deleted them, but then got curious and
restored two. They have some random characters (images of control chars
I suppose) and REM. Have a look if you like. They're unprotected.
I thought they might have something to do with SPYing.
∂17-Feb-82 0152 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 17 February 1982 04:51-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI, llw at SU-AI
Afraid I left wrong impression. First conference call is this
friday and is just me and Mr Vaughn and his friend. Purpose is
simply to determine whether in opinon of experienced fund raiser
and banker there is slightest feasibility. I asked Jimmy Vaughn
some time ago and he consulted best expert he knew (since he has
banking business with him anyway).
Assuming this guy says yes I'll tell you about it; and
the stage may be set for getting something going. Mr Vaughn and
Count Petrini and some other Houston society types will be at
the L-5 Conference, and it might be possible to begin to put
together a fund-raising committee even there--if that's what is
wanted.
Me, I got in this largely because I wanted to see if
this is all smoke or if there's some substance.
I have no objection to your exploring the matter discreetly as you are
planning to do. Lowell may have more information for you.
∂17-Feb-82 0159 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 17 February 1982 04:58-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: JMC at SU-AI
cc: POURNE at MIT-MC, llw at S1-A
Excellent. In first message I had a few queries; any
information I can get before Friday may help, but I can manage
all right without.
After I get views of the banker person, I'll get back to
you two and you can decide on what to do next.
i get distinct impression that banker was interested,
ESPECILLY if there is a possiblity of porfits--even though
profits are long delayed (and possibly that's not a bad feature,
depending on what kind of inheritance structure you're
contemplating...)
∂17-Feb-82 0423 JPM reply to message
[In reply to message rcvd 17-Feb-82 03:12-PT.]
Sure. Is anytime Thursday afternoon good for you? I am OK from 1:00pm
to 3:30pm. Just name a time.
Jim
∂17-Feb-82 0901 JMC*
Pasternak opus
∂17-Feb-82 1141 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
Date: 17 Feb 1982 1122-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
2/22/82 MJ252 Theo Pavlidis
Monday Special Seminar Bell Laboratories, N.J.
3:15p.m. ``Lines and Curves on a Discrete Grid''
2/22/82 Math 380C Mitch Smooke
Monday Numerical Analysis Sandia Laboratories, Livermore
4:15p.m. Seminar ``Numerical Solution of Pre-mixed Flames by an
Efficient Boundary Value Method''
2/23/82 Jordan 041 Scott Kim
Tuesday Computer Science Stanford University
4:15p.m. Colloquium ``Metafont''
2/24/82 Skilling Auditorium Gary Tjaden
Wednesday Computer Systems Cox Cable Communication, Atlanta, Ga.
4:15p.m. Laboratory To be announced
2/26/82 MJ301 To be announced
Friday Database Research
3:30p.m. Seminar
-------
∂17-Feb-82 1315 MDD via NYU SU account
To: JMC at SU-AI
CC: davism.acf1 at NYU
Thanks very much. I've been having trouble because our local system
uses <CONTROL>Z as an escape character - but I think it's
OK now.
∂17-Feb-82 1320 MDD via NYU ATP Prize
To: DCL at SU-AI
CC: JMC at SU-AI, davism.acf1 at NYU
I'd be very pleased if you would agree to serve on a committee which
I have been asked to chair to select the
first recipient of a new prize for "milestone" achievements in
automatic theorem proving. John McCarthy has agreed to serve, and you
would complete the committeee. My ARPA address is DAVISM.ACF1@NYU
and my phone number is (212)-460-7276. -Martin Davis
∂17-Feb-82 1337 LGC Castaneda's Stuff
I'll appreciate it if you'll bring in those materials that I loaned you
dealing with Hector Castaneda's work, if they aren't already available in
your office in MJH (a book, maybe a couple of papers, and a review by Chisholm).
I'll stop in soon to pick them up. -- Lew
P.S. If you'd like an update on the Advice-Taker work, or would like to see
a small demonstration of the conceptual system being constructed,
let me know.
∂17-Feb-82 1535 FFL
To: JMC, FFL
Barbara Moore's office called to remind you of a Context Advisory Council meeting
on Friday, Feb. 26, 9-11 a.m., Polya 152. Would appreciate a confirmation of
your attendance, 7-4378. Would like another representative there if you cannot
attend. 7-4378.
et cal[1,jmc
∂17-Feb-82 2040 ullman@Shasta (SuNet) meeting
Date: 17 Feb 1982 13:46:15-PST
From: ullman at Shasta
To: equip
From: ullman
Subject: meeting
Date: 17 February 1982 13:45:57-PST (Wednesday)
I would like to have a meeting Friday morning 10AM in 252.
The primary topic for discussion is an offer from HP to give us
a network of 5 workstations. If you have any proposed use, either
research or educational, or know anyone who does, be prepared to
tell us at the meeting.
∂17-Feb-82 2051 Paula Bauman <PJB at S1-A> Commute Mileage Reimbursement
Date: 17 Feb 1982 0902-PST
From: Paula Bauman <PJB at S1-A>
Subject: Commute Mileage Reimbursement
To: jmc at SU-AI
Thank you for the information on which route you took. I have submitted
a claim for you for the 12 working days worked, starting with the date
9/21. I tried unsuccessfully to get the exact dates you were paid for
from several sources here at the lab...so, I will keep my fingers crossed
that this claim will be processed for you quickly and with no problems.
Your reimbursement should be mailed to your Lathrop Drive address.-Paula
∂17-Feb-82 2116 Nancy Dorio <CSD.DORIO at SU-SCORE> ADDENDUM TO COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
Date: 17 Feb 1982 1624-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: ADDENDUM TO COLLOQUIUM NOTICE WEEK OF FEBRUARY 22 - 26, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Medical Computing Journal Club - Tuesday, February 23, 1982 at 1:30 p.m.
in MJ301. Ms. Shoko Tsuji from Stanford University, will speak on "Recent
Articles of Interest".
Knowledge Representation Group Meeting - Tuesday, February 23, 1982 at
2:30 p.m. in MJ301. Mr. Paul Cohen from Stanford University, will speak
on "Decision Making for Portfolio Management".
Database Research Seminar - Friday, February 26, 1982 at 3:30 p.m. in MJ301.
Francisco Corella from Stanford University will speak on "The Entity Model".
-------
∂18-Feb-82 0005 Jerry E. Pournelle <POURNE at MIT-MC> film in space
Date: 18 February 1982 02:20-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: film in space
To: JMC at SU-AI
cc: POURNE at MIT-MC
yes.. the lady whose big place you went to (hard to find) is
planning smething of the sort and has financing if she can line
up technical suppot. I hope she'll be at our conference.
∂18-Feb-82 0208 Lowell Wood <LLW at S1-A> Shackleton
Date: 18 Feb 1982 0204-PST
From: Lowell Wood <LLW at S1-A>
Subject: Shackleton
To: pourne at MIT-MC
CC: LLW at S1-A, jmc at SU-AI, RAH at S1-A
∂17-Feb-82 0153 John McCarthy <JMC at SU-AI>
Date: 17 Feb 1982 0155-PST
From: John McCarthy <JMC at SU-AI>
To: pourne at MIT-MC
CC: llw at S1-A
I have no objection to your exploring the matter discreetly as you are
planning to do. Lowell may have more information for you.
[Jerry: Project costs are driven completely by Shuttle costs, which, as
you probably read recently, are currently scheduled to soar 2-3-fold
post-1984, when the STS is going onto a full-cost-recovery accounting
system(!!!). Though a beachhead could apparently be taken and held nearly
indefinitely with a single Shuttle launch, it would probably be of such a
small size as to have mostly symbolic significance. Significant on-site
capability would probably require 3-5 STS lifts, and would still result in
negligible prospects for payback commencement for at least the first
decade, at least from activity of classic mine-this, process-that types.
The only agreed-upon high-reliability ab initio revenue source is from
media feeds, whose magnitude is variously estimated between $10 and 50 M
annually, depending mostly on posited aggressiveness and creativity of
marketing. Present ultra-low-budget planning centers on swapping STS
services for technical data return, using volunteer labor after-hours at
large Gov't. labs and similar outfits for subsystems design, fab and
integration, and Shackleton route for mission staffing. Competitive plan
alternatives are certainly welcomed. Discretion on everyone's part will
be keenly appreciated. Lowell]
∂18-Feb-82 0419 Bill Gosper <RWG at MIT-MC> Stop me if you've heard this one before...
Date: 18 February 1982 07:18-EST
From: Bill Gosper <RWG at MIT-MC>
Subject: Stop me if you've heard this one before...
To: PMF at MIT-MC, MLB at MIT-MC, AMO at MIT-MC, NJAS at MIT-MC,
WGD at MIT-MC, REM at MIT-MC, RZ at MIT-MC, guibas at PARC-MAXC,
ramshaw at PARC-MAXC
cc: ES at MIT-MC, RWG at MIT-MC, dek at SU-AI, jmc at SU-AI
Denote the nth Bernoulli polynomial (BERNPOLY(X,N)) by B[n](x). Then
2*sum(binom(j+2,6*k+2)/(2*k+1) * B[j-6*k](x), k>=0) =
x↑(j+2) + (x-1)↑(j+2) - (x+w)↑(j+2) - (x+w↑2)↑(j+2),
where w↑2+w+1 = 0. (So you only have to compute 1/6 of the "full history"
to get then jth one.)
∂18-Feb-82 0923 Ichiki at SRI-AI TINLUNCH Meeting in Building A Today
Date: 18 Feb 1982 0922-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH Meeting in Building A Today
To: tlgrp:
cc: AIC-Associates:
The TINLUNCH meeting will be today at 12:00 in Building A,
Conference Room A, with Herb Simon.
-------
∂18-Feb-82 1046 RPG@Sail (SuNet) meeting
Date: 18 Feb 1982 1045-PST
From: Dick Gabriel <RPG at SU-AI>
Subject: meeting
To: equip at SHASTA
∂17-Feb-82 2040 ullman@Shasta (SuNet) meeting
Date: 17 Feb 1982 13:46:15-PST
From: ullman at Shasta
To: equip
From: ullman
Subject: meeting
Date: 17 February 1982 13:45:57-PST (Wednesday)
I would like to have a meeting Friday morning 10AM in 252.
The primary topic for discussion is an offer from HP to give us
a network of 5 workstations. If you have any proposed use, either
research or educational, or know anyone who does, be prepared to
tell us at the meeting.
What is each workstation supposed to be? If we have a say in the matter
I suggest not accepting whatever they want to give us, but, instead, we should
hold out for 5 of the Scheme machines that Sussman et al will build
at HP this summer out of 68000 technology. I suggest this an educational use.
∂18-Feb-82 2012 Ingalls at PARC-MAXC Programming exercise
Date: 18 Feb 1982 12:19 PST
From: Ingalls at PARC-MAXC
Subject: Programming exercise
To: Warren@SRI-AI, Pratt@SU-AI, McCarthy@SU-AI
Thought you'd be interested - I wanted to distill things down to a simpler
example than a text editor, and yet keep the stuff about scope, persisting objects,
and user interaction. Forward at will, if you think it's of interest.
---------------------------
Date: 18 Feb. 1982 12:03 pm PST (Thursday)
From: Ingalls.PA
Subject: Programming exercise
To: SmalltalkInterest↑, ComputerResearch↑
Reply-To: Ingalls
I thought others might be interested in this little exercise. I recently had the
experience of not being able to debug a Smalltalk program because I could not
tell how a variable had been set to an incorrect value. This got me back to
daydreaming about real systems vs applicative languages. I tried to think of a
very simple situation which would somehow capture the "real system"-ness of
what we do in Smalltalk, and yet which would be simple enough for me to work
on, at the same time as invent a new system (or at least experiment with an
unfamiliar system such as Prolog). Well, I came up with an example which I
think meets the constraints, and I wanted to share it with you because I think it
will set some of you to thinking. . .
Exercise 1: Write a program which does the following:
If the mouse button is pressed, then create a rectangle whose topLeft
is the mouse location and whose bottomRight is the mouse location
where the button is released.
If the mouse location lies within any of the rectangles so created,
then they should appear highlighted on the screen.
Exercise 2: Amend the above solution so that only one rectangle will be
highlighted at a time.
Exercise 3: Embed the above system recursively so that rectangles can be created
within rectangles.
Naurally, a "good" solution should read very much like the above english
description (and probably shouldn't be much longer). It should not reveal "how"
things are done (control) except where it is a part of the specification. You may
invent any syntax you wish but, of course, you must be prepared to explain it.
You may invent any other mechanism, for that matter, as long as you can
explain it. In fact, a complete solution should cover the details of such
underlying mechanisms as control, scope, storage management, inheritance and
other contraints.
It ought to be simple, right?
I'll collect the responses in about a week and see where things go from there.
------------------------------------------------------------
∂18-Feb-82 2013 reid@Shasta (SuNet) REM
Date: Thursday, 18 Feb 1982 12:30-PST
To: JMC at Sail, Gosper at Parc
Subject: REM
Reply-to: Reid at SU-AI
From: Brian Reid <reid at Shasta>
John, Bill,
I got a phone call today from several female friends at
Carnegie-Mellon, who were extremely upset at the mail that REM has sent
them recently, to the point of wanting to make a formal complaint to
DCA about misuse of the Arpanet and sexual harassment.
I haven't seen the message, but I assume it's just REM up to his usual
tricks. In my memory you two are people he trusts. Is there some way he
can be persuaded to stop this before DCA gets involved? I think that
the CMU women may have already contacted Moses to ask whether his MC
account can be taken away.
Brian Reid
∂18-Feb-82 2014 DWALTZ at BBND AAAI Program Committee
Date: 18 Feb 1982 1723-EST
Sender: DWALTZ at BBND
Subject: AAAI Program Committee
From: DWALTZ at BBND
To: AAAI-OFFICE at SUMEX-AIM, Minsky at MIT-AI, PHW at MIT-AI, TENENBAUM at SRI-KL,
To: Rick at RAND-AI, Reddy at CMU-10B, Newell at CMU-10A, Bobrow at PARC-MAXC,
To: Buchanan at SUMEX-AIM, Englemore at SUMEX-AIM, Feigenbaum at SUMEX-AIM,
To: Bledsoe at UTEXAS-20, Grosz at SRI-AI, Nilsson at SRI-AI, Walker at SRI-AI,
To: Rich at MIT-AI, Feldman at SUMEX-AIM, Balzer at USC-ISIF, McCarthy at SU-AI,
To: Woods at BBND, Webber at BBND, Simon at CMUA
Cc: DWaltz at BBND
Message-ID: <[BBND]18-Feb-82 17:23:35.DWALTZ>
I have now either heard from all members of the old program committee,
or waited long enough to be certain that I will never hear the rest.
The following people have agreed to serve again: Bruce Buchanan, Jon
Bentley, Doug Lenat, Carl Hewitt, Ron Brachman, Saul Amarel,
Bob Balzer, Chuck Rieger, Ira Goldstein, Roger Schank, and Woody
Bledsoe (reluctantly).
In order to be able to properly review papers in all conference
categories, we clearly needed to have people in vision, robotics,
and game playing added to the committee. In addition, Woody Bledsoe
is the only person who reported a willingness to review theorem proving
papers. Therefore, I propose that the following people be added to the
program committee: vision & robotics - Bob Bolles (SRI), Ruszena
Bajcsy (Penn), Dana Ballard (Rochester); theorem proving -
Ken Bowen (Syracuse); game playing - Hans Berliner (CMU).
These people have not been contacted yet, and I ask that you reply to me
as soon as possible if you have any objections to the program
committee staffing. If I have not heard anything by Monday February
22, I will go ahead and notify the prospective committee members.
Thanks,
Dave
∂18-Feb-82 2155 Nilsson at SRI-AI Re: AAAI Program Committee
Date: 18 Feb 1982 1600-PST
From: Nilsson at SRI-AI
Subject: Re: AAAI Program Committee
To: DWALTZ at BBND, AAAI-OFFICE at SUMEX-AIM,
To: Minsky at MIT-AI, PHW at MIT-AI, TENENBAUM at SRI-KL,
To: Rick at RAND-AI, Reddy at CMU-10B, Newell at CMU-10A,
To: Bobrow at PARC-MAXC, Buchanan at SUMEX-AIM,
To: Englemore at SUMEX-AIM, Feigenbaum at SUMEX-AIM,
To: Bledsoe at UTEXAS-20, Grosz at SRI-AI, Nilsson at SRI-AI,
To: Walker at SRI-AI, Rich at MIT-AI, Feldman at SUMEX-AIM,
To: Balzer at USC-ISIF, McCarthy at SU-AI, Woods at BBND,
To: Webber at BBND, Simon at CMUA
cc: NILSSON
Dave, I agree wholeheartedly with Don Walker's reply to you and would
strongly recommend beefing up the language area with some people
whose work covers a broader spectrum of the field. Among these I
might humbly suggest: Barbara Grosz, Bonnie Webber, Bill Woods, Ray
Perrault, Jane Robinson, Ron Kaplan, and Martin Kay (I could name
others also if you need more). Among those left over from last time
(and including the new panel members you mentioned--of whom I am in
favor for their specialities), the area of natural language
processing-- and representation in general--is, to say the least,
idiosyncratic. Good luck! --Nils
-------
∂19-Feb-82 0141 Jerry E. Pournelle <POURNE at MIT-MC> Shackleton
Date: 19 February 1982 04:40-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Shackleton
To: LLW at S1-A
cc: POURNE at MIT-MC, jmc at SU-AI, RAH at S1-A
Conference call put off until Tuesday. Discretion assured. Mr.
Vaughn's interest at least gives a chance that the kinds of
money spoken of could be found.
Having lunch Monday with Gen. Henry.
Date: 18 Feb 1982 0204-PST
From: Lowell Wood <LLW at S1-A>
∂17-Feb-82 0153 John McCarthy <JMC at SU-AI>
Date: 17 Feb 1982 0155-PST
From: John McCarthy <JMC at SU-AI>
To: pourne at MIT-MC
CC: llw at S1-A
I have no objection to your exploring the matter discreetly as you are
planning to do. Lowell may have more information for you.
[Jerry: Project costs are driven completely by Shuttle costs, which, as
you probably read recently, are currently scheduled to soar 2-3-fold
post-1984, when the STS is going onto a full-cost-recovery accounting
system(!!!). Though a beachhead could apparently be taken and held nearly
indefinitely with a single Shuttle launch, it would probably be of such a
small size as to have mostly symbolic significance. Significant on-site
capability would probably require 3-5 STS lifts, and would still result in
negligible prospects for payback commencement for at least the first
decade, at least from activity of classic mine-this, process-that types.
The only agreed-upon high-reliability ab initio revenue source is from
media feeds, whose magnitude is variously estimated between $10 and 50 M
annually, depending mostly on posited aggressiveness and creativity of
marketing. Present ultra-low-budget planning centers on swapping STS
services for technical data return, using volunteer labor after-hours at
large Gov't. labs and similar outfits for subsystems design, fab and
integration, and Shackleton route for mission staffing. Competitive plan
alternatives are certainly welcomed. Discretion on everyone's part will
be keenly appreciated. Lowell]
∂19-Feb-82 0830 Chandra-at-OhioState <Chandrasekaran at RUTGERS> Re: more on simulation
Date: 19 Feb 1982 1109-EST
From: Chandra-at-OhioState <Chandrasekaran at RUTGERS>
Subject: Re: more on simulation
To: JMC at SU-AI
cc: Chandrasekaran at RUTGERS
In-Reply-To: Your message of 19-Feb-82 0209-EST
(1) Thanks for the reference to the report concerning
simulation.
(2) You have convinced me that both tactically as well as from
a theretical viewpoint the interaction between reasoning
and simulation is important. The interesting technical question
is how exactly will the driving of the simulation by the
reasoning program be accomplished. I still think, however,
that the architecture of the reasoning process needs to
be pinned down, before questions of how hook them onto
simulations can be reasonably handled. On the tactical side,
however, you are absolutely right. We have to show a clear
understanding of both the past work in this kind of simulation
as well of how the reasoner and the simulation package will
interact. I am beginning to think about it and discuss it
with nuclear engineering people here at Ohio State.
Thanks for your suggestion.
-------
∂19-Feb-82 0857 WEBBER at BBND Re: AAAI Program Committee
Date: 19 Feb 1982 1115-EST
Sender: WEBBER at BBND
Subject: Re: AAAI Program Committee
From: WEBBER at BBND
To: DWALTZ at BBND
Cc: aaai-office at SUMEX-AIM, minsky at MIT-AI, phw at MIT-AI,
Cc: tenenbaum at SRI-KL, rick at RAND-AI, reddy at CMUB,
Cc: newell at CMUA, bobrow at PARC-MAXC, buchanaen at SUMEX-AIM,
Cc: buchanen at SUMEX-AIM, englemore at SUMEX-AIM,
Cc: feigenbaum at SUMEX-AIM, bledsoe at UTEXAS-20,
Cc: grosz at SRI-AI, nilsson at SRI-AI, walker at SRI-AI,
Cc: rich at MIT-AI, feldman at SUMEX-AIM, balzer at ISIF,
Cc: mccarthy at SU-AI, woods at BBND, simon at CMUA
Message-ID: <[BBND]19-Feb-82 11:15:12.WEBBER>
In-Reply-To: <[BBND]18-Feb-82 17:23:35.DWALTZ>
Dave - I agree with Don's assessments and Nils' suggestions.
Bonnie
∂19-Feb-82 1548 WOODS at BBND Re: AAAI Program Committee
Date: 19 Feb 1982 1312-EST
Sender: WOODS at BBND
Subject: Re: AAAI Program Committee
From: WOODS at BBND
To: DWALTZ at BBND
Cc: AAAI-OFFICE at SUMEX-AIM, Minsky at MIT-AI, PHW at MIT-AI,
Cc: TENENBAUM at SRI-KL, Rick at RAND-AI, Reddy at CMU-10B,
Cc: Newell at CMU-10A, Bobrow at PARC-MAXC, Buchanan at SUMEX-AIM,
Cc: Englemore at SUMEX-AIM, Feigenbaum at SUMEX-AIM,
Cc: Bledsoe at UTEXAS-20, Grosz at SRI-AI, Nilsson at SRI-AI,
Cc: Walker at SRI-AI, Rich at MIT-AI, Feldman at SUMEX-AIM,
Cc: Balzer at USC-ISIF, McCarthy at SU-AI, Webber at BBND,
Cc: Simon at CMUA, Woods at BBND
Message-ID: <[BBND]19-Feb-82 13:12:48.WOODS>
In-Reply-To: <[BBND]18-Feb-82 17:23:35.DWALTZ>
Dave,
I think your proposed additions are fine, and I agree with Nils'
suggestions also. (I don't seem to have gotten the reply by
Walker that Nils refers to.)
-- Bill
∂19-Feb-82 1652 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Commencement
Date: 19 Feb 1982 1204-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Commencement
To: faculty at SU-SCORE
Commencement will take place on Sunday, June 13, 1982. We need one member
of the faculty to serve as a Marshal. I'll be out of town then. Would anyone
like to volunteer? It seems like a pleasant activity, lasting about 2 1/2
hours and it would be nice to see the fruits of our labor.
GENE
-------
∂20-Feb-82 2152 CLT
∂20-Feb-82 2148 JMC iii trips Monday and Wednesday
To: CLT, FFL
I go to L.A. Monday (Feb 22) morning returning monday afternoon
and likewise Wednesday morning returning wednesday afternoon.
Why not just stay, since you don't have to teach Tuesday?
∂20-Feb-82 2207 CLT
I have just succeeded in making the theorem prover go into and infintie
rewrite loop. Saves me from have to prove it total!
See below
DEFN(LEFT (X) (IF (LISTP X) (LEFT (CAR X)) X)))
The lemma CAR.LESSP establishes that (COUNT X) goes down according
to the well-founded function LESSP in each recursive call. Hence, LEFT
is accepted under the principle of definition.
[922 cns / 2.1 s + 0.0 gc + .2 io (= 13 @ 2)]
LEFT
16:
PROVE.LEMMA(LEFT.HACK (REWRITE) (EQUAL (LEFT X) (IF(LISTP X)(LEFT (CAR X))X)))
This conjecture simplifies, expanding the definition of LEFT, to:
T.
Q.E.D.
[647 cns / .9 s + 0.0 gc + .1 io (= 3 @ 1)]
LEFT.HACK
17:
PROVE((EQUAL (LEFT X) X))
***ERROR***
STACK OVERFLOW
NIL
(REWRITE.FNCALL broken)
18:
Oho! Well, most programs aren't tested sufficiently with input where
they aren't expected to succeed. I suppose there may be similar
weaknesses in EKL and FOL. For example, FOL was working for a year
or two before I made it do Russell's paradox.
∂21-Feb-82 0647 Andrew M. Odlyzko <AMO at MIT-MC>
Date: 21 February 1982 09:46-EST
From: Andrew M. Odlyzko <AMO at MIT-MC>
To: RWG at MIT-MC
cc: AMO at MIT-MC, PMF at MIT-MC, MLB at MIT-MC, NJAS at MIT-MC,
WGD at MIT-MC, REM at MIT-MC, RZ at MIT-MC, GUIBAS at MIT-MC,
RAMSHAW at MIT-MC, ES at MIT-MC, dek at SU-AI, jmc at SU-AI
At least some of your Bernoulli polynomial identities can be derived
quite simply form standard results. For example, consider your
sum(binom(n,2*k)*bernpoly(x,n-2*k)/(4↑k*(2*k+1)),k,0,inf)
= bernpoly(x,1)↑n = (x-1/2)↑n.
References will be to teh National Bureau of Standards, Handbook of
Mathematical Functions, Abramowitz & Stegun, eds., Chapter 23. By 23.1.11,
integrate(bernpoly(x+t,n),t,h-1,h) = (bernpoly(x+h,n+1)-bernpoly(x+h-1,n+1))/
(n+1),
which by 23.1.6 equals
(x+h-1)↑n.
On the other hand, by 23.1.7,
bernpoly(x+t,n) = sum(binomial(n,k)*bernpoly(x,k)*h↑(n-k),k, 0, inf).
If we now integrate this last identity with respect to t from h-1 to h,
and set h = 1/2, we obtain you result.
Perhaps something similar could give your 1/6 result? It will
probably have to be somewhat more sophisticated, I think.
∂21-Feb-82 1111 CLT Kriesels Godel obituary
Please return
∂21-Feb-82 1235 Aaai-Office at SUMEX-AIM AAAI-82 conference ...
Date: 21 Feb 1982 1229-PST
From: Aaai-Office at SUMEX-AIM
Subject: AAAI-82 conference ...
To: AAAI-Execs;Minsky at MIT-AI, Feigenbaum at SUMEX-AIM, NILSSON at SRI-AI,
Reddy at CMU-10A, GHS at MIT-AI, PHW at MIT-AI, Scott.Fahlman at CMU-10A,
Mark.Fox at CMU-10A, Sharon.Burks at CMU-10A, Pople at SUMEX-AIM,
Rich at MIT-MC, Webber at BBND, DWaltz at BBND, Tenenbaum at SRI-KL,
Newell at CMU-10A, Walker at SRI-AI, Bobrow at PARC-MAXC,
Buchanan at SUMEX-AIM, Engelmore at SUMEX-AIM, AAAI-OFFICE at SUMEX-AIM,
Bledsoe at UTEXAS-20, Grosz at SRI-AI, McCarthy at SU-AI, Woods at BBND,
Simon at CMU-10A
Marvin,
Thought it would be a good idea to give you an update on what's going on.
CONFERENCE
Things are progressing well for the August 16-20 conference at CMU and
the University of Pittsburgh.
Conference Program: You've been kept abreast of Program Committee developments
by David Waltz. David is now making arrangements for the Program Committee
meeting in April. He also has some good ideas regarding panels and other
non-paper sessions. My feeling is that David would probably welcome
suggestions from you and the rest of the Executive Committee in this area.
David and I will be coordinating the process of turning accepted papers into
conference proceedings. The process of linking conference speakers, topics,
dates, times and lecture halls will be undertaken with the help of Providence
and Scott Fahlman, local arrangements chairman.
Local Arrangements: Because of the scarcity of large auditoriums at CMU,
the majority of conference program activities will take place at the
University of Pittsburgh in Lawrence Hall, Benedum Auditorium and four
University of Pittsburgh Law School classrooms near by. The hope is to
place a good percentage of the socializing at CMU. (A hospitality aside:
The conference banquet -- planned for Thursday, August 19 -- will take
place on board either two or three river boats paddling up the Allegheny,
Ohio and Monongahela rivers. I'm not sure if two boats can do three rivers
or if you need three boats, one for each river. I'm sure Scott Fahlman will
disabuse me of this incredible confusion.) All of the local arrangements
principals -- Scott Fahlman, Harry Pople, Mark Fox, Sharon Burks, et. al. --
are doing an excellent, effective job. It's a pleasure working with them.
Conference Publicity: I've sent out a press release on the conference and
the response to it has been good. We've gotten numerous calls from the
media and the interest there is being piqued. (We're already getting calls
and letters from individuals requesting conference registration forms.) I
expect to send out another release soon; this one coordinated with the CMU
Public Affairs people. We've also completed the poster/flyer on the
conference, and a mass mailing (7,000 pieces) went out last week, mainly
to AAAI and SIGART members. I will be placing some conference ads in
certain publications. You're probably aware of the outrageously high cost
of magazine ads nowadays.
Tutorials: Chuck Rich has done an outstanding job of organizing this year's
tutorial program. I assume you've seen his excellent list of speakers:
Pat Winston, Scott Fahlman, Guy Steele, Tomas Lozano-Perez, Steven Zucker,
Bonnie Lynn Webber, Timothy Finin, Victor Zue, Randy Davis, Charles Rich.
Until signed contracts are returned, these names should not yet be cast
in concrete. The program, in my mind at least, is very strong. As you
know, the tutorial program is planned for August 16 and 17, Monday and
Tuesday.
Odds: A conference registration program has been written and is ready.
The conference brochure -- which will contain registration forms --
is now underway. As far as I'm concerned, the conference plans are
shaping up quite nicely and on schedule.
AI MAGAZINE
The AI Magazine went to the printers on Friday. It should be in the mail
in approximately three weeks.
TIOGA PRIZE
To encourage elegant exposition, Nils Nilsson is proposing a prize of $1,000
for the best paper presented at the AAAI Conference in August. Tioga
Publishing Company will donate the prize money to be awarded by AAAI. An
article on this prize will probably appear in the next issue of the AI
Magazine.
Hmmm, this message is beginning to accrue a sizeable tail so I believe
I'll clip it off right about here if I may.
Thanks, Marvin. Any word on the two letters that have been of interest
to you: the one to the Reagan Administration and the Corporate Beg letter?
I'll help you with these in any way I can. Just do let me know.
Cordially,
Lou Robinson
-------
∂21-Feb-82 1557 RPG ∂21-Feb-82 1409 JMC
If you join the faculty, you'll have to change your PLAN.
You aren't the first to point this out.
-rpg-
∂21-Feb-82 2357 MOON at SCRC-TENEX Fahlman's new new sequence proposal, and an issue of policy
Date: Monday, 22 February 1982 02:50-EST
From: MOON at SCRC-TENEX
To: common-lisp at sail
Subject:Fahlman's new new sequence proposal, and an issue of policy
CMU-20C:<FAHLMAN>NNSEQ.DOC seems to be a reasonable proposal; let's accept
it and move on to something else. A couple nits to pick:
I don't understand the type restrictions for CONCAT. Is (vector fixnum) a
subtype of (vector t)? Is (vector (mod 256.)) a subtype of (vector t)?
Presumably all 3 of these types require different open-coded access
operations on VAXes, so if CONCAT allows them to be concatenated without
explicit coercions then the type restriction is inutile. I would suggest
flushing the type restrictions but retaining the output-type specifier.
After all, the overhead is only a type dispatch for each argument; the
inner loops can be open-coded on machines where that is useful. The
alternative seems to be to have implementation-dependent type restrictions,
something we seem to have decided to avoid totally.
mumble-IF-NOT is equally as useful as mumble-IF, if you look at how they
are used. This is because the predicate argument is rarely a lambda, but
is typically some pre-defined function, and most predicates do not come in
complementary versions. (Myself, I invariably write such things with
LOOP, so I don't have a personal axe to grind.)
REMOVE should take :start/:end (perhaps the omission of these is just a
typo).
A possible other thing to move on to: It's pretty clear that the more
advanced things like the error system, LOOP, the package system, and
possibly the file system aren't going to be reasonable to standardize on
for some time (say, until the summer). As far as packages go, let's say
that there are keywords whose names start with a colon and leave the rest
for later; keywords are the only part of packages that is really pervasive
in the language. As far as errors go, let's adopt the names of the
error-reporting functions in the new Lisp machine error system and leave
the details of error-handling for a later time. I'd like to move down to
some lower-level things. Also I'm getting extremely tired of the large
ratio of hot air to visible results. There are two things that are
important to realize: We don't need to define a complete and comprehensive
Lisp system as the first iteration of Common Lisp for it to be useful. If
the Common Lisp effort doesn't show some fruit soon people are going to
start dropping out.
We should finish defining the real basics like the function-calling
mechanism, evaluation, types, and the declaration mechanism. Then we ought
to work on defining a kernel subset of the language in terms of which the
rest can be written (not necessarily efficiently); the Common Lisp
implementation in terms of itself may not actually be used directly and
completely by any implementation, but will provide a valuable form of
executable documentation as well as an important aid to bringing up of new
implementations. Then some people should be delegated to write such code.
Doing this will also force out any fuzzy thinking in the basic low-level
stuff.
This is, in fact, exactly the way the Lisp machine system is structured.
The only problem is that it wasn't done formally and could certainly
benefit from rethinking now that we have 7 years of experience in building
Lisp systems this way behind us. From what I know of VAX NIL, Spice Lisp,
and S-1 NIL, they are all structured this way also.
Note also that this kernel must include not only things that are in the
language, but some basic tools which ought not to have to be continuously
reinvented; for example the putative declaration system we are assuming
will exist and solve some of our problems, macro-writing tools, a
code-walking tool (which the new syntax for LOOP, for one, will implicitly
assume exists).
∂22-Feb-82 0831 Paul Armer <CSD.ARMER at SU-SCORE> IBM Fellowship
Date: 22 Feb 1982 0830-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: IBM Fellowship
To: CSD-Faculty: ;
Stanford-Phone: (415) 497-2274
Memo to: CSD Faculty
Subject: IBM Fellowships
From: Paul Armer
We have received a letter from IBM announcing that they plan to award
thirty (up from ten in previous years) Fellowships in Computer Science
for the academic year 1982-83. In addition they will award an
unspecified number of Fellowships to female and minority graduate
students. The Fellowships carry a nine month stipend of $9,000 plus
tuition and fees plus $2000 for the department. One of the Fellowships
is assigned to Stanford (meaning they will pick at least one of our
nominees) while there is open competition for the ones not otherwise
assigned. We have received as many as two Fellowships in the past when
there were only ten so we should try for more this year not counting
any female/minority applications we submit.
Would you please give me your nominations by February 26? Gene tells
me that it has been department policy to nominate individuals who are
in the process of finishing a thesis. At the very least, a thesis
proposal is required. They request two letters of recommendation with
one of them to come from the candidate's thesis advisor. Potential
for outstanding research is the most important criterion used in
selection. Preference is given to U.S. citizens or permanent
residents but there is no restriction because of nationality. I
have additional information on submission details.
Lee Hoevel of IBM, who headed the selection committee last year,
tells me that Stanford's submissions could be stronger than they
have been. He says that our letters of recommendation would be
better if we viewed the student nationally rather than in
comparison to other Stanford students.
We have been told that we'll receive a similar offer from General
Electric covering one Fellowship for a Stanford student. However
we are still awaiting formal notification and details.
Paul
-------
∂22-Feb-82 0928 Bob Moore <BMOORE at SRI-AI> Re: disjunctive knowledge
Date: 22 Feb 1982 0927-PST
From: Bob Moore <BMOORE at SRI-AI>
Subject: Re: disjunctive knowledge
To: JMC at SU-AI
cc: BMOORE at SRI-AI
In-Reply-To: Your message of 20-Feb-82 1432-PST
John,
Yes, as far as I know, all the criticisms I made of the PLANNER-type languages
apply equally to the work of Schank and Abelson and their students. If you
view their formalisms from a logical perspective (which is difficult given
the way they describe them), all they allow is the simplest form of modus
ponens: A, A->B |- B where A is a ground literal, and A->B may contain some
free variables.
Some of the other formalisms used in AI are more powerful than this, but to
my knowledge, none solve the disjunction problem except by applying theorem
proving techniques to first-order representations.
--Bob
-------
∂22-Feb-82 0947 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Participation in Educational Technology Study Group
Date: 22 Feb 1982 0945-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Participation in Educational Technology Study Group
To: faculty at SU-SCORE
cc: su-bboard at SU-SCORE,
ADMIN: ;
Decker Walker of the School of Education has written me asking for
nominations fro participants in a study group in educational technology,
part of the Stanford Study of the Schools. Please let me know if you
are interested in involvement with the project. There are different
levels of participation and the total effort will take place between
now and June 1982. The committees are open to faculty, administrators,
and students. A proposal for support and the letter of invitation is in
my office.
GENE GOLUB
-------
∂22-Feb-82 1002 TOB dept equipment buy
To: JMC at SU-AI, csd.golub at SU-SCORE
Where does the department proposal stand for equipment?
∂22-Feb-82 1105 PLATEK at USC-ECLB Visiting Stanford
Date: 22 Feb 1982 1059-PST
From: PLATEK at USC-ECLB
Subject: Visiting Stanford
To: JMC at SU-AI
cc: platek
Prof. McCarthy,
I mentioned to you that the next time I would be in the Stanford area
I would like to see some of the work in AI, automatic theorem proving,
etc. which is of logical interest. I will be there March 5 and
I wondered if you could help me arrange some demos and direct me to
the appropriate people. Besides a general interest I am particularly
curious about anything that can be used for verification.
Richard Platek
-------
∂22-Feb-82 1143 Irmgild Schack <CSD.IRMGILD at SU-SCORE> General Advisors
Date: 22 Feb 1982 1139-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: General Advisors
To: faculty at SU-SCORE
Humanities and Sciences has requested that we provide two general advisors for
freshmen, sophomores and others. I think it is important that we maintain
an active role in undergraduate education even though we do not give a
degree. If you would like to volunteer, please let me know.
GENE
-------
∂22-Feb-82 1206 Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE> equip
Date: 22 Feb 1982 1202-PST
From: Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE>
Subject: equip
To: tob at SU-AI, jmc at SU-AI, csd.golub at SU-SCORE
My understanding, via EAF and Bob Engelmore, is that the proposals
from all the sites are on hold for political reasons.
-------
∂22-Feb-82 1408 HHB chair
To: "@LIST.DIS[1,HHB]" at SU-AI
LIST[1,HHB] contains the list of contributors/contributions
to Tom's chair gift. You will notice that intakes slightly
exceeded outputs ... I will try to devise some way for us
all to enjoy from the surplus.
Harlyn
∂22-Feb-82 1657 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting reminder
Date: 22 Feb 1982 1653-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Meeting reminder
To: Faculty at SU-SCORE
cc: csd.armer at SU-SCORE
This is a reminder that there will be a Gray Tuesday meeting in 252 tomorrow.
The Comp Committee will present a proposed revised reading list. Hope you can
come.
Gene
-------
∂22-Feb-82 1659 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting reminder
Date: 22 Feb 1982 1657-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Meeting reminder
To: Faculty at SU-SCORE
cc: csd.armer at SU-SCORE
This is a reminder that there will be a Gray Tuesday meeting in 252 tomorrow.
The Comp Committee will present a proposed revised reading list. Hope you can
come.
Gene
-------
∂22-Feb-82 1722 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Gene Golub's Birthday (leap-year)
Date: 22 Feb 1982 1718-PST
From: Carolyn Tajnai <CSD.TAJNAI at SU-SCORE>
Subject: Gene Golub's Birthday (leap-year)
To: Su-BBoards at SU-SCORE,
CSD-Faculty: ;
cc: CSD.Berlin at SU-SCORE, JEB at SU-AI
Gene Golub's 50th birthday falls on February 29, 1982 (????)
so Irmgild Schack, Danny Berlin, Jim Boyce, and I invite you to
join us in honoring Gene at the TGIF Friday, Feb. 26, 4:30 p.m.,
third floor lounge of Jacks.
-------
∂23-Feb-82 1006 Ichiki at SRI-AI TINLUNCH Meeting 2/25
Date: 23 Feb 1982 0928-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH Meeting 2/25
To: tlgrp:
This week's TINLUNCH meeting will be at 12:00 in Building A,
Conference Room A. The paper to be discussed is:
Focalizers, the Scoping Problem, and Semantic Interpretation
Rules in Logic Grammars
By: Michael C. McCord
Extra copies of the above paper are on Barbara Grosz's file cabinet.
Stan Rosenschein.
-------
∂23-Feb-82 1216 Paul Armer <CSD.ARMER at SU-SCORE> Compton Fellowships
Date: 23 Feb 1982 0929-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: Compton Fellowships
To: CSD-Faculty: ;
Stanford-Phone: (415) 497-2274
Stanford will award three dissertation Fellowships to minority
students who plan to pursue careers in college or university
teaching. May I have your nominations?
Paul
-------
∂23-Feb-82 1523 RPG Woes
I have been talking to Ted Selker (EJS) who is working on the
DD replacement. He has been telling me tales of woe about the
progress on this project and has some specific complaints about
things that he feels are hampering progress. He has asked me on
several occasions to use whatever influence I had to rectify these
problems. However, the nature of the problem is a person and not a
technical issue, and so I hesitate to act on this.
He says that Roy Hayes has been stalling and doing many incompetent
things on this project. Selker states that the first test of the
replacement could have been made in early december had Roy gotten
the boards wirewrapped, tested, and the parts ordered. I understand that
the first boards have been wirewrapped just recently, they are still
not tested, and only enough parts for 2 channels have just been ordered.
In addition, Selker has complained that Roy replaced an absolute filter
on the DD disk itself with foam rubber, which accounts for the steady
loss of channels. He says that the foam disintegrates and deposits
particles on the surface of the disk.
I understand that Selker has repeatedly complained to Ralph on this
matter but has gotten no response. He feels that unless something
is done to remove Roy from, at least, the DD replacement project,
things will not progress satisfactorally.
If you feel that something should be done about Selker's complaints,
maybe a word to REG is in order.
-rpg-
ge conference
∂24-Feb-82 0108 Jerry E. Pournelle <POURNE at MIT-MC> Council Report
Date: 24 February 1982 04:07-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
Subject: Council Report
To: TAW at S1-A
cc: jmc at SU-AI, llw at SU-AI
General Graham's High Frontier report is done and about
to be circulated. Thus SPACE AND NATIONAL SECURITY, the second
Council report, can now get put together (we were supporting
graham at first since he had the appointments and one voice was
better'n two for just then).
It won't be so long as the first. Major points are
Military and Economic contributions of space to national
security; plus some reiteration and reconfirmation of conclusons
fro the frist report.
If you'll have time to do the printout and such I think
Randy would appreciate that; I will let you know schedule for
sure.
Talked to Mr. Heinlein today; he is definitely coming to the
Conference. I hope we get a good turnout. So far the
professional registration is BIG, and excellent--the sessions
will be very valuable--but the locals and enthusiasts response
is disappointing. However, today the L-5 mailings arrived here
and the new ANALOG column is out this week, and probably things
will pick up.
Date: 21 Feb 1982 1516-PST
From: Tom Wadlow <TAW at S1-A>
Jerry,
Ted and I were wondering if you intend for us to do the
CACNSP report again. Haven't heard anything from you in a while
so we were becoming curious.
--Tom
∂24-Feb-82 0130 Bill Gosper <RWG at MIT-MC> Shop at Bernie's for an 83% rebate
Date: 24 February 1982 04:30-EST
From: Bill Gosper <RWG at MIT-MC>
Subject: Shop at Bernie's for an 83% rebate
To: AMO at MIT-MC
cc: RWG at MIT-MC, PMF at MIT-MC, MLB at MIT-MC, NJAS at MIT-MC,
WGD at MIT-MC, REM at MIT-MC, RZ at MIT-MC, GUIBAS at MIT-MC,
RAMSHAW at MIT-MC, ES at MIT-MC, dek at SU-AI, jmc at SU-AI
AMO's idea indeed extends to the B[n-6k] identity. Define the equation-
valued function
f(w) := (integrate(.,w,w+1) + integrate(.,-w-1,w))
((B(x) + h)↑n = B[n](x+h)) dh
:= int((B(x) + h)↑n + (B(x) - h)↑n,h,w,w+1) = (x+w)↑n
+ (x-w-1)↑n .
Now umbrally expand the integrand, canceling odd terms, integrate
termwise, and take f(0) - f(e↑(2 i pi/3)), as suggested by the
righthand side of the propositus. The summand on the left
vanishes except where k mod 3 = 1, thus preserving only those
k of the original expansion which were 2 mod 6.
The impediment to even sparser recurrences is that, for the
righthand side to be simple, the integrals must be over rational
lengths, and the hexagon may be the last "good" set of cyclo-
tomic roots. Expect no more of this junk mail unless I find
a real surprise.
∂24-Feb-82 0157 Jerry E. Pournelle <POURNE at MIT-MC>
Date: 24 February 1982 04:57-EST
From: Jerry E. Pournelle <POURNE at MIT-MC>
To: jmc at SU-AI
This sat around for months.
Occaision: Pajaro Dunes, Conference on Robotics in Space.
Purpose, to plan "bold new missions made possible by advances in
AI and robotics". After considerale fooferaw, it was admitted
by those present that for 20 or 30 years (certainly for 20)
there was no way to close the loop and make a self-replicating
robot system that could operate in Arizona, much less on the
Moon.
Nevertheless, said I, we can make self-replicating
systems in space: the easiest and the quickest to build is a
lunar colony, replication time about 18 to 20 years, time of
first colony completion depends no whe started; say we begin
with first permanent lunar resident about 1988 (this in 1980).
Associate Administrator: "We can't do that."
"Why?"
"We can't put people on the Moon in 8 years."
"Hmm. We did before. OK, say 1990."
"Cant do it then, either."
Minsky: "Surely we can. We did it before."
Associate Administrator (Administrator listening): "Not
now."
me; 'you mean we won't do it because no one will pay for
it."
NASA official: "I mean we can't do it even if we had the
money."
Minsky: "You mean that we don't have the capability."
NASA "Yes."
Minsky "I wanted to hear a high NASA official say we
could not do now what we did before in 8 years starting from a
lower level."
SILENCE.
POURNE@MIT-MC 02/19/82 04:56:05
Date: 14 Oct 1981 1533-PDT
From: John McCarthy <JMC at S1-A>
As indicated in the draft, I want to contrast the spirit of Shackleton
with that of the previous NASA Administrator if this can legitmately
be done. My source on what Frosch said is you. Could you send me a
message with a memory prompt? Of course, we may end up with a different
approach entirely. Reply to my usual SAIL address rather than S1 which
is at Livermore.
∂24-Feb-82 1038 Feigenbaum at SUMEX-AIM (Response to message)
Date: 24 Feb 1982 1038-PST
From: Feigenbaum at SUMEX-AIM
Subject: (Response to message)
To: JMC at SU-AI
In response to your message sent 24 Feb 1982 0043-PST
John,
thanks for alerting me.....Ed
-------
∂24-Feb-82 1223 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
Date: 24 Feb 1982 1104-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/1/82 Math 380C Johnathan Goodman
Monday Numerical Analysis Stanford University
4:15p.m. Seminar ``Multiple steady states for transonic duct
flow''
3/1/82 McC 240 William E. Foster
Monday Digital Systems Stratus Computer Inc., Natick Massachusetts
4:15p.m. Reliability - ``The Architecture of the Stratus Fault -
Special Seminar Tolerant Computer''
3/2/82 MJ301 Dr. Robert Blum
Tuesday Medical Computing Stanford University
1:30 p.m. Journal Club ``Recent Articles of Interest''
!Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/2/82 MJ301 Mr. Randy Teach
Tuesday Knowledge Stanford University
2:30 p.m. Representaion ``Cinical Explanation - A Sneak Preview''
Group Meeting
3/2/82 Math 380U Barry E. Jacobs
Tuesday Special Seminar University of Maryland
4:15 p.m. ``Applications of Database Logic to the View
Update Problem''
3/2/82 Jordan 041 Professor Edward Feigenbaum
Tuesday Computer Science Stanford University
4:15 p.m. Colloquium ``The Japanese Fifth Generation Computer
Project''
3/3/82 MJ352 Brian Wendell
Wednesday Robotics Seminar Stanford University, Psychology
12:00 p.m ``Physics, Psychology, & Measurement''
3/3/82 Skilling Auditorium Dave Anderson
Wednesday Computer Systems Storage Technology Corp.
4:15 p.m. Laboratory Seminar Title To be Announced
3/5/82 MJ352 Max Wilson
Friday Database Research IBM
3:30 p.m. Seminar ``Relational Database Design using the
Requirements and Design Aids''
!
-------
∂24-Feb-82 1303 FFL
To: JMC, FFL
Barney Oliva called from H-P and says khe needs a list of the people who
attended the SE2 meeting last evening. Please call at 857 205.
∂24-Feb-82 1351 Nancy Dorio <CSD.DORIO at SU-SCORE> CHANGE IN COLLLOQUIUM NOTICE OF MARCH 1 - 5, 1982
Date: 24 Feb 1982 1338-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: CHANGE IN COLLLOQUIUM NOTICE OF MARCH 1 - 5, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
Special Seminar - Tuesday, March 2, 1982 at 3:15 p.m. in MJ 352. Barry E.
Jacobs, from the University of Maryland, will speak on ``Applications of
Database Logic to the View Update Problem''.
-------
∂24-Feb-82 1441 David Warren <WARREN at SRI-AI> Visit by Ehud Shapiro
Date: 24 Feb 1982 1402-PST
From: David Warren <WARREN at SRI-AI>
Subject: Visit by Ehud Shapiro
To: AIC-Associates-et-al: ;
Ehud Shapiro, from Yale, will be visiting the AI Center on Thursday,
March 4, and will give a talk at 11.00 in EK242, followed by
discussion over TINlunch. He's keen to meet people generally; let me
know if you want to fix a particular time to talk to him. An
abstract of his talk follows. --David Warren.
ALGORITHMIC PROGRAM DEBUGGING
The talk lays a theoretical framework for program debugging, with the
goal of partly mechanizing this activity. In particular, we attempt
to formalize and develop algorithmic solutions to the following two
questions:
1. How to identify a bug in a program that behaves incorrectly?
2. How to fix the bug, once it is identified?
We describe interactive diagnosis algorithms, that can identify a bug
in an incorrect program, and a Prolog diagnosis system, based on
these algorithms. Methods for bug correction are developed in two
contexts: program synthesis from examples, and interactive debugging.
Experience with these systems will be described.
-------
∂24-Feb-82 1513 Nancy Dorio <CSD.DORIO at SU-SCORE> ADDENDUM TO COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
Date: 24 Feb 1982 1420-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: ADDENDUM TO COLLOQUIUM NOTICE WEEK OF MARCH 1 - 5, 1982
To: Colloq: ;
Stanford-Phone: (415) 497-2273
AFLB SEMINAR: Thursday, March 4, 1982 at 12:15 p.m. in MJ301. J. Ahrens,
and U. Dieter will speak on "Realistic Versus Abstract Roulette: A Comparison
of Optimal Strategies".
-------
∂25-Feb-82 0902 Betty Scott <CSD.BSCOTT at SU-SCORE> Report Processing
Date: 25 Feb 1982 0900-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
Subject: Report Processing
To: Faculty at SU-SCORE
cc: CSD.BScott at SU-SCORE
I need to know whether your CS Report processing has been satisfactory
over the past six months or so. Specifically, has the turnaround time
from submission to the Publications Office to finished report been
satisfactory, have instructions been followed accurately, etc.
Comments about publications procedures generally are also most welcome.
Your responses ASAP will be appreciated.
Thanks much.
Betty
-------
∂25-Feb-82 0939 Ichiki at SRI-AI TINLUNCH Reminder
Date: 25 Feb 1982 0938-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH Reminder
To: tlgrp:
This week's TINLUNCH meeting will be at 12:00 in Building A,
Conference Room A. The paper to be discussed is:
Focalizers, the Scoping Problem, and Semantic Interpretation
Rules in Logic Grammars
By: Michael C. McCord
Extra copies of the above paper are on Barbara Grosz's file cabinet.
Stan Rosenschein.
-------
∂25-Feb-82 1000 JMC*
phone want
∂25-Feb-82 1042 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
Date: 25 Feb 1982 1039-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: Course Assignments
To: CSD-Faculty: ;
cc: csd.sis at SU-SCORE, csd.keller at SU-SCORE, csd.walker at SU-SCORE
Stanford-Phone: (415) 497-2274
I have discovered that we need to know for the catalog
who is going to teach what courses for 1982-3 by March 1.
Clearly we can't make that date but we must try to come
close. I'm told that the way this has been done in the past
has been for the faculty in each area of the department
to get together and draw up a plan for their area. I'm
therefore asking that the following individuals assume
responsibility for calling a meeting of those who should
be involved:
Area Individual to call meeting
AA Knuth
AI Feigenbaum
NA Oliger
MTC McCarthy
Systems Ullman
Gene tells me that you should assume that each faculty member
teaches four courses in the absence of support and three courses
with 25% support.
If I have your inputs by March 3, we'll only be a few days late.
My abject apologies for not having been aware of this need earlier.
Paul
-------
In the past this has been settled by the Department Chairman or Denny
together with the individual faculty members. Next year as usual, I
will have 50 percent support and propose to teach two courses, CS206
in the Fall and CS226 in the Winter. I alternate CS226 (Epistemology
of AI) and CS258 (mathemtical theory of computation) in alternate years.
Since a given faculty member doesn't always teach in the same area, I
don't see how your scheme is practical. It would be better if you or
Gene spoke individually to the faculty members. I have no idea whom
your are expecting me to call to a meeting.
∂25-Feb-82 1143 RWW
To: FFL, JMC
I have just sent the following message to SSK.
I have just asked the department to cancel your appointment as
research assistent. This means that both RA money and tuition
and machine usage will be canceled. I consider this final from
my point of view. You will have to make other arrangements with
the department. Sorry.
Richard Weyhrauch
Traditionally, graduate students get zapped effective the end of
a quarter, and I suppose you mean to zap Kasturia at the end of
Winter quarter.
Come to think of it, I guess RA's are rarely zapped except at the end of an
academic year. Perhaps you should discuss earlier zapping of Kasturia
with Gene or postpone his zapping till end of Spring.
∂25-Feb-82 1541 Gene Golub <CSD.GOLUB at SU-SCORE> recruiting
Mail-from: ARPANET site SU-SCORE rcvd at 25-Feb-82 1533-PST
Date: 25 Feb 1982 1530-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: recruiting
To: faculty at SU-SCORE
cc: CSD-Administration: ;,
RECRUITMENT: ;
We are now in the midst of the recruiting season. Next week we have two
visitors. On Thursday, March 4 S. Micali will speak at the algorithms
lunch. On Friday, March 5 E. Shapiro of Yale will be visiting.
Please let Irmgild know when you are available. It would be nice if
you could come for lunch or dinner. Hope to hear from you soon,
GENE
PS The following week it is likely we will have a couple of visitors.
-------
∂25-Feb-82 1551 Gene Golub <CSD.GOLUB at SU-SCORE> Visit
Date: 25 Feb 1982 1551-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Visit
To: Shapiro at YALE
cc: Jmc at SU-AI
Please let me know as soon as possible about your visit. If John
McCarthy has not contacted you, please send me a title for your
talk. We are all looking forward to your visit. GENE GOLUB
-------
∂25-Feb-82 1652 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
Date: 25 Feb 1982 1646-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: Course Assignments
To: CSD-Faculty: ;
cc: csd.sis at SU-SCORE, csd.walker at SU-SCORE, csd.keller at SU-SCORE,
csd.armer at SU-SCORE
Stanford-Phone: (415) 497-2274
In sending my memo this morning (copy being resent with this)
I erred in extrapolating from the practice of several groups to
the entire department. If you are not a member of a group which
is making a collective decision will you send me a message
letting me know as soon as possible which courses you would
like to / be willing to / won't refuse to / teach.
AI (Feigenbaum), NA (Oliger) and Systems (Ullman) are taking
the collective approach. If you intend to be on leave for any
part of the year, please let me know about that as well.
Paul
I have discovered that we need to know for the catalog
who is going to teach what courses for 1982-3 by March 1.
Clearly we can't make that date but we must try to come
close. I'm told that the way this has been done in the past
has been for the faculty in each area of the department
to get together and draw up a plan for their area. I'm
therefore asking that the following individuals assume
responsibility for calling a meeting of those who should
be involved:
Area Individual to call meeting
AA Knuth
AI Feigenbaum
NA Oliger
MTC McCarthy
Systems Ullman
Gene tells me that you should assume that each faculty member
teaches four courses in the absence of support and three courses
with 25% support.
If I have your inputs by March 3, we'll only be a few days late.
My abject apologies for not having been aware of this need earlier.
Paul
-------
Next year I plan to teach CS206 (Recursive programming and proving) in
Fall and CS226 (Epistemological problems of artificial intelligence) in
Winter and nothing in Spring (using 50% support). Should this offer any
problems for planning the AI offerings or the systems offerings (assuming
CS206 to fall under systems), please let me know.
∂25-Feb-82 1933 Perlis.EE at UDel-Relay
Date: 25 Feb 82 21:47:32-EST (Thu)
From: Perlis.EE at UDel-Relay
To: jmc at Su-Ai
Via: UDel-EE; 25 Feb 82 21:48-EST
Via: UDel; 25 Feb 82 22:25-EDT
Dear Dr. McCarthy:
Martin Davis suggested to me that you might be interested in some of my
recent work in foundations of knowledge representation and self-reference. I
am therefore sending you several preprints in the mail.
I will also take the liberty of requesting that my year-old application
for a faculty position at Stanford be re-opened. You may recall that you were
away when I wrote to you at that time. Along with the preprints I will
enclose an updated resume.
Best regards,
Donald Perlis
The Department is looking for people, and I will refer your resume to
Professor Golub, the Department Chairman, when it comes. I will write
you if I have a definite reaction to the reprints.
Donald Perlis, whose resume I am forwarding, wishes to apply for a
faculty position.
∂25-Feb-82 2009 RWW
To: JMC, FFL
SSK has spent the last month using SAIL to write a venture capital
proposal (see FIX[1,ssk]) I have sent him repeated messages including
the one I sent you a copy. I have not even had a reply TO A SINGLE
warning message. Todays was one of many. Betty Scott told me that
to cut him off today would cause his existing tuition to disappear,
i.e. this quarter, so I told her to hold off, but not getting any
reply at all seems a bit much. In any event I have no desire for
either myself or CS or SAIL to subsidize his commercial hacking.
He must have seen this coming because Betty Scott said that he has
arranged to TA for someone else in the spring, but I don't know who.
Immediately after I sent him the message that I sent you I also sent
him
PS. this cancelation means canceling THIS quarters tuition.
and I still no reply. To be fair he may not have seen this message
but it still seems that I (? the department?) should do something.
∂25-Feb-82 2011 RWW
PS he was logged on late this afternoon.
From your last two messages, it seems to me that you know what you are doing
and have consulted sufficiently with the authorities.
∂25-Feb-82 2353 HST via SRI-KL lisp history
hallo,prof.mccarthy.bill white writes me that he started the work
on dndral because of a suggestion of a prof.richard watson.do
you know him?in what relation was he to lisp then?herbert stoyan.
There was a Richard Watson at SRI, but he left to work for some
company, maybe an oil company. I don't think he had any relation
to LISP, but he may have had some relation to the chemical structural
questions which dendral tackles. I believe he wrote a book on
time-sharing some time in the sixties. Of course, it may be a
different Richard Watson, so you had better be sure it's the same
one. I think perhaps he was in the Computer Science Department
at Stanford for a year or two.
∂25-Feb-82 2358 Bill Gosper <RWG at MIT-MC>
Date: 26 February 1982 02:57-EST
From: Bill Gosper <RWG at MIT-MC>
To: jmc at SU-AI
i just looked at multip[1,jmc].
Editing glitch (consideration 3.2): "... the program that activated by
the catch."
Consideration 69? Suppose a recursive function expects to be called
with identical args often enough to motivate "memoizing" its values
to avoid recomputation. What's a good mechanism to prevent two
processors from working on the same problem? (Instead of hanging
one process on the outcome of the other, and freeing a processor.)
∂26-Feb-82 0040 Feigenbaum at SUMEX-AIM (Response to message)
Date: 26 Feb 1982 0038-PST
From: Feigenbaum at SUMEX-AIM
Subject: (Response to message)
To: JMC at SU-AI, csd.ullman at SU-SCORE
cc: csd.armer at SU-SCORE
In response to the message sent 25 Feb 1982 1748-PST from JMC at SU-AI
John, that teaching schedule looks fine to me. I'll fold it into the
planning of the "AI guys".
Ed
-------
∂26-Feb-82 0934 Gene Golub <CSD.GOLUB at SU-SCORE> Re: recruiting
Date: 26 Feb 1982 0932-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Re: recruiting
To: Feigenbaum at SUMEX-AIM
cc: jmc at SU-AI
In-Reply-To: Your message of 26-Feb-82 0026-PST
I think it is important that we hear Shapiro. can you setup the
Siglunch talk? Thanks. GENE
-------
∂26-Feb-82 1156 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Recruitment
Date: 26 Feb 1982 1151-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Recruitment
To: faculty at SU-SCORE
cc: csd.dietterich at SU-SCORE
I am setting up the schedules for E. Shapiro (Yale) and S. Micali (Berkeley)
who will be visiting Stanford March 4 and 5. Please let me know whether
you are interested to talk with either of them and if, when.
I appreciate a quick reply.
Irmgild
-------
I certainly want to talk to Shapiro. Late afternoon Thursday is best for me.
∂26-Feb-82 1332 CLT
we could eat out together, but perhaps you don't want to wait
ok, let's do it.
∂26-Feb-82 1938 RPG Selker
Jeff talked to him and the impression he got was that Selker appeared to
be a reasonable person, but has neither had enough time nor access to
information to understand the problems fully, and that Selker should be
allowed some time to do so alone and unconstrained. At that point (in
about a week) a second meeting should occur to discuss the design more
with Jeff.
-rpg-
∂27-Feb-82 1746 YM shapiro visit
I understood that Udi Shapiro is going to give a talk on Friday. Is this
going to be an open talk? If yes when and where is it going to take place
and why it was not announced?
Thanks, Yoni Malachi.
It will be open, and it will be announced. I believe it will be in
Feigenbaum's SIGLUNCH.
∂28-Feb-82 0046 TOB
John
The reporter called me. I said that it was ok to mention my name
and that I agreed with your five points.
Tom
∂24-Feb-82 0043 JMC
To: TOB at SU-AI, feigenbaum at SUMEX-AIM
Partly against my better judgment, I have allowed myself to
be interviewed by the Stanford Daily. I don't think I said much
wrong, and the reporter called me to check, but I still can't be
sure how it will come out. I suppose my main motivation was that
I had better let something happen, since I have been prevented by
other matters from taking any action along the lines we previously
discussed.
I told the Daily reporter to call the rest of you who
took part in the meeting with Dean Lieberman, but, of course, I
can't be sure that he will do so.
Here's hoping it turns out all right.
Sincerely,
∂28-Feb-82 1417 Gene Golub <CSD.GOLUB at SU-SCORE> [ Ehud Shapiro <Shapiro at YALE>: Re: Visit]
Date: 28 Feb 1982 1414-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [ Ehud Shapiro <Shapiro at YALE>: Re: Visit]
To: jmc at SU-AI, feigenbaum at SUMEX-AIM
Has a time been set for his talk? GENE
---------------
Mail-from: ARPANET site YALE rcvd at 27-Feb-82 1136-PST
Date: 27-Feb-82 1434-EST
From: Ehud Shapiro <Shapiro at YALE>
Subject: Re: Visit
To: Csd.Golub at SU-SCORE
In-Reply-To: Your message of 27-Feb-82 1431-EST
I am coming Wednesday evening, March 3rd, and will spend Thursday at
SRI. I plan to return Saturday noon. If there are problems with
covering my stay on Wednesday night I can resolve them with SRI, but
I would like to stay in ne place during the visit.
The title of my talk is "Algorithmic Program Debugging"; an abstract
follows.
The talk lays a theoretical framework for program debugging,
with the goal of partly mechanizing this activity. In
particular, we attempt to formalize and develop algorithmic
solutions to the following two question:
1. How to identify a bug in a program that behaves incorrectly?
2. How to fix a bug, once one is identified?
We describe interactive diagnosis algorithms, that can identify
a bug in an incorrect program, and a Prolog diagnosis system,
based on these algorithms. Methods for bug correction are
developed in two contexts: program synthesis from examples
and interactive debugging. Experience with these systems will
be described.
Thank you for the invitation; I am looking forward to this visit.
-- Udi
p.s. this is a re-mail of the message I have sent February 24.
-------
-------
∂28-Feb-82 1700 Kim.fateman at Berkeley smoking things out of macsyma
Date: 28 Feb 1982 16:35:58-PST
From: Kim.fateman at Berkeley
To: COMMON-LISP@SU-AI, GJC@MIT-MC
Subject: smoking things out of macsyma
I really doubt that all problems are simple
to smoke out; in fact, I suspect that there are still places
where the Lisp Machine version of Macsyma fails for mysterious
reasons. These may be totally unrelated to T vs #T or NIL vs (),
but I do not see how GJC can be so confident.
For example, when we brought Macsyma up on the VAX, (after it
had allegedly been brought up on a CADR) we found
places where property lists were found by computing CAR of atoms;
we found a number of cases of (not)working-by-accident functions whose
non-functionality was noticed only when run on the VAX with a modest
amount of additional error checking. (e.g. programs which should
have bombed out just chugged along on the pdp-10).
GJC claims there is (was?) only one line of Macsyma which legally
depends on other-than truthity of a predicate. I believe this is
false, but in any case, a proof of his claim would require rather
extensive analysis. Whichever way this decision goes (about NIL or ()),
I would be leery of making too much of GJC's note for supporting evidence.
∂28-Feb-82 2048 Brian Reid <reid@Shasta at Sumex-Aim> systems candidate Paul Hilfinger
Mail-from: ARPANET site SUMEX-AIM rcvd at 28-Feb-82 2044-PST
Mail-from: SU-NET host SU-SHASTA rcvd at 28-Feb-82 2042-PST
Date: Sunday, 28 Feb 1982 20:40-PST
To: Faculty at SU-SCORE
Cc: Irmgild at SU-SCORE
Subject: systems candidate Paul Hilfinger
Reply-to: Reid@Shasta at SUMEX-AIM
From: Brian Reid <reid@Shasta at Sumex-Aim>
(I'm not sure if I'm supposed to be sending this or if Gene is, but
it can't hurt to do it twice...)
Paul Hilfinger, a recent Ph.D. from Carnegie-Mellon, will be
interviewing here on Tuesday March 9 and Wednesday March 10. He will
be giving the CS Colloquium on the 9th, and will be here all day on
Wednesday the 10th to talk to faculty and students.
If you would like an appointment with Paul on Wednesday, please call
Linda Kovach, at 7-1440, or send mail to Kovach@Score. Nan should be
sending out an abstract of his talk very soon.
Brian
∂01-Mar-82 0639 Oded Anoaf Feingold <OAF at MIT-MC> another delayed mailing
Date: 1 March 1982 09:33-EST
From: Oded Anoaf Feingold <OAF at MIT-MC>
Subject: another delayed mailing
To: energy
Wow, what a baroque header.
OAF 9:30am Monday, 1 March 1982
--------------------
Redistributed-Date: 15 January 1982 21:51 cst
Redistributed-By: VaughanW.REFLECS at HI-Multics
Redistributed-To: energy at MIT-MC
Date: 8 January 1982 12:41 cst
From: Bibbero.PMSDMKT at HI-Multics
Subject: Power Source Health Hazards
Reply-To: Bibbero.PMSDMKT at HI-Multics
To: {mbx >udd>reflecs>bv>ARPA←fwd} at HI-Multics (energy at MIT-MC)
cc: Bibbero.PMSDMKT at HI-Multics
In-Reply-To: JPM at SU-AI dated 16 Nov 1981 2248-PST
To: Jim McGrath <JPM at SU-AI>
In re your msg Nuclear Power dated 16 Nov 1981 2248-PST
The health costs in terms of human lives cited for coal and nuclear power
in your message are most interesting. The figure for yearly deaths from
air pollution caused by coal seem particularily high to me. I assume
you meant total dependence on coal for all electrical power and the same
assumption for nuclear power. What about industrially-generated power
and process heat? Did your figures just include utilities? I'd appreciate
any details and references.
My interest in these figures stem partly from the fact that I am co-author
of a text on air pollution (Systems Approach to Air Pollution Control,
Wiley Interscience, 1974) and also am currently associate editor for
technological wastes of a new IEEE publication TECHNOLOGY & SOCIETY
MAGAZINE. If you or anyone on this mailing list would like to express
your views on the societal aspects of nuclear vs. coal power from the
viewpoint of probable costs (human or monetary) in this new magazine I
would like to receive a draft of the proposed paper. Please mail to me
directly:
Robert J. Bibbero MS 484
Honeywell Inc.
1100 Virginia Drive
Ft. Washington, PA 19034
∂01-Mar-82 0946 DWALTZ at BBND AAAI Program Committee once again
Date: 1 Mar 1982 1242-EST
Sender: DWALTZ at BBND
Subject: AAAI Program Committee once again
From: DWALTZ at BBND
To: rich at MIT-AI, minsky at MIT-AI, gjs at MIT-AI, phw at MIT-AI, tenenbaum at SRI-KL,
To: reddy at CMUB, newell at CMUA, bobrow at PARC-MAXC, buchanan at SUMEX-AIM,
To: engelmore at SUMEX-AIM, feigenbaum at SUMEX-AIM, aaai-office at SUMEX-AIM,
To: bledsoe at UTEXAS-20, grosz at SRI-AI, nilsson at SRI-AI, walker at SRI-AI,
To: feldman at SUMEX-AIM, mccarthy at SU-AI, woods at BBND, webber at BBND,
To: simon at CMUA
Message-ID: <[BBND] 1-Mar-82 12:42:56.DWALTZ>
Thanks to all who sent back comments on my original message re the
composition of the reviewing committee. I agreed with virtually all.
Since there were very few (=1) papers last time on games, it is probably
not important to have Berliner on the committee -- if there are very
few papers, they could probably be handled by mail review alone.
Therefore, I propose dropping Berliner as a committee
member.
The natural language area was indeed represented
unsatisfactorily, so I propose adding Bill Woods and Barbara Grosz to the
committee. This will also beef up the representation area,
although many of the committee members have indicated willingness to
review representation papers. The committee will then consist
of 18 rather than 17 persons as it did the first time, but 17 was a
rather odd number anyhow.
Unless there are strong and quickly received objections to the revised
list, I will assume that I can go ahead with notification of members, and
planning of the review meeting.
-Dave
∂01-Mar-82 1000 JMC*
Grant about Okner
∂01-Mar-82 1000 JMC*
Chien and John Cocke
∂01-Mar-82 1124 pratt@Shasta (SuNet) Proposed logic course for this Fall
Date: 1 Mar 1982 11:25:04-PST
From: pratt at Shasta
To: jmc@sail
From: pratt
Subject: Proposed logic course for this Fall
Date: 1 March 1982 11:25:03-PST (Monday)
CS???. Logical Algorithms
Algorithms for problems of a logical character. Emphasis on applications to
theorem proving, program verification, deductive computation, heuristic
problem solving, and optimizing compilers. Decision methods for various
quantifier free calculi. Oppen's method of combining such decision methods.
Algorithms for program logics. Unification. Decidable logics of higher type.
3 units, Aut (Pratt)
∂01-Mar-82 1149 pratt@Shasta (SuNet) revised description
Date: 1 Mar 1982 11:39:51-PST
From: pratt at Shasta
To: jmc@sail
From: pratt
Subject: revised description
Date: 1 March 1982 11:39:49-PST (Monday)
At DEK's suggestion I made some changes. Do you think this course should
be an MTC or an AA course? (At MIT we had a theory group which didn't
make this distinction, I have to get used to this categorization.)
-v
To tell the truth, I don't see the import of the categorization either,
and we didn't have such a formal categorization previously. I suspect
it's merely an artifact of Paul Armer's retailing the task of deciding
what courses are to be given next year. The course seems like an advanced
one, and therefore should be whatever you decide. While the elementary
courses need to be fitted together, advanced courses should be at the
discretion of the professor.
∂01-Mar-82 1209 David Warren <WARREN at SRI-AI> Veronica Dahl
Date: 1 Mar 1982 1202-PST
From: David Warren <WARREN at SRI-AI>
Subject: Veronica Dahl
To: JMC at SU-AI
cc: Warren at SRI-AI
Veronica Dahl is currently a visiting professor at U. of Kentucky.
She would like to visit the W Coast after May 10, if any way can be
found to cover her expenses. Would you/Stanford be interested in
inviting her to give a seminar on her work in Prolog and natural
language, and able to pay some expenses. --David Warren.
-------
Unfortunately I will be abroad from the end of March till about the
20th of June, and I can't think of anyone else who would be interested.
If she will still be in the U.S. after June 20, I think we could pay
part of the expenses.
∂01-Mar-82 1403 Gene Golub <CSD.GOLUB at SU-SCORE> Next meeting
Date: 1 Mar 1982 1356-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Next meeting
To: CSD-Tenured-Faculty: ;
Our next regular tenured faculty meeting will take place on Thursday, March 4
at 2:30 in the Conference room next to my office. I want us to discuss
some of the potential candidates for a position and the future directions
of the department. All this in 75 minutes!
Please try to be prompt. GENE
-------
-------
∂01-Mar-82 1404 Gene Golub <CSD.GOLUB at SU-SCORE> Next Meeting
Date: 1 Mar 1982 1357-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Next Meeting
To: CSD-Tenured-Faculty: ;
Our next regular tenured faculty meeting will take place on Thursday, March 4
at 2:30 in the Conference room next to my office. I want us to discuss
some of the potential candidates for a position and the future directions
of the department. All this in 75 minutes!
Please try to be prompt. GENE
-------
-------
∂01-Mar-82 1449 David Warren <WARREN at SRI-AI>
Date: 1 Mar 1982 1430-PST
From: David Warren <WARREN at SRI-AI>
To: JMC at SU-AI
cc: WARREN at SRI-AI
In-Reply-To: Your message of 1-Mar-82 1228-PST
Thanks for your reply; I'll pass the info on to Veronica. --David.
-------
∂01-Mar-82 1527 Ichiki at SRI-AI TINLUNCH meeting 3/4
Date: 1 Mar 1982 1526-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH meeting 3/4
To: tlgrp:
Before the TINLUNCH meeting this Thursday, 3/4, Ehud Shapiro (Yale)
will give a talk at 11:00 pm in EK242.
This week's TINLUNCH meeting will be at 12:00 in Building E,
Conference Room EK242. The paper to be discussed is:
Algorithmic Program Debugging
By: Ehud Shapiro
ALGORITHMIC PROGRAM DEBUGGING
The talk lays a theoretical framework for program debugging, with the
goal of partly mechanizing this activity. In particular, we attempt
to formalize and develop algorithmic solutions to the following two
questions:
1. How to identify a bug in a program that behaves incorrectly?
2. How to fix the bug, once it is identified?
We describe interactive diagnosis algorithms, that can identify a bug
in an incorrect program, and a Prolog diagnosis system, based on
these algorithms. Methods for bug correction are developed in two
contexts: program synthesis from examples, and interactive debugging.
Experience with these systems will be described.
-------
∂01-Mar-82 2000 JMC*
Feigenbaum about Kuo
∂02-Mar-82 0834 Doug Lenat <CSD.LENAT at SU-SCORE> Re: meta-cognition: reasoning about knowledge
Date: 2 Mar 1982 0730-PST
From: Doug Lenat <CSD.LENAT at SU-SCORE>
Subject: Re: meta-cognition: reasoning about knowledge
To: JMC at SU-AI
In-Reply-To: Your message of 1-Mar-82 0300-PST
Will do, John. What worries me is: what if ALL the copies had
that property, and you're just the first person to try to read it!
Doug
-------
∂03-Mar-82 0924 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF FEBRUARY 8 - 12, 1982
Date: 3 Mar 1982 0914-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF FEBRUARY 8 - 12, 1982
To: colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/8/82 Art Building, Rm.4 Charles Bigelow
Monday Special Seminar
3:15 p.m. ``Letterforms: Analog to Digital''
3/8/82 Math 380C George Gross and Paul Gribik
Monday Numerical Analysis P G \&\ E
4:15 p.m. Seminar ``Solution of a large nonlinear programming
problem---the optimal power overflow''
3/9/82 Jordan 041 Paul N. Hilfinger
Tuesday Computer Science CMU
4:15 p.m. Colloquium ``Implementation of Tasking Mechanisms''
3/10/82 Skilling Auditorium Martin Kay
Wednesday Computer Systems Xerox Parc
4:15 p.m. Laboratory Seminar ``Dictionary of the Future and the Future of
the Dictionary''
3/11/82 MJ301 Andy Yao
Thursday AFLB Seminar U.C. Berkeley
12:15 p.m ``The Science of Double Talk''
!
-------
∂03-Mar-82 1227 Nancy Dorio <CSD.DORIO at SU-SCORE> correction to Colloquium notice
Date: 3 Mar 1982 0954-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: correction to Colloquium notice
To: colloq: ;
Stanford-Phone: (415) 497-2273
PLEASE NOTE: COLLOQUIUM NOTICE WEEK 0F FEBRUARY 8 - 12 , Should have read
COLOQUIUM NOTICE WEEK OF MARCH 8 - 12, 1982. Sorry!
-------
∂03-Mar-82 1613 Nancy Dorio <CSD.DORIO at SU-SCORE> conference rooms, spring quarter
Date: 3 Mar 1982 1555-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: conference rooms, spring quarter
To: CSD-Faculty: ;
cc: SU-BBOARD at SU-SCORE
Stanford-Phone: (415) 497-2273
All faculty and staff members who wish to reserve a conference or
seminar room on a regular basis for Spring quarter, please see Nancy and
fill out a reservation form. Due to excessive need of these rooms, priority
will be given to regularly scheduled classes and seminars. Many new classes
and seminars are starting up next quarter. Because of this it is important
that you reserve your space in advance to avoid misunderstanding and
confusion about where your activity will be held and at what time.
Thank you very much!!
-------
∂03-Mar-82 1803 RPG
∂03-Mar-82 1700 JMC Common LISP discussion
Is it technically feasible to put a file on the mailing list for the
Common Lisp discussion messages. It would then be convenient for me
to check it with CKSUM rather than have them in my already large general
mail file. If it were feasible, then it might also be feasible for
several people as SAIL to share the same discussion file.
Common.msg[com,lsp] is such a file. Shall I take you off the list?
-rpg-
Yes. Please take me off the list.
∂04-Mar-82 2152 Purger exceeding your disk quota
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.
READ PURGE for details on purges. Disk allocations and disk usage are
measured in disk blocks (a block is currently one track). If you would
like to buy a bigger allocation, contact Lynn Gotelli (LMG at SAIL).
The unit of allocation, one aliquot, is currently 50 disk blocks.
Your disk quota is: 2000 blocks
Your files occupy 2905 blocks
∂05-Mar-82 1132 Howard Trickey <CSD.TRICKEY at SU-SCORE> Course Evaluations
Date: 5 Mar 1982 1100-PST
From: Howard Trickey <CSD.TRICKEY at SU-SCORE>
Subject: Course Evaluations
To: csd.mayr at SU-SCORE, csl.bkr at SU-SCORE, csl.jlh at SU-SCORE,
csl.fat at SU-SCORE, reg at SU-AI, dek at SU-AI, shortliffe at SUMEX-AIM,
csd.buchanan at SU-SCORE, csd.rww at SU-SCORE, jmc at SU-AI
The Course Evaluation Committee would like to have course evaluations done
in next Thursday's class (March 11). Could you please confirm that this
is all right, or suggest another date. Also, please tell me where and
when the class meets.
-------
∂05-Mar-82 1252 pratt@Shasta (SuNet) dinner with Ehud Shapiro
Date: 5 Mar 1982 12:53:28-PST
From: pratt at Shasta
To: <JMC@SU-AI>, John, McCarthy, RWW@SU-AI, VRP@SU-AI, buchanan@SUMEX-AIM,
csd.geneserth@SU-SCORE, csd.lenat@SU-SCORE
From: pratt
Subject: dinner with Ehud Shapiro
Date: 5 March 1982 12:53:26-PST (Friday)
I have laryngitis, count me out.
-v
∂05-Mar-82 1432 David E. Smith <CSD.SMITH at SU-SCORE> Re: Hayes quote
Date: 5 Mar 1982 1426-PST
From: David E. Smith <CSD.SMITH at SU-SCORE>
Subject: Re: Hayes quote
To: JMC at SU-AI
Stanford Phone: (415)497-1809
In-Reply-To: Your message of 28-Feb-82 1820-PST
The exact reference is
Doyle's thesis (A Model for Deliberation, Action, and Introspection)
section 1.7.1, pg 48.
Doyle is referring to Hayes paper
Some Problems & Non-Problems in Representation Theory, proc. AISB (1974)
pg 63-79.
The relevant part of this paper (section 5) refers mostly to Hayes' GOLUX paper
Computation & Deduction, proc. MFCS symposium (1973, Czech Acadamy of
Sciences). Reprinted in Lecture Notes in CS, Vol 53.
This last paper may be found in the bowels of the library as uncataloged
material #89026.
-- Dave Smith
-------
Thanks for the references.
∂05-Mar-82 1607 Gene Golub <CSD.GOLUB at SU-SCORE> [Mark Crispin <Admin.MRC at SU-SCORE>: SCORE reliability]
Date: 5 Mar 1982 1604-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [Mark Crispin <Admin.MRC at SU-SCORE>: SCORE reliability]
To: faculty at SU-SCORE
cc: csd.admin at SU-SCORE
Here is some information which may be of interest. GENE
---------------
Mail-From: ADMIN.MRC created at 2-Mar-82 14:31:00
Date: 2 Mar 1982 1431-PST
From: Mark Crispin <Admin.MRC at SU-SCORE>
Subject: SCORE reliability
To: CSD.Golub at SU-SCORE
Postal-Address: 725 Mariposa Ave. #103; Mountain View, CA 94041
Phone: (415) 497-1407 (Stanford); (415) 968-1052 (residence)
Gene,
A comment you made about how unreliable SAIL and SCORE were has
been passed on to me. I can't speak for SAIL (other than that it has
always been that way), but your experiences with SCORE are, I hope,
not the normal circumstance. At one time SCORE had a full year without
a software-induced system failure.
In recent months, we have been running a field test of TOPS-20,
which introduced certain problems. Most of these problems have been
fixed already. Other latent problems having to do with ARPANET service
have come up. As the ARPANET changes to TCP/IP protocols, bugs in
ARPANET service having to do with the different schemes of communication
are being tickled more often. I fear that this will continue to be a
problem for the next year until the full changeover to TCP/IP is
completed. I believe, however, that we have a better handle on the
ARPANET software than we previously had, and that the situation will
remain under control.
It has now been over a week since the last software-induced crash,
and the cause of that crash has been fixed (ARPANET service, as usual).
This is at least partially due to the improved release of TOPS-20 we are
now running. It is my sincere hope that once again SCORE will be the
most reliable full-scale timesharing system on campus (SUMEX has a 2020
that never crashes, but only has one or two users on at a time, so it
doesn't count). Most of the credit for SCORE's earlier reliability goes
to my efforts, and I am concentrating on doing the same again.
There are some factors which will make SCORE less reliable in coming
months:
. the addition of Ethernet service on SCORE. The MEIS hardware is still
new and while extensive debugging has gone on with it there is
undoubtably going to be an impact on the system with its addition. The
Ethernet software will be a combination of new software we write (which
will need debugging) and the old PARC Ethernet driver. The PARC code is
known to have serious problems on TOPS-20; CMU is alleged to have fixed
most of these but I'm sure "that last bug" is still there waiting for us.
Since SCORE is the FIRST system ANYWHERE which is going to interface to
Ethernet in the way we have chosen, we can and should expect some of
the less desirable side effects of any experimentation. We expect that
in the long term it will be worth the investment.
. the change in ARPANET from NCP-based protocols to TCP/IP. This is a
major incompatible change, affecting all the network software in a
global manner. We have no choice about doing it; we're being forced to.
I hope that we can make the transition with a minimum of pain, but I am
worried.
. the increased demand in system resources without a corresponding increase
in those resources, specifically, disk and memory space. It's only been
due to superhuman efforts on my part that we have not run out of disk
space before. We are swapping excessively. A heavily-loaded system by
nature is going to be less reliable.
I think that I can say, staking my reputation as a major figure in
TOPS-20 system development in the country (including being the chairman of
a TOPS-20 system programmer's/manager's discussion group), that the field
test of TOPS-20 we are running is no longer a system reliability issue.
With the second field test release (running as of a week ago Monday), the
operating system should be more reliable, with higher performance, than
the current distributed release of TOPS-20 (look at LOTS' reliability if
you don't believe me!).
If you have any questions or feelings about these matters or other
SCORE-related matter that you would care to discuss with me, I am at your
disposal.
-- Mark --
-------
-------
∂05-Mar-82 1621 Mullen at SUMEX-AIM AI quals
Date: 5 Mar 1982 1621-PST
From: Mullen at SUMEX-AIM
Subject: AI quals
To: feigenbaum at SUMEX-AIM, jmc at SU-AI, csd.genesereth at SU-SCORE,
csd.lenat at SU-SCORE, tw at SU-AI
Who is in charge of AI quals this year. Marilyn Walker would
like to know the schedule. Bruce Buchanan
-------
∂05-Mar-82 2009 YM
do you know where I can contact Shapiro?
Telephone him at the Tiki in.
∂06-Mar-82 1134 CLT
I think I will go to the stereo store this afternoon around 3or4
Do you want to come?
∂07-Mar-82 1256 Feigenbaum at SUMEX-AIM Re: AI quals
Date: 7 Mar 1982 1257-PST
From: Feigenbaum at SUMEX-AIM
Subject: Re: AI quals
To: Mullen at SUMEX-AIM, jmc at SU-AI, csd.genesereth at SU-SCORE,
csd.lenat at SU-SCORE, tw at SU-AI
In response to the message sent 5 Mar 1982 1621-PST from Mullen
I dont know, but I dont volunteer. I'm teaching two courses in the
Spring, so have no extra time to organize quals.
Terry, since you will be on sabbatical next year and wont have to do this
next year, could you organize this Spring's qual?
Ed
-------
∂08-Mar-82 0126 Paul Armer <CSD.ARMER at SU-SCORE> Course Assignments
Date: 8 Mar 1982 0118-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: Course Assignments
To: CSD-Faculty: ;
cc: csd.walker at SU-SCORE, csd.keller at SU-SCORE
Stanford-Phone: (415) 497-2274
Here's my first pass at course assignments for 1982-83. There are two
lists - one ordered by course and one ordered by individual. Some new
courses have been added by the Curriculum Committee and some have been
dropped. If you are interested in finding out what description goes
with a given number see <CSD.FILES>BULLETIN on SCORE.
Corrections, additions, deletions and comments are welcome. Even if it
matches your expectations will you please send me a message to that
effect - I want to hear from everyone.
Paul
Created 3/1/82 by CSD.ARMER. This page contains assignments by course for
82-83.
Updated 3/8/82 by CSD.ARMER.
Legend: "Staff" as an entry means that the course will be given,
but the instructor is as yet undecided.
An empty entry or "--" means that the course is not given
that quarter.
"???" means that at this time we don't know whether that course
is being offered this year or not.
A & means tentative assignment.
A * associated with a course means that it is an offering of EE and
that EE pays for the course. Corrections, additions, deletions
comments to CSD.ARMER.
Course Autumn Winter Spring Summer Notes
------------------------------------------------------------------------------
CS75 -- -- Staff
CS101 -- -- Feigenbaum
CS102 &Novak -- --
CS103 Staff Staff Staff(TVI?)
CS104 Staff(TVI) Staff(TVI) Staff(TVI)
CS105-1 Staff Staff Staff Staff
CS105-2 Staff Staff Staff
CS105-3 Staff Staff Staff
CS105-4 Staff Staff Staff
CS105-5 Staff Staff Staff
CS105-6 Staff Staff Staff
CS106-1 Floyd Staff Staff Staff
CS106-2 Floyd Staff Staff Staff
CS106-3 Staff Staff Staff
CS106-4 Staff Staff Staff
CS106-5 Staff Staff Staff
CS106-6 Staff Staff Staff
CS107/EE180 Staff Staff Staff Staff
CS108A Reid -- --
CS108B -- Reid --
CS109A -- -- Erhman
CS109B -- Gorin --
CS111-1/EE181 *Gill *Staff *Williams *Staff
CS111-2 *Staff *Staff
CS111S/EE181S -- *Staff *Staff *Staff
CS112/EE182 *Manning *Banks --
CS121A,B -- Buchanan Buchanan
Shortliffe Shortliffe
CS135 -- -- Schreiber Staff
CS137A Staff -- --
CS137B -- Schreiber --
CS137C -- -- Oliger
CS142/EE285 *Staff *Staff? *Reid?
CS143/EE286 -- *Hennessy --
CS145/EE287 -- -- &Wiederhold
CS146 Lantz Cheriton --
CS147 Tobagi -- --
CS150 -- Pratt --
CS154 -- -- Floyd
CS155 &Mayr -- --
CS156 Manna -- --
CS157 -- -- Manna,Waldinger
CS161 Ullman -- --
CS162 -- Floyd --
CS163 -- -- Pratt
CS192 (by arrangement)
CS193/EE183 *Flynn *Staff *Flynn
CS194/EE281 *Peterson *Gill *Staff
CS199 (by arrangement)
CS200 Armer,Walker -- --
CS204 Knuth -- --
CS206 McCarthy -- --
CS209 (by arrangement)
CS211-1/EE381 McCluskey *Staff --
CS212-1/EE382 -- *Flynn *Staff
CS219 (by arrangement)
CS222 -- -- &Genesereth
CS223 -- &Genesereth --
CS224 -- -- Lenat
CS226 -- McCarthy --
CS227A Roth -- --
CS227B -- Binford --
CS227C -- -- Binford
CS229A -- Feigenbaum --
CS229B -- &Novak --
CS229C -- -- Lenat
CS234 Dantzig -- --
CS235/Stat227 Golub -- --
CS237A,B,C (given 83-84)
CS238A,B,C Schreiber Wilkinson Dahlquist
CS239 (by arrangement)
CS242/EE389 Hennessy -- --
CS243/EE383 -- -- Hennessy
CS244 Cheriton Tobagi --
CS245/EE484 -- Ullman --
CS246 -- -- Cheriton
CS249 (by arrangement)
CS250 (by arrangement)
CS254 Pratt -- --
CS256 -- Manna --
CS257 given 83-84 -- --
CS258 given 83-84 -- --
CS259 (by arrangement)
CS262 -- given 83-84 --
CS263A -- &Mayr --
CS263B -- -- &Mayr
CS265 -- given 83-84 --
CS266 -- -- Staff
CS267 -- -- Ullman
CS269 (by arrangement)
CS275 -- given 83-84 --
CS276 -- Staff --
CS277 Staff -- --
CS293 (by arrangement)
CS300 Staff Staff Staff
CS301 -- -- --
CS310/EE380 *Staff *Staff *Staff
CS311/EE282 *Staff -- *Staff
CS312/EE482 *Flynn
CS315/EE385A *McCluskey *McCluskey *McCluskey
CS316/EE486 *given 83-84
CS317/EE487 *given 83-84
CS318/EE488 *McCluskey
CS319/EE385 (by arrangement any quarter)
CS320 (by arrangement any quarter)
CS321 given 83-84
CS323 (by arrangement) (Any quarter)
CS327 Binford Binford Binford
CS341/EE481 *vanCleemput
CS343/EE483 -- -- *Lantz???
CS344 *given 83-84
CS345/EE385B Staff Staff Staff
CS347/EE384 *Tobagi
CS350 (by arrangement any quarter)
CS355 (by arrangement)
CS370 (by arrangement any quarter)
CS390(by arrangement any quarter)
!
Created 3/6/82 by CSD.ARMER. This page contains assignments by individual for
82-83.
Legend: ??? means that the assignment has not been made for that quarter.
"none" or "leave" means that the person plans no teaching that quarter.
An entry of "--" means that there are no plans for that person for that
quarter.
* associated with a course means that it is a EE course.
* associated with a person means that the person is in the EE
department.
# associated with a person means that the person is not regular
faculty. I.e. lecturer, teaching fellow, visitor, ...
& means a tentative assignment
Name Autumn Winter Spring Summer Notes
------------------------------------------------------------------------------
*#Allison
Baskett ??? ??? ???
Binford CS227B CS227C
CS327 CS327 CS327
Buchanan CS121A CS121B
Cheriton CS244 CS146 CS246/EE386
Dahlquist CS238C
Dantzig CS234 none none
#Ehrman CS109A
Feigenbaum CS229A CS101
Floyd CS106 CS162 CS154
CS106
*Flynn *CS193/EE183 *CS212/EE382 *CS193/EE183
*CS312/EE482
&Genesereth CS223 CS222
*Gill *CS111-1/EE181 *CS194/EE281
Golub CS235/Stat227
#Gorin CS109B
#Guibas
#Hanson,E.
*Hennessy *CS242/EE389 *CS143/EE286 CS243/EE383
Herriot Emeritus
#Hewitt
*Iyer
Knuth CS204 Research leave Research leave
Lantz CS146 CS343/EE483
Lenat CS224
CS229C
*Luckham
#Mairson
Manna CS156 CS256 CS157
*Manning *CS112/EE182
&#Mayr &CS155 CS263A CS263B
McCarthy CS206 CS226
McCluskey CS21/EE381 *CS318/EE488
*CS315/EE385A *CS315/EE385A *CS315/EE385A
Miller none none none
&Novak &CS102 &CS229B
Oliger leave leave CS137C
*Owicki
*Peterson *CS194/EE281
Pratt CS254 CS150 CS163
*Reid CS108A CS108B *CS142/EE285???
Roth CS227A
Schreiber CS238A CS137B CS135
Shortliffe CS121A CS121B ?
*Tobagi CS147 CS244 EE384
Ullman CS161 CS245/EE484 CS267
*vanCleemput *CS341/EE481
*Wakerly
Waldinger CS157
#Walker CS200
#Weyhrauch
&Wiederhold &CS145/EE287
Wilkinson none CS238B none
*Williams *CS111-1/EE181
Winograd leave leave leave
-------
∂08-Mar-82 0838 FFL
To: JMC, FFL
Your final report on NSF Basic Research IN AI is due March 3l, l982.
∂08-Mar-82 1000 JMC*
Find out about deposits.
∂08-Mar-82 1010 Kanerva at SUMEX-AIM Advise
Date: 8 Mar 1982 1012-PST
From: Kanerva at SUMEX-AIM
Subject: Advise
To: McCarthy at SU-AI
I would like to come for advice early this week.
1. Some people at C.I.T. would like to include in Context a VisiCalc-
like program by National Computer Performance, for people to figure
out budgets and the like. The price is $9,500. I am not thrilled
by the idea.
2. Is C.I.T. a healthy place to work? I am discouraged and am trying
to decide whether to continue there. The place is geared toward
managers, and understanding computers is a handicap. They don't
know what to do with you.
I will be busy parts of today and tomorrow afternoons, but mostly I am
available. I would greatly appreciate a reply form you. - Pentti
-------
I don't know about VisiCal; I suppose it simply depends on whether
there is a market for it. I don't know whether C.I.T. is a healthy
place to work; I wouldn't have assumed so. Of course, they shouldn't
be, and presumably aren't, primarily research oriented.
I will be glad to talk with you, and I suggest you phone some morning
after 10 at my home number 857-0672 or some afternoon at my office
number.
∂09-Mar-82 0806 Betty Scott <CSD.BSCOTT at SU-SCORE>
Date: 9 Mar 1982 0804-PST
From: Betty Scott <CSD.BSCOTT at SU-SCORE>
To: JMC at SU-AI
cc: CSD.BSCOTT at SU-SCORE
In-Reply-To: Your message of 5-Mar-82 1419-PST
John, I believe you said you sent the draft to Lowell Wood. Please let
me know when he replies, so I can get the report on its way.
Thanks,
Betty
zzz
-------
∂09-Mar-82 1645 FFL
To: JMC, FFL
Elliott Bloom called to remind you of dinner at the Faculty Club tonight at
6 p.m., SE2.
∂10-Mar-82 0142 RAH via S1-GATEWAY l5 convention
i think its the 2nd thru 4th
∂10-Mar-82 0847 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLLOQUIUM NOTICE WEEK OF MARCH 15 - 19, 1982
Date: 10 Mar 1982 0833-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLLOQUIUM NOTICE WEEK OF MARCH 15 - 19, 1982
To: colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/16/82 MJ301 Dr. Glen Rennels
Tuesday Medical Computing Stanford University
1:30 p.m. Journal Club ``Recent Articles of Interest''
3/16/82 MJ301 Dr. Robert Blum
Tuesday Knowledge Stanford University
2:30 p.m. Representation ``Displaying Time - Oriented Data''
Group Meeting
3/18/82 MJ301 David Matula
Thursday AFLB
12:15 p.m ``The Best Rational Arithmetic Versus Floating
Point Arithmetic for Computer Computation''
!
-------
∂10-Mar-82 1047 Ichiki at SRI-AI TINLUNCH tomorrow
Date: 10 Mar 1982 1045-PST
From: Ichiki at SRI-AI
Subject: TINLUNCH tomorrow
To: tlgrp:
Brian Smith will be TINLUNCH speaker next week. He will give a talk
at 10:30 in EK242 (abstract below), and we will have a general
discussion for TINLUNCH at 12:00.
Reflection and Semantics
in a Procedural Language
(or: 1-, 2-, 3-LISP)
We show how a computational system can be constructed to
"reason", effectively and consequentially, about its own inference
processes. First, we consider the general question of computational
semantics, rejecting traditional approaches, and arguing that the
declarative and procedural aspects of computational symbols (what
they stand for, and what behaviour they engender) should be
analysed independently, in order that they may be coherently
related. Second, we investigate self-referential behaviour in
computational processes, and show how to embed an effective
procedural model of a computational calculus within that calculus
(a model not unlike a meta-circular interpreter, but connected to the
fundamental operations of the machine in such a way as to provide,
at any point in a computation, fully articulated descriptions of the
state of that computation, for inspection and possible modification).
Our claims are three: a) our "rationalised semantics" better captures
the collective tacit understanding of computation than do previous
analyses, b) it is straightforward to add reflective capabilities to
a semantically rationalised formalism, and c) the architecture we
adopt will support arbitrary computational processes able to shift
smoothly between dealing with a given subject domain, and
dealing with their own reasoning processes over that domain.
An instance of the general solution is worked out in the
context of an applicative language. Specifically, we present three
successive dialects of LISP: 1-LISP, a distillation of current practice,
for comparision purposes; 2-LISP, a dialect constructed in terms of
our rationalised semantics, in which the concept of evaluation is
rejected in favour of independent notions of simplification and
reference, and in which the respective categories of notation,
structure, semantics, and behaviour are strictly aligned; and 3-LISP,
an extension of 2-LISP primitively endowed with reflective powers.
-------
∂10-Mar-82 1534 Gene Golub <CSD.GOLUB at SU-SCORE>
Date: 10 Mar 1982 1532-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
To: JMC at SU-AI
In-Reply-To: Your message of 10-Mar-82 1229-PST
Thanks . Is he related to Alan? GENE
-------
I have no idea, but since he's 38, maybe he's too old to be a son.
As you may remember, there was a numerical analyst named Sam Perlis,
I believe also at CMU, perhaps a relative of that one.
∂10-Mar-82 1704 Gene Golub <CSD.GOLUB at SU-SCORE> [Dennis Allison <CSL.LAB.DRA at SU-SCORE>: Gary Lindstrom]
Date: 10 Mar 1982 1701-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [Dennis Allison <CSL.LAB.DRA at SU-SCORE>: Gary Lindstrom]
To: faculty at SU-SCORE
Date: 10 Mar 1982 1656-PST
From: Dennis Allison <CSL.LAB.DRA at SU-SCORE>
Subject: Gary Lindstrom
To: golub at SU-SCORE
Gary Lindstrom from University of Utah will be in town 3/19 and would be
interested in giving a talk to some group (and picking up some small travel
support if we could offer it). He has an interesting talk on parallel
alpha-beta search. Any takers. (It is a rather poor time given SU's schedules
but that's when he is available....)
dra
-------
-------
∂10-Mar-82 1844 Gene Golub <CSD.GOLUB at SU-SCORE> Faculty appointments
Date: 10 Mar 1982 1841-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Faculty appointments
To: Faculty at SU-SCORE
cc: irmgild at SU-SCORE, csd.bscott at SU-SCORE
There will be four persons whom we will consider as assistant professor
appointments at the first faculty meeting on Tuesday, March 30.
They are M. Genesereth, P. Hilfinger, E. Mayr and S. Micali. Their
files are in my office and I urge those persons who are eligible
to vote to inspect these files before the meeting. GENE
-------
∂10-Mar-82 1846 Gene Golub <CSD.GOLUB at SU-SCORE> Files
Date: 10 Mar 1982 1844-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Files
To: faculty at SU-SCORE
cc: irmgild at SU-SCORE, csd.bscott at SU-SCORE
since Irmgild will be on vacation during part of the intersession, the files
will be placed in Betty Scott's office. GENE
-------
∂10-Mar-82 2043 Gene Golub <CSD.GOLUB at SU-SCORE> Shapiro
Date: 10 Mar 1982 2040-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Shapiro
To: jmc at SU-AI
Please call me at your convenience. GENE
-------
∂10-Mar-82 2114 Gene Golub <CSD.GOLUB at SU-SCORE> [Tom Dietterich <CSD.DIETTERICH at SU-SCORE>: Ehud Y. Shapiro student interview]
Date: 10 Mar 1982 2111-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [Tom Dietterich <CSD.DIETTERICH at SU-SCORE>: Ehud Y. Shapiro student interview]
To: jmc at SU-AI
Mail-From: CSD.DIETTERICH created at 6-Mar-82 18:21:42
Date: 6 Mar 1982 1821-PST
From: Tom Dietterich <CSD.DIETTERICH at SU-SCORE>
Subject: Ehud Y. Shapiro student interview
To: Shapiro-interviewers: ;, csd.golub at SU-SCORE, csd.ullman at SU-SCORE,
csl.lantz at SU-SCORE, csl.jlh at SU-SCORE, tob at SU-AI, rwf at SU-AI
cc: CSD.DIETTERICH at SU-SCORE, csd.jf at SU-SCORE
Group student interview with Ehud Shapiro:
People present:
Rich Pattis (REP)
Michael Kenniston (csd.msk)
Yonatan Malachi (YM)
Ken Clarkson (KLC)
Dave Smith (csd.smith)
Tom Spencer (THS)
Tom Dietterich (csd.dietterich)
Shapiro conducted himself well in answering some pretty tough
questions including many detailed questions about his work. The
students present were generally luke warm about him, however. There
was a sense that he was a quick-minded person who would do well in
whatever problems he attacked. However, his area of interest seemed
to be quite narrowly defined to be PROLOG and logic programming. In
response to the question about what he would do if he came to
Stanford, he answered that he would leave debugging for a while and
instead concentrate on parallel execution of PROLOG programs and on
applications of PROLOG to distributed computing. He also expressed a
desire to collaborate with the Japanese Fifth Generation project
(which emphasizes logic programming). This and subsequent answers
gave us the impression that he was more interested in PROLOG than in
any "computer science" problems. In fact, we were not sure exactly
where he would fit within the department. He stated that he is not an
AI person and disagrees with the current AI methodology. On the other
hand, he is not really an MTC person. Nor could we really consider
him a programming languages person. His primary asset for this
department would be that he would be able to teach PROLOG.
One area of particular concern was his answer to the question about
teaching. He indicated that he has never taught and that he was not
very enthusiastic about teaching, although he would "probably try and
do well" on any teaching assignment he was given. He said he would
not want to teach a "core AI" class.
Summary: no one felt particularly strongly that we should (or should
not) hire him. Our main question was what he would bring to Stanford.
We were somewhat concerned that he would not put much energy into
teaching.
--Tom
-------
-------
In my opinion, the student report on Shapiro partly reflects conservatism.
Indeed logic programming does not fit into the Stanford curriculum as in
now exists. However, I think it is an important new direction for computer
science. I am not too enthusiastic about Shapiro's idea of concentrating
on distributed computing, however. But then he made something better out
of debugging than I thought possible.
I have asked Ed Feigenbaum to help me urge reconsideration on Shapiro with
the following:
Gene has eliminated Shapiro, and I think this is a big mistake. The lack
of student enthusiasm seems to reflect a conservative attitude. Shapiro's
interests do seem special given what is currently being taught at Stanford,
but this is how any new field looks at the beginning from the outside. I
think Shapiro has a remarkable number of very wide ranging and rather clear
ideas. I expressed some doubts to him about his interest in distributed
logic programs and got a very cogent reply which I can send you if you
like. I won't be at the March 31 meeting, so I wonder if you are inclined
to urge a reconsideration. I would bet that Stanford is going to look very
silly a few years from now if we have missed a chance to get Shapiro.
∂11-Mar-82 0759 Kanerva at SUMEX-AIM Thanks for the talk
Date: 11 Mar 1982 0801-PST
From: Kanerva at SUMEX-AIM
Subject: Thanks for the talk
To: McCarthy at SU-AI
I cannot say that things are any clearer to me. There certainly is
enough work to do here, so perhaps for now it is best to do it and
not think of other things. - Pentti
-------
∂11-Mar-82 1124 Ehud Shapiro <Shapiro at YALE> Thanks and parallel Prolog
Date: 11-Mar-82 1422-EST
From: Ehud Shapiro <Shapiro at YALE>
Subject: Thanks and parallel Prolog
To: Jmc at SU-AI
cc: Shapiro at YALE
Thanks for your hospitality. I found Stanford an exiting place, but
also quite intense. I still don't know if I what I felt was the tension
around the job interview, or the "objective" nature of the place itself.
I received your letter as I came back to Yale. There are several
reasons why to investigate distributed/parallel Prolog (I do not
distinguish between the two, because I think that at the right level
of abstraction they do not differ). One is that parallel Prolog can
be executed faster on a multi-processor machine, such as the one the
Japanese are planning to build. The other is that I think that a
variant of Prolog may be a very useful for implementing distributed
applications. It took may a while to realize this, but the more I
think about it the more excited I become: in distributed/parallel
Prolog a goal corresponds to a process, and a shared variable between
two goals is a communication channel between processes.
For example, here is how you write a multiplexer, merge(Xs,Ys,Zs), that
merges the two input streams Xs and Ys into one output stream Zs:
merge([],[],[]).
merge([X|Xs],Ys,[X|Zs]) :- merge(Xs,Ys,Zs).
merge(Xs,[Y|Ys],[Y|Zs]) :- merge(Xs,Ys,Zs).
Clearly, Prolog's execution mechanism won't do here: this program is
truly nondeterministic, and the right clause should be invoked as soon
as the input arrives on the channel it reads from. Also, more stringent
assignment of "modes" to arguments should be done, to avoid
multiple-writers conflicts.
Clark and Gregory, in their paper in the ACM conf. on functional
language and machine architecture (from several months ago), give an
example of a simple operating system, with two screens, two keyboards
and one monitor, based on this style of handling streams.
It seems that that the style of programming that results from this
approach is (I think) called object oriented. Consider, for example,
the following programs for stack and queue manipulation. The stack
program receives messages of stack operations, performs the operation
and return the result.
% stack(L,Xs) :- evaluate the stream of messages L on stack Xs.
stack([],←).
stack([pop(X)|L],[X|Xs]) :- stack(L,Xs).
stack([push(X)|L],Xs) :- stack(L,[X|Xs]).
The initialization is: stack(L,[]), where L is the stream of messages.
Although stack is a "global" data structure, the program is functional,
with no explicit side-effects. The side-effect are done to the
variable owened by the process "stack".
The answers to the messages are returned in the variable X for pop(X).
push(X) is just executed, with no response. This can easily be modified
if there is a need for confirmation. For example, one can add the
clause
stack([push(X,ok)|L],Xs) :- stack(L,[X|Xs]).
Which will respond 'ok' in X to the message push(a,X), when the message
received and successfully processed.
It may seem forbiding to invoke a new process for every stack operation
(and, theoretically that's what the stack program does), but here a
new concept comes to our help: tail-recursion optimization of
sequential Prolog maps into "process-optimization" in distributed
Prolog: instead of invoking a new process with the new stack, process
optimization will retain the old process, but will modify its variables,
i.e. its "state", and thus induce an implict side-effect.
Although the approach needs synchronization primitives not available
in current Prolog-10, simple "distributed" programs, such as the one
for stack and queue (below) may still be executed directly in it. For
example:
| ?- stack([push(a),push(b),pop(X),push(c),pop(Y),pop(Z),push(d),pop(W)],[]).
W = d,
X = b,
Y = c,
Z = a
yes
Queues can be implemented with the same ease. Below an example of a queue,
simulated with a "lazy list".
% queue(L,Qhead,Qtail) :- evaluate the list of messages L on queue
% whose head is Qhead and tail is Qtail.
queue([],←,←).
queue([add(X)|L],Qhead,[X|Qtail]) :- queue(L,Qhead,Qtail).
queue([remove(X)|L],[X|Qhead],Qtail) :- queue(L,Qhead,Qtail).
The initial call is queue(L,Qhead,Qhead). For example:
| ?- queue([add(a),add(b),remove(X),remove(Y),add(c),remove(Z)],Q,Q).
X = a,
Y = b,
Z = c,
Q = [a,b,c|←371]
The mode should be queue(+,-,+), so trying to remove an element
from an empty queue will violate the mode and hence fail, or raise an
exception.
However, without modes, remove(X) on an empty queue
will unify X with the appropriate element of the queue when
enough add operations will be done. For example:
| ?- queue([add(a),remove(X),remove(Y),add(c),remove(Z)],Q,Q).
X = a,
Y = c,
Z = ←110,
Q = [a,c,←110|←345]
This is a curious behavior. It means that there is no need to
synchronize add's and remove's, and any sequence of them will unify
the n'th element removed with the n'th element added, even if the
remove occured before the corresponding add.
In summary, I think that all will benefit from distributed/parallel
Prolog: it will enable a richer programming style, faster execution
of programs, and an increased range of applications that can be solved
with a higher level language.
-- Udi.
-------
∂11-Mar-82 1632 Gene Golub <CSD.GOLUB at SU-SCORE> [Buchanan at SUMEX-AIM: statement on Adjunct Professors in the CSD -- from EAF & me]
Date: 11 Mar 1982 1626-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [Buchanan at SUMEX-AIM: statement on Adjunct Professors in the CSD -- from EAF & me]
To: faculty at SU-SCORE
This is a statement which I would like our Faculty endorse. We can
discuss this at our March 30 meeting. Gene
---------------
Mail-from: ARPANET site SUMEX-AIM rcvd at 20-Feb-82 1015-PST
Date: 20 Feb 1982 0944-PST
From: Buchanan at SUMEX-AIM
Subject: statement on Adjunct Professors in the CSD -- from EAF & me
To: csd.golub at SU-SCORE
cc: feigenbaum at SUMEX-AIM
The faculty of the Computer Science Department (CSD) recognize
the contribution of Adjunct Professors to all facets of education
and research at Stanford and would like you to know that we
support the continuation of the Adjunct Professoriate.
We are proud of the fact that the CSD has been recognized as the
best in the country. We receive over 300 applications each year
for 20 places in our PhD program, and we graduate a substantial
fraction of the nation's PhD's -- all of whom are highly sought.
Because of the experimental nature of computer science, our PhD
program depends on a large research staff in the CSD. It is
imperative that we have a vigorous research program at all times,
and the regular faculty alone cannot keep large reseach projects
moving forward.
We agree strongly with the statements made by Dean Kays in the
Campus Report about the programmatic need for adjunct faculty.
Computer Science is more like an Engineering discipline in much of
its research than like the Humanities. The views of Dean Wessells,
as expressed in the Campus Report, are not our own.
The faculty, as well as students, in the CSD benefit by having
Adjunct Professors contributing to the research, publications, and
seminars of the Department. They thereby reduce the demands on the
other faculty, and provide our students with more exposure to
experimental work in Computer Science. But they also provide the
continuity in our research programs that regular faculty cannot
provide because of their numerous other commitments.
As long as the standards under which Adjunct Professors are selected
are as high in practice as in University policy, the education and
research sides of computer science are enhanced with no detrimental
effects. It would be harmful to Computer Science at Stanford to
reduce their numbers or their effectiveness.
-------
-------
∂11-Mar-82 1655 Gene Golub <CSD.GOLUB at SU-SCORE> next Faculty meeting
Date: 11 Mar 1982 1649-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: next Faculty meeting
To: faculty at SU-SCORE
cc: public: ;
The next faculty meeting will be on Tuesday, March 30 at 2:30.
Please send me any agenda items you wish to be considered.
I'll expect all committee chairman to make a report.
We will vote on a number of faculty appointments.
GENE
-------
∂12-Mar-82 0852 FFL
To: JMC, FFL
Guang Zhong called to ask if there were any new information about her husband,
Victor Kao.
∂12-Mar-82 1320 Gene Golub <CSD.GOLUB at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
Date: 12 Mar 1982 1316-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [John McCarthy <JMC at SU-AI>:]
To: RECRUITMENT: ;
cc: jmc at SU-AI, feigenbaum at SUMEX-AIM
Perhaps we should reconsider but I must say his lack of interest in
teaching bothers me greatly. GENE
---------------
Mail-from: ARPANET site SU-AI rcvd at 11-Mar-82 2225-PST
Date: 11 Mar 1982 1924-PST
From: John McCarthy <JMC at SU-AI>
To: csd.golub at SU-SCORE
I have asked Ed Feigenbaum to help me urge reconsideration on Shapiro with
the following:
Gene has eliminated Shapiro, and I think this is a big mistake. The lack
of student enthusiasm seems to reflect a conservative attitude. Shapiro's
interests do seem special given what is currently being taught at Stanford,
but this is how any new field looks at the beginning from the outside. I
think Shapiro has a remarkable number of very wide ranging and rather clear
ideas. I expressed some doubts to him about his interest in distributed
logic programs and got a very cogent reply which I can send you if you
like. I won't be at the March 31 meeting, so I wonder if you are inclined
to urge a reconsideration. I would bet that Stanford is going to look very
silly a few years from now if we have missed a chance to get Shapiro.
-------
As you know, he gives an excellent lecture. I will call him and explore
further the teaching question. In my opinion, he will in fact be one of
our better teachers, because he is so well organized. If his teaching
proves unsatisfactory, then we don't have to re-appoint.
∂12-Mar-82 1341 Gene Golub <CSD.GOLUB at SU-SCORE> Yao appointment
Date: 12 Mar 1982 1340-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: Yao appointment
To: Full-Professors: ;
The Search committee has unanimously approved the appointment
of Andy Yao to a full professorship. I would like to move on this
as quickly as possible though Andy will be at IBM next year.
There will be a meeting of the full professors on Thursday
March 18 in the conference room next to my office. I will
be away then so Jeff Ullman will chair the session. The
file will be in Betty Scott's office.
I will vote yes on this appointment. If you cannot be here,
please let Jeff know your wishes. GENE
-------
∂12-Mar-82 1526 csl.jlh at SU-SCORE (John Hennessy) Re: [John McCarthy <JMC at SU-AI>:]
Date: 12 Mar 1982 1526-PST
From: csl.jlh at SU-SCORE (John Hennessy)
Subject: Re: [John McCarthy <JMC at SU-AI>:]
To: CSD.GOLUB at SU-SCORE
cc: jmc at SU-AI, feigenbaum at SUMEX-AIM
In-Reply-To: Your message of 12-Mar-82 1316-PST
I also believe we should reconsider especially since the AI
people claim to be advisor poor and teaching heavy.
-------
∂12-Mar-82 2123 RAH via S1-GATEWAY saturday visit
Lowell got back, but will be over at Hoover all day Saturday. I'll
be around though.
It looks like I'll have to be at Hoover all day tomorrow also - with
Teller. Perhaps I'll see him there.
∂13-Mar-82 1147 TOB
John
I am interested in the reply you mention below. I am interested in
discussing Shapiro if you want.
Tom
∂12-Mar-82 1320 Gene Golub <CSD.GOLUB at SU-SCORE> [John McCarthy <JMC at SU-AI>:]
Date: 12 Mar 1982 1316-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [John McCarthy <JMC at SU-AI>:]
To: RECRUITMENT: ;
cc: jmc at SU-AI, feigenbaum at SUMEX-AIM
Perhaps we should reconsider but I must say his lack of interest in
teaching bothers me greatly. GENE
---------------
Mail-from: ARPANET site SU-AI rcvd at 11-Mar-82 2225-PST
Date: 11 Mar 1982 1924-PST
From: John McCarthy <JMC at SU-AI>
To: csd.golub at SU-SCORE
I have asked Ed Feigenbaum to help me urge reconsideration on Shapiro with
the following:
Gene has eliminated Shapiro, and I think this is a big mistake. The lack
of student enthusiasm seems to reflect a conservative attitude. Shapiro's
interests do seem special given what is currently being taught at Stanford,
but this is how any new field looks at the beginning from the outside. I
think Shapiro has a remarkable number of very wide ranging and rather clear
ideas. I expressed some doubts to him about his interest in distributed
logic programs and got a very cogent reply which I can send you if you
like. I won't be at the March 31 meeting, so I wonder if you are inclined
to urge a reconsideration. I would bet that Stanford is going to look very
silly a few years from now if we have missed a chance to get Shapiro.
-------
∂13-Mar-82 1935 Oded Anoaf Feingold <OAF at MIT-MC> energy digest #3005.5
Date: 13 March 1982 22:31-EST
From: Oded Anoaf Feingold <OAF at MIT-MC>
Subject: energy digest #3005.5
To: ENERGY-DSN at MIT-MC
A NEW TYPE OF NUCLEAR INITIATIVE IN IDAHO
aluminum power to the people?
Nuclear power cheaper than fossil, evidence at last!
----------------------------------------------------------------------
ES@MIT-MC 01/26/82 22:58:59
Re: A NEW TYPE OF NUCLEAR INITIATIVE IN IDAHO
To: ENERGY at MIT-MC
ACCORDING TO THE FEBRUARY 1982 "ACCESS TO ENERGY", IDAHO WILL BE
VOTING ON A NEW TYPE OF NUCLEAR INITIATIVE: NO LAWS PROHIBITING NUCLEAR
POWER UNLESS APPROVED IN A PLEBISCITE. CONTRIBUTIONS CAN BE SENT TO
CITIZENS FOR ENERGY AND ENVIORNMENT, BOX 575, MERIDIAN, ID 83642.
------------------------------
Date: Thursday, 4 February 1982 16:46-PST
From: KING at KESTREL
Subject: aluminum power to the people?
To: energy at mit-mc
cc: King at KESTREL
I have a half-baked idea.
Lawrence Livermore Labs claims to be on the verge of an
aluminum-air battery that would have a range of 1500 miles in an
electric car, on 71 lb. of aluminum. At current prices, this aluminum
would cost about $55, and it would provide about 500 KWH using
reasonable estimates for the energy efficiency of the car.
My idea is to consider an aluminum battery for the power
source for a house. Peak demand for a house is certainly less then
that for a car. Even a house with electric resistance heat would only
need about 710 lb. of aluminum per month over the coldest winter
months, and would produce about three times that amount of sludge.
The cost of aluminum for a user who is returning sludge (aluminum
hydroxide, I believe) should be a lot less then the 65-70 cents per
poung cost of structural aluminum (which is what I was using, having
no other figures). I suspect that structural aluminum is an expensive
way to buy aluminum NOW, and that it would already be cheaper to buy
just plates of unalloyed and rough-cast aluminum.
Of course, any establishment not big enough to economically
generate its own electricity with its own prime mover could use the
same arrangement.
Every so often (more often during the winter for an electric
heat customer) a truck would come around and service your battery.
(An automatic device that senses battery condition and calls the
company seems not too farfetched.) Electric companies could forget
about peak demand, solar-generated or base-generated electricity could
be used to best advantage. Electricity would cease to be a natural
monopoly. There would not be big power failures leading to huge areas
of New York being burned out. (Individuals might have spare batteries
or pairs of neighbors could make agreements.)
Comments, anyone?
------------------------------
REM@MIT-MC 02/11/82 21:45:39
Re: Nuclear power cheaper than fossil, evidence at last!
To: ENERGY at MIT-MC
On news on TV a minute ago (as I start typing this in)... SMUD, a utility
district in Sacramento County (California), has electricity rates only a third
of PG&E's rates right across the street, because SMUD gets its power from cheap
hydroelectric and nuclear (Rancho Seco) whereas PG&E gets its power from
expensive fossil fuels. (Who are all the people who claim again and again that
nuclear power is too expensive? Do they have any rebuttal to this data?) --
This got on TV because the PG&E customers are now circulating a petition to get
removed from PG&E's service area and annexed to SMUD instead.
END OF ENERGY DIGEST
********************
∂14-Mar-82 1423 Tom Dietterich <CSD.DIETTERICH at SU-SCORE> [Tom Dietterich <CSD.DIETTERICH at SU-SCORE>: Re: [John McCarthy <JMC at SU-AI>:]]
Date: 14 Mar 1982 1359-PST
From: Tom Dietterich <CSD.DIETTERICH at SU-SCORE>
Subject: [Tom Dietterich <CSD.DIETTERICH at SU-SCORE>: Re: [John McCarthy <JMC at SU-AI>:]]
To: jmc at SU-AI
cc: csd.feigenbaum at SU-SCORE
John,
Here is my response to your message about Shapiro.
--Tom
---------------
Mail-From: CSD.DIETTERICH created at 14-Mar-82 13:53:56
Date: 14 Mar 1982 1353-PST
From: Tom Dietterich <CSD.DIETTERICH at SU-SCORE>
Subject: Re: [John McCarthy <JMC at SU-AI>:]
To: csd.golub at SU-SCORE, csd.ullman at SU-SCORE, csl.lantz at SU-SCORE,
csl.jlh at SU-SCORE, csd.jf at SU-SCORE, tob at SU-AI, rwf at SU-AI
cc: CSD.DIETTERICH at SU-SCORE
In-Reply-To: Your message of 12-Mar-82 1316-PST
In response to John McCarthy's motion to reconsider, I'd like to make
some detailed remarks about Shapiro.
Speaking both as a student and as an AI researcher in the area of
inductive inference, I must say that I was very disappointed with
Shapiro. When I saw him at IJCAI, I was impressed with his clarity
and style of presentation, but closer examination of his work shows it
to contain few, if any, new ideas. The idea of doing induction by
exhaustive search of the space of all possible theories is certainly
not a very new--or interesting--idea. The work of Stanford's Tom
Mitchell is certainly superior. Slightly more interesting is his
method for intelligently selecting questions to ask of the user, but
there are few real world domains where this approach would be
applicable. In the context of grammatical inference, for example,
Shapiro's algorithm requires that the language user be able to state
the grammaticality of any sentential form (including forms containing
non-terminals). As another example, in the area of hardware debugging
Shapiro's algorithm would require that EVERY part of the device be
observable. The DART project here at Stanford takes the much more
realistic approach of assuming that not all parts of the device are
visible, and hence, the inputs of the device must be manipulated to
drive it into the appropriate state where the failure is exhibited.
Shapiro's "observable state" requirement makes it impossible to apply
his induction algorithm to most problems in scientific inference,
since virtually all science is limited in its ability to examine the
intermediate states of the phenomena of interest.
Aside from the merits of his previous work, I am concerned about his
research methodology. He does not seem to be driven by problems, but
rather by his desire to apply logic programming. He does not consider
himself an AI researcher, and he is not concerned with how we might
improve our methods for problem solving, planning, learning, etc.
Instead, his research is tool-based. Now it is important for computer
science to have toolsmiths--the whole systems area involves building
better tools--but Shapiro does not seem to fit into the systems area
either. He claims not to be a systems person, and his remarks about
applying PROLOG to distributed computing bear this out. Again, it
appears that his chief concern is not "How can we do distributed
computing better?" but "How can I apply PROLOG is this general area
of computer science?" If he follows his present course, I doubt that
he will have much impact on computer science.
In conclusion, I think his lack of outstanding research work, his
questionable research methodology, and his expressed lack of interest
in teaching all indicate that we should not hire him at this time.
--Tom
-------
-------
∂14-Mar-82 1925 Solomon at MIT-MULTICS (Richard Jay Solomon) account on SAIL
Redistributed-Date: 14 March 1982 22:26 est
Redistributed-By: Solomon.Datanet at MIT-MULTICS
Redistributed-To: jmc at SU-AI
Date: 14 March 1982 22:26 est
From: Solomon at MIT-MULTICS (Richard Jay Solomon)
Subject: account on SAIL
Sender: Solomon.Datanet at MIT-MULTICS
To: me at SU-AI
cc: reg at SU-AI, Neuman.ComRes at MIT-MULTICS,
Pool.Datanet at MIT-MULTICS,
Solomon.Datanet at MIT-MULTICS (hold.sv)
Acknowledge-To: Solomon.Datanet at MIT-MULTICS
We received your note about a $1000 deposit for use of SAIL a few weeks
ago. Unfortunately, since we only need access to one quad, which I
remeber is about $30 per month, we cannot justify prepaying for access
to SAIL for some 33 months in advance. We would be happy to prepay six
months in advance if such arrangements can be made.
∂14-Mar-82 2009 Lowell Wood <LLW at S1-A> Missed Connections
Date: 14 Mar 1982 2005-PST
From: Lowell Wood <LLW at S1-A>
Subject: Missed Connections
To: jmc at SU-AI
CC: LLW at S1-A, RAH at S1-A
∂09-Mar-82 0047 John McCarthy <JMC at SU-AI> coming out Saturday
Date: 09 Mar 1982 0049-PST
From: John McCarthy <JMC at SU-AI>
Subject: coming out Saturday
To: llw at S1-A
If it is convenient, I would come out Saturday for Shackleton and S-1.
It is a convenient time for Rod, and I would come out around 1pm and
stay into the evening.
[John, I'm sorry, but I didn't see this until now. When else would be good
for you? Lowell]
∂14-Mar-82 2058 CLT
I'd rather do taxes Tuesday evening if thats ok.
I have a meeting with Schwictenberg Tuesday and
I'll probably want to continue preparing on Monday evening.
∂14-Mar-82 2323 Lowell Wood <LLW at S1-A> Aluminum *Energy* Costs
Date: 14 Mar 1982 2034-PST
From: Lowell Wood <LLW at S1-A>
Subject: Aluminum *Energy* Costs
To: jmc at SU-AI
CC: LLW at S1-A
∂13-Mar-82 2046 John McCarthy <JMC at SU-AI> aluminum power to the people?
Date: 13 Mar 1982 2042-PST
From: John McCarthy <JMC at SU-AI>
Subject: aluminum power to the people?
To: llw at S1-A
Can this be??
Date: Thursday, 4 February 1982 16:46-PST
From: KING at KESTREL
Subject: aluminum power to the people?
To: energy at mit-mc
cc: King at KESTREL
I have a half-baked idea.
Lawrence Livermore Labs claims to be on the verge of an
aluminum-air battery that would have a range of 1500 miles in an
electric car, on 71 lb. of aluminum. At current prices, this aluminum
would cost about $55, and it would provide about 500 KWH using
reasonable estimates for the energy efficiency of the car.
My idea is to consider an aluminum battery for the power
source for a house. Peak demand for a house is certainly less then
[John: Basically, yes. However, it's marginal economically--the clown
didn't consider capital costs (Standard Blunder #4). Lowell]
∂14-Mar-82 2355 Feigenbaum at SUMEX-AIM confidential
Date: 14 Mar 1982 2354-PST
From: Feigenbaum at SUMEX-AIM
Subject: confidential
To: jmc at SU-AI
John,
Re Shapiro:
I have a very high respect for your views on where to go in AI
research, so I would be glad to join you in pressing for a
reconsideration of the Shapiro application.
My personal view (which I have expressed to nobody, and express
here for the first time) is that Shapiro impressed me as
brilliant but narrow as a computer scientist. He seemed to be
interested in nothing but Prolog--not even in the AI issues it
would help address. My preference in computer scientists is more
along the lines of the McCarthy model, i.e. interested and
creative along a fairly broad front of computer science.
Perhaps, if it is true that he has a disdain for teaching, the
job-title of Research Associate is better than Assistant
Professor. If he were to accept that job, I would be willing to
pick up half of his research salary. In fact, since I am
currently writing my ARPA renewal, I would be willing to ask
ARPA to pick up the entire salary (dont know if they would go
for it, but I could try).
Final note: I too will not be able to be at that March 30
faculty meeting, so can only help press for reconsideration
before the fact, so to speak.
Let me know what you think of all this.
Ed
-------
∂15-Mar-82 1440 AMAREL at RUTGERS Japanese AIers
Date: 15 Mar 1982 1738-EST
From: AMAREL at RUTGERS
Subject: Japanese AIers
To: JMC at SU-AI
cc: Amarel at RUTGERS
John, As you may know, I am General Chairman of IJCAI-83 (which is to
be held in Karlsruhe, W Germany); and I am now trying to assemble an
Exec Committee for the Conference. I would like to have a representative
from Japan, and the names of Fuchi (member of main committee for 5th genera
tion computer) and Nagao were suggested to me. I would appreciate having
your views about them, and any ideas you may have about other candidates.
Thanks, Saul
-------
∂15-Mar-82 1442 Don Walker <WALKER at SRI-AI> Westinghouse Science Talent Search Winner
Date: 15 Mar 1982 1157-PST
From: Don Walker <WALKER at SRI-AI>
Subject: Westinghouse Science Talent Search Winner
To: tlgrp: ;
The following item appeared in the New York Times Week in Review Section
on March 7th:
Doublespeak
Intrigued by the symbols used to represent ideas, Reena Beth Gordon
began a project last july to determine how English was understood when
spoken. Relying on research in mathematical linguistics, the 16-year-old
senior at Midwood High School in Brooklyn designed a system that predicted
how ambiguous adverb-verb sentence structures should be resolved. Last
week she won the first place award, a $12,000 scholarship, in the
Westinghouse Science Talent Search, one of the most prestigious competitions
for teen-aged scientists. Miss Gordon will attend Harvard in the fall.
Anyone have any further information about it?
-------
∂15-Mar-82 1501 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Dennis Heimbigner
Date: 15 Mar 1982 1128-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Dennis Heimbigner
To: faculty at SU-SCORE
cc: csd.bscott at SU-SCORE
Dennis Heimbigner will be interviewing for a faculty position on April 2.
Please let me know by Wednesday, March 17 whether you are interested in
talking with him.
Thanks,
Irmgild
-------
∂15-Mar-82 1538 Ichiki at SRI-AI 3/18 TINLUNCH meeting
Date: 15 Mar 1982 1539-PST
From: Ichiki at SRI-AI
Subject: 3/18 TINLUNCH meeting
To: tlgrp:
TINLUNCH meeting 3/18/82
By: Mabry Tyson
In: Conference Room EK242
Chapter out of Mabry's dissertation,
"APRVR: A Priority-Ordered Agenda Theorem Prover"
This week's Tinlunch will be on a chapter out of my dissertation,
"APRVR: A Priority-Ordered Agenda Theorem Prover" which deals with the
specifics of the agenda mechanism used. In particular, the chapter
discusses two techniques that were possible using agendas that were
not possible in an earlier recursive theorem prover. One is an
improvement in the handling of splitting on conjunctive goals while
the other is the use of non-local case splits. If certain conditions
are found in trying to prove some (sub)goal, APRVR will examine the
ancestors of that goal and find an appropriate ancestor on which to
attempt an indicated case-split.
Also included in the handout will be a small section describing one of
the theorems proved and a description of the proof.
-------
∂15-Mar-82 1629 FFL
To: JMC, FFL
Ghang Zhang (Kao) called and said she would call again Tuesday about 3:15.
∂15-Mar-82 1647 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Faculty Lunch
Date: 15 Mar 1982 1643-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Faculty Lunch
To: faculty at SU-SCORE
cc: csd.bscott at SU-SCORE
This is a reminder that there will be a faculty lunch tomorrow at 12:15
at the Boys Town conference room.
Irmgild
-------
∂15-Mar-82 2227 RPG
To: JMC at SU-AI, GHG at SU-AI
John and Gene:
Martin Griss is planning to visit the area next week, in part to
discuss sabbatical plans with HP, LLNL, and Stanford. Would either or
both of you like to talk to next tuesday (March 23)?
I, personally, would like to see him take his leave in the area,
and at Stanford in particular. Aside from seminars and the usual sabbatical
duties, I believe that he and I could put a good Lisp up on the Sun workstation
during his stay.
-rpg-
∂15-Mar-82 2241 Arthur Keller <CSD.KELLER at SU-SCORE> Comprehensive Reading List
Date: 15 Mar 1982 2236-PST
From: Arthur Keller <CSD.KELLER at SU-SCORE>
Subject: Comprehensive Reading List
To: Faculty at SU-SCORE
cc: ark at SU-AI, CSD.MWalker at SU-SCORE, CSD.Armer at SU-SCORE,
CSD.BScott at SU-SCORE
Reply-to: ARK@SU-AI
On Tuesday, 23 February 1982, the Comprehensive Committee proposed a revised
reading list that was tentatively accepted by the faculty by voice vote
pending a period of comments on the proposal. Terry Winograd, the chairman
of the Comprehensive Committee, reports that no comments were received
from faculty members. Consequently, the revised reading list will be
adopted and will be effective for the Spring 1982 exam. I will have
copies of the reading list printed and made available to the faculty and
students. If you have any comments or suggestions regarding changes
that should be made to the comprehensive exam or the syllabus, please
send them to me, and I will accumulate them and make them available to
the new Comprehensive Committee and other interested parties.
Thank you.
Arthur
-------
∂16-Mar-82 1139 LES
∂14-Mar-82 2304 JMC
What's your business address, so I can use you as clearance reference?
-------
Imagen Corp.; 12769 Dianne Drive; Los Altos Hills, CA 94022
∂16-Mar-82 1159 Paula Bauman <PJB at S1-A> PSQ
Date: 16 Mar 1982 1151-PST
From: Paula Bauman <PJB at S1-A>
Subject: PSQ
To: jmc at SU-AI
CC: PJB at S1-A
John, I will be happy to deliver the fingerprints to whomever--
I was asked if you have completed the necessary paperwork for your
Q Clearance, and if so, when and to whom were they submitted? I need
this information before anyone will accept your fingerprints here.
Paula, I don't want to officially submit the fingerprints yet, because I
haven't completed the other papework. However, ask them if they are
willing to look at the fingerprints and say whether they are technically
satisfactory or whether I have to have them done again. They have to be
clear enough for classification. If they are technically satisfactory,
then please keep them for me until I have the paperwork done. This may
take a while longer, because I have never applied for a clearance before,
and I have made many foreign trips, and it isn't easy to be sure the list
is complete.
∂16-Mar-82 1215 FFL
To: JMC, FFL
Dr. Mock of the Army Research office called. Will call you again Wednesday,
about 1:30 p.m.
∂16-Mar-82 1301 CLT Seminar at SRI
To: "@LOGIC.DIS[1,CLT]" at SU-AI
SPEAKER: David Plaisted from University of Illinois
TIME: Tuesday March 23, 10:30am
PLACE: AIC Conference Room EK242 (SRI Building E)
Visitors will be met in the lobby of building E
(host = Richard Waldinger)
Title and abstract:
Applications of Term-Rewriting Systems
Term-rewriting systems are mathematically simple structures with
many applications to computer science. This talk will discuss termination,
confluence, and partial correctness of term-rewriting systems, and their
application to theorem proving, programming languages, and program
generation. Among the areas considered will be the recursive path ordering,
the Knuth-Bendix method, rewriting strategies, normal forms,
paramodulation strategies in theorem provers, inductive proofs, and the
OBJ system. This latter system is a term-rewriting system based programming
language of comparable elegance to PROLOG, and has a uniform method of
error handling. Some open problems will be mentioned.
∂16-Mar-82 2230 LGC Advice-Taker Discussion
It turns out that next Monday or Tuesday would be a better time than this
Thursday for our first advice-taker discussion. RPG is pushing hard just now
to get some LISP done while Rod Brooks is in town, and he and I need some time
to talk before our 3-way meeting -- time which probably won't be available
till late Thursday or Friday. Would Monday or Tuesday be ok with you?
-- Lew
Let's make it Monday at 2pm to talk about the Advice Taker.
∂16-Mar-82 2235 TOB
John
Would you call Dr. Mock? Might mention TOB's long experience in stereo and
constraints for OTV (cube corners) and other special surface features,
that the proposal had very much of my input.
That we will do well in spite of Liebes's leaving and will staff.
Tom
Dr. Steve Moxk 919/549-0641 ARO
∂11-Mar-82 1213 SL ARO proposal!
Tom, hard to believe but true: Steve Mock from ARO just called to say that
he was interested in supporting the proposal we sent to him and had
written off, though more to the tune of $100k than the $200+k that we had
requested. I told him that I was leaving and that the decision was not
likely reversable. He reacted that that was too bad, as I had been a key
person in the OTV part, which seemed to interest him. He said that since
McCarthy had been PI, he wanted to confer with him, given the situation.
What if he asks me what OTV stands for?
I phoned Mock. He said that the proposal would have to be resubmitted,
so I explained about the success of the long campaign to make you an
Adjunct Professor, so we agreed that it when it is resubmitted, you should
be PI. No action is necessary until you get something in the mail, but
I think it might be worth your while to talk to him on the phone.
∂17-Mar-82 0853 FFL
To: JMC, FFL
Mark Mills called. He can be reached today at 703 691 1746.
la times, sd gas and e rates, 11.3 cents, smud 3.5 cents
full time energy consultant
∂17-Mar-82 0902 CL.BOYER at UTEXAS-20 String Search and Syllabus
Date: Wednesday, 17 March 1982 10:57-CST
From: CL.BOYER at UTEXAS-20
To: John McCarthy <JMC at SU-AI>
Subject: String Search and Syllabus
Cc: Moore at UTEXAS-20
Thank you for the note on what you teach in the theory of computation.
Concerning your suggested "improvement" to our algorithm, let me first
say that analyzing the complexity of that algorithm is over my head.
However, I will give you my intuitive perception. The key
to efficient implementation of the algorithm is getting the
inner loop of the algorithm short. As we described in our
ACM paper, we got that inner loop down to 3 instructions.
That was enough to fetch a character from the string, look
it up in the table, and decide what to do. It would be my
guess that coding the inner loop to consider search numbers
would increase the number of instructions to 4. Thus the
algorithm would run about 30% slower except for the
reduction in the initialization. But that initialization
only requires a BLT over a 128 word table. So I doubt there
would be a savings except in short searches.
∂17-Mar-82 0903 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 22 - 26, 1982
Date: 17 Mar 1982 0851-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF MARCH 22 - 26, 1982
To: colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/23/82 MJ301 Mr. Randy Teach
Tuesday Medical Computing School of Education
1:30 p.m. Journal Club ``Recent Articles of Interest''
3/23/82 MJ301 Avron Barr
Tuesday Knowledge Stanford University
2:30 p.m. Representation ``Interactive Acquisition of Knowledge about
Group Meeting Cases''
3/25/82 MJ301 Donald Johnson
Thursday AFLB U.C. Berkeley
12:15 p.m ``Solving Problems by searching Implicit
Matrices''
!
-------
∂17-Mar-82 1005 TOB
∂17-Mar-82 0137 JMC
John
OTV stands for orthogonal trihedral vertex, i.e. cube corners.
that is only a part of the proposal.
Tom
What if he asks me what OTV stands for?
∂17-Mar-82 1119 RPG
To: JMC, LGC
∂17-Mar-82 0140 JMC
To: LGC, RPG
Let's make it Monday at 2pm to talk about the Advice Taker.
This isn't good either. I have to attend some meeting at LLL that day (it's
something I've put off for months and cannot get out of).
-rpg-
∂17-Mar-82 1143 CLT truth
where are the papers by the author of the TRUTH paper I read?
I want to make a copy for SGF and also put the author on the mailing list
∂17-Mar-82 1205 CLT truth again
I remembered the name and Fran found the papers
∂17-Mar-82 1213 RPG via S1-A
Griss
∂17-Mar-82 1213 RPG via S1-A Stupid mailer
About Griss. Will you be able to talk to him tuesday (march 23) about
his sabbatical? Golub will be out of town.
-rpg-
∂17-Mar-82 1332 CLT truth again
The paper I read is missing from the Perlis file
Perhaps you could bring it to the office?
∂18-Mar-82 0843 REG
re: NYT. I agree. I'm going to leave the letter in our files to indicate
that we at least thought about it.
∂18-Mar-82 1237 Colmerauer.TextMgt at MIT-MULTICS
Date: 18 March 1982 15:38 est
From: Colmerauer.TextMgt at MIT-MULTICS
To: John McCarthy <JMC at SU-AI>
In-Reply-To: Msg of 03/18/82 13:58 from John McCarthy
test ok
Your test message also received ok. I have begun to think about alternatives,
and I'll send you another message soon.
∂18-Mar-82 1707 David Warren <WARREN at SRI-AI> Re: Colmerauer is on ARPAnet.
Date: 18 Mar 1982 1703-PST
From: David Warren <WARREN at SRI-AI>
Subject: Re: Colmerauer is on ARPAnet.
To: JMC at SU-AI
cc: WARREN at SRI-AI
In-Reply-To: Your message of 18-Mar-82 1600-PST
Many thanks for the information about Colmerauer's net address.
On a completely different matter, I am currently writing a fairly
major proposal for DEC, Tewksbury, to assist them in developing computer
architectures to support Prolog, using the VAX as a starting point.
The main SRI people involved would be me and Fernando Pereira (subject
to him accepting an offer from SRI). Digital would be very keen to
have you associated with the project, to the extent of a few days of
your time per year, to monitor the project and give them a broader
prospective on its progress (and also, I suspect, just for prestige
purposes). If you are willing to consider doing this, I would be
very grateful, and would be happy to give you more details.
--David.
PS. I will be away in Sweden all next week.
-------
It would be interesting to be associated with the VAX Prolog project.
I didn't read carefully enough, logic programming architecture is even
more interesting.
∂18-Mar-82 1736 David Warren <WARREN at SRI-AI>
Date: 18 Mar 1982 1725-PST
From: David Warren <WARREN at SRI-AI>
To: JMC at SU-AI
cc: WARREN at SRI-AI
In-Reply-To: Your message of 18-Mar-82 1720-PST
I'm very glad you're interested. I'll get in touch with you
about this when I get back from Sweden. --David.
-------
∂18-Mar-82 1835 CLT
as I just got here, I think I won't go for supper
∂19-Mar-82 0201 Peter G. Neumann <NEUMANN at SRI-AI> VERkshop III?
Date: 19 Mar 1982 0146-PST
From: Peter G. Neumann <NEUMANN at SRI-AI>
Subject: VERkshop III?
To: VERKshop: ;
Dear VERkshoppers,
As spring begins to emerge, I am getting increasingly many inquiries
about whether this April will shower upon us another VERkshop. The
answer is NO, although a VERkshop III is now possible in the fall.
The "intimacy" of VERkshops I and II seems difficult to maintain
without drastically pruning attendance, as the community of interested
parties has blossomed considerably. So, perhaps the next one will
have to have a different format. (A much smaller implementer's
meeting has been suggested for a long time now. It is certainly time
for that.) There is significant progress to report from many fronts
(despite apparent tight times in funding), and so some sort of meeting
is probably in order. (Please let me know if you have any irrevocable
time constraints in the fall, or any preferences in date and place.
It might help in planning.) And don't construe this as an inviation!
I am just trying to be helpful and to stave off a lot more queries.
More whenever.
Peter
-------
∂22-Mar-82 0657 PJH via ROCHESTER golux
good to hear from you. No., golux was never implemented. Indeed
it got caught in conceptual problems before reaching the stage of
implementation. Id be delighted to talk to you about these if you are
interested, by the way.
The correct reference is
Hayes, P. "Computation and Deduction", Proc. 3rd M.F.C.S.
Conference, 1973. Czech.Acad.Sci.
(thats Math Foundations Computer Science, by the way)
What are you working on these days? I hope to get West in August, will
you be there thw then??
Pat
∂22-Mar-82 0810 FFL
To: JMC, FFL
The number for Clark or Kowalski is 01 589 5111.
∂22-Mar-82 0939 MILTON at SRI-AI Database Seminar, Friday, 26 March, by Umesh Dayal
Mail-from: ARPANET site SRI-AI rcvd at 22-Mar-82 0903-PST
Date: 22 Mar 1982 0854-PST
From: MILTON at SRI-AI
Subject: Database Seminar, Friday, 26 March, by Umesh Dayal
To: CS345: ;
Remailed-date: 22 Mar 1982 0914-PST
Remailed-from: MILTON at SRI-AI
Remailed-to: ark-score at SU-SCORE
Remailed-date: 22 Mar 1982 0910-PST
Remailed-from: Arthur Keller <CSD.KELLER at SU-SCORE>
Remailed-to: colloq: ;
Database Seminar
Friday, 26 March 1982 at 3:15 p.m. (not 3:30)
MJH 352, Stanford
Umeshwar Dayal
Computer Corporation of America (CCA)
Title: Local Query Optimization in MULTIBASE
Abstract:
MULTIBASE is a system for integrated access to heterogeneous
distributed databases. After a brief introduction to
MULTIBASE, the talk will focus on the issue of local query
optimization. One of the tasks of MULTIBASE is to present a
high-level query interface to access path-oriented systems
such as CODASYL. the interface compilesqueries into efficient
programs for processing the queries. The principal problem
with constructing such an interface is access path optimiza-
tion. We solve this problem for the class of tree queries.
We characterize the strategies for processing a tree query;
develop a model for estimating the cost of a strategy; and
use this model to find the optimal strategy for a given query.
-------
∂22-Mar-82 1000 JMC*
roughgarden
∂22-Mar-82 1135 Jeff Rubin <JBR at S1-A>
Date: 22 Mar 1982 1126-PST
From: Jeff Rubin <JBR at S1-A>
To: jmc at SU-AI
∂22-Mar-82 0415 John McCarthy <JMC at SU-AI>
Date: 22 Mar 1982 0419-PST
From: John McCarthy <JMC at SU-AI>
To: jbr at S1-A
Try "help login" on the s1 foonly.
Thanks for pointing that out, John. I deleted the offending section. --jeff
∂22-Mar-82 1345 Sharon A. Aglito <CSD.AGLITO at SU-SCORE> Interviews with Dennis Heimbigner on April 2.
Date: 22 Mar 1982 1327-PST
From: Sharon A. Aglito <CSD.AGLITO at SU-SCORE>
Subject: Interviews with Dennis Heimbigner on April 2.
To: Buchanan at SUMEX-AIM, csd.feigenbaum at SU-SCORE, rwf at SU-AI,
csd.genesereth at SU-SCORE, jgh at SU-AI, phy at SU-AI,
csd.lenat at SU-SCORE, csd.mayr at SU-SCORE, jmc at SU-AI,
csl.ejm at SU-SCORE, jeo at SU-AI, csd.pratt at SU-SCORE, bkr at SU-AI,
wiederhold at SUMEX-AIM, tw at SU-AI
Dennis Heimbigner, who is applying for a position with this department,
will be here on April 2. Gene would like him interviewed by as many CSD
people as possible. If anyone receiving this message would like to talk
to Mr. Heimbigner for 20-30 minutes during one of the time slots listed
below, please let me know.
ll:30 - noon
2:00 - 2:30
2:30 - 3:00
3:00 - 3:30
Sharon Aglito
-------
∂23-Mar-82 0847 RPG Griss
How about 2pm? Turns out his plane leaves earlier than I thought.
-rpg-
∂23-Mar-82 0903 FFL
To: JMC, FFL
Carl Jansson, the man from Sweden who called about seeing you when he
is at Stanford today, is coming from San Francisco to see you about l p.m.
He called this morning. You told me you would see him this afternoon.
∂23-Mar-82 1220 Don Walker <WALKER at SRI-AI> David McDonald, Thursday, 3-25, at 10:30 and for TINLUNCH
Date: 23 Mar 1982 0904-PST
From: Don Walker <WALKER at SRI-AI>
Subject: David McDonald, Thursday, 3-25, at 10:30 and for TINLUNCH
To: tlgrp: ;
cc: bboard at SRI-AI
Talk in EK242 at 10:30, Thursday, 25 March, followed by TINLUNCH discussion.
There just might be some papers available for reading--if the USPO delivers
today; will notify if so.
"Natural Language Generation as a Computational Problem"
David D. McDonald
University of Massachusetts at Amherst
ABSTRACT
For my Ph.D. thesis I developed a "linguistics engine" that generates
a coherent natural language text as the realization of the goals concepts,
and relations (collectively called a "message") passed to it by a planning
system. This program, MUMBLE, is the first of its kind to be independent
of the choice of knowledge representation used in the planning system
and has worked from "messages" given in the predicate calculus, FRL,
KL-ONE, and PLANNER-style assertions. It embodies several psychologically
plausible limitations on its computational power: strictly left-to-right
production of texts, linguistically motivated limitations on the examinable
buffer, indelible decisions (no backtracking), and a structural distinction
between the treatment of function words and content words. It is a
data-directed system, driven first by the message to be expressed (rather
than its grammar), and then by the linguistic structure of the text
planned for production. This explicit representation of the structure
of the text and its alternative realizations allows the rules of the
grammar to be exceptionally flexible and general.
The talk will describe MUMBLE's architecture and the motivation behind
it, and will try to give a feeling for how a planner would use it drawing
from the example domain that is presently running on the Lispmachines.
-------
∂23-Mar-82 1441 Perlis.EE at UDel-Relay faculty position
Date: 23 Mar 82 17:23:09-EST (Tue)
From: Perlis.EE at UDel-Relay
To: jmc at Su-Ai
cc: lantz at Su-Score
Subject: faculty position
Via: UDel-EE; 23 Mar 82 17:33-EST
Via: UDel; 23 Mar 82 17:40-EDT
Just a reminder to pass on my resume to the
search committee; Keith Lantz says he hasn't
seen it yet.
Incidentally, I've been working
on an application of my truth definition to
program semantics--essentially an approach
to combinatory logic in which functions can
apply to a universal domain which includes them
as elements. So far it seems to be doing well.
This isn't surprising, since Russell's Paradox
has been noted before as the culprit behind
functional semantics. I'll send you a copy
of whatever I get written on this.
∂23-Mar-82 2105 CLT question
do you know if EIK vol. 16 (1980) might refer to a russian journal
and if so which one?
I don't know. Is there any other information?
Can't help then. Librarians might be able to help.
∂23-Mar-82 2143 CLT
Elektronische Informationsverarbeitung und Kybernetik
∂24-Mar-82 1113 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF MARCH 29 - APRIL 2, 1982
Date: 24 Mar 1982 1101-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF MARCH 29 - APRIL 2, 1982
To: colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
3/30/82 Jordan 041 Terry Winograd
Tuesday Computer Science Stanford University
4:15 p.m. Colloquium ``A Language for the Specification of Discrete
Temporal Systems''
3/31/82 MJ352 Oussama Khatib \&\ John Craig
Wednesday Robotics Seminar Stanford University
12 p.m. ``Manipulator Dynamics and their use in Control''
4/1/82 MJ301 J. Makowsky
Thursday AFLB Technion
12:15 p.m ``A Unifying Approach to Specification of
Abstract Data Types''
4/2/82 MJ352 Dennis Heimbigner
Friday Database Research University of Southern California ECL
3:30 p.m. Seminar ``A Federated Architecture for Database Systems''
!
-------
∂24-Mar-82 1219 LGC On Sortal Predicates in Quantifiers
To: JMC
CC: RPG
There's an important consideration in favor of including sortal predicates
in quantifiers that I forgot to mention in the discussion last Monday. It's
that quantifiers such as 'many', 'most', 'almost all' '3/4 of', etc., which
do play a role in common-sense reasoning (though not in mathematical logic),
REQUIRE a sortal predicate for their correct formal representation.
And it seems worthwhile to have a quantifier construction that is general
enough from the beginning to handle such quantifiers naturally.
∂24-Mar-82 1357 TOB
John
Thanks for calling.
Tom
mail mas82 1144 JMC
I phoned Mock. He said that the proposal would have to be resubmitted,
so I explained about the success of the long campaign to make you an
Adjunct Professor, so we agreed that it when it is resubmitted, you should
be PI. No action is necessary until you get something in the mail, but
I think it might be worth your while to talk to him on the phone.
∂25-Mar-82 1000 JMC*
get shapiro co-ordinates
∂25-Mar-82 1100 JMC*
take stuff out of bag
∂25-Mar-82 1101 Carolyn Tajnai <CSD.TAJNAI at SU-SCORE> Forum Update
Date: 25 Mar 1982 1054-PST
From: Carolyn Tajnai <CSD.TAJNAI at SU-SCORE>
Subject: Forum Update
To: CSD-Faculty: ;
This week has been a good week for the Forum. On Monday Mike Flynn brought
in International Computers Ltd, and today Jim Clark brought in Shell
development.
Membership now is 31.
Keep up the good work!
Carolyn
-------
∂25-Mar-82 1634 FFL
To: JMC, FFL
Cuthbert Hurd called,wanted to set a lunch date. Will call you tonight
at home, he said. 854-1901
∂26-Mar-82 0119 CSD.ULLMAN@SU-SCORE (SuNet) EQUIPMENT
Mail-from: ARPANET host SU-SCORE rcvd at 26-Mar-82 0111-PST
Date: 25 Mar 1982 1623-PST
From: Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE at SUMEX-AIM>
Subject: EQUIPMENT
To: equip at SU-SHASTA
cc: csd.golub at SU-SCORE at SUMEX-AIM, csd.bscott at SU-SCORE at SUMEX-AIM
I just got a message from Bob Kahn to the effect that our proposal has been
signed out of his office. I assume that means that we can expect some money
in a few months, and the PI's ought to get together to come up with a
concrete list totalling about $750K for fiscal 82.
-------
∂26-Mar-82 0839 RPG
∂26-Mar-82 0327 JMC
(if nil 'a 'b) works at SAIL but not at S1.
It looks like they work equally well both places.
∂26-Mar-82 0846 RPG EQUIPMENT
∂26-Mar-82 0119 CSD.ULLMAN@SU-SCORE (SuNet) EQUIPMENT
Mail-from: ARPANET host SU-SCORE rcvd at 26-Mar-82 0111-PST
Date: 25 Mar 1982 1623-PST
From: Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE at SUMEX-AIM>
Subject: EQUIPMENT
To: equip at SU-SHASTA
cc: csd.golub at SU-SCORE at SUMEX-AIM, csd.bscott at SU-SCORE at SUMEX-AIM
I just got a message from Bob Kahn to the effect that our proposal has been
signed out of his office. I assume that means that we can expect some money
in a few months, and the PI's ought to get together to come up with a
concrete list totalling about $750K for fiscal 82.
-------
Since I'm not a PI and cannot represent myself, the Advice Taker project will
need a Lisp Machine this year (or else I'll have to spend almost all my
time trying to get around it). There are already sets of formulae that
wil not fit into the 510 pages Plisp allows. I believe that Lisp Machines
are on the 1982 list, so I would like to request that you insist on a
non-Dolphin for one (or two) of them.
-rpg-
∂26-Mar-82 1000 JMC*
hurd
∂26-Mar-82 1022 FFL
To: JMC, FFL
Mrs. Kuo will be in to see you on Monday, 1:30 p.m., as you requested.
∂26-Mar-82 1445 ullman@Shasta (SuNet) meeting
Date: 26 Mar 1982 14:09:39-PST
From: ullman at Shasta
To: equip
From: ullman
Subject: meeting
Date: 26 March 1982 14:09:24-PST (Friday)
I would like to have a meeting of the committee next Friday (4/2) at
10:00AM in 252MJH (assuming that is free). The principal matter to discuss
is plans for the initial purchase of equipment, totalling about $750K.
Oh by the way, I got a call from DEC that our SUVAXes are here, and
where would we like them deleivered next week. Any ideas, gang?
∂26-Mar-82 1602 CSD.ULLMAN@SU-SCORE (SuNet) [KAHN at USC-ISI: Re: RE Compiler Project]
Mail-from: ARPANET host SU-SCORE rcvd at 26-Mar-82 1601-PST
Date: 26 Mar 1982 1521-PST
From: Jeffrey D. Ullman <CSD.ULLMAN at SU-SCORE at SUMEX-AIM>
Subject: [KAHN at USC-ISI: Re: RE Compiler Project]
To: equip at SU-SHASTA
Mail-from: ARPANET site USC-ISI rcvd at 26-Mar-82 1132-PST
Date: 26 Mar 1982 1137-PST
Sender: KAHN at USC-ISI
Subject: Re: RE Compiler Project
From: KAHN at USC-ISI
To: CSD.ULLMAN at SU-SCORE
Cc: KAHN at USC-ISI
Message-ID: <[USC-ISI]26-Mar-82 11:37:28.KAHN>
In-Reply-To: Your message of 25 Mar 1982 1616-PST
Jeff,
The answer to your question is essentially yes. I cannot
personally authorize you to spend anything - only a contracting
officer can. However, we have sent the proposal forward exactly
as you submitted it. It is possible the legal folks or
procurement types will still have questions or that we will need
some cost data as backup. Until they tell us we need to provide
it, we'll assume its not needed.
In order to release any specific amount of funds, we will of
course have to agree with you how the funds will be spent. But
we can do that on the net or anyway thats convenient.
The funds should be available in the next few months, and
certainly by summer. Probably about 1.4M between now and
December to cover the end of fy2 and fy83. The remainder will be
provided afterwards.
Congress doesnt know anything about the equipment effort yet. It
was reprogrammed in fy82. It shows up for the first time in
their budget submission in fy83 but they havent held hearings for
fy83 yet, much less issued comments. As for bobs comments, they
must have been relative to internal issues of no import outside
the office. In the category of not telling a user of a computer
that interrupts were inhibited for 30 ms while a real-time task
was completed.
Hope that helps.
bob
-------
∂26-Mar-82 1637 RINDFLEISCH@SUMEX-AIM (SuNet) Re: meeting
Date: 26 Mar 1982 1637-PST
From: T. C. Rindfleisch <Rindfleisch at SUMEX-AIM>
Subject: Re: meeting
To: jef at SU-AI, equip at SU-SHASTA
cc: Rindfleisch at SUMEX-AIM
In-Reply-To: Your message of 26-Mar-82 1409-PST
I would like one of the SUVAX's in S-101 of the Med Center. Tom R.
-------
∂26-Mar-82 1740 Hobbs at SRI-AI Stanford Course Announcement
Date: 26 Mar 1982 1739-PST
From: Hobbs at SRI-AI
Subject: Stanford Course Announcement
To: AIC-Associates:, tinlunchers:
Linguistics 243:
Topics in Semantics and Pragmatics:
Coherence in Discourse
Spring Quarter, 1982
Taught by:
Jerry R. Hobbs Geoffrey Nunberg
Artificial Intelligence Center Department of Linguistics
SRI International Stanford University
Tues 1:15-3:05
Location: E229
This seminar will examine the problem of what it means for a discourse
to be coherent, from various perspectives, such as those of linguistics,
artificial intelligence, psychology, philosophy, ethnomethodology, and
literary theory. The course will be divided into two parts, the first
investigating the structure of monologues, the second the structure of
dialogues. In the first part we will examine several views of the
notion of coherence in discourse: as an intrinsic property of the text
itself, as arising from the structure of what is being talked about,
and as reflecting cognitive structures and communicative strategies.
In the second part we will look at ethnomethodological and
sociolinguistic work on the "framework" or form of conversation. We
will then examine structure in conversation from the point of view of
content, moving from question-answer pairs to larger stretches of
conversation, viewed as planned or purposeful behavior.
Students registered will be expected to write a paper on some aspect of
the course relevant to their interests.
A reading packet will be available from Kinko's in north Menlo Park,
opposite Shepherd Cadillac. Packet includes *-ed papers, which are
required. Other papers are merely recommended and will be available in
the Linguistics Reading Room.
March 30 - Overview of Course: Perspectives on Discourse.
April 6 - Coherence as Reference.
* David Rumelhart, "Notes on a Schema for Stories"
* E.M. Forster, <Aspects of the Novel>, chapters 2, 5.
April 13 - Coherence as Communicative Strategy.
* David Levy, "Communicative Goals and Strategies: Between
Discourse and Syntax", S&S12: 183-212.
Jerry Morgan, "Toward a Rational Model of Discourse
Comprehension", TINLAP2: 109-14.
* S. Jay Keyser, "Wallace Stevens: Form and Meaning in Four
Poems", excerpts.
April 20 - Coherence as a Property of Texts.
* M. Halliday and R. Hasan, <Cohesion in English>, chapter 5
on conjunction.
Robert Longacre, "The Paragraph as a Grammatical Unit",
S&S12: 115-34.
* Jerry Hobbs, "Why is Discourse Coherent?"
April 27 - The Framework of Dialogue.
* Harvey Sacks, First Lecture.
* Frederick Erikson, "Money Tree, Lasagna Bush, Salt and
Pepper: Social Construction of Topical Cohesion in
a Conversation among Italian-Americans", Tannen: 43-70.
Gail Jefferson, "Side Sequences", Sudnow: 294-338.
May 4 - Sequences of Speech Acts.
Stephen Levinson, "The Essential Inadequacies of Speech Act
Models of Dialogue"
* William Labov, "Speech Actions and Reactions in Personal
Narrative", Tannen: 219-47.
* William Shakespeare, <King Lear>, Act 1, Scene 1.
May 11 - Pairwise Coherence.
* Erving Goffman, "Replies and Responses", <Forms of Talk>: 5-77.
Jerry Hobbs and Jane Robinson, "Why Ask?", <Discourse Processes>
2:311-8.
Evangeline Marlos, "Why Answer? A Goal-based Analysis of
a Speech Event"
Anita Pomerantz, "Compliment Responses: Notes on the
Co-operation of Multiple Constraints", Schenkin: 79-112.
May 18 - Global Coherence of Conversations I.
* Charlotte Linde and Joe Goguen, "The Structure of Planning
Discourse"
May 25 - Global Coherence of Conversations II.
* William Labov and David Fanshel, <Therapeutic Discourse>,
excerpts.
* Jerry Hobbs and David Evans, "Conversation as Planned Behavior",
<Cognitive Science> 4: 349-77.
John Dore and Ray McDermott, "Linguistic Indeterminacy and
the Social Context in Utterance Interpretation".
June 1 - Review and Discussion.
References:
Givon, T. (Ed.) (1979) <Syntax and Semantics>, Vol. 12: <Discourse and>
<Syntax>. New York: Academic Press.
Goffman, E. (1981) <Forms of Talk>. Philadelphia: University of
Pennsylvania Press.
Labov, W. and D. Fanshel, <Therapeutic Discourse>. New York: Academic
Press.
Schenkin, J. (Ed.) (1978) <Studies in the Organization of Conversational>
<Interaction>. New York: Academic Press.
Sudnow, D. (Ed.) (1972) <Studies in Social Interaction>. New York:
Free Press.
Tannen, D. (Ed.) (1982) <Analyzing Discourse: Text and Talk>,
Georgetown University Press.
Waltz, D. (Ed.) (1978) <Theoretical Issues in Natural Language>
<Processing-2>.
-------
∂27-Mar-82 1742 CLT
I'm off to the concert. See you later tonight
∂27-Mar-82 2333 RWG Cheshire binders
To: BS at SU-AI
CC: JMC at SU-AI
come in two flavors, hairy and simple. The hairy one takes binder tape
cassettes in three widths and cuts them to the length you specify.
The simple one uses precut 11" tapes and cannot bind books more than
about 1.5 times the thickness of JMC's flavor papers. But it is
simpler to operate, harder to jam, and has colored tapes. We have
both flavors at PARC. (Xerox bought the company.) My phone
(494 4491) will eventually ring at Linda Williams's switchboard,
which is right next to a hairy Cheshire. I suspect she could tell
you a lot.
There's also a marvelous paper joggler into which you can put your
Dover output prior to Cheshiring, which Symbolics Inc just had to
have, until they learned it cost $600. I think it's by FMC. Call
the main Xerox number (494 4000) and ask for Jeannie Treichel or
Janet Moreland in Bldg 32.
∂28-Mar-82 0217 REM EMACLS.RPG[UP,DOC] last written by JMC
To: RPG, JMC
I'm not sure who's currently maintaining this, so I'm sending to both.
I tried to start up a PLISP (MacLISP) and E (ETV) from the instructions, but
I guess it didn't work. First I got my usual LISP.INI out of the way and
copied the one from [MAC,LSP]. Then I started up PLISP and then detached
it in running mode so it could receive mail from E. Then I started up E
and said <ctrl>15X SLISP and it said it was establishing communication with
job 15 (the detached running PLISP job). Then I tried various commands for
sending s-expressions from E to PLISP but never got anything back.
Finally after about 4 tries, E complained that the LISP job's mailbox
was full and not emptying.
What is the correct procedure for starting up a fresh E&PLISP?
∂28-Mar-82 1828 Purger exceeding your disk quota
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.
READ PURGE for details on purges. Disk allocations and disk usage are
measured in disk blocks (a block is currently one track). If you would
like to buy a bigger allocation, contact Lynn Gotelli (LMG at SAIL).
The unit of allocation, one aliquot, is currently 50 disk blocks.
Your disk quota is: 2000 blocks
Your files occupy 2958 blocks
∂28-Mar-82 1844 GOSPER at PARC-MAXC [RWG: Life.]
Date: SUNDAY, 28 MARCH 1982 18:15-PST
From: GOSPER at PARC-MAXC
To: jmc at sail
Subject: [RWG: Life.]
Date: WEDNESDAY, 10 MARCH 1982 21:25-PST
From: Bill Gosper <RWG at MIT-MC>
To: RWG at MIT-MC, DLW5 at MIT-AI
cc: SWOLF at MIT-MC, HQM at MIT-MC, REM at MIT-MC, PMF at MIT-MC,
HIC at MIT-MC, EF at MIT-MC, MINSKY at MIT-MC, woods, taylor,
gosper, kaehler, jbr at S1-A
DLW5@MIT-AI 03/10/82 15:39:04
Hi. I was told that you have written a Life program for the Lispm.
Is this true? Where can I FTP the code from? Thanks.
Around New Year's, I received six separate Life queries within a few
days. Finally, in response to hqm's request for an algorithm to
avoid recomputing the stable stuff, I began experimenting with a
somewhat radical iterator, anticipating installing a spiffy, multiwindow
interface, file-io, etc.
The main features of this algorithm are large space ((2↑23)↑2 is the
default, but it's a run-time variable), and repetition skipping.
The main misfeatures have turned out to be a confusing display, and
a need for about a factor of 10 more physical (vs virtual) memory
than johan has. (About 2↑22 words would do nicely--he has < 2↑19).
So, on the good side, a lone glider in a (2↑23)↑2 space starts slowly
but accelerates to generation 2↑21 in about half a minute, and a
glider gun doubles its speed every 15 secs or so. On the bad side,
an r pentomino takes about an hour of disk thrashing to reach gen
2↑10, and less than a minute more to reach 2↑21.
Also on the bad side, I have yet to implement a comprehensible
displaying algorithm (tho there remains one intriguing idea).
The problem is that time does not run at the same rate in
different places, nor does it necessarily even run forwards.
In fact, the algorithm is indifferent to the x, y, and time
coordinates (and even the SIZE) of the cells it is running,
and that it is Life (vs some other rule) is known to only one
four-line function. One might argue that, without a more convincing
display, there is some question of whether it is running Life at
all. For example, the most straightforward display alg I thought
of was to artificially maintain cell coordinates, and paint
whatever was the current result in its proper position. The
effect was so bizarre that i spent three days looking for non-
existent bugs. Gliders appeared as unrecognizable splotches
that occasionally flew backwards. Pulsars would come to rest
with three quadrants in the wrong phase . . . . In an attempt
to rescue causality, I put in hair to inhibit "flashbacks", but
this had the completely unexpected effect of updating the
center of the universe only a fractal subset of the time (i.e.,
almost never, in the sense of measure theory). Finally, I
introduced an artificial time coordinate and "ratcheted" each
1 by 1 cell (thus sacrificing the God's eye view of all 2↑46
cells). So now the worst thing that happens is that sometimes
a glider will disappear completely, and sometime later reappear
in three places at once.
The iterator algorithm is not very complicated. For every n up to,
say, 23 (only to avoid bignum conses), there are cells of size
2↑n. Actually, there is only one cell of size 2↑23 (the universe)
and at most four of size 2↑22 (its quadrants), fewer if any
quadrants are identical. The cells, even down to 1by1s, are
uniquified flavor objects. That is, there is a unique object
corresponding to the 1by1 "off" cell, and another for "on".
Similarly, there are at most 2↑4 2by2 cells, etc. Thus an
empty universe of size (2↑23)↑2 is represented with only 24 objects,
and likewise for a totally "on" universe. Each 2by2 and larger
has four instance variables which are its quadrant cells. Each
4by4 and larger has a fifth instance variable, its RESULT, which
if known, is the same size as a quadrant, and is what the central
region of the cell will be in the time it takes "light" (i.e.
ignorance of outside influences) to traverse 1/4 of the cell.
You can't (except for 4by4's) ask a cell to take 1 step, you
ask for its result in 2↑(n-2). If it doesn't know, it enters
a thirteenfold(!) recursion involving the RESULTS of its quadrants,
and quadrants formed from rearranging their quadrants, and
quadrants made by gluing together those results.
The only source of new cells is the gluing of quadrants, which is
accomplished with a hashtable uniquifier. This is the key to both
the economies and diseconomies of the algorithm. It astronomically
compresses repetitive spaces by sharing, and it often returns
a "smart" cell which already knows its RESULT. But the very
nonlocality of a smooth hashing algorithm foredooms disk-limited
speed as soon as the hashtable exceeds physical memory size.
One other bonus, which may rescue the display problem, is that
the algorithm is effectively time reversible, in that aborting
out of a run-the-universe-2↑21-steps method hardly loses any
state, and restarting from generation 0 incurs only log(n)
penalty to zoom up to the state you were in when you aborted.
I propose to run a separate display process which just runs
to an arbitrary generation number, and then aborts, artifically
recursing if a RESULT step would overshoot the target frame.
But it's not a demo yet.
Re: Life p.s.
P.S. You might think that a glider flying off to the horizon would
create a huge number of garbage cells, but fortunately, due to
strobing, any object with uniform velocity generates about the same, fixed
number of cells of each size. And for large sizes, a whole salvo of
(parallel) gliders looks the same as one glider!
∂29-Mar-82 0219 Bill Gosper <RWG at MIT-MC> hic on h's ifu design, etc.
Date: 29 March 1982 05:19-EST
From: Bill Gosper <RWG at MIT-MC>
Subject: hic on h's ifu design, etc.
To: jmc at SU-AI
Well, not a concrete one, maybe in his head...at this stage, we're gonna
take it out of the critical path as far as the software goes by doing a simple
memory map that's "plug compatible" with the IFU...he'll proby
have it done in a month, since sudsing has started
∨
[MESSAGE FROM HIC at MIT-MC 2:39am]
Though without the IFU, it may be about A machine speed.
∨
[MESSAGE FROM HIC at MIT-MC 2:39am]
well, when we get the thing running for real with a TV, we'll hvae
to try it...won't it be nice being 3 times faster too?
∨
[MESSAGE FROM HIC at MIT-MC 2:38am]
We do have 1.5 meg of 36 bit words plugged in, though (3 boards!)
∨
[MESSAGE FROM HIC at MIT-MC 2:37am]
Hmmm....some signs, but they are seen only through 8" telescope...
we are going to proceed with a grass-roots memory map so we can stick a disk
on the thing.
∨
∂29-Mar-82 0820 FFL
To: JMC, FFL
You sent a letter to Charles Bachman, recommending Boyer and Moore for the
Turing Award. Is this the letter you want sent to Bledsoe?
Yes, the letter to Bachman.
∂29-Mar-82 1029 David Warren <WARREN at SRI-AI>
Date: 29 Mar 1982 1027-PST
From: David Warren <WARREN at SRI-AI>
To: JMC at SU-AI
cc: WARREN at SRI-AI
In-Reply-To: Your message of 19-Mar-82 1256-PST
(I just got back from Europe and I'll be in touch soon).
Bob Kowalski: 01-947-0419
Keith Clark: 01-262-8486 (weekdays)
049-12-2277 (weekends).
I forgot to ask whether they can be reached on ARPAnet.
∂29-Mar-82 1434 David Warren <WARREN at SRI-AI>
Date: 29 Mar 1982 1220-PST
From: David Warren <WARREN at SRI-AI>
To: JMC at SU-AI
cc: WARREN at SRI-AI
In-Reply-To: Your message of 29-Mar-82 1159-PST
No, Bob and Keith are not accessible as yet on the Arpanet. --David.
-------
∂29-Mar-82 1440 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Faculty Lunch and Meeting
Date: 29 Mar 1982 1440-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Faculty Lunch and Meeting
To: faculty at SU-SCORE
cc: csd.bscott at SU-SCORE, csd.armer at SU-SCORE
Tomorrow, Tuesday 3/30, there will be the usual faculty lunch
in the Boys Town conference room.
At 2:30, as previously announced, there will be a faculty meeting
in room 252.
GENE
-------
∂29-Mar-82 1451 TOB
John: Thanks for calling. This may be a repeat message, but I wasn't sure that
I got finished on th system before system crash.
Tom
∂24-Mar-82 1358 MAS aro
∂24-Mar-82 1356 TOB
mail mas82 1144 JMC
I phoned Mock. He said that the proposal would have to be resubmitted,
so I explained about the success of the long campaign to make you an
Adjunct Professor, so we agreed that it when it is resubmitted, you should
be PI. No action is necessary until you get something in the mail, but
I think it might be worth your while to talk to him on the phone.
MAS - You got something from them in the mail with comments from the
reviewers.
∂29-Mar-82 1628 David Warren <WARREN at SRI-AI> Prolog Architectures / DEC
Date: 29 Mar 1982 1623-PST
From: David Warren <WARREN at SRI-AI>
Subject: Prolog Architectures / DEC
To: JMC at SU-AI
cc: Warren at SRI-AI
The next message from me will contain a copy of a draft proposal
to Digital for the project I mentioned to you briefly a few days back.
This draft has already been sent to Peter Jessel, and I gather he
is keen for it to proceed, but is holding fire temporarily for
internal political reasons (basically he's waiting to win over the
body of opinion within DEC that thinks DEC should look first at
architectures for Lisp).
I'd appreciate any comments from you, and hearing whether or
not you're still interested in being involved (to the very small
extent suggested). Please keep this reasonably confidential.
Thanks for your interest.
David Warren.
-------
∂29-Mar-82 1629 David Warren <WARREN at SRI-AI> Draft DEC Proposal
Date: 29 Mar 1982 1625-PST
From: David Warren <WARREN at SRI-AI>
Subject: Draft DEC Proposal
To: JMC at SU-AI
ARCHITECTURES FOR KNOWLEDGE BASED SYSTEMS
[*DRAFT*] Technical Proposal
by David Warren, Computer Scientist,
Artificial Intelligence Center, SRI International
for Peter Jessel, [*title*]
Distributed and Mid-Range Advance Development,
Digital Equipment Corporation, Tewksbury, MA
INTRODUCTION
SRI International (SRI) hereby proposes to assist Digital Equipment
Corporation (DEC) in a project to develop computer architectures to
support Prolog, using the VAX as starting point. This research and
advanced development project will cover the complete hardware,
firmware and software spectrum.
BACKGROUND
- KNOWLEDGE BASED SYSTEMS
[* Refer to Sanjai Narain's dissertation: Mycin -> Loglisp *]
[* Refer to reported use of Prolog for knowledge based system at IBM,
San Jose *]
- PROLOG
Prolog is a general purpose programming language based on logic.
It can be viewed either as an extension of pure Lisp, or as an
extension of a relational database query language. It was first
conceived in 1972, by Alain Colmerauer at the University of
Marseille. Since then, it has been used, mainly in Europe, for a
wide variety of applications, including natural language processing,
algebraic symbol manipulation, compiler writing, architectural
design, and expert systems. It has recently been chosen as the
kernel language for the Japanese Fifth Generation Computer Systems
project.
The Edinburgh DEC-10 Prolog system includes a compiler which
generates code comparable in efficiency with that produced by current
Lisp compilers. Other work [Warren: VLDB-81] indicates that, for
queries over small databases, DEC-10 Prolog's speed is comparable
with or better than current relational database systems.
[* Connections between Prolog and KB systems, Lisp, relational DBs,
NL and dataflow (single assignment languages). *]
- CURRENT USE OF PROLOG BY DEC CUSTOMERS
DEC-10 Prolog has been sent on request to at least 70 separate
DEC-10 and -20 installations, of which at least 32 are in the USA.
An unknown number of other installations, particularly in the USA,
have obtained DEC-10 Prolog indirectly. In addition, versions of
PDP-11 Prolog (also developed at Edinburgh) have been installed at a
probably comparable number of PDP-11 sites. A derivative of PDP-11
Prolog, called Prolog-1, is marketed by Expert Systems Ltd, of
Oxford, England. The Hungarians have developed an excellent portable
implementation of Prolog, called M-Prolog. It is believed they are
contemplating marketing a VAX version.
- THE JAPANESE FIFTH GENERATION PROJECT
The Japanese Government, after a two year preliminary study, is
about to embark on an ambitious project to develop computer systems
for the 1990s. Their plans are documented in the proceedings of the
"International Conference on Fifth Generation Computer Systems",
which they convened. They have chosen Prolog as the kernel language
for this project. Their reasons, as described in the paper by Fuchi
in the proceedings, are very similar to those that motivate this
proposal.
The first step they will take, which they say will serve as a
"springboard" for the rest of the project, will be to develop a high
performance personal Prolog machine (see in particular the paper by
Yokoi et al in the proceedings). A surprisingly specific description
of this machine is given. The fact that the word size will be 36
bits suggests that some form of emulation of DEC-10 Prolog is
intended.
Ultimately the Japanese are aiming to achieve, using advanced
parallel architectures, a performance of what they call 1 GIGA-LIPS,
ie. one billion logical inferences per second. This would be
equivalent to something like 10,000 to 100,000 times the power of
a DEC KL-10.
- SUMMARY
A number of separate developments point to knowledge based
systems in general, and Prolog in particular, becoming important
forces in the not too distant future. The Japanese aim to capture
a major share of this market. This proposal aims to give DEC the
ability to resist any challenge to its current market, and equally
to take advantage of any new market these developments stimulate.
Although the proposed project is speculative, to the extent that a
large market for Prolog does not currently exist, the costs will be
small compared with the potential gains if the future of computing is
anything like what many people, including the Japanese, predict. If
DEC wishes to respond to these developments, now is the time to act.
WHAT IS BEING PROPOSED
The overall objective is to develop the best possible architectures
for running Prolog programs, and thereby to provide especial support
for knowledge based systems. It is expected that the architectures
will also provide good support for Lisp, as a side benefit.
The main insight behind the DEC-10 Prolog compiler was that it is
relatively easy to translate a Prolog clause into a sequence of
fairly simple machine-oriented operations. These operations were
implemented as a mixture of in-line DEC-10 instructions and calls to
out-of-line subroutines. Each operation can be considered to be an
instruction of an "abstract Prolog machine".
Such instructions could be encoded very compactly on a
byte-oriented machine (cf. the compactness of the VAX instruction
set). Although such an encoding requires some degree of
interpretation overhead, the overhead is not in fact that great, even
for a purely software implementation. The reason is that the Prolog
instructions correspond to quite a few conventional machine
instructions, while the interpretation overhead can be reduced to
only a couple of instructions by a technique similar to "threaded
code". If some or all of the important operations were to be
implemented in microcode or even hardware, there would be great scope
for speed improvements. It would also be easier to simplify the
abstract instruction set without suffering unacceptable speed
penalties. This, then, would be the proposed route for developing a
basic Prolog machine. The VAX appears particularly suitable as a
starting point.
Such a basic Prolog machine could be developed in two directions:
(1) It could form one component of a conventional computer system,
serving to improve the performance of knowledge based systems. In
this case it would be important to develop an adequate systems
programming interface between the Prolog machine and the host machine
(the VAX), to provide access to other software. Such an interface
is provided in DEC-10 Prolog through certain rather ad hoc extensions
called "pseudo-Prolog".
(2) It could form the basis of a self-contained personal Prolog
machine, where Prolog would effectively be the machine language, cf.
Lisp Machines, for example. In this case, too, there might be a
requirement for lower level extensions to Prolog, but of a more
limited kind. In addition, a personal Prolog machine would
present the challenge of developing a total environment for Prolog,
comparable with that available on Lisp machines, Xerox Dolphins, etc.
An important ingredient of Prolog is what might be termed its
"relational database subset". In principle, it is possible to use
Prolog exactly as a relational database management system. However
current Prolog implementations do not fully exploit this possibility.
They are not able to cope with really large volumes of data, and they
do not support both fast access to and fast updating of large
relations. However there is no reason why a Prolog system should not
be extended to achieve this - the technology already exists in
relational database systems and it is compatible with current Prolog
implementation practice. Hence this improvement would be a part of
this proposal. Incidentally, the potential identity of Prolog with a
relational DBMS does not seem to be fully appreciated by the
Japanese. To begin with, at least, they will separate Prolog from
databases.
When executing database queries, Prolog exhibits a high degree of
potential parallelism, which would appear to lend itself to dataflow
techniques. A very preliminary study on this has already been done
by Warren in connection with the Manchester dataflow machine. This
kind of parallelism is a special case of Prolog's OR-parallelism.
Prolog also exhibits another kind of potential parallelism, called
AND-parellelism, which however is less easy to recognise in advance
and perhaps to realise. It too may be amenable to dataflow
techniques or perhaps to other more conventional techniques. A
special case of AND-parallelism is the parallelism inherent in
unification, the basic pattern matching operation of Prolog. There
is no reason why this shouldn't be exploited by special purpose
hardware. Indeed, a complete unification chip has already been
produced in an experimental study by a student of Jim Kajiya at
Caltech. All these possibilities should be investigated.
WHAT THE RESULTS WILL BE
To summarise, different strands of this proposal, if carried
through to conclusion, are intended to lead to the following results:
(1) A high quality, high performance software implementation of
Prolog for the VAX, comparable to and compatible with DEC-10 Prolog.
(2) Hardware or firmware extensions to the VAX to support Prolog.
(3) A personal Prolog machine, analogous to current Lisp machines,
where Prolog (possibly with some extensions) will be the basic
machine language.
(4) Advanced parallel architectures (eg. dataflow) supporting Prolog.
(5) A Prolog system capable of serving as an efficient relational
database management system for moderate sized databases (of the order
of, say, tens of megabytes).
WORK STATEMENT
SRI will exert its best efforts to assist DEC in the design of the
proposed software, firmware and hardware. SRI will supply general
expertise on Prolog, and will assist in keeping the project abreast
of Prolog developments elsewhere. DEC will supply general expertise
on hardware and firmware. SRI will be responsible for implementing
agreed items of software, but will not be responsible for
implementing hardware or firmware.
The work will cover some yet to be agreed subset of the following
tasks. A suitable initial study might comprise tasks (1) through
(4). Each task is followed by an estimate of the effort it will
require on the part of SRI, in person months.
(1) Design an abstract Prolog machine, suitable for implementation on
a large virtual memory, byte-oriented machine, with a view to some of
the basic operations being implemented in hardware.
[6 months].
(2) Implement the core of (1) in software for the VAX.
[9 months].
(3) Investigate to what extent (1) can be implemented in hardware,
modifying the design and software implementation accordingly.
[at least 6 months].
(4) Investigate the possibilities for exploiting parallelism in
Prolog, whether by dataflow or by more conventional techniques.
[at least 6 months].
(5) Extend the results of (2) or (3) to provide a systems programming
interface to the VAX architecture itself and to other VAX software.
[9 months].
(6) Extend the results of (5) to a full Prolog implementation
comparable to DEC-10 Prolog.
[24 months].
(7) In the course of (6), design and implement suitable indexing
techniques etc., to cater for moderate sized relational databases
within Prolog.
[9 months].
(8) In the course of (6), investigate what extensions would be
required to provide an environment suitable for a personal Prolog
machine.
[at least 12 months].
PERSONNEL
The SRI personnel who would perform the proposed work are David
Warren, Fernando Pereira (subject to his accepting a current offer
from SRI), and possibly other new hires with experience of Prolog.
Warren and Pereira were the principal implementors of DEC-10 Prolog.
Amongst other things, David Warren designed and implemented the
Prolog compiler which forms the foundation of DEC-10 Prolog.
Fernando Pereira was responsible for the system's interfaces to
Tops-10 and Tops-20, and has wide experience of DEC software.
Biographies of Warren and Pereira are attached. Both of them have
research interests in natural language processing, as well as other
areas of artificial intelligence. They jointly implemented, in
Prolog, a domain-independent natural language question-answering
system, called "Chat".
[*It is hoped that*] Professor John McCarthy, of Stanford
University, will monitor the project's progress for DEC, to the
extent of a few days of his time per year. [* He has indicated
that he is interested. *]
Apart from the individuals mentioned, SRI and neighboring
institutions provide an unrivalled concentration of people working in
areas relevant to this project. Of particularly importance here are
the areas of knowledge based systems (SRI, Stanford), high
performance personal computers (Xerox, Stanford), and VLSI (Stanford,
SRI, etc.). In addition, SRI is at this moment taking delivery of a
quantity of Lisp Machines, hands on experience of which will be
extremely valuable.
-------
I don't think Prolog is an extension of pure Lisp, because it doesn't
have composition of functions. Colmerauer was still at Grenoble
when he invented Prolog. Perhaps you would be willing to make
the architecture suitable for Lisp as one of the direct objectives and
not merely as a side effect.
∂29-Mar-82 1757 Chandra-at-OhioState <Chandrasekaran at RUTGERS> visit to SU-AI
Date: 29 Mar 1982 2026-EST
From: Chandra-at-OhioState <Chandrasekaran at RUTGERS>
Subject: visit to SU-AI
To: JMC at SU-AI
cc: Chandrasekaran at RUTGERS
I have tentative plans yto be in the Bay area during
May 3-7. If you have some free time during that time, I'd
like to visit you to discuss some of the ideas in the proposal
in some depth, especially concerning the interaction of
simulation and reasoning programs.
-------
Unfortunately, I'll be in France May 3-7.
∂29-Mar-82 1756 Chandra-at-OhioState <Chandrasekaran at RUTGERS> help on funding
Date: 29 Mar 1982 2022-EST
From: Chandra-at-OhioState <Chandrasekaran at RUTGERS>
Subject: help on funding
To: JMC at SU-AI
cc: CHANDRASEKARAN at RUTGERS
Sorry to bother you, but I was wondering if you have
had a chance to discuss with Dr Teller what sort of things
may be done to help me obtain funding for the AI application
to nuclear safety. A possibility that you had suggested was
that he might write to NRC Chairman Palladino, or to smooth the
way with DOE people. I am getting quite concerned
that I get going on the project.
I hear rumors about some other
AI projects being talked about for nuclear app;lications,
and I'd like not to be left out.
I also hear that there is an ARPA division on nuclear reactor
studies. Do you know anybody there?
Thanks for any help you can give.
-------
∂29-Mar-82 2110 TOB
To: JMC, "@AL.DIS[DIS,TOB]"
Brain Carlislie and I talked about this some time ago
and he talked about our getting one. I rellly want it
much better than the vendig machine for fun and public
relations. Sorry for not talking about it, but I did
not know what Unimation intended about publicity.
∂22-Mar-82 1751 RV in case you havent seen it:
To: "@RTI.DIS[1,JJC]"
n048 1227 19 Mar 82
BC-ROBOTS Adv21 2takes
(FOR RELEASE SUN MAR 21)
(FINANCIAL)
By BARNABY J. FEDER
c. 1982 N.Y. Times News Service
DANBURY, Conn. - Four decades ago, science fiction writer Isaac
Asimov's robot stories caught the imagination of a Columbia
University physics student named Joseph F. Engelberger. Sometime in
1985, a robot named in Asimov's honor is likely to be serving
Engelberger and other directors the nation's first and largest
industrial robot manufacturer coffee at their board meetings here.
Now a prototype in the company's research laboratory, Isaac the
Robot is being designed to do more than traverse the boardroom
serving coffee. Engelberger also wants Isaac to provide snacks
prepared in the adjoining kitchen's microwave oven and wash dishes.
Engelberger's company, Unimation Inc., has no plans to market Isaac,
or similar robots, but Isaac is more than just a whimsical tribute to
Asimov. Engelberger envisions Isaac - a mobile, improved version of
the PUMA robot the company already sells - as the forerunner of a new
generation of domestic and commercial service robots that Unimation
and other robotics companies will begin selling during the 1990s.
The right to be an out-of-the-closet visionary is one of the
relished and hard-won benefits that the 56-year-old Engelberger has
earned for his pivotal role in bringing the robot industry to life,
both in the United States and abroad.
Actually, it was George C. Devol, not Engelberger, who developed and
patented the basic technology on which the industry is founded. But
since meeting Devol in 1956, Engelberger has preached the gospel that
''smart'' machines were the key to getting people out of dangerous or
tedious production jobs and a key to improving productivity. And his
company, a subsidiary of the Condec Corp. of Old Greenwich, Conn.,
turned out the first robots that industry was willing to buy.
As a result, no robotics gathering today would be considered
complete without the presence of the crew-cut, bow-tied Engelberger
and his blunt observations about competitors, customers and robots
themselves. ''He is as important to the industry as he is to the
company, in some respects more so,'' said Laura Conigliaro, the Bache
Halsey Stuart Shields analyst who is Wall Street's best-known
robotics expert. ''He is a spokesman and a showman, and he is good at
it.''
''He was the one that listened,'' said Devol, who now runs a robot
leasing and consulting business from his home in Fort Lauderdale,
Fla. Devol recalls numerous efforts to interest established companies
in his work, including some, such as IBM, that have recently entered
the now rapidly growing robotics field.
''George Devol was unable to restrain himself from spilling the
whole dream out, which scared most businessmen off,'' said
Engelberger during an interview last week at Unimation's
headquarters. ''I kept myself from talking about some of the things
that have happened, which he envisioned.''
The ''whole dream'' is emerging now that robots have achieved
acceptance in an increasing variety of industrial tasks - from
materials handling to painting and welding - and are rapidly being
improved to the point that more difficult jobs, such as assembly,
will be economically feasible. More important, as computer-machine
tool hybrids capable of being reprogrammed to adapt to changing
conditions, they have been recognized as a key building block in the
flexible, highly-automated factory of the future.
It took American industry a long time to catch on. The Japanese, by
contrast, were fascinated from the outset. When the Japanese
government invited Engelberger to lecture in 1967, he was startled to
find an audience of more than 700 executives and engineers, and
subsequently overwhelmed by five hours of questions from the floor.
''I still haven't seen anything outstanding technologically over
there,'' Engelberger said, but he agreed with the general consensus
that Japan, with some 15,000 installed robots, is far ahead of the
United States in experience with the devices. The Japanese robotics
industry has already gone through its first major shakeout, he said,
with Kawaskai Heavy Machine, Unimation's licensee, still the market
leader.
Although slowed by the recession, the United States may be catching
up. Domestic sales of robots, which passed the $100 million level for
the first time in 1980, soared to $155 million last year. Moreover,
giants like General Electric, Westinghouse, IBM, Textron and Bendix
have plunged into the competitive fray on the assumption that sales
of both robots themselves and related computer-controlled
manufacturing systems will soon run into billions of dollars annually.
For Unimation, which first turned a profit in 1975 after 19 years of
research, development and market building, the sudden surge in
competition has marked the beginning of a tough new challenge. The
first established competitors to enter the field - ASEA of Sweden,
Cincinnati Milacron, and a slew of major Japanese companies - all
helped Unimation by legitimizing robotics for many dubious customers.
Now the field is crowded. Unimation is already being pressed for
market leadership by Milacron. In preparation for both the growing
market and the ineviatable shakeout, the company is expanding
production capability and its management team, spending heavily to
develop new robots, and marshaling financial resources in preparation
for the looming shakeout.
Engelberger has turned over operating responsibility to Paul
Allegretto, executive vice president, and now concentrates on
research and development, ''big ticket sales,'' as he puts it, and
finances.
The new challenges seem no worse than those Engelberger faced
earlier. In 1956, as the 31-year-old chief engineer of the airplane
products division of a small industrial conglomerate later acquired
by Dresser Industries, he was asked to liquidate the division in
return for a promotion. The offer of a raise was especially tempting
because his wife, Marge, was pregnant with their second child.
''Instead, I got my Masters degree in business in one week-
end,'' he said. ''I bought five business books at Barnes & Noble,
read them, went in the following Monday and got four months to look
for money to buy the division.'' These developments, an encounter
with Devol at a cocktail party, and the financial backing of Norman
Schafler, founder of the Consolidated Diesel Electric Co., now
Condec, were the foundation stones for Unimation.
There were a lot of unpleasant surprises. Engelberger said that his
new company, Consolidated Controls, became profitable in seven
months, but the plunge into robotics by its offspring, Unimation, was
another matter. ''I was wrong on the money, wrong on the time, and
right on the concept,'' he said. In addition, he discovered that the
very word ''robot'' was ominous enough to scare away some otherwise
likely customers and that some of the best early applications, like
spot welding, were overlooked until pointed out by users like General
Motors.
The financial burdens of financing Unimation's maturation induced
Condec to sell control of Unimation to Pullman Inc. in 1962 to bring
in more development money. Condec bought Pullman's interest back in
1974. Last fall, 20 percent of Unimation's shares were offered
publicly, for about $24 million, to sustain the company's current
growth and give it liquid equity that could be used in acquisitions
or compensation plans for key employees.
''Most entrepeneurs wouldn't go into it if they really understood
how long it takes,'' Engelberger said. Of course, there was more than
money to keep Engelberger going. ''There are very few people
fortunate enough to get started early in something that turns out to
be really big,'' he conceded. ''It's strange in a way, but I started
getting awards before Unimation managed to become profitable.''
nyt-03-19-82 1533est
***************
∂30-Mar-82 1007 RPG@SU-AI (SuNet) Vaxes
Date: 30 Mar 1982 1006-PST
From: Dick Gabriel <RPG at SU-AI>
Subject: Vaxes
To: equip at SHASTA
Since the last meeting a number of Lisps have become available on the Vax. I
would like to request that one of them be placed in a location available to
me for evaluation and possible use.
-rpg-
∂30-Mar-82 1327 RPG@Sail (SuNet) Clarification
Date: 30 Mar 1982 1326-PST
From: Dick Gabriel <RPG at SU-AI>
Subject: Clarification
To: equip at SHASTA
My previous message concerning Lisp and Vaxes was not sufficiently
unambiguous due, in part, to a non-contiguity in time with the
question to which it was a response. To be clear:
I would like to be able to have access to one of the DEC Suvaxes
when they arrive. In recent times the following Lisps have become
available on Vaxes: NIL, InterLisp, and T (a variant of Scheme). In addition
Franz is available. I would like to evaluate both the performance of
these Lisps with respect to each other and the performance of the 11/750s
as vehicles for Lisp machines. I understand that this is in concert with
our proposal to DEC for use of these machines. I believe this implies
that I would like an ``account'' on one of these machines and that one
of them should be located in MJH. I will need to run VMS and Unix variously
to explore these Lisps.
Secondly, NIL is potentially a useful Lisp for my ARPA sponsored research
(note also that the above-mentioned evaluation is an ARPA funded project),
and I would possibly find the continued use of one of these machines for
my work desirable. I understand that this is in concert with our proposal
to ARPA for use of these machines.
Therefore, I propose that one of these Suvaxes be installled in the
MJH machine room.
-rpg-
∂30-Mar-82 1429 Gene Golub <CSD.GOLUB at SU-SCORE>
Date: 30 Mar 1982 1428-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
To: faculty at SU-SCORE
cc: csd.irmgild at SU-SCORE
Hans-J Boehm will speak at the colloquium on Tuesday, April 13
on Logic for Expressions with Side Effects. He is interested in a faculty
appointment. Please let Irmgild know if you wish to see him. Gene
-------
∂30-Mar-82 1704 Ichiki at SRI-AI 4/1 TINLUNCH MEETING
Date: 30 Mar 1982 1703-PST
From: Ichiki at SRI-AI
Subject: 4/1 TINLUNCH MEETING
To: tlgrp:
Doug Appelt will be presenting the following paper for TINLUNCH
on Thursday, 1 April, from 12:00-1:00 in Conference Room EK242.
WHAT'S NECESSARY TO HIDE?
Modeling Action Verbs
By: James F. Allen
-------
∂31-Mar-82 0743 Gene Golub <CSD.GOLUB at SU-SCORE> [Irmgild Schack <CSD.IRMGILD at SU-SCORE>: Dennis Heimbigner]
Date: 31 Mar 1982 0739-PST
From: Gene Golub <CSD.GOLUB at SU-SCORE>
Subject: [Irmgild Schack <CSD.IRMGILD at SU-SCORE>: Dennis Heimbigner]
To: faculty at SU-SCORE
Just a reminder! GENE
---------------
Mail-From: CSD.IRMGILD created at 15-Mar-82 11:28:49
Date: 15 Mar 1982 1128-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Dennis Heimbigner
To: faculty at SU-SCORE
cc: csd.bscott at SU-SCORE
Dennis Heimbigner will be interviewing for a faculty position on April 2.
Please let me know by Wednesday, March 17 whether you are interested in
talking with him.
Thanks,
Irmgild
-------
-------
∂31-Mar-82 0918 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Dennis Heimbigner
Date: 31 Mar 1982 0918-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Dennis Heimbigner
To: faculty at SU-SCORE
Please let me know whether you are interested to join Gene and Dennis
Heimbigner for lunch at the faculty club. Dinner arrangements are still
being made; please let me know who wants to come.
thanks,
Irmgild
-------
∂31-Mar-82 1135 Nancy Dorio <CSD.DORIO at SU-SCORE> COLLOQUIUM NOTICE WEEK OF APRIL 5 - 9, 1982
Date: 31 Mar 1982 1116-PST
From: Nancy Dorio <CSD.DORIO at SU-SCORE>
Subject: COLLOQUIUM NOTICE WEEK OF APRIL 5 - 9, 1982
To: colloq: ;
Stanford-Phone: (415) 497-2273
Date Place Person
Day Event From
Time Title
-------------------------------------------------------------------------------
4/5/82 Math 380C James Sethian
Monday Numerical Analysis U.C. Berkeley
4:15 p.m. Seminar ``Numerical Study of Turbulent Combustion in a
Square Region''
4/6/82 MJ301 Dr. Glenn Rennels
Tuesday Medical Computing Stanford University
1:30 p.m. Journal Club ``Recent Articles of Interest''
4/6/82 MJ301 Mr. John Kunz
Tuesday Knowledge Stanford University
2:30 p.m. Representation ``Medical Diagnosis using AI, Mathematics, and a
Group Meeting Physiologic Model''
4/6/82 Jordan 041 To be Announced
Tuesday Computer Science
4:15 p.m. Colloquium
4/8/82 MJ301 Joachim Ahrens \&\ Ulrich Dieter
Thursday AFLB
12:15 p.m ``Realistic Versus Abstract Roulette: A
Comparison of Optimal Strategies''
4/8/82 To be Announced Nils Nilsson
Thursday SRI Seminar SRI International
3:00 p.m. ``An Overview of AI Research at SRI''
4/9/82 MJ352 Michael Lesk
Friday Database Research Bell Labs, Murray Hill, N.J.
3:15 p.m. Seminar ``Helping the Lost and Confused: Human Factors of
Database Access to Books and Maps''
!
-------
∂31-Mar-82 1307 C.S./Math Library <ADMIN.LIBRARY at SU-SCORE> Journal of Information Processing
Date: 31 Mar 1982 1308-PST
From: C.S./Math Library <ADMIN.LIBRARY at SU-SCORE>
Subject: Journal of Information Processing
To: jmc at SU-AI
We do receive this quarterly from the Information Processing Society of
Japan.
Harry
-------
∂31-Mar-82 1325 JK nsf proposal
Have you heard anything more about it? I recall you saying something
about finding out more in a few weeks a month ago.
∂31-Mar-82 1942 Irmgild Schack <CSD.IRMGILD at SU-SCORE> Meeting Reminder
Date: 31 Mar 1982 1658-PST
From: Irmgild Schack <CSD.IRMGILD at SU-SCORE>
Subject: Meeting Reminder
To: CSD-Tenured-Faculty: ;
cc: csd.bscott at SU-SCORE
This is just a reminder that there will be a tenured faculty meeting
tomorrow, Thursday April 1, at 2:30 in room 220.
GENE
-------
∂01-Apr-82 0832 MDD via NYU reference
I used your name as a reference again. I hope this is OK.
See you in June.
I'm glad to be a reference, but I'll be going to France on the 10th. If
you tell me to whom the reference should be addressed, I'll do the letter
before I leave, and my secretary can send it when the inquiry arrives.
∂01-Apr-82 0849 Ichiki at SRI-AI Remember TINLUNCH
Date: 1 Apr 1982 0849-PST
From: Ichiki at SRI-AI
Subject: Remember TINLUNCH
To: tlgrp:
Doug Appelt will be presenting the following paper for TINLUNCH
on Thursday, 1 April, from 12:00-1:00 in Conference Room EK242.
WHAT'S NECESSARY TO HIDE?
Modeling Action Verbs
By: James F. Allen
-------
∂01-Apr-82 0858 Paul Armer <CSD.ARMER at SU-SCORE> Award Nominations
Date: 1 Apr 1982 0857-PST
From: Paul Armer <CSD.ARMER at SU-SCORE>
Subject: Award Nominations
To: CSD-Faculty: ;
cc: csd.mwalker at SU-SCORE, csd.keller at SU-SCORE
Stanford-Phone: (415) 497-2274
We have been asked to nominate Stanford faculty for the Walter J.
Gores award for excellence in teaching. One is awarded each year
to a senior faculty member, one to a junior faculty member and
one to a teaching assistant. If you are interested in nominating
someone, I have nomination forms. I also have forms for the
Kenneth M. Cuthbertson Award for Exceptional Service to Stanford.
Unfortunately both nominations close very soon--April 9.
Paul
-------
∂01-Apr-82 0858 FFL
To: JMC, FFL
Teller's office called to ask if we had sent a letter re Wood and the
Marconi Fdn. As I recall, we only sent a draft to Teller. Did you do
anything more? The office seemed very interested in having the letter
sent.
∂01-Apr-82 0859 FFL
To: JMC, FFL
Prof. Golub's office has made a tentative appointment for you with
Dennis Heimbigner for Friday at 2:30 p.m. You are also invited for
either lunch or dinner on that day with him. Let me know and I will
make the reservation.
∂01-Apr-82 0909 FFL
To: JMC, FFL
Prof. Golub has asked me to remind you of the tenured Faculty meeting
this afternoon. 1:30 p.m.
∂01-Apr-82 0914 FFL
To: JMC, FFL
Sorry -- an error in previous message. The tenured Faculty meeting is
at 2:30 this afternoon.
∂01-Apr-82 0959 ullman@Shasta (SuNet) SUVAXES
Date: 1 Apr 1982 10:00:41-PST
From: ullman at Shasta
To: equip
From: ullman
Subject: SUVAXES
Date: 1 April 1982 10:00:30-PST (Thursday)
I have received a number of requests for the location of the SUVAXES.
We'll have to decide what to do at the meeting Friday. A much more critical
issue is how to pay for them, as we do not, apparently, have any money on
hand.
∂01-Apr-82 1010 Ted Anderson <OTA at S1-A> On the road again!
Date: 01 Apr 1982 1006-PST
From: Ted Anderson <OTA at S1-A>
Subject: On the road again!
To: TAW at S1-A, jmc at SU-AI, RAH at S1-A
CC: OTA at S1-A
I heard today on the radio that I-5 south to LA was closed due to snow. I would
expect this to be cleared up by friday, but it may be closed again by then. So
we may have to take 101 to LA instead. This will take more time so it would
be prudent to leave by 10AM on friday, at that point we'll call the CHP about
raod conditions on I-5 and 101 and make a decision.
Ted Anderson
I can't leave Stanford till 11:30 Friday, because I have a meeting at 10
concerned with using our ARPA computer money. If it turns out you have
to leave early, perhaps I'd better go separately. If I leave at 11:30
from Stanford, I can be at the Lab a few minutes after 12:30 depending
on traffic. Judging by our previous trips, this should get us there
in time for the 7pm ceremonies.