perm filename F83.IN[LET,JMC] blob
sn#864707 filedate 1984-01-08 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00546 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00058 00002 ∂28-Sep-83 1159 PEREIRA@SRI-AI.ARPA APPROACHES TO COMPUTER LANGUAGES Seminar
C00061 00003 ∂28-Sep-83 1229 @SRI-AI.ARPA:BRESNAN.PA@PARC-MAXC.ARPA course rescheduled
C00065 00004 ∂28-Sep-83 1407 Bossu.GIA@MIT-MULTICS.ARPA
C00066 00005 ∂28-Sep-83 1442 ULLMAN@SU-SCORE.ARPA CS440
C00067 00006 ∂28-Sep-83 1452 RV AI Qual
C00068 00007 ∂28-Sep-83 1634 HANS@SRI-AI.ARPA FTP of the associations file
C00070 00008 ∂28-Sep-83 1655 DFH CS206 Information Sheets
C00071 00009 ∂28-Sep-83 1800 WALKER@SRI-AI.ARPA online access to AI abstracts
C00075 00010 ∂28-Sep-83 1941 GENESERETH@SUMEX-AIM Re: meeting
C00076 00011 ∂28-Sep-83 2101 RESTIVO@SU-SCORE.ARPA
C00077 00012 ∂28-Sep-83 2201 lantz%SU-HNV.ARPA@SU-SCORE.ARPA FYI
C00083 00013
C00085 00014 ∂29-Sep-83 0921 Raj.Reddy@CMU-CS-A Electronic Library Proposal
C00116 00015 ∂29-Sep-83 0949 WALKER@SRI-AI.ARPA
C00117 00016 ∂29-Sep-83 1003 AAAI-OFFICE@SUMEX-AIM Electronic Library Service meeting
C00119 00017 ∂29-Sep-83 1039 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
C00120 00018 ∂29-Sep-83 1149 DFH ginsbe.1
C00121 00019 ∂29-Sep-83 1153 RINDFLEISCH@SUMEX-AIM Symbolics 3600 Deliveries
C00125 00020 ∂29-Sep-83 1351 RINDFLEISCH@SUMEX-AIM Re: Symbolics 3600 Deliveries
C00127 00021 ∂29-Sep-83 1437 RPG CL Directory
C00128 00022 ∂29-Sep-83 1735 WALKER@SRI-AI.ARPA Fikes as AAAI Secretary-Treasurer
C00131 00023 ∂29-Sep-83 1626 PATTERMANN@SUMEX-AIM Re: Symbolics 3600 Deliveries
C00134 00024 ∂29-Sep-83 1612 ULLMAN@SU-SCORE.ARPA apology
C00135 00025 ∂29-Sep-83 1538 WALKER@SRI-AI.ARPA [Don Walker <WALKER@SRI-AI>: [Derek Sleeman <SLEEMAN@SUMEX-AIM.ARPA>: AI Research Notes series]]
C00144 00026 ∂29-Sep-83 1524 DFH Telex re Brazil conference
C00146 00027 ∂29-Sep-83 1509 croft@Safe Re: Symbolics 3600 Deliveries
C00150 00028 ∂29-Sep-83 2018 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
C00152 00029 ∂29-Sep-83 2153 YEAGER@SUMEX-AIM Re: Symbolics 3600 Deliveries
C00154 00030 ∂29-Sep-83 2207 WALKER@SRI-AI.ARPA COLING 84 CALL FOR PAPERS
C00159 00031 ∂29-Sep-83 2259 GOLUB@SU-SCORE.ARPA Appointment
C00160 00032 ∂30-Sep-83 0702 BUCHANAN@SUMEX-AIM.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
C00162 00033 ∂30-Sep-83 0834 LIBRARY@SU-SCORE.ARPA Socrates
C00163 00034 ∂30-Sep-83 0900 RINDFLEISCH@SUMEX-AIM Re: Symbolics 3600 Deliveries
C00167 00035 ∂30-Sep-83 0926 WALKER@SRI-AI.ARPA
C00170 00036 ∂30-Sep-83 0948 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
C00172 00037 ∂30-Sep-83 1008 RPG Time for a change
C00173 00038 ∂30-Sep-83 1059 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS
C00175 00039 ∂30-Sep-83 1208 WALKER@SRI-AI.ARPA Re: mailing list
C00176 00040 ∂30-Sep-83 1256 JRP@SRI-AI.ARPA Meeting
C00181 00041 ∂30-Sep-83 1308 Kay.PA@PARC-MAXC.ARPA Re: Meeting
C00182 00042 ∂30-Sep-83 1331 AAAI-OFFICE@SUMEX-AIM
C00184 00043 ∂30-Sep-83 1339 JANET@KESTREL maclisp manual
C00186 00044 ∂30-Sep-83 1336 DKANERVA@SRI-AI.ARPA Newsletter, September 29, 1983
C00203 00045 ∂30-Sep-83 1901 @SU-SCORE.ARPA:FEIGENBAUM@SUMEX-AIM.ARPA Re: Appointment
C00204 00046 ∂01-Oct-83 0000 JMC*
C00205 00047 ∂01-Oct-83 0624 Colmerauer.GIA@MIT-MULTICS.ARPA Paris
C00206 00048 ∂01-Oct-83 1745 ULLMAN@SU-SCORE.ARPA CS440
C00207 00049 ∂01-Oct-83 1954 Pool.Datanet@MIT-MULTICS.ARPA CATV paper
C00209 00050 ∂02-Oct-83 0744 Pool.Datanet@MIT-MULTICS.ARPA Misaddressed mail
C00210 00051 ∂02-Oct-83 1618 Janet.Asbury@CMU-CS-A Electronic Library
C00269 00052 ∂02-Oct-83 1702 JJW EKL manual
C00270 00053 ∂03-Oct-83 1002 LUNDSTROM@SU-SCORE.ARPA [weeks at ames-vmsb: ACM SIGBIG OCT 5th MEETING]
C00273 00054 ∂03-Oct-83 1105 KJB@SRI-AI.ARPA Todays meeting in C1
C00278 00055 ∂03-Oct-83 1212 @MIT-MC:DAVIS@MIT-OZ
C00279 00056 ∂03-Oct-83 1234 AAAI@SU-SCORE.ARPA Quarterly Presidential message
C00281 00057 ∂03-Oct-83 1351 AAAI-OFFICE@SUMEX-AIM Change in the AAAI Secretariat
C00284 00058 ∂03-Oct-83 1501 ullman%SU-HNV.ARPA@SU-SCORE.ARPA CS440
C00285 00059 ∂03-Oct-83 1502 PETERS@SRI-AI.ARPA
C00286 00060 ∂03-Oct-83 1633 DFH Barwise
C00287 00061 ∂03-Oct-83 1849 LEP traffic lights
C00288 00062 ∂03-Oct-83 1853 ALMOG@SRI-AI.ARPA thanks
C00289 00063 ∂03-Oct-83 2110 PETERS@SRI-AI.ARPA Project B1, Extensions of semantic theories
C00291 00064 ∂03-Oct-83 2142 RV qual.
C00294 00065 ∂04-Oct-83 0910 ULLMAN@SU-SCORE.ARPA panel show
C00296 00066 ∂04-Oct-83 1114 @SRI-AI.ARPA:halvorsen.pa@PARC-MAXC.ARPA Reminder: Research Seminar, Natural Language, Thur., Oct. 6
C00298 00067 ∂04-Oct-83 1323 LENAT@SU-SCORE.ARPA
C00299 00068 ∂04-Oct-83 1336 ULLMAN@SU-SCORE.ARPA seminar
C00302 00069 ∂04-Oct-83 1549 WALDINGER@SRI-AI.ARPA circumscription paper.
C00304 00070 ∂04-Oct-83 1636 DFH Chris Goad called
C00305 00071 ∂04-Oct-83 1658 RIGGS@SRI-AI.ARPA Afternoon Colloquium Schedule
C00310 00072 ∂04-Oct-83 1745 RV Procrastination
C00312 00073 ∂04-Oct-83 1939 ALMOG@SRI-AI.ARPA talk
C00314 00074 ∂04-Oct-83 2017 PEREIRA@SRI-AI.ARPA REMINDER: Research Seminar, Programming Languages, Thur., Oct. 6
C00315 00075 ∂04-Oct-83 2044 FEIGENBAUM@SUMEX-AIM nils nilsson
C00317 00076 ∂05-Oct-83 0103 TOB
C00318 00077 ∂05-Oct-83 0832 AAAI-OFFICE@SUMEX-AIM meeting
C00319 00078 ∂05-Oct-83 0853 RPG
C00321 00079 ∂05-Oct-83 0854 RPG Binford
C00322 00080 ∂05-Oct-83 0908 BMACKEN@SRI-AI.ARPA associations
C00323 00081 ∂05-Oct-83 0919 CLT lamb
C00324 00082 ∂05-Oct-83 0956 ullman%SU-HNV.ARPA@SU-SCORE.ARPA meetings
C00326 00083 ∂05-Oct-83 1018 DFH Prof. Tratatenbrot
C00327 00084 ∂05-Oct-83 1032 LUNDSTROM@SU-SCORE.ARPA [weeks at ames-vmsb: ACM SIGBIG Oct 5th Meeting]
C00331 00085 ∂05-Oct-83 1124 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 6
C00334 00086 ∂05-Oct-83 1136 DFH Psychology today article
C00337 00087 ∂05-Oct-83 1141 Janet.Asbury@CMU-CS-A Electronic Library
C00339 00088 ∂05-Oct-83 1302 YM computer facilities committee
C00348 00089 ∂05-Oct-83 1401 GOLUB@SU-SCORE.ARPA [John McCarthy <JMC@SU-AI>:]
C00350 00090 ∂05-Oct-83 1435 LIBRARY@SU-SCORE.ARPA Access to SOCRATES the public access catalog to library materials
C00354 00091 ∂05-Oct-83 1442 HANS@SRI-AI.ARPA CSLI addresses
C00356 00092 ∂05-Oct-83 1449 EENGELMORE@SUMEX-AIM NILS MTG.
C00357 00093 ∂05-Oct-83 1558 DELAGI%SUMEX-AIM.ARPA@SU-SCORE.ARPA Re: meetings
C00358 00094 ∂05-Oct-83 1654 DFH travel
C00360 00095 ∂05-Oct-83 1737 DKANERVA@SRI-AI.ARPA CSLI Newsletter, No. 3
C00383 00096 ∂05-Oct-83 2023 @SU-SCORE.ARPA:slndstrm@Shasta Re: meetings
C00385 00097 ∂06-Oct-83 0007 HST visit in januar 84
C00386 00098 ∂06-Oct-83 0829 RIGGS@SRI-AI.ARPA CSLI Education Committee Meeting
C00387 00099 ∂06-Oct-83 1001 BMACKEN@SRI-AI.ARPA meeting room
C00389 00100 ∂06-Oct-83 1150 YM computer facilities committee
C00390 00101 ∂06-Oct-83 1330 DFH Kuo
C00391 00102 ∂06-Oct-83 1342 DFH Livermore security form
C00392 00103 ∂06-Oct-83 1345 DFH more travel
C00393 00104 ∂06-Oct-83 1623 DFH AI Qual for R. Vistnes
C00394 00105 ∂06-Oct-83 2111 JMC
C00395 00106 ∂07-Oct-83 0029 FEIGENBAUM@SUMEX-AIM Re: committee assignments
C00398 00107 ∂07-Oct-83 0828 AAAI-OFFICE@SUMEX-AIM Re: President's message
C00399 00108 ∂07-Oct-83 1150 AAAI-OFFICE@SUMEX-AIM Next meeting
C00401 00109 ∂07-Oct-83 1217 AAAI-OFFICE@SUMEX-AIM Fredkin
C00403 00110 ∂07-Oct-83 1303 PACK@SU-SCORE.ARPA blocks axioms
C00407 00111 ∂07-Oct-83 1332 oliger%SU-NAVAJO.ARPA@SU-SCORE.ARPA Re: meetings
C00408 00112 ∂07-Oct-83 1335 ULLMAN@SU-SCORE.ARPA Next talk
C00413 00113 ∂08-Oct-83 0925 CLT mud
C00414 00114 ∂08-Oct-83 1546 FEIGENBAUM@SUMEX-AIM
C00415 00115 ∂08-Oct-83 2359 POURNE@MIT-MC
C00416 00116 ∂09-Oct-83 1342 ALMOG@SRI-AI.ARPA reminder on why context wont go away
C00419 00117 ∂09-Oct-83 2023 ULLMAN@SU-SCORE.ARPA cs440
C00424 00118 ∂10-Oct-83 0648 DOYLE@CMU-CS-C.ARPA
C00425 00119 ∂10-Oct-83 0817 DFH appointment
C00426 00120 ∂10-Oct-83 0900 CLT*
C00427 00121 ∂10-Oct-83 1018 DFH RPG talk
C00428 00122 ∂10-Oct-83 1038 DFH appointment
C00429 00123 ∂10-Oct-83 1127 DFH RPG talk
C00430 00124 ∂10-Oct-83 1509 RESTIVO@SU-SCORE.ARPA
C00431 00125 ∂10-Oct-83 1521 CLT
C00433 00126 ∂10-Oct-83 1533 AMSLER@SRI-AI.ARPA Access to Stanford Libraries
C00437 00127 ∂10-Oct-83 1713 JRP@SRI-AI.ARPA This Thursday's approaches to natural language seminar
C00438 00128 ∂10-Oct-83 2242 @SU-SCORE.ARPA:HDG@SU-AI chess clocks
C00440 00129 ∂10-Oct-83 2353 @SRI-AI.ARPA:PULLUM%HP-HULK.HP-Labs@Rand-Relay Job available at HP Labs.
C00443 00130 ∂11-Oct-83 0905 JMC*
C00444 00131 ∂11-Oct-83 0917 GROSZ@SRI-AI.ARPA use of CSLI mailing lists
C00445 00132 ∂11-Oct-83 0944 100 (from: Doug Ferguson) Inforλ λλ$[D$[D¬¬λλλ$[D$[D$[D
C00446 00133 ∂11-Oct-83 1119 DFH office keys
C00447 00134 ∂11-Oct-83 1400 AAAI-OFFICE@SUMEX-AIM Confirmation of R. Fikes
C00449 00135 ∂11-Oct-83 1438 PEREIRA@SRI-AI.ARPA Approaches to Computer Languages Seminar -- reminder
C00450 00136 ∂11-Oct-83 1454 @SU-SCORE.ARPA:HDG@SU-AI chess clocks addendum
C00451 00137 ∂11-Oct-83 1506 @SRI-AI.ARPA:SAG%HP-HULK.HP-Labs@Rand-Relay this week's colloquium
C00453 00138 ∂11-Oct-83 2225 JMC*
C00454 00139 ∂11-Oct-83 2326 GOLUB@SU-SCORE.ARPA Industrial Lectureships
C00457 00140 ∂11-Oct-83 2336 GOLUB@SU-SCORE.ARPA
C00459 00141 ∂11-Oct-83 2347 GOLUB@SU-SCORE.ARPA
C00460 00142 ∂11-Oct-83 2354 GOLUB@SU-SCORE.ARPA teaching schedule
C00462 00143 ∂12-Oct-83 0900 JMC*
C00463 00144 ∂12-Oct-83 1055 DFH Brazil ticket
C00464 00145 ∂12-Oct-83 1200 @SRI-AI.ARPA:TW@SU-AI Talkware seminar schedule (and current abstract)
C00467 00146 ∂12-Oct-83 1211 GOLUB@SU-SCORE.ARPA Visiting Scholar
C00471 00147 ∂12-Oct-83 1212 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 13
C00474 00148 ∂12-Oct-83 1227 GOLUB@SU-SCORE.ARPA
C00475 00149 ∂12-Oct-83 1401 YOM Manuals
C00476 00150 ∂12-Oct-83 1523 JJW EKL at LOTS
C00477 00151 ∂12-Oct-83 1539 MRC@SU-SCORE.ARPA primes
C00479 00152 ∂12-Oct-83 1546 MRC@SU-SCORE.ARPA
C00481 00153 ∂12-Oct-83 1614 BSCOTT@SU-SCORE.ARPA New ARPA Contract
C00484 00154 ∂12-Oct-83 1619 PEREIRA@SRI-AI.ARPA Approaches to Comp. Lang., Oct 27
C00486 00155 ∂12-Oct-83 1702 KJB@SRI-AI.ARPA Executive committee meetings
C00487 00156 ∂12-Oct-83 1759 DKANERVA@SRI-AI.ARPA October 13 Newsletter (No. 4)
C00498 00157 ∂13-Oct-83 1049 DELAGI%SUMEX-AIM.ARPA@SU-SCORE.ARPA SuperLunch - PreSeminar Meeting
C00501 00158 ∂13-Oct-83 1227 oliger%SU-NAVAJO.ARPA@SU-SCORE.ARPA Re: SuperLunch - PreSeminar Meeting
C00502 00159 ∂13-Oct-83 1229 ULLMAN@SU-SCORE.ARPA future talks
C00504 00160 ∂13-Oct-83 1433 DFH
C00505 00161 ∂13-Oct-83 1458 GOLUB@SU-SCORE.ARPA appointments
C00507 00162 ∂13-Oct-83 1617 CASLEY@SU-SCORE.ARPA Research Assistantship
C00508 00163 ∂13-Oct-83 1644 ROODE@SRI-NIC MCI Mail and disk storage
C00511 00164 ∂13-Oct-83 1806 pratt%SU-NAVAJO.ARPA@SU-SCORE.ARPA appointments
C00512 00165 ∂14-Oct-83 0948 GOLUB@SU-SCORE.ARPA Meeting on Tuesday
C00513 00166 ∂14-Oct-83 1123 KJB@SRI-AI.ARPA Seminar for next quarter
C00514 00167 ∂14-Oct-83 1319 DFH schedule conflict
C00515 00168 ∂15-Oct-83 1533 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS
C00517 00169 ∂15-Oct-83 1919 BrianSmith.pa@PARC-MAXC.ARPA Lisp as Language
C00522 00170 ∂15-Oct-83 1934 BrianSmith.pa@PARC-MAXC.ARPA [BrianSmith.pa: Lisp as Language]
C00528 00171 ∂17-Oct-83 0937 ALMOG@SRI-AI.ARPA reminder on why-context-wont-go-away
C00531 00172 ∂17-Oct-83 1113 HANS@SRI-AI.ARPA Mailing addresses for the individual research projects
C00534 00173 ∂17-Oct-83 1501 @SU-SCORE.ARPA:slndstrm@Shasta Visit Schedule - Burton Smith, Denelcor
C00537 00174 ∂17-Oct-83 1506 @SU-SCORE.ARPA:RPG@SU-AI Queue-based Multiprocessing Lisp
C00539 00175 ∂17-Oct-83 1510 ULLMAN@SU-SCORE.ARPA [Dick Gabriel <RPG@SU-AI>: Queue-based Multiprocessing Lisp ]
C00541 00176 ∂17-Oct-83 1521 KONOLIGE@SRI-AI.ARPA thesis defense
C00542 00177 ∂17-Oct-83 1701 ULLMAN@SU-SCORE.ARPA New Room
C00543 00178 ∂17-Oct-83 1936 CLT car
C00544 00179 ∂18-Oct-83 0908 DFH phone messages
C00545 00180 ∂18-Oct-83 1542 RINDFLEISCH@SUMEX-AIM.ARPA More Symbolics Machines Coming
C00547 00181 ∂18-Oct-83 1543 DFH Security form
C00548 00182 ∂18-Oct-83 1726 PETERS@SRI-AI.ARPA Thursday's Approaches to Natural Languages seminar
C00549 00183 ∂18-Oct-83 1741 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 20th
C00552 00184 ∂18-Oct-83 1831 CHANDRASEKARAN@RUTGERS.ARPA Simulation and Reasoning
C00556 00185 ∂19-Oct-83 0818 LIBRARY@SU-SCORE.ARPA Reply to McCarthy and Keller concerning Library Services
C00565 00186 ∂19-Oct-83 0924 @SRI-AI.ARPA:TW@SU-AI Abstract for Brian Reid talkware seminar today --NEW ROOM 380Y
C00568 00187 ∂19-Oct-83 1004 cheriton@Diablo Re: Reply to McCarthy and Keller concerning Library Services
C00570 00188 ∂19-Oct-83 1211 KONOLIGE@SRI-AI.ARPA Thesis Orals
C00571 00189 ∂19-Oct-83 1318 TW
C00572 00190 ∂19-Oct-83 1320 DFH Travel (Austin/Urbana)
C00573 00191 ∂19-Oct-83 1536 KJB@SRI-AI.ARPA visits
C00574 00192 ∂19-Oct-83 1750 BMACKEN@SRI-AI.ARPA budget
C00576 00193 ∂19-Oct-83 1903 @SRI-AI.ARPA:BRESNAN.PA@PARC-MAXC.ARPA A2 meeting
C00581 00194 ∂19-Oct-83 2051 DKANERVA@SRI-AI.ARPA Newsletter No. 5, October 20, 1983
C00608 00195 ∂20-Oct-83 0843 BSCOTT@SU-SCORE.ARPA Re: Martin Brooks expenses
C00609 00196 ∂20-Oct-83 0901 KJB@SRI-AI.ARPA Putting up the advisory panel
C00611 00197 ∂20-Oct-83 0944 DKANERVA@SRI-AI.ARPA Correction for Newsletter
C00612 00198 ∂20-Oct-83 0950 BMACKEN@SRI-AI.ARPA Associations
C00614 00199 ∂20-Oct-83 1013 DKANERVA@SRI-AI.ARPA Re: mailing lists
C00615 00200 ∂20-Oct-83 1126 @SU-SCORE.ARPA:TRATTNIG@SU-SIERRA.ARPA Today's lunch with Burton Smith
C00617 00201 ∂20-Oct-83 1605 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
C00619 00202 ∂21-Oct-83 0011 ARK A New Income Source for CSD-CF?
C00621 00203 ∂21-Oct-83 0950 @SU-SCORE.ARPA:GEORGEFF@SRI-AI.ARPA LISP 1980 Conference
C00623 00204 ∂21-Oct-83 1033 ME selling the news
C00625 00205 ∂21-Oct-83 1227 Winograd.PA@PARC-MAXC.ARPA Re: Lisp as Language
C00627 00206 ∂22-Oct-83 1143 AAAI-OFFICE@SUMEX-AIM.ARPA workshop
C00629 00207 ∂23-Oct-83 2117 ALMOG@SRI-AI.ARPA reminder on why context wont go away
C00641 00208 ∂24-Oct-83 1354 @SRI-AI.ARPA:BrianSmith.pa@PARC-MAXC.ARPA LISP as Language Course, Winter Quarter
C00648 00209 ∂24-Oct-83 1633 DFH
C00650 00210 ∂24-Oct-83 1706 ULLMAN@SU-SCORE.ARPA Next Seminar
C00653 00211 ∂25-Oct-83 0918 BOSACK@SU-SCORE.ARPA
C00655 00212 ∂25-Oct-83 1125 AAAI-OFFICE@SUMEX-AIM.ARPA fellowships
C00657 00213 ∂25-Oct-83 1157 @SU-SCORE.ARPA:DELAGI@SUMEX-AIM.ARPA Visit by PPA Developers
C00660 00214 ∂25-Oct-83 1410 TW
C00661 00215 ∂26-Oct-83 0018 GOLUB@SU-SCORE.ARPA Re: library keys
C00662 00216 ∂26-Oct-83 0505 CL.BOYER@UTEXAS-20.ARPA Re: visit
C00663 00217 ∂26-Oct-83 1138 AAAI-OFFICE@SUMEX-AIM.ARPA ACS
C00665 00218 ∂26-Oct-83 1545 GOLUB@SU-SCORE.ARPA
C00666 00219 ∂26-Oct-83 1748 PACK@SU-SCORE.ARPA Circumscription etc
C00671 00220 ∂27-Oct-83 1020 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doctoral fellowships
C00675 00221 ∂27-Oct-83 1046 BUCHANAN@SUMEX-AIM.ARPA Re: Post-Doctoral fellowships
C00676 00222 ∂28-Oct-83 0929 AAAI-OFFICE@SUMEX-AIM.ARPA vacation
C00677 00223 ∂29-Oct-83 0719 ATP.BLEDSOE@UTEXAS-20.ARPA Re: visit
C00679 00224 ∂30-Oct-83 1732 CLT Sarah
C00680 00225 ∂31-Oct-83 0916 DFH K. Clark Parlog paper
C00681 00226 ∂31-Oct-83 0955 DFH Claudia Mazzetti
C00682 00227 ∂31-Oct-83 1058 DFH D. Chudnovsky called
C00683 00228 ∂31-Oct-83 1250 ULLMAN@SU-SCORE.ARPA Next Talk
C00684 00229 ∂31-Oct-83 1545 @SU-SCORE.ARPA:DELAGI@SUMEX-AIM.ARPA PPA Lunch Location
C00685 00230 ∂31-Oct-83 2020 PACK@SU-SCORE.ARPA towers
C00689 00231 ∂01-Nov-83 1009 LGC H-P recommendation
C00693 00232 ∂01-Nov-83 1932 TREITEL@SUMEX-AIM.ARPA Security
C00695 00233 ∂01-Nov-83 2004 KJB@SRI-AI.ARPA Lunch?
C00697 00234 ∂01-Nov-83 2045 @MIT-MULTICS.ARPA:"Roy Goodman Amdahl(EX)."@ODEN.MAILNET Dear John,
C00700 00235 ∂01-Nov-83 2047 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET Testing connection from University of Stockholm
C00703 00236 ∂01-Nov-83 2258 CLT Calendar items
C00705 00237 ∂01-Nov-83 2313 Mailer failed mail returned
C00706 00238 ∂01-Nov-83 2343 ARK
C00708 00239 ∂02-Nov-83 0645 OP things being stolen
C00709 00240 ∂02-Nov-83 0813 DFH Vistnes AI Qual
C00710 00241 ∂02-Nov-83 0820 DFH Personnel Meeting
C00711 00242 ∂02-Nov-83 0932 TREITEL@SUMEX-AIM.ARPA
C00712 00243 ∂02-Nov-83 1016 LGC Where's Keith Clark?
C00713 00244 ∂02-Nov-83 1103 NILSSON@SRI-AI.ARPA interview
C00716 00245 ∂03-Nov-83 1537 PACK@SU-SCORE.ARPA Towers
C00717 00246 ∂04-Nov-83 0900 JMC*
C00718 00247 ∂04-Nov-83 1026 PACK@SU-SCORE.ARPA
C00719 00248 ∂04-Nov-83 1059 AAAI-OFFICE@SUMEX-AIM.ARPA '86 site selection
C00725 00249 ∂04-Nov-83 2355 long@csnet-cic Mail problem at CSNET-RELAY
C00727 00250 ∂05-Nov-83 1129 ALMOG@SRI-AI.ARPA talk
C00741 00251 ∂05-Nov-83 1459 AMAREL@RUTGERS.ARPA Re: '86 site selection
C00749 00252 ∂06-Nov-83 0032 RPG
C00756 00253 ∂06-Nov-83 1340 Fikes.pa@PARC-MAXC.ARPA Re: '86 site selection
C00758 00254 ∂07-Nov-83 0852 Hans.Berliner@CMU-CS-A your presidential msg
C00768 00255 ∂07-Nov-83 0900 JMC*
C00769 00256 ∂07-Nov-83 1031 DFH
C00770 00257 ∂07-Nov-83 1037 AAAI-OFFICE@SUMEX-AIM.ARPA Next meeting
C00772 00258 ∂07-Nov-83 1100 JMC*
C00773 00259 ∂07-Nov-83 1118 BUCHANAN@SUMEX-AIM.ARPA Re: Next meeting
C00774 00260 ∂07-Nov-83 1141 DFH
C00775 00261 ∂07-Nov-83 1214 ULLMAN@SU-SCORE.ARPA Next Seminar
C00778 00262 ∂07-Nov-83 1232 pratt@Navajo
C00786 00263 ∂07-Nov-83 1731 YOM Midterm
C00787 00264 ∂07-Nov-83 1905 YOM CS206
C00790 00265 ∂07-Nov-83 2227 YOM
C00791 00266 ∂08-Nov-83 1010 ullman%SU-HNV.ARPA@SU-SCORE.ARPA Browne workshop
C00793 00267 ∂08-Nov-83 1546 RESTIVO@SU-SCORE.ARPA
C00794 00268 ∂08-Nov-83 1619 RESTIVO@SU-SCORE.ARPA
C00796 00269 ∂08-Nov-83 1719 CLT
C00797 00270 ∂09-Nov-83 0943 STAN@SRI-AI.ARPA Mind & Action meeting Friday afternoon
C00802 00271 ∂09-Nov-83 1120 TW
C00803 00272 ∂09-Nov-83 1048 Raj.Reddy@CMU-CS-A Paris Trip
C00805 00273 ∂09-Nov-83 0958 AAAI-OFFICE@SUMEX-AIM.ARPA Future Conference sites
C00808 00274 ∂09-Nov-83 1705 ME Testing connection from University of Stockholm
C00811 00275 ∂09-Nov-83 2034 CLT project j
C00812 00276 ∂10-Nov-83 0955 KONOLIGE@SRI-AI.ARPA Oral exam
C00814 00277 ∂10-Nov-83 1500 AAAI-OFFICE@SUMEX-AIM.ARPA revised date for on-line abstract meeting
C00816 00278 ∂10-Nov-83 2117 ullman@Shasta panel
C00817 00279 ∂11-Nov-83 0132 HST VISIT
C00818 00280 ∂11-Nov-83 0824 ALMOG@SRI-AI.ARPA thanks
C00820 00281 ∂11-Nov-83 1153 sap@Shasta friday testing
C00821 00282 ∂11-Nov-83 1218 sap@Shasta
C00822 00283 ∂11-Nov-83 1440 LENAT@SU-SCORE.ARPA ...if you haven't heard it before
C00824 00284 ∂11-Nov-83 1622 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET Acknowledgement
C00826 00285 ∂13-Nov-83 0005 ALMOG@SRI-AI.ARPA baby's comment on friday
C00838 00286 ∂13-Nov-83 1613 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET addresses
C00841 00287 ∂13-Nov-83 2312 HST VISIT,LISP-CONF,ZEHE
C00842 00288 ∂13-Nov-83 2344 POURNE@MIT-MC upcoming meeting
C00843 00289 ∂14-Nov-83 1000 JMC*
C00844 00290 ∂14-Nov-83 1000 JMC*
C00845 00291 ∂14-Nov-83 1000 JMC*
C00846 00292 ∂14-Nov-83 1029 RPG
C00848 00293 ∂14-Nov-83 1139 DFH student appointment
C00849 00294 ∂14-Nov-83 1450 DFH meeting
C00850 00295 ∂14-Nov-83 1503 MWALKER@SU-SCORE.ARPA Kurt Konolige Oral Exam
C00851 00296 ∂14-Nov-83 1647 ULLMAN@SU-SCORE.ARPA seminar
C00854 00297 ∂14-Nov-83 1656 AAAI-OFFICE@SUMEX-AIM.ARPA Keynote speaker
C00856 00298 ∂15-Nov-83 0212 YM Reminder: CSD-CF town meeting
C00858 00299 ∂15-Nov-83 0219 YM
C00859 00300 ∂15-Nov-83 1059 DEK search committee
C00860 00301 ∂15-Nov-83 1324 BOSACK@SU-SCORE.ARPA
C00861 00302 ∂15-Nov-83 1410 BMOORE@SRI-AI.ARPA Re: baby's comment on friday
C00868 00303 ∂15-Nov-83 1428 ALMOG@SRI-AI.ARPA meeting
C00869 00304 ∂15-Nov-83 1428 ALMOG@SRI-AI.ARPA sorry
C00871 00305 ∂15-Nov-83 1521 DFH Dental appt.
C00872 00306 ∂15-Nov-83 1652 DFH Brooklyn college
C00873 00307 ∂15-Nov-83 1656 DFH flight reservations
C00874 00308 ∂15-Nov-83 2126 ALMOG@SRI-AI.ARPA msg
C00875 00309 ∂15-Nov-83 2324 mailer@diablo add knowledge
C00876 00310 ∂16-Nov-83 0928 JK
C00878 00311 ∂16-Nov-83 2353 YOM Ignorance
C00880 00312 ∂17-Nov-83 0005 YOM ignorance
C00885 00313 ∂17-Nov-83 0657 STAN@SRI-AI.ARPA Tinlunch & CBCL
C00887 00314 ∂17-Nov-83 0848 DFH
C00888 00315 ∂17-Nov-83 0932 @MIT-MC:Hewitt%MIT-OZ@MIT-MC limitations of logic
C00893 00316 ∂17-Nov-83 1137 YOM
C00894 00317 ∂17-Nov-83 1448 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doc Program
C00902 00318 ∂17-Nov-83 1455 @SU-SCORE.ARPA,@MIT-MC:RICH%MIT-OZ@MIT-MC AAAI
C00904 00319 ∂17-Nov-83 1619 DEK thanks for your thoughts
C00905 00320 ∂17-Nov-83 1638 DFH Inference Corp.
C00906 00321 ∂17-Nov-83 2152 KUO message
C00907 00322 ∂17-Nov-83 2157 KUO
C00908 00323 ∂17-Nov-83 2218 KUO message
C00909 00324 ∂17-Nov-83 2228 KUO
C00910 00325 ∂17-Nov-83 2233 KUO
C00911 00326 ∂18-Nov-83 0920 BSCOTT@SU-SCORE.ARPA Comtex Agreement
C00913 00327 ∂18-Nov-83 1001 PATASHNIK@SU-SCORE.ARPA csd-cf town meeting
C00915 00328 ∂18-Nov-83 1027 YM
C00916 00329 ∂18-Nov-83 1028 JF@SU-SCORE.ARPA finding the room
C00918 00330 Betty, I apologize for the misprint.
C00919 00331 ∂18-Nov-83 1632 DFH
C00920 00332 ∂18-Nov-83 1647 Brachman@SRI-KL.ARPA Keynote speakers for AAAI-84
C00924 00333 ∂18-Nov-83 2128 LLW@S1-A Missed Connections Regrets
C00931 00334 ∂19-Nov-83 2201 RPG Blocks
C00933 00335 ∂19-Nov-83 2316 ALMOG@SRI-AI.ARPA meeting
C00934 00336 ∂20-Nov-83 0230 POURNE@MIT-MC Upcoming meeting
C00935 00337 ∂20-Nov-83 0949 NILSSON@SRI-AI.ARPA (Response to message)
C00936 00338 ∂20-Nov-83 1018 BSCOTT@SU-SCORE.ARPA
C00937 00339 ∂20-Nov-83 1501 CLT
C00938 00340 ∂21-Nov-83 0543 WONG@SU-SCORE.ARPA Rotation number
C00940 00341 ∂21-Nov-83 1033 AAAI-OFFICE@SUMEX-AIM.ARPA Reminder
C00942 00342 ∂21-Nov-83 1216 RPG
C00943 00343 ∂21-Nov-83 1235 GENESERETH@SUMEX-AIM.ARPA rough draft
C00962 00344 ∂21-Nov-83 1443 @MIT-MC:RICKL%MIT-OZ@MIT-MC Re: limitations of logic
C00975 00345 ∂21-Nov-83 1547 @SU-SCORE.ARPA:oliger@navajo Visit of Ben Gerber of Cray Res. and NSF Initiative
C00977 00346 ∂21-Nov-83 1644 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doctorate program
C00987 00347 ∂21-Nov-83 2034 @USC-ECL:FIKES%USC-ECLD%USC-ECL@SRI-NIC Re: Post-Doctorate program
C00990 00348 ∂21-Nov-83 2204 BUCHANAN@SUMEX-AIM.ARPA Re: rough draft
C00992 00349 ∂21-Nov-83 2239 LEP Tower building in Prolog
C00994 00350 ∂22-Nov-83 1020 vardi@Diablo Re: knowledge seminar
C00995 00351 ∂22-Nov-83 1434 AMSLER@SRI-AI.ARPA Re: rough draft
C01018 00352 ∂22-Nov-83 1541 JJW Thesis topic
C01021 00353 ∂22-Nov-83 1703 DFH Alex Jacobson
C01022 00354 ∂23-Nov-83 1045 CLT car
C01023 00355 ∂23-Nov-83 1439 NILSSON@SRI-AI.ARPA Strips
C01026 00356 ∂23-Nov-83 2110 ULLMAN@SU-SCORE.ARPA Winter CS440
C01028 00357 ∂25-Nov-83 1518 @SU-SCORE.ARPA:cheriton@diablo POPL Paper on Concurrent Prolog
C01029 00358 ∂26-Nov-83 0153 ME reply addr
C01031 00359 ∂26-Nov-83 1305 perlis%umcp-cs@CSNet-Relay Re: logic.pro[f83,jmc] Proposal for logic in AI mailing list
C01032 00360 ∂26-Nov-83 1537 @MIT-MC:DAM%MIT-OZ@MIT-MC Edited Mailing List
C01040 00361 ∂26-Nov-83 2109 @MIT-MC:KDF%MIT-OZ@MIT-MC Logic in AI list
C01052 00362 ∂28-Nov-83 0151 HST VIXIT
C01053 00363 ∂28-Nov-83 0813 KJB@SRI-AI.ARPA talking
C01054 00364 ∂28-Nov-83 1000 JMC*
C01055 00365 ∂28-Nov-83 1017 GENESERETH@SUMEX-AIM.ARPA
C01056 00366 ∂28-Nov-83 1133 HAKEN@SUMEX-AIM.ARPA Resolution Theorem Proving
C01057 00367 ∂28-Nov-83 1212 ULLMAN@SU-SCORE.ARPA Next Meeting
C01059 00368 ∂28-Nov-83 1546 CLT supper tonight
C01060 00369 ∂28-Nov-83 1705 LGC
C01061 00370 ∂28-Nov-83 2300 JMC*
C01062 00371 ∂29-Nov-83 0939 @SU-SCORE.ARPA:CHAPPELL@SRI-AI.ARPA Tinlunch
C01063 00372 ∂29-Nov-83 1309 JK
C01066 00373 ∂29-Nov-83 1524 DFH appointment
C01067 00374 ∂29-Nov-83 1614 CHAPPELL@SRI-AI.ARPA Tinlunch 12/15
C01068 00375 ∂29-Nov-83 1733 SJG British public opinion
C01069 00376 ∂30-Nov-83 0006 100 (from: golu}) address
C01070 00377 ∂30-Nov-83 0423 @MIT-MC:JCMA%MIT-OZ@MIT-MC Edited Sub-List on philosophy-of-logic
C01078 00378 ∂30-Nov-83 0939 @MIT-MC:DAM%MIT-OZ@MIT-MC Edited Sub-List on philosophy-of-logic
C01082 00379 ∂30-Nov-83 1338 BOSACK@SU-SCORE.ARPA
C01090 00380 ∂30-Nov-83 1423 JJW CS 206
C01091 00381 ∂30-Nov-83 1423 DFH Woody Bledsoe
C01092 00382 ∂30-Nov-83 1425 BOSACK@SU-SCORE.ARPA
C01093 00383 ∂30-Nov-83 1514 WORLEY@SU-SCORE.ARPA
C01095 00384 ∂30-Nov-83 1548 DEW
C01096 00385 ∂30-Nov-83 1619 LIBRARY@SU-SCORE.ARPA Every day....
C01097 00386 ∂30-Nov-83 1627 SJG office
C01098 00387 ∂30-Nov-83 1630 ANDY@SU-SCORE.ARPA
C01101 00388 ∂30-Nov-83 1632 LIBRARY@SU-SCORE.ARPA The Artificial Intelligence Report
C01104 00389 ∂30-Nov-83 1701 BUCHANAN@SUMEX-AIM.ARPA Re: The Artificial Intelligence Report
C01106 00390 ∂30-Nov-83 2205 GROSOF@SUMEX-AIM.ARPA quote
C01107 00391 ∂30-Nov-83 2255 AAAI@SU-SCORE.ARPA
C01109 00392 ∂01-Dec-83 0046 LLW@S1-A Weekend Meeting RSVP
C01111 00393 ∂01-Dec-83 0429 JLH.BRADFORD@SU-SIERRA.ARPA
C01113 00394 ∂01-Dec-83 0747 LIBRARY@SU-SCORE.ARPA another version of quote
C01114 00395 ∂01-Dec-83 0840 TAJNAI@SU-SCORE.ARPA
C01115 00396 ∂01-Dec-83 0911 CC.CLIVE@UTEXAS-20.ARPA quotation
C01117 00397 ∂01-Dec-83 0912 RJW symetrizing
C01119 00398 ∂01-Dec-83 1008 WATROUS@RUTGERS.ARPA [COMEAU@RU-BLUE.ARPA: Re: [John McCarthy <JMC@SU-AI>:]]
C01121 00399 ∂01-Dec-83 1119 JJW
C01122 00400 ∂01-Dec-83 1421 @MIT-MC:DAM%MIT-OZ@MIT-MC LOGIC-IN-AI edited mailing list.
C01133 00401 ∂01-Dec-83 1654 DFH Fri. schedule
C01134 00402 ∂01-Dec-83 1753 WALDINGER@SRI-AI.ARPA
C01135 00403 ∂01-Dec-83 2327 vardi@Diablo Knowledge Seminar
C01137 00404 ∂02-Dec-83 1135 LUNDSTROM@SU-SCORE.ARPA [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
C01140 00405 ∂03-Dec-83 0900 JMC*
C01141 00406 ∂03-Dec-83 1326 SJG general
C01143 00407 ∂03-Dec-83 1420 SJG travel
C01144 00408 ∂03-Dec-83 1706 CHAPPELL@SRI-AI.ARPA TINLunch
C01145 00409 ∂03-Dec-83 1750 TREVOR@SU-SCORE.ARPA
C01146 00410 ∂03-Dec-83 2232 FEIGENBAUM@SUMEX-AIM.ARPA Re: The Artificial Intelligence Report
C01148 00411 ∂04-Dec-83 1104 @MIT-MC:BATALI%MIT-OZ@MIT-MC Comments on your Charter
C01152 00412 ∂04-Dec-83 1453 KONOLIGE@SRI-AI.ARPA Re: ignorance
C01154 00413 ∂04-Dec-83 2309 HST visit and compilation
C01156 00414 ∂04-Dec-83 2343 HST visit
C01157 00415 ∂05-Dec-83 0754 @MIT-MC:DAM%MIT-OZ@MIT-MC Comments on your Charter
C01161 00416 ∂05-Dec-83 0847 RPG Capital Equipment
C01162 00417 ∂05-Dec-83 0900 JMC*
C01163 00418 ∂05-Dec-83 0900 JMC*
C01164 00419 ∂05-Dec-83 1026 DFH Tues. a.m. off
C01165 00420 ∂05-Dec-83 1103 CLT Sato and proposal
C01167 00421 ∂05-Dec-83 1953 YOM
C01168 00422 ∂05-Dec-83 2141 POURNE@MIT-MC maximum effort
C01171 00423 ∂06-Dec-83 1410 ME news wires
C01172 00424 ∂06-Dec-83 1600 JMC*
C01173 00425 ∂06-Dec-83 1707 HALPERN.SJRLVM1.IBM-SJ@Rand-Relay knowledge seminar
C01177 00426 ∂06-Dec-83 1842 SJG contact
C01179 00427 ∂07-Dec-83 0008 BACH@SUMEX-AIM.ARPA re: To the person with the red pencil and nasty temperament: (from SAIL's BBOARD)
C01182 00428 ∂07-Dec-83 0900 JMC*
C01183 00429 ∂07-Dec-83 1017 DFH Lee Douglas
C01184 00430 ∂07-Dec-83 1123 DFH flight reservations
C01185 00431 ∂07-Dec-83 1458 SJG right track
C01186 00432 ∂07-Dec-83 1557 BSCOTT@SU-SCORE.ARPA Len Bosack
C01188 00433 ∂07-Dec-83 1651 BSCOTT@SU-SCORE.ARPA
C01189 00434 ∂08-Dec-83 0851 CLT
C01190 00435 ∂08-Dec-83 0914 @MIT-MC:DAM%MIT-OZ@MIT-MC Comments
C01191 00436 ∂08-Dec-83 1241 @SU-SCORE.ARPA:VLSI.SLNDSTRM@SU-SIERRA.ARPA CS440 TODAY - Bob Keller -Rediflow
C01195 00437 ∂08-Dec-83 1340 REGES@SU-SCORE.ARPA TA for 258?
C01196 00438 ∂08-Dec-83 1412 TW
C01199 00439 ∂08-Dec-83 1458 SHORTLIFFE@SUMEX-AIM.ARPA new AAAI-M liaison
C01201 00440 ∂08-Dec-83 1855 @MIT-MC:MACQUEEN@USC-ISIF bibliography and McCarty paper
C01203 00441 ∂08-Dec-83 1900 @MIT-MC:MACQUEEN@USC-ISIF oops...
C01204 00442 ∂09-Dec-83 0614 RMS@MIT-MC
C01206 00443 ∂09-Dec-83 1034 uucp@Shasta Lisp for IBM
C01207 00444 ∂09-Dec-83 1048 avg@diablo
C01208 00445 ∂09-Dec-83 1403 JJW Janet Lee's incomplete
C01209 00446 ∂09-Dec-83 1625 minker%umcp-cs@CSNet-Relay Gregory Minc
C01212 00447 ∂10-Dec-83 1015 JK
C01214 00448 ∂10-Dec-83 1427 CLT alice
C01215 00449 ∂11-Dec-83 1201 JJW Janet Lee
C01217 00450 ∂11-Dec-83 2025 RESTIVO@SU-SCORE.ARPA Re: What is the moral of this story?
C01218 00451 ∂11-Dec-83 2345 EJH@SU-SCORE.ARPA Re: What is the moral of this story?
C01219 00452 ∂12-Dec-83 1008 TW
C01221 00453 ∂12-Dec-83 1409 AAAI@SRI-AI.ARPA next Presidential message
C01222 00454 ∂12-Dec-83 1507 AAAI@SRI-AI.ARPA
C01223 00455 ∂12-Dec-83 1847 MACQUEEN@USC-ISIF
C01224 00456 ∂13-Dec-83 1610 BSCOTT@SU-SCORE.ARPA ARPA Funding
C01228 00457 ∂13-Dec-83 1818 Newman.ES@PARC-MAXC.ARPA Re: New Generation computing: Japanese and U.S. views
C01234 00458 ∂13-Dec-83 2032 TREITEL@SUMEX-AIM.ARPA Comparable quotes
C01236 00459 ∂13-Dec-83 2204 KJB@SRI-AI.ARPA Semantics of computer languages seminar
C01238 00460 ∂14-Dec-83 1038 vardi@diablo Knowledge Seminar
C01239 00461 ∂14-Dec-83 1050 JK proposals for japan
C01240 00462 ∂14-Dec-83 1055 JK
C01241 00463 ∂14-Dec-83 1206 MOGUL@SU-SCORE.ARPA Re: Comparable quotes
C01244 00464 ∂14-Dec-83 1357 TREITEL@SUMEX-AIM.ARPA Re: Comparable quotes
C01248 00465 ∂14-Dec-83 1401 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
C01250 00466 ∂14-Dec-83 1458 WALKER@SRI-AI.ARPA Re: next meeting
C01251 00467 ∂14-Dec-83 2308 JJW Let-by-need
C01253 00468 ∂16-Dec-83 0413 YOM Finals
C01256 00469 ∂16-Dec-83 0742 MCCORDUCK@COLUMBIA-20.ARPA Re: Doctor's dilemma
C01258 00470 ∂16-Dec-83 0910 WALKER@SRI-AI.ARPA Tutorials at IJCAI-85
C01260 00471 ∂16-Dec-83 1438 LENAT@SU-SCORE.ARPA Re: Tutorials at IJCAI-85
C01261 00472 ∂16-Dec-83 1648 DFH Messages
C01262 00473 ∂16-Dec-83 2231 POURNE@MIT-MC
C01265 00474 ∂17-Dec-83 1951 PYLYSHYN@CMU-CS-C.ARPA The Frame Problem and other enigmas
C01268 00475 ∂19-Dec-83 0900 JMC*
C01269 00476 ∂19-Dec-83 0924 PYLYSHYN@CMU-CS-C.ARPA Re: frame problem
C01271 00477 ∂19-Dec-83 1051 BSCOTT@SU-SCORE.ARPA Incentive Funds for 81-82
C01272 00478 ∂19-Dec-83 1529 MCCORDUCK@COLUMBIA-20.ARPA
C01274 00479 ∂19-Dec-83 2200 JMC*
C01275 00480 ∂20-Dec-83 0439 HST visit
C01277 00481 ∂20-Dec-83 1359 ZM RE: Consulting Profs
C01279 00482 ∂20-Dec-83 1532 KJB@SRI-AI.ARPA Lunch?
C01280 00483 ∂20-Dec-83 1538 KJB@SRI-AI.ARPA
C01281 00484 ∂20-Dec-83 2226 CLT visit
C01282 00485 ∂21-Dec-83 0913 DFH Room 252
C01283 00486 ∂21-Dec-83 1039 STAN@SRI-AI.ARPA Ehud Shapiro
C01285 00487 ∂21-Dec-83 1159 Nix.pa@PARC-MAXC.ARPA Re: Ehud Shapiro
C01287 00488 ∂21-Dec-83 1248 TW Consulting professorships
C01296 00489 ∂21-Dec-83 1440 YEARWOOD@SU-SCORE.ARPA Grades for 206
C01298 00490 ∂21-Dec-83 1423 KJB@SRI-AI.ARPA room
C01299 00491 ∂21-Dec-83 1515 Nix.pa@PARC-MAXC.ARPA Re: Ehud Shapiro
C01301 00492 ∂21-Dec-83 2220 ENGELMORE@SUMEX-AIM.ARPA Re: Winter 1983 or Winter 1984
C01303 00493 ∂22-Dec-83 1007 STAN@SRI-AI.ARPA Udi's abstract
C01307 00494 ∂22-Dec-83 1046 STAN@SRI-AI.ARPA Ehud Shapiro
C01308 00495 ∂22-Dec-83 1117 TW Joint appointments
C01314 00496 ∂22-Dec-83 1125 DMC meeting about partial pass of AI qual.
C01315 00497 ∂22-Dec-83 1147 KJB@SRI-AI.ARPA McCarthy's talks
C01316 00498 ∂22-Dec-83 1147 KJB@SRI-AI.ARPA oops
C01317 00499 ∂22-Dec-83 1152 CLT tonight
C01318 00500 Here are the flights my secretary, Diana Hall, found. I have decided
C01320 00501 ∂22-Dec-83 1754 DEK change my vote
C01321 00502 ∂23-Dec-83 0442 HALPERN.SJRLVM1.IBM-SJ@Rand-Relay knowledge seminar
C01324 00503 ∂24-Dec-83 1021 CLT turkey
C01325 00504 ∂24-Dec-83 1700 JMC*
C01326 00505 ∂24-Dec-83 1700 CLT*
C01327 00506 ∂26-Dec-83 1000 JMC*
C01328 00507 ∂26-Dec-83 1000 JMC*
C01329 00508 ∂27-Dec-83 0900 JMC*
C01330 00509 ∂27-Dec-83 1000 JMC*
C01331 00510 ∂27-Dec-83 1022 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
C01333 00511 ∂27-Dec-83 2323 LLW@S1-A Regrets and Greetings
C01335 00512 ∂28-Dec-83 0020 @SU-SCORE.ARPA:PARK@SRI-AI.ARPA ISIR14
C01337 00513 ∂28-Dec-83 0338 CL.BOYER@UTEXAS-20.ARPA Space
C01340 00514 ∂28-Dec-83 0903 MEYER@MIT-MC dirty Lisp
C01345 00515 ∂28-Dec-83 1032 CLT stoyan
C01346 00516 ∂28-Dec-83 1607 decvax!mulga!yorick.moncskermit@Berkeley Received: from
C01350 00517 ∂28-Dec-83 1802 WALKER@SRI-AI.ARPA Re: next meeting
C01351 00518 ∂29-Dec-83 1000 JMC*
C01352 00519 ∂29-Dec-83 1000 JMC*
C01353 00520 ∂29-Dec-83 1427 CLT tonight
C01354 00521 ∂30-Dec-83 0057 LGC Terminal Return
C01355 00522 ∂30-Dec-83 1145 ME long mail addresses
C01359 00523 ∂30-Dec-83 1325 BSCOTT@SU-SCORE.ARPA Indirect Cost Rebate
C01361 00524 ∂31-Dec-83 0430 ME
C01362 00525 ∂31-Dec-83 2110 CLT ?
C01370 00526 ∂31-Dec-83 2149 CLT calendar item
C01371 00527 ∂01-Jan-84 0054 WIEDERHOLD@SRI-AI.ARPA
C01372 00528 ∂01-Jan-84 1743 ME host name rand-cs
C01373 00529 ∂02-Jan-84 1827 TARNLUND@SRI-AI.ARPA a visit to Palo Alto
C01374 00530 ∂03-Jan-84 1130 FIKES%USC-ECLD%USC-ECL%MARYLAND.ARPA@USC-ECL.ARPA Executive Committee Meeting Minutes
C01382 00531 ∂03-Jan-84 1159 LUNDSTROM@SU-SCORE.ARPA [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
C01385 00532 ∂03-Jan-84 1357 GOTELLI@SU-SCORE.ARPA Gavan Duffy
C01386 00533 ∂04-Jan-84 0929 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
C01388 00534 ∂04-Jan-84 1027 ROD Shapiro.
C01389 00535 ∂04-Jan-84 1110 WALKER@SRI-AI.ARPA Re: next meeting
C01390 00536 ∂04-Jan-84 1152 TW Consulting professorship committee meeting
C01394 00537 ∂04-Jan-84 1622 ASHOK@SU-SCORE.ARPA CS 206 grades
C01396 00538 ∂04-Jan-84 1704 LUNDSTROM@SU-SCORE.ARPA [olken@LBL-CSAM (Frank Olken [csam]): Re: SIGBIG meeting]
C01401 00539 ∂05-Jan-84 2053 KJB@SRI-AI.ARPA Thomason paper
C01402 00540 ∂05-Jan-84 2053 KJB@SRI-AI.ARPA Thomason paper
C01403 00541 ∂05-Jan-84 2200 DEK proxy
C01404 00542 ∂05-Jan-84 2314 @SU-SCORE.ARPA:cheriton@diablo Distributed Prolog article
C01405 00543 ∂06-Jan-84 1543 DFH Stoyan accommodations
C01406 00544 ∂06-Jan-84 1646 DFH Messages
C01408 00545 ∂07-Jan-84 1442 BSCOTT@SU-SCORE.ARPA ARPA Umbrella Contract
C01411 00546 ∂07-Jan-84 2239 RWW STOYAN
C01412 ENDMK
C⊗;
∂28-Sep-83 1159 PEREIRA@SRI-AI.ARPA APPROACHES TO COMPUTER LANGUAGES Seminar
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Sep 83 11:59:45 PDT
Date: Wed 28 Sep 83 11:58:03-PDT
From: PEREIRA@SRI-AI.ARPA
Subject: APPROACHES TO COMPUTER LANGUAGES Seminar
To: csli-friends@SRI-AI.ARPA
cc: kjb@SRI-AI.ARPA
The first meeting will take place on Sep. 29 from 2:00 PM to 3:30 in
Redwood Hall room G19, when Brian Smith will introduce the seminar
with a discussion of the background of CSLI's work on computer
languages and their connections to human languages.
On Oct. 6, Fernando Pereira will speak on "Logic as a Programming
Language".
On Oct. 13, Terry Winograd will speak on "Specification Languages".
Any questions on this seminar to Fernando Pereira (tel. (415)859-5494,
net. pereira@sri-ai).
-------
∂28-Sep-83 1229 @SRI-AI.ARPA:BRESNAN.PA@PARC-MAXC.ARPA course rescheduled
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Sep 83 12:28:49 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Wed 28 Sep 83 12:29:32-PDT
Date: Wed, 28 Sep 83 12:21 PDT
From: BRESNAN.PA@PARC-MAXC.ARPA
Subject: course rescheduled
To: csli-people@sri-ai.ARPA
cc: BRESNAN.PA@PARC-MAXC.ARPA
The Linguistics Department has changed Linguistics 233 "Topics in
Syntactic Theory" from Monday to Friday from 12:15 to 3:05 in building
30, room 30B. A few members of CSLI may be interested in attending this
course, which I will be teaching. It is an advanced course that
presupposes familiarity with current syntactic theory. The course will
focus on linguistic evidence bearing on substantive hypotheses rather
than on mathematical or computational issues. I plan to lecture on the
following topics:
1) anaphoric binding systems
Anaphoric binding systems--that is, sets of reflexive and nonreflexive
pronouns with their specific properties of interpretation--cannot be
universally characterized in terms of the complementary distribution of
'bound' and 'free' pronouns. Study of Norwegian, Icelandic, and
English reveals three independent dimensions of anaphoric binding
systems: the subjective dimension (whether or not the antecedent is a
subject), the nuclear dimension (whether or not the anaphor lies in the
same clause nucleus as the anaphor), and the logophoric dimension
(whether or not the anaphor lies in an indirect discourse structure with
respect to the antecedent). These three dimensions define a space of
possible anaphors in terms of which one can locate the actual anaphors
of various languages. The structure of this theory is extremely simple,
and should throw light on the nature of the linguistic representations
that are semantically interpreted.
2) control and complementation
Control concerns the interpretation of infinitives and other predicative
expressions of natural language. The general theory of Bresnan's
(1982) "Control and Complementation" will be discussed and applied to
problems of control in Russian, Icelandic, and Warlpiri, as well as
English nominalizations and predicative phrasal constructions (so-called
"small clauses").
∂28-Sep-83 1407 Bossu.GIA@MIT-MULTICS.ARPA
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 28 Sep 83 14:07:20 PDT
Date: 28 September 1983 17:03 edt
From: Bossu.GIA at MIT-MULTICS
To: John McCarthy <JMC at SU-AI>
In-Reply-To: Message of 28 September 1983 00:18 edt from John McCarthy
thank you for Pat Hayes person←id.
i have sent a mail for him last week. (succesfully) but i have no
answer.(of course if he doesn't log in). so i will write or phone (it
was about the reception of a paper to review for AI).
sincerely yours G. Bossu
∂28-Sep-83 1442 ULLMAN@SU-SCORE.ARPA CS440
Received: from SU-SCORE by SU-AI with TCP/SMTP; 28 Sep 83 14:42:29 PDT
Date: Wed 28 Sep 83 14:40:23-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: CS440
To: super@SU-SCORE.ARPA
The first meeting of the public seminar (CS440) is tomorrow,
Thursday, Sept. 29, 4:15PM in 301 MJH.
Note that this is AFTER the Wilson seminar at 3PM in the CIS conference room.
By the way, I still need a speaker for Oct. 20th.
I'd volunteer myself, but will not be here.
-------
∂28-Sep-83 1452 RV AI Qual
To: JMC@SU-AI
CC: DFH@SU-AI, DBL@SU-AI, buchanan@SUMEX-AIM
As you may remember, at the beginning of the summer we made arrangements
that I retake the AI Qual the first week in October. Since that's next
week, I'm just reminding you. Shall I contact your secretary or you
to arrange for a date and time?
Thanks,
Rick
∂28-Sep-83 1634 HANS@SRI-AI.ARPA FTP of the associations file
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Sep 83 16:34:41 PDT
Date: Wed 28 Sep 83 16:25:18-PDT
From: Hans Uszkoreit <Hans@SRI-AI.ARPA>
Subject: FTP of the associations file
To: csli-folks@SRI-AI.ARPA
This will be of concern to people without SRI-AI accounts:
I have changed the protection for the directory <CSLI> and for the file
<CSLI>associations. Now you can get this file, which lists most members
of all research projects, via FTP by logging in as ANONYMOUS while in
FTP. (Give any password you like.)
Please, report problems to REQUESTS@SRI-AI.
-------
∂28-Sep-83 1655 DFH CS206 Information Sheets
Could you leave me a message indicating if these are
alright as is or what changes Yoram should make so I
can make copies before class tomorrow. Thanks.
∂28-Sep-83 1800 WALKER@SRI-AI.ARPA online access to AI abstracts
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Sep 83 18:00:24 PDT
Date: Wed 28 Sep 83 17:53:23-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: online access to AI abstracts
To: Genesereth@SU-SCORE.ARPA
cc: JMC@SU-AI.ARPA, Amsler@SRI-AI.ARPA, Feinler@SRI-NIC.ARPA, Fikes@SRI-AI.ARPA,
Nilsson@SRI-AI.ARPA, AAAI@SRI-AI.ARPA
Mike,
Claudia Mazzetti tells me that you are in the process of creating
an RFP with the intention of soliciting commercial vendors who would be
interested in establishing an online database of AI abstracts. If so, I
would like to argue strongly against such an approach on the grounds that
the 20-year-old technology reflected in DIALOG, ORBIT, and their brethren
is quite inappropriate for use by the AI community. You may not be aware
of the discussions that took place with the ARPA Network Information Center
last year; Lou Robinson was one of those involved from AAAI. Our concern
in these discussions was how to set up something that would build more on
what AI can be expected to deliver. It may be that the ARPANET may not be
an appropriate location, but I am concerned that any investment in the
existing commercial systems would preclude anything interesting ever being
done on the problem. That is precisely the situation now with the whole
field of bibliographic retrieval. We should be the people challenging the
current monopoly, not giving in to it.
I am not aware of this topic having been discussed in AAAI circles,
although I did miss the Council meeting in Washington. In any case, much
more extensive exploration of alternatives is indicated before the strategy
you are reported to be following is accepted. Since this field is in fact
my area of specialization, I may be more aware both of the problems with
current systems and of the prospects for much more interesting developments.
In any case, I would appreciate knowing more about the steps you have taken
up to this point. You may well have satisfied yourself that you have no
alternatives, but I believe there are some that AAAI should be considering.
Don
-------
We are not talking about storing abstracts but about storing the complete
documents. Are you willing to be on a committee that will make a proposal
to the Council?
∂28-Sep-83 1941 GENESERETH@SUMEX-AIM Re: meeting
Received: from SUMEX-AIM by SU-AI with PUP; 28-Sep-83 19:41 PDT
Date: Wed 28 Sep 83 19:44:09-PDT
From: Michael Genesereth <GENESERETH@SUMEX-AIM>
Subject: Re: meeting
To: JMC@SAIL
In-Reply-To: Your message of Wed 28 Sep 83 12:54:00-PDT
John,
I have some scattered meetings tomorrow, and I'll try to find you
in one of the cracks. Will you be in at any particular time?
mrg
-------
∂28-Sep-83 2101 RESTIVO@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 28 Sep 83 21:00:54 PDT
Date: Wed 28 Sep 83 21:02:28-PDT
From: Chuck Restivo <RESTIVO@SU-SCORE.ARPA>
To: jmc@SU-AI.ARPA
cc: restivo@SU-SCORE.ARPA
I request another appointment with you, please. Name the time and place.
-------
∂28-Sep-83 2201 lantz%SU-HNV.ARPA@SU-SCORE.ARPA FYI
Received: from SU-SCORE by SU-AI with TCP/SMTP; 28 Sep 83 22:00:44 PDT
Received: from Diablo by Score with Pup; Wed 28 Sep 83 21:58:03-PDT
Date: Wed, 28 Sep 83 21:57:47 PDT
From: Keith Lantz <lantz@Diablo>
Subject: FYI
To: super@Score
The University of Maryland was recently awarded 4.2 million dollars
by the National Science Foundation to develop the hardware and
software for a parallel processing laboratory. More than half of
the award amount is going directly for hardware acquisition, and
this money is also being leveraged through substantial vendor
discounts and joint research programs now being negotiated. We
will be buying things like lots of Vaxes, Sun's, Lisp Machines,
etc., to augment our current 2 780's, ethernet, etc. system.
Several new permanent positions are being created in the Computer
Science Department for this laboratory. The two permanent positions
now open are Laboratory Director and Hardware Specialist. Salaries
are negotiable. Working conditions are informal academic research.
Detailed position descriptions follow.
LABORATORY DIRECTOR. Duties: Manage and oversee the development of
hardware and software for a network of research computers; work
closely with faculty and graduate students on research projects;
administer and operate computer facility; plan, specify, and acquire
equipment to maintain a first class laboratory; conduct research
in hardware and software. Additional Factors: Extensive knowledge
of either software engineering or digital design; knowledge of
computer market and vendors; 3 years experience in related work.
Staffing: Applicant will hire and direct staff of 4 professional
programmers and engineers, 12 graduate and undergraduate workers,
and an administrative assistant. Remarks: Maryland already has
several Vaxes and a good complement of wizards and hardware hackers.
This is a good position for a Unix or hardware wizard with a flair
for adminstration who is about to get a PhD and wants to continue
the good life in academia.
HARDWARE SPECIALIST. Duties: Work with computer science researchers
in design and fabrication of innovative architectures; supervise
graduate assistant technicians; evaluate and recommend new hardware
acquisition for state-of-the-art research laboratory. Additional
Factors: Two years experience in design, fabrication, debugging,
and testing of microprocessor digital ciruits. Home computer
hardware hacking a plus. Remarks: We are currently building a
256 Z-80 machine designed here called the ZMOB. Completion of
this, upgrading it to faster busses and better chips (68000),
converting it to VLSI, and participating in the design of a
follow-on parallel architecture will be the work of the hardware specialist.
Anyone interested should make initial inquiries, send resumes, etc.
to Mark Weiser at one of the addresses below:
Mark Weiser
Computer Science Department
University of Maryland
College Park, MD 20742
(301) 454-6790/4251/6291 (in that order).
UUCP: {seismo,allegra,brl-bmd}!umcp-cs!mark
CSNet: mark@umcp-cs
ARPA: mark.umcp-cs@UDel-Relay
Expires: Oct 18, 1983
;;
;; Autologout
Kjob
.
∂29-Sep-83 0828 KJB@SRI-AI.ARPA rooms for today
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 08:28:26 PDT
Date: Thu 29 Sep 83 08:28:27-PDT
From: KJB@SRI-AI.ARPA
Subject: rooms for today
To: csli-friends@SRI-AI.ARPA
We found out late yesterday that the room in redwood is not
ready for use yet. For today we will meet
In Turing Auditorium for the morning seminar.
Here at Ventura for TINLunch.
In room 163E for the p.m. seminar and colloquium. This room is in
Building 160 at the front of the Quad.
It may be hard to make it back here from the seminar for tea, and then
back again for the colloquium. Still, just to emphasize how reliable
tea is, it will be served today for those who turn up.
Sorry for this confusion, and any inconvenience it causes. Hopefully
we will be in Redwood next week, and if not, know further in advance.
Jon
-------
∂29-Sep-83 0921 Raj.Reddy@CMU-CS-A Electronic Library Proposal
Received: from CMU-CS-A by SU-AI with TCP/SMTP; 29 Sep 83 09:20:06 PDT
Received: from [128.2.254.192] by CMU-CS-PT with CMUFTP; 29 Sep 83 12:08:50 EDT
Date: 29 Sep 83 1202 EDT (Thursday)
From: Raj.Reddy@CMU-CS-A
To: JMC@SU-AI
Subject: Electronic Library Proposal
CC: Raj.Reddy@CMU-CS-A
Message-Id: <29Sep83.120255.RR29@CMU-CS-A>
Dear John,
Here is an incomplete draft of the research plan. Could you go through
and make changes (put the changes between asterisks) and we will proceed
to change it and send it.
A PILOT PROJECT IN ELECTRONIC LIBRARIES
M. Griffith, J. McCarthy, R. Reddy
1. SUMMARY
2. PROBLEM
An Electronic Library, as the name suggests, is the electronic equivalent of
a Physical Library. It is expected that when a system capable of fulfilling
all the functions of a real library is developed, it can provide many new
functions that are currently unavailable: instantaneous access, multiple
access, reduced cost, and language translation aids.
Constructing an operational Electronic Library poses a number of technical,
social and legal problems at present. The purpose of the Pilot Project is to
bypass many of these problems for now, and concentrate on a few core problems
which do not require any new technical breakthroughs. The key problems to be
studied will involve the questions of how to acquire, represent, transmit, and
present Nineteenth Century French Literature. By selecting NCFL, we finesse
the issue of copyright, the problems of formulas and drawings, and ensure
widespread availability of French Literature.
3. BACKGROUND AND NEED
For some time it has been cost-effective to put the entire Library of
Congress into a computer file and make all its resources available to anyone in
the country with a computer terminal. There is no need to argue that all
printed paper will be abolished, but I would certainly get rid of ninety
percent of my books and magazines if I could access it from my terminal at
home.
It is now possible to get a gigabyte disk module for under $20,000. If we
count a book as 500,000 bytes, then this module can store 2000 books. The
space occupied by the module would store about 300 books on shelves. The cost
comes to $10 per book. Recent word information compression would give another
factor of four in storage density, reducing the cost to approximately $2.50 per
book and reducing the storage volume to one twenty-fourth of that required to
store books on shelves.
Books cost much more these days and so does the space required to store them,
although the cost of cataloging the books is apparently larger than either.
Recently IBM (3380) announced disk files storing 2.52 gigabytes per unit
which would store 20,000 compressed books in the space taken by 300 on shelves.
The Library of Congress would then require between 1000 and 2000 such disk
units.
Digital videodisks storing much more are predicted for the reasonably near
future, but the project is practical with technology now in hand. It is time
to begin.
Consider the following system. In addition to existing paper libraries,
there would exist one or more computerized libraries containing everything that
has ever been published, i.e. a computerized version of the Library of
Congress. This library would be accessible over the telephone network from any
computer terminal in the country. A reader could browse through the library
catalog and various bibliographies just as though he were physically present.
He could read any book by calling it page by page onto his terminal's screen or
he could have it transmitted to a local printer. At present, there are a
number of laser printers that print arbitrary fonts for less than $10,000, but
we can envisage cheaper printers in the future.
Most office workers would have terminals on their desks, and many people
would have them at home. At present a good enough terminal costs about $800,
and high quality terminals should cost about $2,000 if manufactured in moderate
quantity. Most offices can afford a high quality printer.
Of course, yet better terminals may eventually be available. We can imagine
a pocket terminal consisting of a rolled up plastic screen with a 1024 by 1024
array of liquid crystal dots accompanied by another rolled up pressure
sensitive keyboard and a pocket computer with enough memory to store a book.
Suppose that it has a modular jack that can plug into any telephone so that the
user can call the library, scan it for a while and then reload his book memory.
This would be nicer than the technology now available, but the available
technology is good enough to justify a start.
From the user's point of view, the advantages of the computerized library are
the following:
1. All books, magazines and newspapers are available.
2. Anything can be obtained in a few seconds.
3. Nothing is ever out.
4. The library is open 24 hours a day 365.2425 days a year.
Many paper libraries would be found unnecessary. In particular, university
libraries could carry out their functions with much less money and manpower,
since their users would switch to the electronic library for much of their
work.
The establishment of such a system involves many problems and will take some
years, but we will mention some facilities that can and should be started right
away. Moreover, people who don't have terminals at home now or on their desks
or don't use them at all may be difficult to convince of the advantages of such
a library.
4. PROBLEMS
It is expensive to convert the books to computer readable form. Equipment
for reading special type fonts is available and reliable. The recent Kurzweil
equipment reads arbitrary fonts with training but is reported to rely to a
substantial degree on a blind person's ability to know when something was
garbled and try again and on his ability to understand imperfectly read
material. The lowest error rates are apparently those obtained by the
Information International Grafix I system. This machine is very expensive,
mainly because it uses obsolete computer hardware, but the company would update
it if the market existed. Even if much of the material had to be retyped by
hand, the project would be worth what it would cost.
Of course, much new material is generated in computer-readable form, but many
forms are used, and as yet no-one has developed a system for putting all this
material into a common form.
The copyright law requires permission to put copyrighted material into
computer form. In my opinion, copyrights should be respected and suitable
financial arrangements based on readership should be negotiated. Once a
computerized library exists, it will be so much more accessible than other
libraries that authors and publishers will find it to their advantage to
negotiate suitable deals.
The best arrangement might be that the copyright owner could set whatever
price he pleased for reading his material. The reader could decide whether or
not to pay it.
There is a problem of unauthorized copying. The problem exists whether a
national library exists or not, and the temptations will increase as copying
machines get more convenient and cheaper and when a general purpose machine for
reading documents from paper to computer files becomes available.
At present an author gets ten to twenty percent of the retail price of his
books, except that he gets nothing for unsold books and less for mass market
paperbacks. An electronic publishing system could afford to give the author
eighty percent of the price paid by the readers, because there would be no
physical production or distribution costs. This would permit increased income
for authors and reduced prices for the readers. Presumably there is some price
elasticity for reading that would produce more reading with reduced prices and
greater convenience. This would greatly reduce the temptation to copy
illegally, since the reader would find it less burdensome to pay the writer his
due.
It is likely that the amount of illegal copying would be low enough so that
the system would survive. If not, we will eventually have to go to a system
where reading is essentially free and writers are paid according to a formula
by the Government. This would have many disadvantages, since no formula could
take fully into account the fact that different writers have different
abilities and put different amounts of work into books of different kinds. Of
course, the present system doesn't take this into account very well either, but
there are some works now that charge very high prices, i.e. newsletters.
These could still operate outside the standard system.
5. GETTING STARTED
Already there exist numerous databases available by telephone from anywhere
in the country. Some of them contain bibliographic information, i.e. abstracts
and references, but others contain the texts of the material. Some of them are
subsidized by government grants, e.g. many of the medical databases, and
others, e.g. the legal databases and the "New York Times" Databank, are profit
making businesses. The charges for using them range from $25 to $200 per hour
except for subsidized customers.
One important step could be taken by the Federal Government. It is required
by the Freedom of Information Act and other laws to make very large amounts of
information available to the public. This information would be much more
conveniently available if it were in a database accessible from anywhere in the
country. This especially includes the Federal Register where all new laws,
regulations, announcements of hearings and requests for comments are published.
6. TECHNICAL PROBLEMS
While it is easy to compute the costs of the storage media, which are already
cheaper than paper, it is harder to calculate the costs of the computers. This
is because present systems have not really been optimized for handling very
large numbers of users. It will also be necessary to optimize telephone
access. For this there are many possibilities.
A daytime cross-country call costs 54 cents for one minute. In a minute
36,000 bytes can be transmitted at 4800 bits/second. This means from $7.50 to
$15.00 to transmit a book uncompressed or from $1.87 to $3.75 with a
compression of 4. We can imagine a terminal that could store a minute's worth
of text and could decompress it for reading. These costs are unpleasantly
high, but they can be reduced in various ways. First, technology permits
substantially lower long distance transmission costs. Indeed the one minute
transcontinental charge late at night is 16 cents making our compressed book
cost from $.56 to $1.12 if transmitted all at once. This is probably less than
the cost of a trip to a library if one's time is worth much. The independent
long distance telephone companies are often 40 percent below AT&T, which brings
our optimistic number down to 33 cents, which is reminiscent of the days when
pocket books were a quarter.
We can suppose that the terminal would remember the telephone number and
catalog number and automatically phone for another minute's transmission when
the reader is close to the end of what it has in storage. These costs are even
less attractive when browsing is wanted. A solution for that is to use the
European telephone charging system which allows calls as short as 4 seconds.
Current networks keep the cost for maintaining a connection down by
time-sharing lines, but this doesn't reduce the cost of straight data
transmission.
An obvious possible saving is to have local libraries with frequently
consulted books and magazines. With optical fibers and other new means of
transmission, the transmission costs can be brought down to the point that
local libraries will be unnecessary.
7. FRENCH ELECTRONIC LIBRARY
The time is ripe for it to be socially worthwhile and economically feasible
to put the world literature in the French language into computer form and make
it available world wide.
Image the following system. The French language literature is put into
computer form, either by optical character recognition machines or by
keyboarding in low wage countries. A central computer library in France keeps
this literature on the equivalent of about 1000 IBM 3380 disk files. Three
large bandwidth satellites are put up to provide worldwide transmission
facilities. Reading rooms with suitable terminals are located in every place
where there is sufficient interest. A reader can call up any book or other
document from any terminal. When he does so, the first two pages are
transmitted via the satellite to the reading room computer and the first page
is displayed on his terminal. Perhaps the library catalog and other currently
popular documents are kept in local file.
8. CURRENT STATUS
<Mike Griffith to provide>
9. PLAN FOR RESEARCH
We propose to undertake the following pilot project.
1. A few RA81 disks are acquired from Digital Equipment Corporation and
attached to a VAX computer. This is currently the most
cost-effective disk file available.
2. A request for proposals for a few hundred thousand dollars worth of
book input is sent both to keyboarding companies and those that do
optical character recognition. In addition existing computerized
text is solicited from those who have it for experimental use. The
initial reading list is taken from the public domain literature.
3. About 20 telephone lines are attached to the VAX, so that the
library is available from existing terminals and micro-computers in
the Paris area.
4. The necessary programs are written and installed.
At this point a technical demonstration is feasible. An attempt is
made to determine what is most attractive to the users of the
library within the budget available.
5. An experimental terminal cluster is installed in a reading room in
the Paris area. It should be a place that is open for a large
number of hours.
If the results are encouraging, the second phase includes:
1. Giving the computerized library its own computer.
2. More books.
3. Obtaining the co-operation of publishers of current books, magazines
and newspapers for an expanded program. An experimental financial
arrangement should be adopted.
4. Design of a reading terminal that can be used in connection with the
French telephone system's electronic yellow pages.
5. An experimental reading room in an underdeveloped country using
existing satellite transmission channnels.
6. Developing an optical character recognition system optimized toward
reading books.
The pilot project is intended to lead to a demonstration by the end of 1984
with several thousand books on line.
1. EQUIPMENT PLAN - We expect to start with a VAX with a gigabyte of
memory as the EL Machine located at CMIRH in Paris. This machine
will have at least 32 lines permitting anyone in the Paris region
with a terminal, personal computer or a Minitel to be able to use
it. By 1985 we hope to extend the service throughout France using
the CMIRH network.
2. ACQUISITON - There are already several thousand books available at
"----- Le Langru Francais" at Nancy. We hope to acquire these. In
addition we hope to acquire a similar collection from Britain and
the USA. Also we will have about 1000 books manually entered in
Third World countries. This is expected to be quite inexpensive,
about 2000 FFr per book.
All these different books will probably come in different formats. We will
develop format conversion programs to put them in CMIRH standard format.
10. REPRESENTATION
Information on the disk will be stored in a compact form with frequently
occurring words coded and formatting information bracketed approximately.
Terminals and personal computers with local processing capability will
receive a decoding program followed by coded text which is expected to also
reduce the transmission time and cost. Dumb terminals will receive fully
decoded text. Decoding time should be less than 1 second per 10 words in
sequence.
11. TRANSMISSION
Initally only serial line transmission will be considered. VAX will support
up to 19.2 kiloband transmission. Terminals and personal computers with local
processing will be able to correct transmission error using Kermit-like
programs. They can also accept data at much higher rates for later
presentation at user specified rates.
12. PRESENTATION
It will be possible to access information from the on-line library from
almost all commmonly available terminals and personal computers.
However, from an ergonomic (human factors) point of view, high resolution
bit-mapped displays (equivalent in resolution to the FAX standard) with a
powerful personal computer with at least 2 megabytes of memory would be highly
desirable. Low cost versions (<$1000) of such terminals should be available by
the end of the decade. It is expected to take at least that long to acquire
and represent a substantial collection of books, reports and newspapers in
electronic form.
13. SELECTION
<What books will be on-line in the first year. Mike Griffiths to approach
Academe Francais.>
Table of Contents
1. SUMMARY 0
2. PROBLEM 0
3. BACKGROUND AND NEED 0
4. PROBLEMS 0
5. GETTING STARTED 0
6. TECHNICAL PROBLEMS 0
7. FRENCH ELECTRONIC LIBRARY 1
8. CURRENT STATUS 1
9. PLAN FOR RESEARCH 1
10. REPRESENTATION 1
11. TRANSMISSION 1
12. PRESENTATION 1
13. SELECTION 1
Raj
∂29-Sep-83 0949 WALKER@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 09:49:29 PDT
Date: Thu 29 Sep 83 09:41:41-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
cc: Amsler@SRI-AI.ARPA
In-Reply-To: Your message of Wed 28 Sep 83 18:16:00-PDT
For storing complete documents, my arguments are even more relevant.
I would be willing to be on a such a committee and would suggest that
my colleague Bob Amsler would also be extremely valuable on that
committee.
Don
-------
∂29-Sep-83 1003 AAAI-OFFICE@SUMEX-AIM Electronic Library Service meeting
Received: from SUMEX-AIM by SU-AI with PUP; 29-Sep-83 10:03 PDT
Date: Thu 29 Sep 83 10:06:04-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Electronic Library Service meeting
To: Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA, Nilsson@SRI-AI.ARPA, JMC@SAIL,
Buchanan@SUMEX-AIM, Genesereth@SUMEX-AIM
cc: aaai-office@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John has asked me to organize a meeting to discuss the proposed AAAI
sponsored electronic absract/library service. The planned meeting
will be held in Margaret Jacks Hall, CS Dept, Rm 252 on Wednesday,
October 5 at 11:00 am.
If you do not respond to this message, I can assume that you will be
attending the meeting.
--- Claudia
-------
∂29-Sep-83 1039 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 10:38:59 PDT
Date: Thu 29 Sep 83 10:39:28-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: 10/5, 11am, MJH252, Electronic Library Service meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: Amsler@SRI-AI.ARPA, Nilsson@SRI-AI.ARPA, JMC@SU-AI.ARPA,
Buchanan@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA
In-Reply-To: Your message of Thu 29 Sep 83 10:05:42-PDT
I plan to be there.
-------
∂29-Sep-83 1149 DFH ginsbe.1
Appears to be incomplete.
∂29-Sep-83 1153 RINDFLEISCH@SUMEX-AIM Symbolics 3600 Deliveries
Received: from SUMEX-AIM by SU-AI with PUP; 29-Sep-83 11:53 PDT
Date: Thu 29 Sep 83 11:53:30-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM>
Subject: Symbolics 3600 Deliveries
To: Pattermann@SUMEX-AIM, Yeager@SUMEX-AIM, Schmidt@SUMEX-AIM,
Genesereth@SUMEX-AIM, Grinberg@SUMEX-AIM, JMC@SAIL, RPG@SAIL, Bosack@Score,
Veizades@SUMEX-AIM, Spurgeon@SUMEX-AIM
cc: Rindfleisch@SUMEX-AIM, Feigenbaum@SUMEX-AIM, Buchanan@SUMEX-AIM,
CMiller@SUMEX-AIM, BScott@Score
This is to summarize the status of our Symbolics 3600 order and to
coordinate the plans for delivery. We have ordered 9 machines total
over the next year with 5 confirmed on funding from the DARPA Equipment
Contract for immediate delivery and 4 more delayed pending future
funding and Stanford's authorization to deliver. Of the initial 5, 2
have been shipped for arrival at MJH within about the next week. One
of these machines will be for the Formal Reasoning (FR) group and one
temporarily for the HPP. The remaining 3 should ship by end of October.
2 of these will go to Welch Road for the HPP and 1 will come to MJH for
FR. In addition, at that time, the 3600 temporarily used by HPP in MJH
from the first 2 will be turned over to FR. Thus, HPP will end up with
2 machines and FR with 3 and we will not have to move any of the
machines after initial installation.
So, the immediate need is to make arrangements for installing the first
machines and integrating them with the VAX's. Recall that this must
be over a 10 MBPS Ethernet and so we will need a board for the MJH VAX
and in the near future, one for the SUMEX file server. Symbolics
software is set up for the Interlan board. The contact at Symbolics
is Janet Perry for installation. Could Ed Pattermann and Len Bosack
coordinate with her over installation and acceptance testing and also
take care of ordering compatible 10 MBPS Ethernet boards for the VAX's
and working out integrating UNIX software support for the 3600.
Thanks, Tom R.
-------
∂29-Sep-83 1351 RINDFLEISCH@SUMEX-AIM Re: Symbolics 3600 Deliveries
Received: from SUMEX-AIM by SU-AI with PUP; 29-Sep-83 13:51 PDT
Date: Thu 29 Sep 83 13:48:53-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM>
Subject: Re: Symbolics 3600 Deliveries
To: Pattermann@SUMEX-AIM, Yeager@SUMEX-AIM, Schmidt@SUMEX-AIM,
Genesereth@SUMEX-AIM, Grinberg@SUMEX-AIM, JMC@SAIL, RPG@SAIL, Bosack@Score,
Veizades@SUMEX-AIM, Spurgeon@SUMEX-AIM
cc: Feigenbaum@SUMEX-AIM, Buchanan@SUMEX-AIM, CMiller@SUMEX-AIM, BScott@Score,
Rindfleisch@SUMEX-AIM, Mogul@Diablo
In-Reply-To: Your message of Thu 29 Sep 83 11:53:50-PDT
PS: Jeff Mogul reminds me that we will be moving to UNIX 4.2 in a
"short" while and that Symbolics had a hard time integrating CHAOSNET
support into 4.1. We need to understand when Symbolics will support
3600 file service under 4.2. Maybe they believe they will have IP/TCP
implemented by then and won't have to worry about CHAOSNET anymore?
Tom R.
-------
∂29-Sep-83 1437 RPG CL Directory
I think we should have some [*,C(ommon)L(isp)] directories now. We're
importing code from CMU now. I will arrange it with REG .
-rpg-
I suppose that if you arrange it today or tomorrow it can be with REG -
otherwise with LB.
∂29-Sep-83 1735 WALKER@SRI-AI.ARPA Fikes as AAAI Secretary-Treasurer
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 17:35:50 PDT
Date: Thu 29 Sep 83 17:36:34-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Fikes as AAAI Secretary-Treasurer
To: JMC@SU-AI.ARPA
cc: Fikes@PARC-MAXC.ARPA, Nilsson@SRI-AI.ARPA, AAAI@SRI-AI.ARPA,
Reddy@CMU-CS-A.ARPA
John,
As you know Richard Fikes is in line as a replacement for me as
AAAI secretary-treasurer. The appropriate procedure at this point is
to get Council approval (presuming that you as a key node in the process
have no objections). Accordingly, I suggest that you send out such a
message--or I will if you prefer--stating just that and indicating that
unless you here to the contrary within 7 (n?) days, the appointment is
considered confirmed.
I would like to retain my position on the Finance Committee, of
which Raj is Chairman, and which will then consist of Raj, me, you (as
President), and Richard (as Sec-Treas).
I will of course support Richard during the transitional phase
(and probably for some time after), transferring files, etc., as
appropriate.
Let me know just how you would like this handled.
Don
-------
I have asked Claudia to send a message to the Executive Committee on
my behalf proposing the change of secretary treasurers, and I agree
to the proposals in your message.
∂29-Sep-83 1626 PATTERMANN@SUMEX-AIM Re: Symbolics 3600 Deliveries
Received: from SUMEX-AIM by SU-AI with PUP; 29-Sep-83 16:26 PDT
Date: Thu 29 Sep 83 16:28:27-PDT
From: Ed Pattermann <PATTERMANN@SUMEX-AIM>
Subject: Re: Symbolics 3600 Deliveries
To: Rindfleisch@SUMEX-AIM
cc: Yeager@SUMEX-AIM, Schmidt@SUMEX-AIM, Genesereth@SUMEX-AIM,
Grinberg@SUMEX-AIM, JMC@SAIL, RPG@SAIL, Bosack@Score, Veizades@SUMEX-AIM,
Spurgeon@SUMEX-AIM, Feigenbaum@SUMEX-AIM, Buchanan@SUMEX-AIM,
CMiller@SUMEX-AIM, BScott@Score
In-Reply-To: Your message of Thu 29 Sep 83 11:53:52-PDT
Tom,
Why are we bothering with Chaosnet support anyway? If HPP is getting two
machines, can't they boot off of each other? Will HPP have a 9-track tape
drive for one of the units? Wouldn't the cost of a tape drive be a better bet
than putting resources into bringing up Chaosnet support? Maybe we should just
hang on till IP/TCP support arrives.
Also, you stated Symbolics software is setup for the Interlan board. You are
then buying the SYMBOLICS UNIX support software I gather. I hope this software
comes up better than the software came up for the LM-2 on Diablo. Also, I'm
not sure Bill Yeager's Welch gateway code will encapsulate Chaos packets at
this time. Bill, care to comment?
-- Ed
-------
∂29-Sep-83 1612 ULLMAN@SU-SCORE.ARPA apology
Received: from SU-SCORE by SU-AI with TCP/SMTP; 29 Sep 83 16:12:19 PDT
Date: Thu 29 Sep 83 16:06:26-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: apology
To: super@SU-SCORE.ARPA
My apologies to everyone who was sent on a wild-goose chase to
a nonexistent seminar at 3PM today.
-------
∂29-Sep-83 1538 WALKER@SRI-AI.ARPA [Don Walker <WALKER@SRI-AI>: [Derek Sleeman <SLEEMAN@SUMEX-AIM.ARPA>: AI Research Notes series]]
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 15:37:25 PDT
Date: Thu 29 Sep 83 15:28:18-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: [Don Walker <WALKER@SRI-AI>: [Derek Sleeman <SLEEMAN@SUMEX-AIM.ARPA>: AI Research Notes series]]
To: JMC@SU-AI.ARPA, Fikes@PARC-MAXC.ARPA, mack.ubc@RAND-RELAY.ARPA,
joshi.upenn@UDEL-RELAY.ARPA, amarel@RUTGERS.ARPA, hayes@ROCHESTER.ARPA,
raj.reddy@CMU-CS-A.ARPA
I sent the attached messages to the noted distribution list some time ago,
but have had no responses from AAAI and IJCAI; as noted, ACL has made other
commitments. Sleeman has sent several messages since to find out whether
I had gotten any reactions to his proposal. We can either tell him the
issue is under consideration--and appoint someone to consider it, or suggest
that we intend to handle such activities ourselves through our own publishing
connection--if in fact we do. Please respond with some guidance or a point
of view. My own preference is to work with someone else, and most likely
Kaufmann, but I would like to be able to say that some deliberation has
taken place.
---------------
Date: Mon 19 Sep 83 11:21:52-PDT
From: Don Walker <WALKER@SRI-AI>
Subject: [Derek Sleeman <SLEEMAN@SUMEX-AIM.ARPA>: AI Research Notes series]
To: rperrault, evens, mann@USC-ISIB, lauri@UTEXAS-20, weischedel@UDEL-RELAY,
acl@ROCHESTER, joshi.upenn@UDEL-RELAY, bonnie.upenn@UDEL-RELAY,
sondheimer@USC-ISIF, jrobinson, amarel@RUTGERS, bledsoe@UTEXAS-20,
brachman@SRI-KL, buchanan@SUMEX-AIM, krd@MIT-MC, bengelmore@SRI-KL,
lerman@SRI-KL, csd.genesereth@SU-SCORE, grosz@SRI-AI, hart@SRI-KL,
jmc@SU-AI, nilsson@SRI-AI, reddy@CMU-CS-A, stan@SRI-AI,
stefik@PARC-MAXC, gjs@MIT-MC, tenenbaum@SRI-KL, walker@SRI-AI,
bonnie.upenn@UDEL-RELAY, aaai-office@SUMEX-AIM, uw-beaver!ubc-vision!mack@LBL-CSAM,
ubc-vision!mack@UW-BEAVER, joshi.upenn@UDEL-RELAY, amarel@RUTGERS,
hayes@ROCHESTER, raj.reddy@CMU-CS-A
This message is being sent to the relevant people in ACL, AAAI, and IJCAI.
As noted, ACL has recently set up its own monograph series with Cambridge
University Press. It may be reasonable for AAAI and IJCAI to consider
working out similar arrangements, perhaps with Kaufmann. Reactions???
Don
---------------
1) 18-Sep Derek Sleeman AI Research Notes series
2) 19-Sep To: SLEEMAN@SUMEX-AI Re: AI Research Notes series
Message 1 -- ************************
Return-Path: <@SRI-KL.ARPA:SLEEMAN@SUMEX-AIM.ARPA>
Received: from SRI-KL.ARPA by SRI-AI.ARPA with TCP; Sun 18 Sep 83 15:10:23-PDT
Received: from SUMEX-AIM.ARPA by SRI-KL.ARPA with TCP; Sun 18 Sep 83 15:10:15-PDT
Date: Sun 18 Sep 83 15:10:36-PDT
From: Derek Sleeman <SLEEMAN@SUMEX-AIM.ARPA>
Subject: AI Research Notes series
To: walker@SRI-KL.ARPA
cc: sleeman@SUMEX-AIM.ARPA, sridharan@RUTGERS.ARPA
As you may have heard Pitman's will shortly start publishing an AI Research Notes
series. Sridharan and I are currently the main editors. The plan is to publish:
a) monographs b) collections of companion papers c) polished versions of theses.
Under b), we include selected, and usually enhanced, papers from conferences.
I would like to discuss with you the possibility of doing just that for IJCAI,
AAAI and ACL conferences.
I have put several copies of the prospectus in the mail to you at SRI, and hope
we can get in touch shortly,
Derek Sleeman
[Now physically at Stanford]
-------
Message 2 -- ************************
Date: Mon 19 Sep 83 10:27:23-PDT
From: Don Walker <WALKER@SRI-AI>
Subject: Re: AI Research Notes series
To: SLEEMAN@SUMEX-AIM
cc: sridharan@RUTGERS
In-Reply-To: Your message of Sun 18 Sep 83 15:10:26-PDT
Derek,
Thanks for the information about your new series. In direct
competition with you is the new series ACL has established with
Cambridge University Press, with the title "Studies in Natural Language
Processing." However, AAAI and IJCAI are less well organized, although
some discussions have taken place with William Kaufmann, Inc., the
publisher that handles distribution for proceedings from both organizations.
I will forward your message to the appropriate people in both AAAI and
IJCAI; I will also let people in ACL know of your plans. It is likely
that there is more than enough material to satisfy a range of publication
strategies. One policy position likely to be taken by everyone is that
if a substantial portion of the papers for a volume come from proceedings,
there should be some financial reimbursement to the organization. I'll
be glad to discuss these matters further following some feedback.
In any case, good luck,
Don
-------
-------
-------
Perhaps this is naive in some way, but it seems to me that there are
adequate avenues for publishing books and collections of papers in
AI and no need for either IJCAI or AAAI to take a hand. Let the
authors reap what financial rewards there are. Also I'd rather
AAAI used what initiative the Executive Committee can summon in
other ways.
∂29-Sep-83 1524 DFH Telex re Brazil conference
There is a telex from Renteria (I think) requesting
that we cable your acceptance of their invitation to
speak at Informatica 83--apparently their central bank
needs this in order to authorize issuance of your air
ticket. Before I send the cable, would you confirm that
what I told him about your return reservations is correct?
Also see his telex of 9/13. Essentially: Leave SFO evening
of Oct 12, arrive Oct 13; and I told him to make your return
reservation so that you arrived back here on Tues, Oct. 18,
which means leaving Sao Paulo either the night of Oct 17
or early on Oct 18. I would like to let him know if there
are any changes to this in the same telex.
∂29-Sep-83 1509 croft@Safe Re: Symbolics 3600 Deliveries
Received: from SU-SAFE by SU-AI with PUP; 29-Sep-83 15:09 PDT
Date: Thu, 29 Sep 83 15:11 PDT
From: Bill Croft <croft@Safe>
Subject: Re: Symbolics 3600 Deliveries
To: Bosack@SU-Score, Genesereth@Sumex-Aim, Grinberg@Sumex-Aim, JMC@Sail,
Pattermann@Sumex-Aim, RPG@Sail, Rindfleisch@Sumex-Aim,
Schmidt@Sumex-Aim, Spurgeon@Sumex-Aim, Veizades@Sumex-Aim,
Yeager@Sumex-Aim
Cc: BScott@SU-Score, Buchanan@Sumex-Aim, CMiller@Sumex-Aim,
Feigenbaum@Sumex-Aim, Mogul@Diablo
Regarding Symbolics support of TCP/IP, here is a note
I just received from Dave Plummer, who is implementing it.
They are not as far along as I had hoped, since they were
busy bringing up X.25.
Plummer also wishes us "good luck" in installing chaosnet
in VMUNIX 4.1c or 4.2, since they "have not found time" to
do it:
---- forwarded message
Date: 27 September 1983 21:36 EDT
From: David C. Plummer <DCP@MIT-MC>
Subject: TCP/IP on Symbolics 3600
To: croft%Safe@SU-SCORE
Cc: pattermann@SUMEX-AIM, schmidt@SUMEX-AIM
We decided HDLC/X.25 support was more important and realistic as
a 5.0 goal than TCP/IP was. Therefore, TCP/IP has been pushed
back to 5.1 and we really plan on doing it for 5.1 since there is
nothing on the horizon to get in our way this time. All the
multi-network support are in the 5.0 system and work.
I expect MIT will be the pre-alpha test site since they have a
pretty complex IP network which will find a lot of subtleties
(and because we have a lot of friends over there). Your site
would probably be an excellent alpha/beta test site. The person
you should talk to is Pete Miller (617-864-4660, Cambridge
timezone). I can probably give him adequate persuasion if he
sounds dubious.
Time scale: Not quite sure yet. We still haven't tied up all the
loose ends in preparation for 5.0 beta test. I would think we
will have gotten something that talks by the December holidays.
Then it is cleaning up loose ends and making it efficient. If
you are allowed to use the ARPAnet for such purposes, you can
probably get one or two day turnaround on bug reports, which
would help us as well.
Good luck on bring up chaos on your vax. Jim Kulp hasn't found
the time yet to make it work in 4.1c/4.2 at our site.
---- end forwarded message
∂29-Sep-83 2018 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 20:18:37 PDT
Date: Thu 29 Sep 83 20:12:45-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: 10/5, 11am, MJH252, Electronic Library Service meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: Amsler@SRI-AI.ARPA, Nilsson@SRI-AI.ARPA, JMC@SU-AI.ARPA,
Buchanan@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA
In-Reply-To: Your message of Thu 29 Sep 83 10:05:42-PDT
Does anyone have any objections to my asking Jake Feinler, director of
the ARPA Network Information Center to come to the meeting? Her group
is addressing questions that entail some of the same technology we will
be talking about, particularly for control of the literature in the NIC.
I am not sure that she will be available, because she has been in Europe
for the past two weeks and will be returning Monday, but I think she would
be good to have present, and her charter from DCA does encourage her to look
at some of these issues.
Don
-------
Sure. Invite Jake.
∂29-Sep-83 2153 YEAGER@SUMEX-AIM Re: Symbolics 3600 Deliveries
Received: from SUMEX-AIM by SU-AI with PUP; 29-Sep-83 21:52 PDT
Date: Thu 29 Sep 83 21:55:42-PDT
From: Bill <Yeager@SUMEX-AIM>
Subject: Re: Symbolics 3600 Deliveries
To: PATTERMANN@SUMEX-AIM
cc: Rindfleisch@SUMEX-AIM, Schmidt@SUMEX-AIM, Genesereth@SUMEX-AIM,
Grinberg@SUMEX-AIM, JMC@SAIL, RPG@SAIL, Bosack@Score, Veizades@SUMEX-AIM,
Spurgeon@SUMEX-AIM, Feigenbaum@SUMEX-AIM, Buchanan@SUMEX-AIM,
CMiller@SUMEX-AIM, BScott@Score
In-Reply-To: Your message of Thu 29 Sep 83 16:28:50-PDT
Well, if we must encapsulate and forward chaos net packets from the HPP-net
to the SUMEX-net to the SUMEX-10MB net, I certainly can put that in the
gateway. It won't happen next week but by the end of the month or early
October is certainly feasible.
This kind of thing is usually quite simple to impliment once the addressing
of the chaos net stuff is understood.
Bill
-------
∂29-Sep-83 2207 WALKER@SRI-AI.ARPA COLING 84 CALL FOR PAPERS
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Sep 83 22:07:40 PDT
Date: Thu 29 Sep 83 22:05:48-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: COLING 84 CALL FOR PAPERS
To: csli-friends@SRI-AI.ARPA
cc: Kay@PARC-MAXC.ARPA
******************************* PLEASE POST ********************************
CALL FOR PAPERS
COLING 84, TENTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL LINGUISTICS
COLING 84 is scheduled for 2-6 July 1984 at Stanford University, Stanford,
California. It will also constitute the 22nd Annual Meeting of the
Association for Computational Linguistics, which will host the conference.
Papers for the meeting are solicited on linguistically and computationally
significant topics, including but not limited to the following:
o Machine translation and machine-aided translation.
o Computational applications in syntax, semantics, anaphora, and
discourse.
o Knowledge representation.
o Speech analysis, synthesis, recognition, and understanding.
o Phonological and morpho-syntactic analysis.
o Algorithms.
o Computational models of linguistic theories.
o Parsing and generation.
o Lexicology and lexicography.
Authors wishing to present a paper should submit five copies of a summary
not more than eight double-spaced pages long, by 9 January 1984 to: Prof.
Yorick Wilks, Languages and Linguistics, University of Essex, Colchester,
Essex, CO4 3SQ, ENGLAND [phone: 44-(206)862 286; telex 98440 (UNILIB G)].
It is important that the summary contain sufficient information, including
references to relevant literature, to convey the new ideas and allow the
program committee to determine the scope of the work. Authors should clearly
indicate to what extent the work is complete and, if relevant, to what
extent it has been implemented. A summary exceeding eight double-spaced
pages in length may not receive the attention it deserves.
Authors will be notified of the acceptance of their papers by 2 April 1984.
Full length versions of accepted papers should be sent by 14 May 1984 to
Dr. Donald Walker, COLING 84, SRI International, Menlo Park, California,
94025, USA [phone: 1-(415)859-3071; arpanet: walker@sri-ai].
Other requests for information should be addressed to Dr. Martin Kay, Xerox
PARC, 3333 Coyote Hill Road, Palo Alto, California 94304, USA
[phone: 1-(415)494-4428; arpanet: kay@parc].
******************************* PLEASE POST ********************************
-------
∂29-Sep-83 2259 GOLUB@SU-SCORE.ARPA Appointment
Received: from SU-SCORE by SU-AI with TCP/SMTP; 29 Sep 83 22:58:57 PDT
Date: Thu 29 Sep 83 23:00:24-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: Appointment
To: feigenbaum@SU-SCORE.ARPA, jmc@SU-SCORE.ARPA
The two of you could make a strong case for Nels if you went to
the Dean together. I would be pleased to join you. GENE
-------
∂30-Sep-83 0702 BUCHANAN@SUMEX-AIM.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 30 Sep 83 07:02:34 PDT
Date: Fri 30 Sep 83 07:05:23-PDT
From: Bruce Buchanan <BUCHANAN@SUMEX-AIM.ARPA>
Subject: Re: 10/5, 11am, MJH252, Electronic Library Service meeting
To: WALKER@SRI-AI.ARPA
cc: AAAI-OFFICE@SUMEX-AIM.ARPA, Amsler@SRI-AI.ARPA, Nilsson@SRI-AI.ARPA,
JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA
In-Reply-To: Your message of Thu 29 Sep 83 20:21:57-PDT
sounds like a good idea to invite Jake Feinler -- the more information
we have the better.
bgb
-------
∂30-Sep-83 0834 LIBRARY@SU-SCORE.ARPA Socrates
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Sep 83 08:33:52 PDT
Date: Fri 30 Sep 83 08:35:13-PDT
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: Socrates
To: jmc@SU-AI.ARPA
Prof. McCarth,
It is my undetstanding that when the Stanford network is completed you will
be able to use Socrates through CIT. At this point, I would assume that you
would need an account with CIT. I have a call in to John Sack to verify
what I am telling you and will get back to you with verification or more
information.
Harry
-------
∂30-Sep-83 0900 RINDFLEISCH@SUMEX-AIM Re: Symbolics 3600 Deliveries
Received: from SUMEX-AIM by SU-AI with PUP; 30-Sep-83 09:00 PDT
Date: Fri 30 Sep 83 08:58:42-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM>
Subject: Re: Symbolics 3600 Deliveries
To: PATTERMANN@SUMEX-AIM
cc: Yeager@SUMEX-AIM, Schmidt@SUMEX-AIM, Genesereth@SUMEX-AIM,
Grinberg@SUMEX-AIM, JMC@SAIL, RPG@SAIL, Bosack@Score, Veizades@SUMEX-AIM,
Spurgeon@SUMEX-AIM, Feigenbaum@SUMEX-AIM, Buchanan@SUMEX-AIM,
CMiller@SUMEX-AIM, BScott@Score, Rindfleisch@SUMEX-AIM
In-Reply-To: Your message of Thu 29 Sep 83 16:28:51-PDT
Ed,
"Why are we bothering with Chaosnet support anyway?" I agree about the
long term but the LM-2 and 3600's currently use CHAOSNET protocols and
CHAOSNET is already installed in our UNIX 4.1. Why not use them so we
have flexibility for storing and moving files around between machines
to maximize the immediate usefulness of the Lisp Machines? We will need
10 MBPS connections for the 3600's whether we use CHAOSNET or IP/TCP
so it seems to me we have all the pieces and should make them play
together. We may well delay bringing up UNIX 4.2 on some of our
machines in order to retain CHAOSNET support until Symbolics gets IP/TCP
done. We will have a streamer tape for the 3600's to import Symbolics
software but that is not a generally available tape format around here.
"Also, you stated Symbolics software is setup for the Interlan board.
You are then buying the SYMBOLICS UNIX support software I gather." We
already have the SYMBOLICS UNIX support software installed and will get
an updated version with the InterLan 10 MBPS driver.
"Also, I'm not sure Bill Yeager's Welch gateway code will encapsulate
Chaos packets at this time." Since our move to Welch Road is likely to
be before Symbolics gets reliable IP/TCP support, it is highly desirable
for Bill's gateway to be able to route CHAOSNET packets out of Welch
Road to appropriate hosts and servers. MIT gateways already handle
CHAOS packets so we should be able to get what's involved from them.
The machines at Welch Road will connect to the 10 MBPS net modulo the
LM-2 which I'm not sure where it will end up.
Tom R.
-------
∂30-Sep-83 0926 WALKER@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 09:26:41 PDT
Date: Fri 30 Sep 83 09:27:41-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Your message of Fri 30 Sep 83 00:20:00-PDT
John,
Given that position, Sleeman, as one of those with initiative, needs to
know how we will handle a mass cannibalizing of the proceedings into
volumes that he can peddle through his publisher. AAAI has done all
the work. Are we willing to allow him free reign over use of the
results? Or do we ask for 50% of whatever royalties are offered, or
75% or 100%? Books like that will cut into sales of the proceedings,
which are an important income item. If we work through Kaufmann, for
example, we can at least compensate for lost proceedings income with
the sale of collections. Alternatively, we could limit the number of
papers someone could assemble from a particular set of proceedings.
Our current practice has been to allow people to reprint individual
papers in a collection without asking for reimbursement. So far,
requests have only been for a few papers within a volume, and I have
taken to requiring the credit line to include the information that
Kaufmann distributes the proceedings, which may provide some financial
return. Another way would be to establish some fee per article.
I would have preferred that Sleeman and Sridharan had not asked, but their
asking provokes some policy issues that have to be dealt with in some way.
Don
-------
∂30-Sep-83 0948 WALKER@SRI-AI.ARPA Re: 10/5, 11am, MJH252, Electronic Library Service meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 09:48:01 PDT
Date: Fri 30 Sep 83 09:45:25-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: 10/5, 11am, MJH252, Electronic Library Service meeting
To: BUCHANAN@SUMEX-AIM.ARPA
cc: AAAI-OFFICE@SUMEX-AIM.ARPA, Amsler@SRI-AI.ARPA, Nilsson@SRI-AI.ARPA,
JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA
In-Reply-To: Your message of Fri 30 Sep 83 07:04:28-PDT
have invited Jake following John's earlier reply; her presence of course
depends on being able to break away so soon after two week's absence, but I
will follow up by phone on my message to her just to make sure she knows
about it
-------
∂30-Sep-83 1008 RPG Time for a change
When you got autologged-out last night, your password remained visible on
the then unhidden screen. I erased that screen at 10am; it was visible from
18:54 yesterday.
-rpg-
∂30-Sep-83 1059 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 10:56:52 PDT
Received: from SU-AI.ARPA by SRI-AI.ARPA with TCP; Fri 30 Sep 83 10:55:55-PDT
Date: 30 Sep 83 1049 PDT
From: Carolyn Talcott <CLT@SU-AI>
Subject: SEMINAR IN LOGIC AND FOUNDATIONS
To: "@DIS.DIS[1,CLT]"@SU-AI
Organizational and First Meeting
Time: Wednesday, Oct. 5, 4:15-5:30 PM
Place: Mathematics Dept. Faculty Lounge, 383N Stanford
Speaker: Ian Mason
Title: Undecidability of the metatheory of the propositional calculus.
Before the talk there will be a discussion of plans for the seminar
this fall.
S. Feferman
[PS - about distribution lists -
I have added CSLI-folks@SRI-AI to my logic distribution list,
if you receive this notice twice it is probably because you were
already on the orginal distribution list. Send me a message and I
will remove the redundancy. If you read this notice on a bboard
and would like to be on the distribution list send me a message.
I you received this message as electronic mail and would like to
be deleted from the list also send me a message.
- CLT@SU-AI]
∂30-Sep-83 1208 WALKER@SRI-AI.ARPA Re: mailing list
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 12:07:48 PDT
Date: Fri 30 Sep 83 12:06:40-PDT
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: mailing list
To: JMC@SU-AI.ARPA
In-Reply-To: Your message of Fri 30 Sep 83 10:55:00-PDT
I'll certainly answer Raj; AAAI is an alternative address for the
AAAI office; it is easier to type than the sumex one
-------
∂30-Sep-83 1256 JRP@SRI-AI.ARPA Meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 12:56:05 PDT
Date: 30 Sep 1983 1254-PDT
From: Jrp at SRI-AI
Subject: Meeting
To: JMC at SAIL, WASOW%SU-PSYCH at SCORE, WASOW at PARC,
To: KAY at PARC, STAN, RIGGS, jrp
You all are the education committee. We need to meet. Some of
the items on the agenda.
Status of students at CSLI until we get a program.
We have given some students support. Right now, thats
the only status we have to confer on students. But
some students who dont need support should be encouraged
to have a connection with CSLI, and should be granted some
privileges, such as computer time. How formal should
this be?
How do we arrive at what our program should be?
We described a program in the proposal. It will have
to take a more definite shape before it gets implemented.
Do we want to get some student input for this? Enlarge
the committee? Appoint a subcommittee? Discuss it
with departments? All of the above? None of the above?
Advisors for the students we are supporting.
I told each of the students we are giving RAships to
that in addition to their regular advisor, CSLI would
appoint them a counterpoint advisor, someone whose
interests overlap withtheirs, but whose background
is quite different. If this is a good idea, we need
to make some decisions.
Other items that come to mind before the meeting.
----------
In order to have this meeting, we need to settle on a time. Let me
propose that the meeting be in the afternnon next week, and be
scheduled for one hour, and be held at Ventura Hall. Would all
of you then send a message to RIGGS
@SRI-AI (My secretary), telling which of the following hours
are possibilities for you:
Wednesday 1;15, 2;15, 3;15; 4;15
Thurs, from after tin lunch to 2;15
Friday, 1;15, 2;15, 3;15, 4;15
----------
To prepare for the meeting, you might i) read the relevant parts of
the proposal. I'll try to put the relevant parts online, if I can
find them.
ii) If you are at Stanford, look at your own depts program
and think about how are program might fit in with it, consult with
the people in your dept. who might have ideas; think;
iii) If not at Stanford, do the above anyway with respect
to the depts. with whom, we hope, you will soon be associated;
consult your wider experience, etc.
-----------
I will request that educ.comm be a Csli mailing list, to facilitate
commication, but until then you can invoke the file <jrp>educ.committee
if you want to exchange idea.
-------
∂30-Sep-83 1308 Kay.PA@PARC-MAXC.ARPA Re: Meeting
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 30 Sep 83 13:08:30 PDT
Date: Fri, 30 Sep 83 13:09 PDT
From: Kay.PA@PARC-MAXC.ARPA
Subject: Re: Meeting
In-reply-to: "Jrp@SRI-AI.ARPA's message of 30 Sep 83 12:54 PDT"
To: RIGGS@SRI-AI.ARPA
cc: JMC@SAIL.ARPA, WASOW%SU-PSYCH@SCORE.ARPA, WASOW.PA@PARC-MAXC.ARPA,
KAY.PA@PARC-MAXC.ARPA, STAN@SRI-AI.ARPA
I think I can manage any of the proposed times. My preference, which I
was not asked for, would be for Thurs, from after tinlunch to 2;15.
--Martin
∂30-Sep-83 1331 AAAI-OFFICE@SUMEX-AIM
Received: from SUMEX-AIM by SU-AI with PUP; 30-Sep-83 13:31 PDT
Date: Fri 30 Sep 83 13:34:34-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
To: JMC@SAIL
cc: AAAI-OFFICE@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
In-Reply-To: Your message of Fri 30 Sep 83 10:54:00-PDT
John,
Although it is unofficial yet, how about including R. Fikes (Fikes.PARC)
in our discussions. Otherwise, those names above are the only names I know
that have expressed an interest in this proposal.
-- Claudia
P.S. It is "Feinler@SRI-NIC." So far, Nils is the only one not able to
attend the meeting.
-------
∂30-Sep-83 1339 JANET@KESTREL maclisp manual
Received: from KESTREL by SU-AI with TCP/SMTP; 30 Sep 83 13:39:11 PDT
Date: 30 Sep 1983 1339-PDT
From: Janet Coursey <Janet at KESTREL>
Subject: maclisp manual
To: jmc at SU-AI
Reply-To: jvc@score
The bookstore is sold out of the new Pittman Maclisp manual.
They will not order more until they are told by a professor how many
more to order. I think that students of Lenat's lisp course, cs102,
probably purchased many of them, although it was not required.
Also hackers at large from CSD and Lots probably purchased many---
a decent manual has been long awaited. Maybe an unsatisfied demand
count can be taken in cs102 and cs206, and the sum plus spares ordered.
The bookstore said the time of arrival would be about a month after the
order; MIT is slow.
Janet Coursey
-------
∂30-Sep-83 1336 DKANERVA@SRI-AI.ARPA Newsletter, September 29, 1983
Received: from SRI-AI by SU-AI with TCP/SMTP; 30 Sep 83 13:36:00 PDT
Date: Fri 30 Sep 83 13:34:45-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: Newsletter, September 29, 1983
To: CSLI-friends@SRI-AI.ARPA
This page is just to get the Diablo to work!
! CSLI Newsletter
September 29, 1983 * * * Number 2
The first issue of CSLI Newsletter was well received, if demand for
copies is any indication, so here is a second installment. A few of
the items will be a bit out of date by the time this reaches you, but
it might help remind you to get your announcements in early.
Starting with this issue, I'll be putting the newsletter together, so
please send announcements and other material you want included in the
newsletter to me, DKanerva@SRI-AI.
- Dianne Kanerva
* * * * * *
Research Seminar: Approaches to Natural Language
The first CSLI research seminar in natural language will be given by
Joan Bresnan, who will speak on issues in syntactic theory.
Place: Redwood Hall, classroom
Time: Thursday, September 29, 10:00 am
The next seminar, to be held on Thursday, October 6, will be given by
Bob Moore, and will have as its topic "Problems in Semantic Analysis
of Natural Language."
* * * * * *
Research Seminar: Approaches to Computer Languages
The first meeting will take place on September 29, 2:00-3:30, in
Redwood Hall , Room G19, when Brian Smith will introduce the seminar
with a discussion of the background of CSLI 's work on computer
languages and their connections to human languages.
The next seminar, to be held on Thursday, October 6, Fernando
Pereira will speak on "Logic as a Programming Language." On
October 13, Terry Winograd will speak on "Specification Languages."
Questions about this seminar can be referred to Fernando Pereira
(phone 415-859-5494, net Pereira@SRI-AI).
* * * * * *
TINLunch Schedule
TINLunch will be held on each Thursday at Ventura Hall on the Stanford
University campus as a part of CSLI activities. Copies of TINLunch
papers will be available in EJ251 at SRI and in Ventura Hall at Stanford
University.
!
October 6: "The Role of Grammars in Models of Language Use"
by
Robert C. Berwick and Amy S. Weinberg
Schedule of people in charge of discussions:
October 6 Marilyn Ford
October 13 Michael Georgeff
October 20 Per-Christian Halvorsen
October 27 Jerry Hobbs
* * * * * *
CSLI Research Program in Situated Language
The description of the research program will be available for
distribution early next week. Don Walker suggested that we publish
it in the ACL Newsletter, which seems like a good idea. Of course,
he has not seen how long it is, so it might have to be shortened for
there.
* * * * * *
CSLI and COLING
As many of you know, COLING (an international COmputational
LINGuistics organization) will be meeting at Stanford next July. CSLI
has agreed to help out by sponsoring a series of invited lectures.
Ray Perrault is serving as CSLI Liaison with COLING. If you have any
suggestions for good speakers, or other suggestions for helping to
make the meeting valuable for everyone, let him know.
* * * * * *
Inner Colloquium
Our second colloquium will be this Thursday, Sept. 29, from 4:00-5:00
in the classroom of Redwood Hall (please note this is a change from
the previous building, Polya Hall.) Our speaker will be Robert C.
Moore of SRI. An abstract of his presentation follows:
DEDUCTIVE METHODS FOR COMMONSENSE REASONING
by Robert C. Moore
AI Center, SRI International
Automatically drawing conclusions from a base of commonsense,
factual information has always been considered one of the central
problems of artificial intelligence. This talk is an overview of a
broad spectrum of issues and ideas that have arisen in attempting to
solve this problem. It will attempt to sketch out some of the
connections between various notions, including: rule-based systems,
forward and backward chaining, first-order logic, resolution
theorem-proving, natural deduction, pattern matching, unification,
horn-clause logic, the closed-world assumption, reasoning from
incomplete knowledge, connection graphs, logic programming,
multi-sorted logic, property inheritance, higher-order logic, modal
and intensional logic.
! * * * * * *
Outer Colloquium
Our first outer colloquium will be held Thursday, Oct. 5, 1983
(contrary to the mistaken announcement in last week's newsletter). The
speaker will be Dr. Geoffrey Nunberg of the Stanford Linguistics
Department. The title is "Prescriptive Grammar."
* * * * * *
Remote Affiliates
The Executive Committee decided at its September 29 meeting not to
begin a remote-affiliate program at this time.
* * * * * *
Organizational Meeting for Project A2:
"Syntactic Constraints on Discourse"
A short organizational meeting will be held on Tuesday, October 4, at
3:00 p.m. in Ventura (the conference room).
* * * * * *
Logic Seminar
Sol Feferman's Fall Quarter Logic Seminar will have an organizational
meeting Wednesday, October 5, 4:15-5:30, at the Math Corner. Ian Mason
will then speak about some of his work.
* * * * *
New Seminar in Computer Science Department
Terry Winograd, Wednesdays at 2:15, Jacks Hall 301
Terry Winograd will be organizing a weekly seminar in the computer science
Department this fall on a new area he is currently developing as a research
topic: the theory of "talkware." This area deals with the design and
analysis of languages that are used in computing but are not programming
languages. These include specification languages, representation languages,
command languages, protocols, hardware description languages, data base
query languages, etc. There is currently a lot of ad hoc but sophisticated
practice for which a more coherent and general framework needs to be
developed. The situation is analogous to the development of principles of
programming languages from the diversity of "coding" languages and methods
that existed in the early fifties.
The seminar will include outside speakers and student presentations of
relevant literature, emphasizing how the technical issues dealt with in
current projects fit into the development talkware theory. It will meet
at 2:15 every Wednesday in Jacks 301. The first meeting will be Wednesday,
September 28. All CSLI people are welcome to attend, and urged to volunteer
as a speaker if they are doing relevant work. For a more extensive
description, see {SCORE or MAXC}<WINOGRAD>TALKWARE or {SAIL}TALKWA[1,TW].
!
* * * * *
Linguistics Course Rescheduled
The Linguistics Department has changed Linguistics 233 "Topics in
Syntactic Theory" from Monday to Friday, 12:15-3:05 p.m., Bldg. 30,
Room 30B. A few members of CSLI may be interested in attending this
course, which Joan Bresnan will be teaching. It is an advanced course that
presupposes familiarity with current syntactic theory. The course will
focus on linguistic evidence bearing on substantive hypotheses rather
than on mathematical or computational issues. The following topics will
be included:
1. Anaphoric Binding Systems
Anaphoric binding systems--that is, sets of reflexive and nonreflexive
pronouns with their specific properties of interpretation--cannot be
universally characterized in terms of the complementary distribution of
`bound' and `free' pronouns. Study of Norwegian, Icelandic, and
English reveals three independent dimensions of anaphoric binding
systems: the subjective dimension (whether or not the antecedent is a
subject), the nuclear dimension (whether or not the anaphor lies in the
same clause nucleus as the anaphor), and the logophoric dimension
(whether or not the anaphor lies in an indirect discourse structure with
respect to the antecedent). These three dimensions define a space of
possible anaphors in terms of which one can locate the actual anaphors
of various languages. The structure of this theory is extremely simple,
and should throw light on the nature of the linguistic representations
that are semantically interpreted.
2. Control and Complementation
Control concerns the interpretation of infinitives and other predicative
expressions of natural language. The general theory of Bresnan's
(1982) "Control and Complementation" will be discussed and applied to
problems of control in Russian, Icelandic, and Warlpiri, as well as
English nominalizations and predicative phrasal constructions (so-called
"small clauses").
* * * * *
West Coast Conference on Formal Linguistics, 1984
The third annual West Coast Conference on Formal Linguistics (WCCFL 3) has
now been scheduled for Friday March 16 - Sunday March 18, 1984, at the
University of California, Santa Cruz. This conference is the premier West
Coast forum for work relating to generative grammar (including phonology)
and formal semantics. A call for abstracts will go out by mail soon.
Anyone interested in being on the mailing list should write to WCCFL 3,
Syntax Research Center, Cowell College, UCSC, Santa Cruz, CA 95064. Further
information will be provided as and when further arrangements are made.
- Geoff Pullum
! * * * * *
Report from the Computing Committee
An Imagen printer is being ordered for Ventura, to be installed within
a few weeks. Eight 9600-baud lines are being ordered between Ventura and
SRI to improve access to the SRI 2060. It will eventually be possible
to use these lines in the other direction to provide access for SRI
people to processors installed on the Stanford Campus.
* * * * *
-------
∂30-Sep-83 1901 @SU-SCORE.ARPA:FEIGENBAUM@SUMEX-AIM.ARPA Re: Appointment
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Sep 83 19:01:19 PDT
Received: from SUMEX-AIM.ARPA by SU-SCORE.ARPA with TCP; Fri 30 Sep 83 18:17:06-PDT
Date: Fri 30 Sep 83 17:52:18-PDT
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM.ARPA>
Subject: Re: Appointment
To: GOLUB@SU-SCORE.ARPA, jmc@SU-SCORE.ARPA
In-Reply-To: Your message of Thu 29 Sep 83 23:01:56-PDT
Re the Nilsson appointment and the Dean's Office:
OK, John, why don't we do it? We would have to get togehter first to
plan our case. Do you want to?
Ed
-------
∂01-Oct-83 0000 JMC*
Remember! No term project and take home final.
∂01-Oct-83 0624 Colmerauer.GIA@MIT-MULTICS.ARPA Paris
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 1 Oct 83 06:24:43 PDT
Date: 1 October 1983 09:12 edt
From: Colmerauer.GIA at MIT-MULTICS
Subject: Paris
To: JMC at SU-AI
How ws was the meeting with Jacques Lang?
Lang said he would support the project, but we were not in a position
to ask him for anything concrete for lack of a local manager.
How was the meeting with Jacques Lang?
∂01-Oct-83 1745 ULLMAN@SU-SCORE.ARPA CS440
Received: from SU-SCORE by SU-AI with TCP/SMTP; 1 Oct 83 17:44:58 PDT
Date: Sat 1 Oct 83 17:44:17-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: CS440
To: super@SU-SCORE.ARPA
My schedule shows Ed Ashcroft as the speaker.
Unfortunately, I have no recollection of who made the contact,
and my attempts to reach Ed A. have been unsuccessful.
Anybody have some help on this?
-------
∂01-Oct-83 1954 Pool.Datanet@MIT-MULTICS.ARPA CATV paper
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 1 Oct 83 19:54:27 PDT
Date: 1 October 1983 22:55 edt
From: Pool.Datanet at MIT-MULTICS
Subject: CATV paper
To: JMC at SU-AI
I'm as puzzled as you are? US mail, net mail, how. I'm not sure what
was sent out and why?
It was net mail and sent to JMC@MC and forwarded to me. Here's a sample
∂20-Sep-83 1212 @MIT-MC:Pool@MIT-MULTICS Jackson paper
Received: from MIT-MC by SU-AI with TCP/SMTP; 20 Sep 83 12:12:13 PDT
Date: 20 September 1983 1441-edt
From: Ithiel de Sola Pool <Pool @ MIT-MULTICS>
Subject: Jackson paper
To: JMC @ MIT-MC
.nf
Jackson's is a superb analysis of the available technologies
of communications, which at the end falls into an ideological
trap. What this paper does well, is something that needs to
be done more often. Too much of our public policy is made by
lawyers and economists who do not understand the technological
alternatives with which they are playing. As Jackson notes,
the history of broadcasting regulation is a star example. The
∂02-Oct-83 0744 Pool.Datanet@MIT-MULTICS.ARPA Misaddressed mail
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 2 Oct 83 07:44:42 PDT
Date: 2 October 1983 10:41 edt
From: Pool.Datanet at MIT-MULTICS
Subject: Misaddressed mail
To: JMC at SU-AI
Human error can confound anything. The comments you got were to go to
CLJ -at MIT-MC. How JMC got written on them I cannot imagine. No harm
done except that Chuck Jackson gets his immediate electronic mail a week
late.
∂02-Oct-83 1618 Janet.Asbury@CMU-CS-A Electronic Library
Received: from CMU-CS-A by SU-AI with TCP/SMTP; 2 Oct 83 16:15:01 PDT
Received: from [128.2.254.192] by CMU-CS-PT with CMUFTP; 2 Oct 83 19:01:48 EDT
Date: 2 Oct 83 1911 EDT (Sunday)
From: Jan.Asbury <Janet.Asbury@CMU-CS-A>
To: JMC@SU-AI
Subject: Electronic Library
CC: Janet.Asbury@CMU-CS-A
Message-Id: <02Oct83.191101.JA61@CMU-CS-A>
Dear Dr. McCarthy,
I just checked my electronic mail and discovered that the message I sent
you on Friday never reached you because I mistakenly addressed it to JCM,
not JMC. I am so sorry for the error. Professor Reddy had said he was
hoping to receive a revised version of the proposal from you by tomorrow
so he can send it on. As you will see, there are only minor changes between
this version and the first version we sent you. The word 'Physical' in the
first line has been changed to 'conventional', some section headings have
been changed to subsections, and Section 3.3 reads 'Technical Issues' instead
of 'Technical Problems'. I am sending an MSS file and a DOC file. Please
make any revisions on the MSS file.
Jan Asbury
Secretary to Professor Reddy
@device(dover)
@make(article)
@heading(A PILOT PROJECT IN ELECTRONIC LIBRARIES)
@center(J. McCarthy, M. Griffith, R. Reddy))
@section(SUMMARY)
@section(PROBLEM)
An Electronic Library, as the name suggests, is the electronic
equivalent of a conventional library. It is expected that when a
system capable of fulfilling all the functions of a real library
is developed, it can provide many new functions that are currently
unavailable: instantaneous access, multiple access, reduced
cost, and language translation aids.
Constructing an operational Electronic Library poses a number of
technical, social and legal problems at present. The purpose of the
Pilot Project is to bypass many of these problems for now, and
concentrate on a few core problems which do not require any new
technical breakthroughs. The key problems to be studied will
involve the questions of how to acquire, represent, transmit, and
present Nineteenth Century French Literature. By selecting NCFL,
we finesse the issue of copyright, the problems of formulas and
drawings, and ensure widespread availability of French Literature.
@section(BACKGROUND AND NEED)
For some time it has been cost-effective to put the entire
Library of Congress into a computer file and make all its resources
available to anyone in the country with a computer terminal. There
is no need to argue that all printed paper will be abolished, but
I would certainly get rid of ninety percent of my books and
magazines if I could access it from my terminal at home.
It is now possible to get a gigabyte disk module for
under $20,000. If we count a book as 500,000 bytes, then this
module can store 2000 books. The space occupied by the module
would store about 300 books on shelves. The cost comes to $10 per
book. Recent word information compression would give another factor
of four in storage density, reducing the cost to approximately
$2.50 per book and reducing the storage volume to @i(one twenty-fourth)
of that required to store books on shelves.
Books cost much more these days and so does the space required to store
them, although the cost of cataloging the books is apparently larger
than either.
Recently IBM (3380) announced disk files storing 2.52 gigabytes per
unit which would store 20,000 compressed books in the space taken
by 300 on shelves. The Library of Congress would then require between
1000 and 2000 such disk units.
Digital videodisks storing much more are predicted for
the reasonably near future, but the project is practical with technology
now in hand. It is time to begin.
Consider the following system. In addition to existing paper
libraries, there would exist one or more computerized libraries containing
everything that has ever been published, i.e. a computerized version
of the Library of Congress. This library would be accessible
over the telephone network from any computer terminal in the country.
A reader could browse through the library catalog and various
bibliographies just as though he were physically present. He could read
any book by calling it page by page onto his terminal's screen or he
could have it transmitted to a local printer. At present, there are
a number of laser printers that print arbitrary fonts for less than
$10,000, but we can envisage cheaper printers in the future.
Most office workers would have terminals on their desks, and many
people would have them at home. At present a good enough terminal
costs about $800, and high quality terminals should cost about
$2,000 if manufactured in moderate quantity. Most offices can afford
a high quality printer.
Of course, yet better terminals may eventually be available. We can
imagine a pocket terminal consisting of a rolled up plastic screen
with a 1024 by 1024 array of liquid crystal dots accompanied by
another rolled up pressure sensitive keyboard and a pocket computer
with enough memory to store a book. Suppose that it has a modular
jack that can plug into any telephone so that the user can call the
library, scan it for a while and then reload his book memory. This
would be nicer than the technology now available, but the available
technology is good enough to justify a start.
>From the user's point of view, the advantages of the computerized
library are the following:
@begin(enumerate)
All books, magazines and newspapers are available.
Anything can be obtained in a few seconds.
Nothing is ever out.
The library is open 24 hours a day 365.2425 days a year.
@end(enumerate)
Many paper libraries would be found unnecessary. In particular,
university libraries could carry out their functions with much less
money and manpower, since their users would switch to the electronic
library for much of their work.
The establishment of such a system involves many problems and will take
some years, but we will mention some facilities that can and should be
started right away. Moreover, people who don't have terminals at home now
or on their desks or don't use them at all may be difficult to convince
of the advantages of such a library.
@subsection(Problems)
@begin(itemize)
It is expensive to convert the books to computer readable form. Equipment
for reading special type fonts is available and reliable. The recent
Kurzweil equipment reads arbitrary fonts with training but is reported
to rely to a substantial degree on a blind person's ability to know when
something was garbled and try again and on his ability to understand imperfectly
read material. The lowest error rates are apparently those obtained
by the Information International Grafix I system. This machine is very
expensive, mainly because it uses obsolete computer hardware, but
the company would update it if the market existed. Even if much of
the material had to be retyped by hand, the project would be worth what
it would cost.
Of course, much new material is generated in computer-readable form,
but many forms are used, and as yet no-one has developed a system
for putting all this material into a common form.
The copyright law requires permission to put copyrighted material into
computer form. In my opinion, copyrights should be respected and
suitable financial arrangements based on readership should be negotiated.
Once a computerized library exists, it will be so much more accessible than
other libraries that authors and publishers will find it to their
advantage to negotiate suitable deals.
The best arrangement might be that the copyright owner could set whatever
price he pleased for reading his material. The reader could decide whether
or not to pay it.
There is a problem of unauthorized copying. The problem exists whether a
national library exists or not, and the temptations will increase as
copying machines get more convenient and cheaper and when a general
purpose machine for reading documents from paper to computer files
becomes available.
At present an author gets ten to twenty percent of the retail price
of his books, except that he gets nothing for unsold books and less
for mass market paperbacks. An electronic publishing system could
afford to give the author eighty percent of the price paid by the
readers, because there would be no physical production or distribution
costs. This would permit increased income for authors and reduced
prices for the readers. Presumably there is some price elasticity for
reading that would produce more reading with reduced prices and greater
convenience. This would greatly reduce the temptation to copy illegally,
since the reader would find it less burdensome to pay the writer his due.
It is likely that the amount of illegal copying would be low enough
so that the system would survive. If not, we will eventually have
to go to a system where reading is essentially free and writers are
paid according to a formula by the Government. This would have many
disadvantages, since no formula could take fully into account the
fact that different writers have different abilities and put different
amounts of work into books of different kinds. Of course,
the present system doesn't take this into account very well either,
but there are some works now that charge very high prices, i.e.
newsletters. These could still operate outside the standard system.
@end(itemize)
@subsection(Getting Started)
Already there exist numerous databases available by telephone from
anywhere in the country. Some of them contain bibliographic
information, i.e. abstracts and references, but others contain the
texts of the material. Some of them are subsidized by government
grants, e.g. many of the medical databases, and others, e.g. the
legal databases and the "New York Times" Databank, are profit making
businesses. The charges for using them range from $25 to $200 per
hour except for subsidized customers.
One important step could be taken by the Federal Government. It
is required by the Freedom of Information Act and other laws to
make very large amounts of information available to the public.
This information would be much more conveniently available
if it were in a database accessible from anywhere in the country.
This especially includes the Federal Register where all new laws,
regulations, announcements of hearings and requests for comments
are published.
@subsection(Technical Issues)
While it is easy to compute the costs of the storage media, which
are already cheaper than paper, it is harder to calculate the
costs of the computers. This is because present systems have not
really been optimized for handling very large numbers of users.
It will also be necessary to optimize telephone access. For this
there are many possibilities.
A daytime cross-country call costs 54 cents for one minute.
In a minute 36,000 bytes can be transmitted at 4800 bits/second.
This means from $7.50 to $15.00 to transmit a book uncompressed
or from $1.87 to $3.75 with a compression of 4. We can imagine a
terminal that could store a minute's worth of text and could decompress
it for reading. These costs are unpleasantly high, but they can be
reduced in various ways. First, technology permits substantially
lower long distance transmission costs. Indeed the one minute
transcontinental charge late at night is 16 cents making our
compressed book cost from $.56 to $1.12 if transmitted all at once.
This is probably less than the cost of a trip to a library if one's time
is worth much. The independent long distance telephone companies are
often 40 percent below AT&T, which brings our optimistic number down to
33 cents, which is reminiscent of the days when pocket books were a
quarter.
We can suppose that the terminal would remember the telephone number and
catalog number and automatically phone for another minute's transmission when the
reader is close to the end of what it has in storage. These costs
are even less attractive when browsing is wanted. A solution for that
is to use the European telephone charging system which allows calls
as short as 4 seconds. Current networks keep the cost for maintaining
a connection down by time-sharing lines, but this doesn't reduce the cost
of straight data transmission.
An obvious possible saving is to have local libraries with frequently
consulted books and magazines. With optical fibers and other new means of
transmission, the transmission costs can be brought down to the
point that local libraries will be unnecessary.
@subsection(French Electronic Library)
The time is ripe for it to be socially worthwhile and economically
feasible to put the world literature in the French language into
computer form and make it available world wide.
Image the following system. The French language literature is put
into computer form, either by optical character recognition machines
or by keyboarding in low wage countries. A central computer library
in France keeps this literature on the equivalent of about 1000 IBM
3380 disk files. Three large bandwidth satellites are put up
to provide worldwide transmission facilities. Reading rooms with
suitable terminals are located in every place where there is
sufficient interest. A reader can call up any book or other document
from any terminal. When he does so, the first two pages are
transmitted via the satellite to the reading room computer and the first
page is displayed on his terminal. Perhaps the library catalog and
other currently popular documents are kept in local file.
@section(CURRENT STATUS)
<Mike Griffith to provide>
@section(PLAN FOR RESEARCH)
We propose to undertake the following pilot project.
@begin(enumerate)
A few RA81 disks are acquired from Digital Equipment Corporation and
attached to a VAX computer. This is currently the most cost-effective
disk file available.
A request for proposals for a few hundred thousand dollars worth of
book input is sent both to keyboarding companies and those that do
optical character recognition. In addition existing computerized
text is solicited from those who have it for experimental use.
The initial reading list is taken from the public domain literature.
About 20 telephone lines are attached to the VAX, so that the library
is available from existing terminals and micro-computers in the Paris
area.
@begin(multiple)
The necessary programs are written and installed.
At this point a technical demonstration is feasible. An attempt
is made to determine what is most attractive to the users of the
library within the budget available.
@end(multiple)
An experimental terminal cluster is installed in a reading room in the
Paris area. It should be a place that is open for a large number of hours.
@end(enumerate)
If the results are encouraging, the second phase includes:
@begin(enumerate)
Giving the computerized library its own computer.
More books.
Obtaining the co-operation of publishers of current books, magazines
and newspapers for an expanded program. An experimental financial
arrangement should be adopted.
Design of a reading terminal that can be used in connection with the
French telephone system's electronic yellow pages.
An experimental reading room in an underdeveloped country using
existing satellite transmission channnels.
Developing an optical character recognition system optimized toward
reading books.
@end(enumerate)
The pilot project is intended to lead to a demonstration by
the end of 1984 with several thousand books on line.
@begin(enumerate)
EQUIPMENT PLAN - We expect to start with a VAX with a gigabyte of
memory as the EL Machine located at CMIRH in Paris. This machine
will have at least 32 lines permitting anyone in the Paris region
with a terminal, personal computer or a Minitel to be able to use it.
By 1985 we hope to extend the service throughout France using the
CMIRH network.
ACQUISITON - There are already several thousand books available at
"----- Le Langru Francais" at Nancy. We hope to acquire these.
In addition we hope to acquire a similar collection from Britain and
the USA. Also we will have about 1000 books manually entered in Third
World countries. This is expected to be quite inexpensive, about 2000
FFr per book.
@end(enumerate)
All these different books will probably come in different formats.
We will develop format conversion programs to put them in CMIRH
standard format.
@i(Representation.) Information on the disk will be stored in a compact form with
frequently occurring words coded and formatting information bracketed
approximately.
Terminals and personal computers with local processing capability will
receive a decoding program followed by coded text which is expected to
also reduce the transmission time and cost. Dumb terminals will receive
fully decoded text. Decoding time should be less than 1 second per
10 words in sequence.
@i(Transmission.) Initially only serial line transmission will be considered. VAX will
support up to 19.2 kiloband transmission. Terminals and personal
computers with local processing will be able to correct transmission
error using Kermit-like programs. They can also accept data at much
higher rates for later presentation at user specified rates.
@i(Presentation.) It will be possible to access information from the on-line library
from almost all commmonly available terminals and personal computers.
However, from an ergonomic (human factors) point of view, high
resolution bit-mapped displays (equivalent in resolution to the FAX
standard) with a powerful personal computer with at least 2 megabytes
of memory would be highly desirable. Low cost versions (<$1000)
of such terminals should be available by the end of the decade.
It is expected to take at least that long to acquire and represent
a substantial collection of books, reports and newspapers in electronic
form.
@i(Selection.) <What books will be on-line in the first year. Mike Griffiths to
approach Academe Francais.>
A PILOT PROJECT IN ELECTRONIC LIBRARIES
J. McCarthy, M. Griffith, R. Reddy
)
1. SUMMARY
2. PROBLEM
An Electronic Library, as the name suggests, is the electronic equivalent of
a conventional library. It is expected that when a system capable of
fulfilling all the functions of a real library is developed, it can provide
many new functions that are currently unavailable: instantaneous access,
multiple access, reduced cost, and language translation aids.
Constructing an operational Electronic Library poses a number of technical,
social and legal problems at present. The purpose of the Pilot Project is to
bypass many of these problems for now, and concentrate on a few core problems
which do not require any new technical breakthroughs. The key problems to be
studied will involve the questions of how to acquire, represent, transmit, and
present Nineteenth Century French Literature. By selecting NCFL, we finesse
the issue of copyright, the problems of formulas and drawings, and ensure
widespread availability of French Literature.
3. BACKGROUND AND NEED
For some time it has been cost-effective to put the entire Library of
Congress into a computer file and make all its resources available to anyone in
the country with a computer terminal. There is no need to argue that all
printed paper will be abolished, but I would certainly get rid of ninety
percent of my books and magazines if I could access it from my terminal at
home.
It is now possible to get a gigabyte disk module for under $20,000. If we
count a book as 500,000 bytes, then this module can store 2000 books. The
space occupied by the module would store about 300 books on shelves. The cost
comes to $10 per book. Recent word information compression would give another
factor of four in storage density, reducing the cost to approximately $2.50 per
book and reducing the storage volume to one twenty-fourth of that required to
store books on shelves.
Books cost much more these days and so does the space required to store them,
although the cost of cataloging the books is apparently larger than either.
Recently IBM (3380) announced disk files storing 2.52 gigabytes per unit
which would store 20,000 compressed books in the space taken by 300 on shelves.
The Library of Congress would then require between 1000 and 2000 such disk
units.
Digital videodisks storing much more are predicted for the reasonably near
future, but the project is practical with technology now in hand. It is time
to begin.
Consider the following system. In addition to existing paper libraries,
there would exist one or more computerized libraries containing everything that
has ever been published, i.e. a computerized version of the Library of
Congress. This library would be accessible over the telephone network from any
computer terminal in the country. A reader could browse through the library
catalog and various bibliographies just as though he were physically present.
He could read any book by calling it page by page onto his terminal's screen or
he could have it transmitted to a local printer. At present, there are a
number of laser printers that print arbitrary fonts for less than $10,000, but
we can envisage cheaper printers in the future.
Most office workers would have terminals on their desks, and many people
would have them at home. At present a good enough terminal costs about $800,
and high quality terminals should cost about $2,000 if manufactured in moderate
quantity. Most offices can afford a high quality printer.
Of course, yet better terminals may eventually be available. We can imagine
a pocket terminal consisting of a rolled up plastic screen with a 1024 by 1024
array of liquid crystal dots accompanied by another rolled up pressure
sensitive keyboard and a pocket computer with enough memory to store a book.
Suppose that it has a modular jack that can plug into any telephone so that the
user can call the library, scan it for a while and then reload his book memory.
This would be nicer than the technology now available, but the available
technology is good enough to justify a start.
From the user's point of view, the advantages of the computerized library are
the following:
1. All books, magazines and newspapers are available.
2. Anything can be obtained in a few seconds.
3. Nothing is ever out.
4. The library is open 24 hours a day 365.2425 days a year.
Many paper libraries would be found unnecessary. In particular, university
libraries could carry out their functions with much less money and manpower,
since their users would switch to the electronic library for much of their
work.
The establishment of such a system involves many problems and will take some
years, but we will mention some facilities that can and should be started right
away. Moreover, people who don't have terminals at home now or on their desks
or don't use them at all may be difficult to convince of the advantages of such
a library.
3.1. Problems
- It is expensive to convert the books to computer readable form.
Equipment for reading special type fonts is available and reliable.
The recent Kurzweil equipment reads arbitrary fonts with training but
is reported to rely to a substantial degree on a blind person's
ability to know when something was garbled and try again and on his
ability to understand imperfectly read material. The lowest error
rates are apparently those obtained by the Information International
Grafix I system. This machine is very expensive, mainly because it
uses obsolete computer hardware, but the company would update it if
the market existed. Even if much of the material had to be retyped
by hand, the project would be worth what it would cost.
- Of course, much new material is generated in computer-readable form,
but many forms are used, and as yet no-one has developed a system for
putting all this material into a common form.
- The copyright law requires permission to put copyrighted material
into computer form. In my opinion, copyrights should be respected
and suitable financial arrangements based on readership should be
negotiated. Once a computerized library exists, it will be so much
more accessible than other libraries that authors and publishers will
find it to their advantage to negotiate suitable deals.
- The best arrangement might be that the copyright owner could set
whatever price he pleased for reading his material. The reader could
decide whether or not to pay it.
- There is a problem of unauthorized copying. The problem exists
whether a national library exists or not, and the temptations will
increase as copying machines get more convenient and cheaper and when
a general purpose machine for reading documents from paper to
computer files becomes available.
- At present an author gets ten to twenty percent of the retail price
of his books, except that he gets nothing for unsold books and less
for mass market paperbacks. An electronic publishing system could
afford to give the author eighty percent of the price paid by the
readers, because there would be no physical production or
distribution costs. This would permit increased income for authors
and reduced prices for the readers. Presumably there is some price
elasticity for reading that would produce more reading with reduced
prices and greater convenience. This would greatly reduce the
temptation to copy illegally, since the reader would find it less
burdensome to pay the writer his due.
- It is likely that the amount of illegal copying would be low enough
so that the system would survive. If not, we will eventually have to
go to a system where reading is essentially free and writers are paid
according to a formula by the Government. This would have many
disadvantages, since no formula could take fully into account the
fact that different writers have different abilities and put
different amounts of work into books of different kinds. Of course,
the present system doesn't take this into account very well either,
but there are some works now that charge very high prices, i.e.
newsletters. These could still operate outside the standard system.
3.2. Getting Started
Already there exist numerous databases available by telephone from anywhere
in the country. Some of them contain bibliographic information, i.e. abstracts
and references, but others contain the texts of the material. Some of them are
subsidized by government grants, e.g. many of the medical databases, and
others, e.g. the legal databases and the "New York Times" Databank, are profit
making businesses. The charges for using them range from $25 to $200 per hour
except for subsidized customers.
One important step could be taken by the Federal Government. It is required
by the Freedom of Information Act and other laws to make very large amounts of
information available to the public. This information would be much more
conveniently available if it were in a database accessible from anywhere in the
country. This especially includes the Federal Register where all new laws,
regulations, announcements of hearings and requests for comments are published.
3.3. Technical Issues
While it is easy to compute the costs of the storage media, which are already
cheaper than paper, it is harder to calculate the costs of the computers. This
is because present systems have not really been optimized for handling very
large numbers of users. It will also be necessary to optimize telephone
access. For this there are many possibilities.
A daytime cross-country call costs 54 cents for one minute. In a minute
36,000 bytes can be transmitted at 4800 bits/second. This means from $7.50 to
$15.00 to transmit a book uncompressed or from $1.87 to $3.75 with a
compression of 4. We can imagine a terminal that could store a minute's worth
of text and could decompress it for reading. These costs are unpleasantly
high, but they can be reduced in various ways. First, technology permits
substantially lower long distance transmission costs. Indeed the one minute
transcontinental charge late at night is 16 cents making our compressed book
cost from $.56 to $1.12 if transmitted all at once. This is probably less than
the cost of a trip to a library if one's time is worth much. The independent
long distance telephone companies are often 40 percent below AT&T, which brings
our optimistic number down to 33 cents, which is reminiscent of the days when
pocket books were a quarter.
We can suppose that the terminal would remember the telephone number and
catalog number and automatically phone for another minute's transmission when
the reader is close to the end of what it has in storage. These costs are even
less attractive when browsing is wanted. A solution for that is to use the
European telephone charging system which allows calls as short as 4 seconds.
Current networks keep the cost for maintaining a connection down by
time-sharing lines, but this doesn't reduce the cost of straight data
transmission.
An obvious possible saving is to have local libraries with frequently
consulted books and magazines. With optical fibers and other new means of
transmission, the transmission costs can be brought down to the point that
local libraries will be unnecessary.
3.4. French Electronic Library
The time is ripe for it to be socially worthwhile and economically feasible
to put the world literature in the French language into computer form and make
it available world wide.
Image the following system. The French language literature is put into
computer form, either by optical character recognition machines or by
keyboarding in low wage countries. A central computer library in France keeps
this literature on the equivalent of about 1000 IBM 3380 disk files. Three
large bandwidth satellites are put up to provide worldwide transmission
facilities. Reading rooms with suitable terminals are located in every place
where there is sufficient interest. A reader can call up any book or other
document from any terminal. When he does so, the first two pages are
transmitted via the satellite to the reading room computer and the first page
is displayed on his terminal. Perhaps the library catalog and other currently
popular documents are kept in local file.
4. CURRENT STATUS
<Mike Griffith to provide>
5. PLAN FOR RESEARCH
We propose to undertake the following pilot project.
1. A few RA81 disks are acquired from Digital Equipment Corporation and
attached to a VAX computer. This is currently the most
cost-effective disk file available.
2. A request for proposals for a few hundred thousand dollars worth of
book input is sent both to keyboarding companies and those that do
optical character recognition. In addition existing computerized
text is solicited from those who have it for experimental use. The
initial reading list is taken from the public domain literature.
3. About 20 telephone lines are attached to the VAX, so that the
library is available from existing terminals and micro-computers in
the Paris area.
4. The necessary programs are written and installed.
At this point a technical demonstration is feasible. An attempt is
made to determine what is most attractive to the users of the
library within the budget available.
5. An experimental terminal cluster is installed in a reading room in
the Paris area. It should be a place that is open for a large
number of hours.
If the results are encouraging, the second phase includes:
1. Giving the computerized library its own computer.
2. More books.
3. Obtaining the co-operation of publishers of current books, magazines
and newspapers for an expanded program. An experimental financial
arrangement should be adopted.
4. Design of a reading terminal that can be used in connection with the
French telephone system's electronic yellow pages.
5. An experimental reading room in an underdeveloped country using
existing satellite transmission channnels.
6. Developing an optical character recognition system optimized toward
reading books.
The pilot project is intended to lead to a demonstration by the end of 1984
with several thousand books on line.
1. EQUIPMENT PLAN - We expect to start with a VAX with a gigabyte of
memory as the EL Machine located at CMIRH in Paris. This machine
will have at least 32 lines permitting anyone in the Paris region
with a terminal, personal computer or a Minitel to be able to use
it. By 1985 we hope to extend the service throughout France using
the CMIRH network.
2. ACQUISITON - There are already several thousand books available at
"----- Le Langru Francais" at Nancy. We hope to acquire these. In
addition we hope to acquire a similar collection from Britain and
the USA. Also we will have about 1000 books manually entered in
Third World countries. This is expected to be quite inexpensive,
about 2000 FFr per book.
All these different books will probably come in different formats. We will
develop format conversion programs to put them in CMIRH standard format.
Representation. Information on the disk will be stored in a compact form with
frequently occurring words coded and formatting information bracketed
approximately.
Terminals and personal computers with local processing capability will
receive a decoding program followed by coded text which is expected to also
reduce the transmission time and cost. Dumb terminals will receive fully
decoded text. Decoding time should be less than 1 second per 10 words in
sequence.
Transmission. Initially only serial line transmission will be considered.
VAX will support up to 19.2 kiloband transmission. Terminals and personal
computers with local processing will be able to correct transmission error
using Kermit-like programs. They can also accept data at much higher rates for
later presentation at user specified rates.
Presentation. It will be possible to access information from the on-line
library from almost all commmonly available terminals and personal computers.
However, from an ergonomic (human factors) point of view, high resolution
bit-mapped displays (equivalent in resolution to the FAX standard) with a
powerful personal computer with at least 2 megabytes of memory would be highly
desirable. Low cost versions (<$1000) of such terminals should be available by
the end of the decade. It is expected to take at least that long to acquire
and represent a substantial collection of books, reports and newspapers in
electronic form.
Selection. <What books will be on-line in the first year. Mike Griffiths to
approach Academe Francais.>
Table of Contents
1. SUMMARY 0
2. PROBLEM 0
3. BACKGROUND AND NEED 0
3.1. Problems 0
3.2. Getting Started 0
3.3. Technical Issues 0
3.4. French Electronic Library 1
4. CURRENT STATUS 1
5. PLAN FOR RESEARCH 1
∂02-Oct-83 1702 JJW EKL manual
To: JMC, YOM, JK, GLB
The EKL manual is now about ready for the class. If you would like to proof-
read it before it goes out, Dover the file EKLMAN.PRE[EKL,JJW]. (The table of
contents will be at the end of the file; the index hasn't been produced yet.)
Joe
∂03-Oct-83 1002 LUNDSTROM@SU-SCORE.ARPA [weeks at ames-vmsb: ACM SIGBIG OCT 5th MEETING]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Oct 83 10:02:31 PDT
Date: Mon 3 Oct 83 09:17:02-PDT
From: Stephen Lundstrom <LUNDSTROM@SU-SCORE.ARPA>
Subject: [weeks at ames-vmsb: ACM SIGBIG OCT 5th MEETING]
To: super@SU-SCORE.ARPA
FYI - Steve Lundstrom
---------------
Return-Path: <weeks@ames-vmsb>
Received: from ames-vmsb by SU-SCORE.ARPA with TCP; Thu 29 Sep 83 15:38:59-PDT
Date: Thu, 29 Sep 83 15:30:34 PDT
From: weeks at ames-vmsb
Subject: ACM SIGBIG OCT 5th MEETING
To: LUNDSTROM at SU-SCORE
ASSOCIATION FOR COMPUTING MACHINERY
SAN FRANCISCO GOLDEN GATE CHAPTER
SPECIAL INTEREST COMMITEE, "SIGBIG"
For Large High Speed Computers
Wednesday, October 5, 1983
Frank Kuo
SRI
"SPREAD, A Supercomputer Network"
7:00 Business & Progress Reports
7:30 Speaker
At SRI, 333 Ravenswood, Menlo Park
(More information on room location will follow. )
Please let Mary Fowler know if you are planning to attend.
Due to the short notice of this meeting the speaker's
presentation will be given at a later meeting if many
people are unable to attend.
For more information,
and to reply, contact Mary Fowler, (415)965-6515
Ames Decnet CER::fowler
Arpanet Fowler @ ames-vmsa
Ride Sharing, contact Frank Olken, LBL
ARPANET olken at lbl-csam
(415)486-5891
-------
∂03-Oct-83 1105 KJB@SRI-AI.ARPA Todays meeting in C1
Received: from SRI-AI by SU-AI with TCP/SMTP; 3 Oct 83 11:05:46 PDT
Date: Mon 3 Oct 83 10:59:55-PDT
From: KJB@SRI-AI.ARPA
Subject: Todays meeting in C1
To: csli-friends@SRI-AI.ARPA
This is a reminder of today's working meeting at PARC on the
semantics of computer languages. It will be in room 1613
(unless my trick for remembering it failed me: 7 years before
the Mayflower) at 4 p.m. Stanford and SRI people meet
at 3:55 at the visitor's entrance.
I will have copies of various things we might want to work through we
me, so that we can decide which of several paths to pursue.
Note: For now, projects C1 and D1 are combined, so this is
also a project D1 meeting. {Carl: Sorry I forgot to tell you
this sooner.)
-------
∂03-Oct-83 1212 @MIT-MC:DAVIS@MIT-OZ
Received: from MIT-MC by SU-AI with TCP/SMTP; 3 Oct 83 12:12:37 PDT
Date: Mon 3 Oct 83 14:59:17-EDT
From: Randy Davis <DAVIS@MIT-OZ>
To: JMC%SU-AI@MIT-MC.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sat 1 Oct 83 22:46:20-EDT
I agree with your comments re: Dowdy, having given it more thought.
Also, I'm told that OTA is virtually powerless politically. It would
be something of a wheel spinning exercise. The OTA meeting
has been scheduled for 31 Oct, and I've declined, based on the
advice about their lack of clout.
R.
-------
∂03-Oct-83 1234 AAAI@SU-SCORE.ARPA Quarterly Presidential message
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Oct 83 12:34:18 PDT
Date: Mon 3 Oct 83 12:35:23-PDT
From: AAAI-OFFICE <AAAI@SU-SCORE.ARPA>
Subject: Quarterly Presidential message
To: JMC@SU-AI.ARPA
John,
We're trying to send the AI Magazine to the printer in two weeks.
Can you pls send me your Presidential message this week? Thanks!
-- Claudia
-------
The presidential message will be ready this week.
On another subject, the message asking for approval of replacing
Don Walker by Rich Fikes should have referred to my requesting approval
rather than you. No correction is necessary unless you feel like it.
∂03-Oct-83 1351 AAAI-OFFICE@SUMEX-AIM Change in the AAAI Secretariat
Received: from SUMEX-AIM by SU-AI with PUP; 03-Oct-83 13:51 PDT
Date: Mon 3 Oct 83 13:50:49-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Change in the AAAI Secretariat
To: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BUCHANAN@SUMEX-AIM,
DAVIS@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA, LERMAN@SRI-KL.ARPA,
GENESERETH@SUMEX-AIM, GROSZ@SRI-AI.ARPA, HART@SRI-KL.ARPA, JMC@SAIL,
MINSKY@MIT-MC.ARPA, NILSSSON@SRI-AI.ARPA, REDDY@CMU-CS-A.ARPA,
STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA, GJS@MIT-MC.ARPA,
TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA, BONNIE.UPENN@UDEL-RELAY.ARPA,
BLEDSOE@UTEXAS-20.ARPA, RICH@UTEXAS-20.ARPA, AAAI-OFFICE@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
As some of you already know, Don Walker, the AAAI's
Secretary/Treasurer, has asked to be released from this position. It
has been recommended that Richard Fikes replace Don as the
Secretary/Treasurer.
If you disagree with the appointment of Rich as Secretary/Treasurer,
please inform me within the next (7) seven days. Otherwise, we will
assume you concur with this recommendation.
Lastly, the Executive Council sincerely thanks Don for the time and
effort he has contributed to the organization. Without him, the AAAI
might have faltered in its development as the premier AI scientific
society in the US.
Cordially,
Claudia
-------
∂03-Oct-83 1501 ullman%SU-HNV.ARPA@SU-SCORE.ARPA CS440
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Oct 83 15:00:50 PDT
Received: from Diablo by Score with Pup; Mon 3 Oct 83 14:58:31-PDT
Date: Mon, 3 Oct 83 14:58 PDT
From: Jeff Ullman <ullman@Diablo>
Subject: CS440
To: su-bboards@Diablo, super@score
The new room for the CS440 supercomputer seminar is 380-X.
This Thursday the speaker will be Ed Ashcroft of SRI speaking
on "The SRI Dataflow project".
∂03-Oct-83 1502 PETERS@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 3 Oct 83 15:02:19 PDT
Date: Mon 3 Oct 83 15:02:56-PDT
From: Stanley Peters <PETERS@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 3 Oct 83 10:55:00-PDT
Oops. Sorry. You aren't on the Building Committee, and so
shouldn't have gotten even one copy of it. I'll look into the
address file and see what the trouble is. Thanks for alerting
me.
-------
∂03-Oct-83 1633 DFH Barwise
phone no. at CSLI is 7-1202
∂03-Oct-83 1849 LEP traffic lights
Thursday at 4 is fine.
∂03-Oct-83 1853 ALMOG@SRI-AI.ARPA thanks
Received: from SRI-AI by SU-AI with TCP/SMTP; 3 Oct 83 18:53:21 PDT
Date: 3 Oct 1983 1849-PDT
From: Almog at SRI-AI
Subject: thanks
To: jmc at SU-AI
hello,
I really ought to apologize. Not only for stopping you in the middle,
but because it was really exciting for me. In fact, I really had
some ideas on how to go on with "normally" and I hope you'll
have time to talk. But thanks for the interesting time, I really
had to go or face execution. Joseph almog
-------
∂03-Oct-83 2110 PETERS@SRI-AI.ARPA Project B1, Extensions of semantic theories
Received: from SRI-AI by SU-AI with TCP/SMTP; 3 Oct 83 21:09:54 PDT
Date: Mon 3 Oct 83 21:09:07-PDT
From: Stanley Peters <PETERS@SRI-AI.ARPA>
Subject: Project B1, Extensions of semantic theories
To: csli-folks@SRI-AI.ARPA
If you are a Principal or Associate of CSLI and intend to work on
this project during the current year, please send me a message
(with a cc: to BMacken) saying what percentage of your time at
CSLI you expect to devote to B1.
Some preliminary nosing around indicates that among the many
topics that this project can and should concern itself with,
popular candidates for starters are the semantics of plurals,
collectives, and mass terms. Anyone who wishes to work on them
should please also indicate this and let me know whether you can
attend an initial meeting next Monday afternoon at 3:00.
Thanks.
-------
∂03-Oct-83 2142 RV qual.
∂03-Oct-83 1153 DFH AI Qual
Buchanan is not available any time this week, and will then
be gone to Europe until 10/28. I can get Lenat and McCarthy
at 2:45 pm Thurs. Buchanan's suggestion via his secretary is
that you talk to Lenat about getting someone else to take his
place on the committee.
- - - - - - - - -
Prof. McCarthy:
I talked to Lenat this afternoon for a while about the qual. I mentioned
to him, as I did to you in May, that I think I do not perform well in oral
test situations; he said he knew of several others who had this problem.
He suggested that I might do better taking a written qual instead. What
do you think of this idea? If you are agreeable with it, you might be
able to get together with Lenat on Thurs. at 2:45 (as planned above) to
discuss the format and content of the exam.
If possible, I would like to take the written test Friday, if it is
ready by then; if you want me to take it orally, then I'd like to do it
on Thursday. In that case, we'd need to find another faculty member.
I will try to find you tomorrow to talk about this.
Rick
A written exam for one student is much too much work. I propose that
we delay until early November and then proceed as before.
∂04-Oct-83 0910 ULLMAN@SU-SCORE.ARPA panel show
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Oct 83 09:09:59 PDT
Date: Tue 4 Oct 83 09:11:06-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: panel show
To: genesereth@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA, sso.owicki@SU-SIERRA.ARPA,
pratt@SU-SCORE.ARPA, reid@SU-GLACIER.ARPA, trattnig@SU-SCORE.ARPA
cc: meindl@SU-SIERRA.ARPA
Thank you all for agreeing to serve on the "software in the 1990's" panel.
The plan is for it to run 10:30-noon on Weds., Nov. 9, place to be determined.
A "speech" of about 10 minutes per person is about right, leaving half an
hour for questions. I suggest that an appropriate topic is how are things
going to be different in 10 years, and what should we be doing now to
make those things happen, but feel free to flame about anything.
-------
∂04-Oct-83 1114 @SRI-AI.ARPA:halvorsen.pa@PARC-MAXC.ARPA Reminder: Research Seminar, Natural Language, Thur., Oct. 6
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Oct 83 11:13:54 PDT
Delivery-Notice: While sending this message to SU-AI.ARPA, the
SRI-AI.ARPA mailer was obliged to send this message in 50-byte
individually Pushed segments because normal TCP stream transmission
timed out. This probably indicates a problem with the receiving TCP
or SMTP server. See your site's software support if you have any questions.
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Tue 4 Oct 83 11:13:15-PDT
Date: 4 Oct 83 11:13 PDT
From: halvorsen.pa@PARC-MAXC.ARPA
Subject: Reminder: Research Seminar, Natural Language, Thur., Oct. 6
To: csli-friends@sri-ai.ARPA
The CSLI research seminar on natural language will
this week be given by Bob Moore, SRI & CSLI.
Title: "Problems in Semantic Analysis of Natural Language"
Place: Redwood Hall, classroom
Time: Thursday, October 6, 10:00 am
Redwood Hall is close to Ventura Hall on the Stanford Campus. It can
be reached from Campus Drive or Panama Street. From Campus Drive follow
the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
∂04-Oct-83 1323 LENAT@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Oct 83 13:23:23 PDT
Date: Tue 4 Oct 83 13:23:00-PDT
From: Doug Lenat <LENAT@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
cc: DBL@SU-AI.ARPA, buchanan@SUMEX-AIM.ARPA, RV@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 4 Oct 83 10:28:00-PDT
I am amenable to either plan (oral or written), though I firmly
believe that oral exams allow students to show what they know more fully.
Doug
-------
∂04-Oct-83 1336 ULLMAN@SU-SCORE.ARPA seminar
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Oct 83 13:32:53 PDT
Delivery-Notice: While sending this message to SU-AI.ARPA, the
SU-SCORE.ARPA mailer was obliged to send this message in 50-byte
individually Pushed segments because normal TCP stream transmission
timed out. This probably indicates a problem with the receiving TCP
or SMTP server. See your site's software support if you have any questions.
Date: Tue 4 Oct 83 13:31:06-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: seminar
To: super@SU-SCORE.ARPA,
CS440: ;
We meet 4:15PM Thursday. The room has been changed to 380-X
(In the Math. building).
The speaker is Ed Ashcroft of SRI International; his
title and abstract is:
The proposed SRI Demand-driven Tagged Dataflow Machine.
SRI proposes to design, and possibly build, a dataflow machine which
will be a natural extension of the working dataflow machine at
Manchester (England). The novel features of the machine will be that
it will be demand-driven, it will be programmed in Lucid (but other
languages can be used also), and it will use 128 processors, at least
initially. (It will be designed to be extendible to 256 processors.)
It will use conventional off-the-shelf components for the most part,
and is conservatively estimated to be a 16 megaflop machine.
This talk will discuss the machine design, pointing out how it builds
on the experience at Manchester, and will discuss the novel ways in
which tags are used to implement Lucid.
-------
∂04-Oct-83 1549 WALDINGER@SRI-AI.ARPA circumscription paper.
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Oct 83 15:49:43 PDT
Date: 4 Oct 1983 1548-PDT
From: Waldinger at SRI-AI
Subject: circumscription paper.
To: jmc at SAIL
john:
would you be willing to referee a paper "applications of protected
circumscription" for possible inclusion in the Conference on
Automated Deduction next year? papers are submitted anonymously;
this one is about 5 1/2 fully packed pages. they need a decision
mid-november, but it does not require a full-blown referee report.
richard
-------
I'll look at "applications of protected circumscription", and if it
looks too hard for me to follow, I'll send it back. Don't send me
the only copy.
∂04-Oct-83 1636 DFH Chris Goad called
Wants to talk to you. Daytime phone is 493-0145
∂04-Oct-83 1658 RIGGS@SRI-AI.ARPA Afternoon Colloquium Schedule
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Oct 83 16:58:15 PDT
Date: Tue 4 Oct 83 16:57:24-PDT
From: RIGGS@SRI-AI.ARPA
Subject: Afternoon Colloquium Schedule
To: csli-friends@SRI-AI.ARPA
Center for the Study of Language and Information
COLLOQUIUM SCHEDULE
October
* * * * * * * * * * * * * * * * *
Thursday, October 6, 1983
GEOFFREY NUNBERG
Department of Linguistics
Stanford University
"Prescriptive Grammar"
* * * * * * * * * * * * * * * * *
Thursday, October 13, 1983
LAURI KARTTUNEN STAN ROSENCHEIN
Department of Linguistics Artificial Intelligence Lab
University of Texas, Austin S.R.I.
Symposium on "Computational" and "Theoretical" Linguistics
* * * * * * * * * * * * * * * * *
Thursday, October 20, 1983
JERRY HOBBS
Artificial Intelligence Laboratory
S.R.I.
"Discourse Coherence and Discourse Structure"
* * * * * * * * * * * * * * * * *
Thursday, October 27, 1983
JAY M. TENENBAUM
Laboratory for Artificial Intelligence Research
Fairchild
"A.I. Research at Fairchild"
* * * * * * * * * * * * * * * * *
ALL COLLOQUIA MEET AT 4:15 P.M., IN ROOM G-19, REDWOOD HALL
Center for the Study of Language and Information
COLLOQUIUM SCHEDULE
November (tentative)
* * * * * * * * * * * * * * * * *
Thursday, November 3, 1983
GLYNN WINSKEL
Department of Mathematics
Carnegie-Mellon University
"Denotational Semantics: An Introduction"
* * * * * * * * * * * * * * * * *
Thursday, November 10, 1983
MICHAEL BEESON
Department of Mathematics and Computer Science
San Jose State University
"Computational Aspects of Intuitionistic Logic"
* * * * * * * * * * * * * * * * *
Thursday, November 17, 1983
CHARLES FILLMORE
Department of Linguistics
University of California, Berkeley
Title to be announced
* * * * * * * * * * * * * * * * *
ALL COLLOQUIA MEET AT 4:15 P.M., IN ROOM G-19, REDWOOD HALL
-------
∂04-Oct-83 1745 RV Procrastination
To: JMC@SU-AI, DBL@SU-AI, buchanan@SUMEX-AIM
∂04-Oct-83 1028 JMC
To: DBL@SU-AI, buchanan@SUMEX-AIM, RV@SU-AI
A written exam for one student is much too much work. I propose that
we delay until early November and then proceed as before.
- - - - - - - -
I'm agreeable. This time, though, I'd like to arrange the date a couple
of weeks (at last) in advance so things don't get messed up like they have
been the last couple of days. I'll get in touch with the right
secretaries.
Rick
The second week in November would suit me best, because I'll be back
only on the 1st from two weeks of travelling (my last for some time).
However, the first is possible if necessary. The right secretary to
arrange a time is Diana Hall DFH.
∂04-Oct-83 1939 ALMOG@SRI-AI.ARPA talk
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Oct 83 19:39:23 PDT
Date: 4 Oct 1983 1937-PDT
From: Almog at SRI-AI
Subject: talk
To: jmc at SU-AI
here is an idea of pat and mine. In view of the interest that your
remarks generated today and in view of the distinctive nature
of your work in AI, and the special way you pose the problems(different
from Grosz), and in view of the very different target issues you are
positing, and(i could go on forever)..., will it not be extremely
useful if you gave, just after Grosz, that is on 10.26.83 a talk
on your view of things. This will be most important,we think, to
the REAL interaction between philosophers and AI researchers.
I hope you'll agree, it will be our pleasure. Many thanks
in advance, Joseph
-------
I will be away for the second half of the month. Any time in November
would be fine, however.
∂04-Oct-83 2017 PEREIRA@SRI-AI.ARPA REMINDER: Research Seminar, Programming Languages, Thur., Oct. 6
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Oct 83 20:17:46 PDT
Date: Tue 4 Oct 83 20:14:38-PDT
From: PEREIRA@SRI-AI.ARPA
Subject: REMINDER: Research Seminar, Programming Languages, Thur., Oct. 6
To: csli-friends@SRI-AI.ARPA
The CSLI research seminar on approaches to computer languages will this
week be given by Fernando Pereira, SRI & CSLI.
Title: "Logic as a Programming Language"
Place: Redwood Hall, classroom
Time: Thursday, October 6, 2:00 PM
-------
∂04-Oct-83 2044 FEIGENBAUM@SUMEX-AIM nils nilsson
Received: from SUMEX-AIM by SU-AI with PUP; 04-Oct-83 20:44 PDT
Date: Tue 4 Oct 83 20:38:49-PDT
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM>
Subject: nils nilsson
To: eengelmore@SUMEX-AIM
cc: jmc@SAIL
Ellie,
would you see if you could get an appointment with Dean Gordon Bower
for John mcCarthy and me on Tuesday Oct 11, after 10am and before 3:30pm
(John, are all those times good for you?). If not, try Monday late
afternoon.
Please try tro get a reasonably up-to-date resume of Nils from Golub's
office and send one to john and give one to me.
John, any comments/
Ed
-------
I have a class from 1:15 to 2:30. Otherwise, it's ok.
∂05-Oct-83 0103 TOB
To: JMC@SU-AI, RPG@SU-AI, rindfleisch@SUMEX-AIM
This certainly does not correspond to my impression of events.
I tried several times to get representation on the committee
and was unsuccessful in even finding out what was going on. If
he talked with me, it was not clear to me what it was about, and
that it was planning, as opposed to talk.
∂31-Aug-83 1732 JMC
To: TOB@SU-AI
CC: RPG@SU-AI, rindfleisch@SUMEX-AIM
RPG tells me that he talked to you while the proposal was being prepared,
and you expressed no interest at the time.
∂05-Oct-83 0832 AAAI-OFFICE@SUMEX-AIM meeting
Received: from SUMEX-AIM by SU-AI with PUP; 05-Oct-83 08:32 PDT
Date: Wed 5 Oct 83 08:35:21-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: meeting
To: JMC@SAIL
cc: aaai-office@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
Do you have some time before the 11:00 meeting to discuss a possible
fellowship program and an agenda for the meeting?
Claudia
-------
∂05-Oct-83 0853 RPG
To: TOB
CC: JMC
∂05-Oct-83 0103 TOB
To: JMC@SU-AI, RPG@SU-AI, rindfleisch@SUMEX-AIM
This certainly does not correspond to my impression of events.
I tried several times to get representation on the committee
and was unsuccessful in even finding out what was going on. If
he talked with me, it was not clear to me what it was about, and
that it was planning, as opposed to talk.
∂31-Aug-83 1732 JMC
To: TOB@SU-AI
CC: RPG@SU-AI, rindfleisch@SUMEX-AIM
RPG tells me that he talked to you while the proposal was being prepared,
and you expressed no interest at the time.
The matter has been resolved, thanks to calm negotiations with Rod Brooks.
He didn't seem to have any difficulty understanding what I was saying.
∂05-Oct-83 0854 RPG Binford
I hope I wasn't out of line with him, but I find dealing with him
difficult.
I've asked Dianna to arrange for my office, Carolyn's, and Wieneke's
to have a common key so that the Lisp machine consoles will be available
to all.
Don't forget the senior RA promotion.
-rpg-
∂05-Oct-83 0908 BMACKEN@SRI-AI.ARPA associations
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Oct 83 09:07:56 PDT
Date: Wed 5 Oct 83 09:07:32-PDT
From: BMACKEN@SRI-AI.ARPA
Subject: associations
To: csli-friends@SRI-AI.ARPA
I have updated the associations file on <CSLI> with your latest
comments. Are there any more changes or additions?
B.
-------
∂05-Oct-83 0919 CLT lamb
should be put in the oven at 5pm at 350.
I think the shelf needs to be lowered.
It should be one up from the bottom.
(Estimated time of eating = 7:30)
∂05-Oct-83 0956 ullman%SU-HNV.ARPA@SU-SCORE.ARPA meetings
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 09:56:14 PDT
Received: from Diablo by Score with Pup; Wed 5 Oct 83 09:54:58-PDT
Date: Wed, 5 Oct 83 09:54 PDT
From: Jeff Ullman <ullman@Diablo>
Subject: meetings
To: rfn@sail, super@score
It apppears that our Thursday-at-3 meetings are defunct,
except that we have a scheduled talk for 3PM on the 19th of Oct.
The 3PM time conflicts with John H.'s class and probably a few
other things. However, there are still a few things we might
talk about, e.g., Steve still promises to report on the Los Alamos
meeting, and I could report on IFIP.
I'd like to try to agree on a time, so would you please mail
Rosemary Napier (rfn@sail) the times when you can/cannot meet.
I suspect Sunday Brunch will be the only possibility, but let's
try anyway.
∂05-Oct-83 1018 DFH Prof. Tratatenbrot
called. He is at 325-7539.
Also, Victor Kuo just arrived.
∂05-Oct-83 1032 LUNDSTROM@SU-SCORE.ARPA [weeks at ames-vmsb: ACM SIGBIG Oct 5th Meeting]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 10:31:51 PDT
Date: Wed 5 Oct 83 10:27:30-PDT
From: Stephen Lundstrom <LUNDSTROM@SU-SCORE.ARPA>
Subject: [weeks at ames-vmsb: ACM SIGBIG Oct 5th Meeting]
To: Super@SU-SCORE.ARPA
This meeting is tonight. Steve
---------------
Return-Path: <weeks@ames-vmsb>
Received: from ames-vmsb by SU-SCORE.ARPA with TCP; Wed 5 Oct 83 09:36:45-PDT
Date: Wed, 05 Oct 83 09:35:43 PDT
From: weeks at ames-vmsb
Subject: ACM SIGBIG Oct 5th Meeting
To: LUNDSTROM at SU-SCORE
ASSOCIATION FOR COMPUTING MACHINERY
SAN FRANCISCO GOLDEN GATE CHAPTER
SPECIAL INTEREST COMMITEE, "SIGBIG"
For Large High Speed Computers
Wednesday, October 5, 1983
Frank Kuo
SRI
"SPREAD, A Supercomputer Network"
7:00 Business & Progress Reports
7:30 Speaker
At SRI, 333 Ravenswood, Menlo Park,
in the main Administration Building, room # AD 119.
The security guard in the lobby will be able to give directions.
Please let Mary Fowler know if you are planning to attend.
Due to the short notice of this meeting the speaker's
presentation will be given at a later meeting if many
people are unable to attend.
For more information,
and to reply, contact Mary Fowler, (415)965-6515
Ames Decnet CER::fowler
Arpanet Fowler @ ames-vmsa
Ride Sharing, contact Frank Olken, LBL
ARPANET olken at lbl-csam
(415)486-5891
To be sure that the electronic mail is reaching everyone would you please
let me know if you received the original meeting announcement or the
additional information message or this combined meeting announcement.
Thank you,
Cindy Weeks (ARPAnet: weeks@ames-vmsb)(AMES DECnet: JUP::weeks)(415-965-6015)
-------
∂05-Oct-83 1124 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 6
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Oct 83 11:23:52 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Wed 5 Oct 83 11:22:31-PDT
Date: Wed, 5 Oct 83 11:22 PDT
From: desRivieres.PA@PARC-MAXC.ARPA
Subject: CSLI Activities for Thursday Oct. 6
To: csli-friends@SRI-AI.ARPA
CSLI Schedule for Thursday, October 6, 1983
10:00 Research seminar on natural language
Speaker: Bob Moore (SRI & CSLI)
Title: "Problems in Semantic Analysis of Natural Language"
Place: Redwood Hall, classroom
12:00 TINLunch
Discussion leader: Marilyn Ford
Place: Ventura Hall (as usual)
Paper for discussion:
THE ROLE OF GRAMMARS IN MODELS OF LANGUAGE USE
BY
Robert C. Berwick and Amy S. Weinberg
Pages 1-25
2:00 Research seminar on approaches to computer languages
Speaker: Fernando Pereira (SRI & CSLI)
Title: "Logic as a Programming Language"
Place: Redwood Hall, classroom
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speaker: Geoffrey Nunberg (Dept. of Linguistics, Stanford)
Title: "Prescriptive Grammar"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It can
be reached from Campus Drive or Panama Street. From Campus Drive follow
the sign for Jordan Quad. Parking is in the C-lot between Ventura
and Jordan Quad.
∂05-Oct-83 1136 DFH Psychology today article
Alfred Meyer called. He wants me to read the
changes to him over the phone today. In looking
at the two versions, I was uncertain which was the
one to go by. If you are not going to be here
around 1, could you leave me a message as to which
version I should read from or how they relate to
each other? Thanks.
∂05-Oct-83 1141 Janet.Asbury@CMU-CS-A Electronic Library
Received: from CMU-CS-A by SU-AI with TCP/SMTP; 5 Oct 83 11:41:06 PDT
Received: from [128.2.254.192] by CMU-CS-PT with CMUFTP; 5 Oct 83 14:28:22 EDT
Date: 5 Oct 83 1414 EDT (Wednesday)
From: Jan.Asbury <Janet.Asbury@CMU-CS-A>
To: JMC@SU-AI
Subject: Electronic Library
CC: Janet.Asbury@CMU-CS-A
Message-Id: <05Oct83.141435.JA61@CMU-CS-A>
Dear Dr. McCarthy,
Thanks for the note saying you received my netmail. Would you send a
descriptive biography to be included with the Electronic Library proposal?
Raj asked me to inquire about the revised version. He is anxious to send it
off. Could you let us know when we might expect it?
Thanks much.
Jan
∂05-Oct-83 1302 YM computer facilities committee
Is there going to be only one committee this year, as oppose to a current
facilities and planning committees?
Thanks, Yoni (student bureaucrat)
I see no need for two committees, and Gene hasn't suggested anything else.
∂05-Oct-83 1401 GOLUB@SU-SCORE.ARPA [John McCarthy <JMC@SU-AI>:]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 14:01:43 PDT
Date: Wed 5 Oct 83 14:03:03-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: [John McCarthy <JMC@SU-AI>:]
To: JMC@SU-AI.ARPA
What is this about? GENE
---------------
Return-Path: <JMC@SU-AI>
Received: from SU-AI.ARPA by SU-SCORE.ARPA with TCP; Wed 5 Oct 83 14:00:26-PDT
Date: 05 Oct 83 1358 PDT
From: John McCarthy <JMC@SU-AI>
To: YM@SU-AI, golub@SU-SCORE
I see no need for two committees, and Gene hasn't suggested anything else.
-------
Sorry, Gene. The question was whether there are separate planning and
current computer facilities committees.
∂05-Oct-83 1435 LIBRARY@SU-SCORE.ARPA Access to SOCRATES the public access catalog to library materials
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 14:35:37 PDT
Date: Wed 5 Oct 83 14:27:02-PDT
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: Access to SOCRATES the public access catalog to library materials
To: su-bboards@SU-SCORE.ARPA
cc: mccarthy@SU-AI.ARPA
Several people in the computer science department have been asking how they
can get access to SOCRATES from their own terminals. SOCRATES is the online
catalog for the Stanford University Libraries and will be available free
from terminals located in Green, Meyer, Engineering, and Biology. Eventually
all libraries will have terminals including Math/CS. In addition, the file
is available to anyone who has a CIT account. Those of you in CS who can
dial up right from your terminal can access SOCRATES as long as you have
a CIT account. In order to be able to get onto CIT without logging off,
a program needs to be written for the ethernet tip to go in that direction
(CIT is able to go to CS computers).
How much interest is there within CS to have access to SOCRATES through your
terminal without logging off? If there is a lot of interest, I can relay
this information to John Sack. Any other questions or suggestions you
have about SORATES may also be sent to me and I will try to answer them
or pass them on to John.
Harry
-------
∂05-Oct-83 1442 HANS@SRI-AI.ARPA CSLI addresses
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Oct 83 14:42:24 PDT
Date: Wed 5 Oct 83 14:34:21-PDT
From: Hans Uszkoreit <Hans@SRI-AI.ARPA>
Subject: CSLI addresses
To: csli-folks@SRI-AI.ARPA
cc: tyson@SRI-AI.ARPA
I have noticed that several CSLI-internal (not confidential) messages
were sent to CSLI-FRIENDS.
Please try to remember that CSLI-FRIENDS will not only distribute mail
to the people affiliated with CSLI (which can be reached through CSLI-FOLKS
or CSLI-PEOPLE) but also to many other nice people in the bay area who
we would like to invite to talks, seminars, etc.
Let us try not to swamp them with lots of CSLI related mail that doesn't
mean anything to them. Otherwise CSLI could get a reputation of being
another one of those notorious sources of net junk mail.
Also, if you reply to messages that have been sent via a mailing list to
a group of people, be careful when you intend to reply to the sender.
Check if your mailing program will not reply to everybody mentioned in the
header of the original message. You might be able to set the option of
whom replies should be sent to (you can do that in MM). -- Hans
-------
∂05-Oct-83 1449 EENGELMORE@SUMEX-AIM NILS MTG.
Received: from SUMEX-AIM by SU-AI with PUP; 05-Oct-83 14:48 PDT
Date: Wed 5 Oct 83 14:52:07-PDT
From: Ellie Engelmore <EENGELMORE@SUMEX-AIM>
Subject: NILS MTG.
To: Feigenbaum@SUMEX-AIM, JMC@SAIL
cc: DFH@SAIL, EENGELMORE@SUMEX-AIM
Dean Gordon Bower is expecting to meet with you from 10:00 to 11:00,
Tuesday, October 11 at "The Deanery".
I put a copy of Nils' CV in JMC's mailbox and a copy in EAF's purple
folder.
Ellie
-------
∂05-Oct-83 1558 DELAGI%SUMEX-AIM.ARPA@SU-SCORE.ARPA Re: meetings
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 15:58:48 PDT
Received: from SUMEX-AIM by Score with Pup; Wed 5 Oct 83 15:55:02-PDT
Date: Wed 5 Oct 83 15:56:41-PDT
From: Bruce Delagi <DELAGI@SUMEX-AIM>
Subject: Re: meetings
To: ullman@Diablo
cc: rfn@SAIL, super@Score, DELAGI@SUMEX-AIM
In-Reply-To: Your message of Wed 5 Oct 83 09:57:31-PDT
Sunday Brunch is taken but noon Tuesday thru Thursday is good (more productive
than lunch).
/bruce
-------
∂05-Oct-83 1654 DFH travel
Your prepaid ticket is here for the Brazil trip -- I
think there are some schedule problems but their computer
was down, so will tell you what I find out tomorrow morning.
The real problem is that you need a visa for both Brazil and
Korea. If you don't have any more passport size photos, get
6 or so done. I will get visa applications from Dina Bolla
tomorrow, at least for Brazil -- the Korean consulate is mailing
me one. Also bring your passport in.
∂05-Oct-83 1737 DKANERVA@SRI-AI.ARPA CSLI Newsletter, No. 3
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Oct 83 17:34:34 PDT
Date: Wed 5 Oct 83 17:33:20-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: CSLI Newsletter, No. 3
To: csli-friends@SRI-AI.ARPA
! CSLI Newsletter
October 6, 1983 * * * Number 3
Starting with this issue, the CSLI Newsletter will go out each
Thursday morning. Please try to get anything you want included in the
week's issue in to me by Wednesday morning. The Newsletter serves
also as a record of CSLI activities, so we want to include as many as
possible of the events of each week, even though they may have been
announced through the mail or otherwise.
- Dianne Kanerva
* * * * * * *
ADVISORY PANEL
Our first advisory panel meeting will probably be November 10-12.
So far, Barbara Partee, George Miller, and Jerry Fodor have agreed to
serve on the panel. Acceptances are coming in daily. I hope that all
the CSLI folk will be around for that meeting.
- Jon Barwise
* * * * * * *
THE SEMANTICS OF COMPUTER LANGUAGES GROUP
The working group on the semantics of computer languages met at
Xerox PARC on Monday, October 3, and planned activities for the fall.
There was a lot of discussion as to whether to work on a single topic
or to delve into a number of different topics. It was decided to do
both--to work through Gordon Plotkin's new book on denotational
semantics on alternate weeks, with talks on a diversity of topics
related to the semantics of computer languages the other weeks.
The group will meet on Tuesday mornings at Xerox PARC,
9:30-11:30. On October 11, Ian Mason will lead the discussion of
Chapter 1 of Plotkin's book, and on October 18, Henson Graves will
discuss category theory and computer languages.
* * * * * * *
ACKNOWLEDGMENTS FOR PAPERS FOR CSLI PROJECTS
In preparing papers and technical reports to be published by CSLI
or SRI, the appropriate funding acknowledgments are as follows.
1. For CSLI reports and papers:
"The research reported in this paper has been made possible
by a gift from the System Development Foundation."
2. For SRI reports and papers:
"The research reported in this paper has been made possible
by a gift from the System Development Foundation and was
conducted as part of a coordinated research effort with
the Center for the Study of Language and Information,
Stanford University."
! * * * * * * *
CSLI SCHEDULE FOR THURSDAY, OCTOBER 6, 1983
10:00 Research Seminar on Natural Language
Speaker: Bob Moore (SRI & CSLI)
Title: "Problems in Semantic Analysis of Natural Language"
Place: Redwood Hall, classroom
12:00 TINLunch
Discussion leader: Marilyn Ford
Paper for discussion: "The Role of Grammars in Models of Language Use"
by Robert C. Berwick and Amy S. Weinberg,
Pages 1-25
Place: Ventura Hall (as usual)
2:00 Research Seminar on Approaches to Computer Languages
Speaker: Fernando Pereira (SRI & CSLI)
Title: "Logic as a Programming Language"
Place: Redwood Hall, classroom
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speaker: Geoffrey Nunberg (Dept. of Linguistics, Stanford)
Title: "Prescriptive Grammar"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It
can be reached from Campus Drive or Panama Street. From Campus Drive
follow the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
* * * * * * *
!
* * * * * * *
Center for the Study of Language and Information
AFTERNOON COLLOQUIUM SCHEDULE
October
* * * * *
Thursday, October 6, 1983
GEOFFREY NUNBERG
Department of Linguistics
Stanford University
"Prescriptive Grammar"
* * * * *
Thursday, October 13, 1983
LAURI KARTTUNEN STAN ROSENCHEIN
Department of Linguistics Artificial Intelligence Lab
University of Texas, Austin S.R.I.
Symposium on "Computational" and "Theoretical" Linguistics
* * * * *
Thursday, October 20, 1983
JERRY HOBBS
Artificial Intelligence Laboratory
S.R.I.
"Discourse Coherence and Discourse Structure"
* * * * *
Thursday, October 27, 1983
JAY M. TENENBAUM
Laboratory for Artificial Intelligence Research
Fairchild
"A.I. Research at Fairchild"
* * * * *
ALL COLLOQUIA MEET AT 4:15 P.M., IN ROOM G-19, REDWOOD HALL
!
* * * * * * *
Center for the Study of Language and Information
AFTERNOON COLLOQUIUM SCHEDULE
November (tentative)
* * * * *
Thursday, November 3, 1983
GLYNN WINSKEL
Department of Mathematics
Carnegie-Mellon University
"Denotational Semantics: An Introduction"
* * * * *
Thursday, November 10, 1983
MICHAEL BEESON
Department of Mathematics and Computer Science
San Jose State University
"Computational Aspects of Intuitionistic Logic"
* * * * *
Thursday, November 17, 1983
CHARLES FILLMORE
Department of Linguistics
University of California, Berkeley
Title to be announced
* * * * *
ALL COLLOQUIA MEET AT 4:15 P.M., IN ROOM G-19, REDWOOD HALL
* * * * * *
!
* * * * * *
TINLUNCH SCHEDULE
TINLunch will be held on each Thursday at Ventura Hall on the
Stanford University campus as a part of CSLI activities. Copies of
TINLunch papers will be at SRI in EJ251 and at Stanford University
in Ventura Hall.
October 6 Marilyn Ford
October 13 Kurt Konolige
October 20 Per-Christian Halvorsen
NEXT WEEK: "On Inheritance Hierarchies with Exceptions"
by David W. Etherington and Raymond Reiter,
discussion led by Kurt Konolige
* * * * * * *
WEST COAST CONFERENCE ON FORMAL LINGUISTICS
March 16-18, 1984
The deadline for abstracts of papers to be submitted for the
Third West Coast Conference on Formal Linguistics will be December 15,
1983. They should be prepared in the usual format: one side maximum,
8.5" by 11" paper, eight copies, no identification of author,
accompanied by 3" by 5" card bearing author's name , affiliation,
January mailing address, and phone number. They should be sent to
WCCFL 3
Syntax Research Center
Cowell College
UCSC
SANTA CRUZ, CA 95064
An official announcement in hard copy will be mailed out soon. Anyone
at a linguistics department or language research institution west of
the Rockies or in Canada west of Ontario should see one on their
bulletin board. People who believe their location is so obscure that
they might not be found should write to the address given above
requesting that circulars be sent to them.
* * * * * * *
!
* * * * * * *
CSCSI-84 CALL FOR PAPERS
May 18-20, 1984
Canadian Society for Computational Studies of Intelligence
The Fifth National Conference of the CSCSI will be held at the
University of Western Ontario in London, Canada. Papers are requested
in all areas of AI research, particularly those listed below. The
Program Committee members responsible for these areas are included.
Knowledge Representation:
Ron Brachman (Fairchild R & D), John Mylopoulos (U of Toronto)
Learning:
Tom Mitchell (Rutgers U), Jaime Carbonell (CMU)
Natural Language:
Bonnie Weber (U of Pennsylvania), Ray Perrault (SRI)
Computer Vision:
Bob Woodham (U of British Columbia), Allen Hanson (U Mass)
Robotics:
Takeo Kanade (CMU), John Hollerbach (MIT)
Expert Systems and Applications:
Harry Pople (U of Pittsburgh), Victor Lesser (U Mass)
Logic Programming:
Randy Goebel (U of Waterloo), Veronica Dahl (Simon Fraser U)
Cognitive Modelling:
Zenon Pylyshyn, Ed Stabler (U of Western Ontario)
Problem Solving and Planning:
Stan Rosenschein (SRI), Drew McDermott (Yale)
!
Authors are requested to prepare Full papers, of no more
than 4000 words in length, or Short papers of no more than 2000 words
in length. A full page of clear diagrams counts as 1000 words. When
submitting, authors must supply the word count as well as the area in
which they wish their paper reviewed. (Combinations of the above
areas are acceptable). The Full paper classification is intended for
well-developed ideas, with significant demonstration of validity,
while the Short paper classification is intended for descriptions of
research in progress. Authors must ensure that their papers describe
original contributions to or novel applications of Artificial
Intelligence, regardless of length classification, and that the
research is properly compared and contrasted with relevant
literature.
Three copies of each submitted paper must be in the hands of the
Program Chairman by December 7, 1983. Papers arriving after that date
will be returned unopened, and papers lacking word count and
classifications will also be returned. Papers will be fully reviewed
by appropriate members of the program committee. Notice of acceptance
will be sent on February 28, 1984, and final camera ready versions are
due on March 31, 1984. All accepted papers will appear in the
conference proceedings.
Correspondence should be addressed to either the General Chairman
or the Program Chairman, as appropriate.
General Chairman Program Chairman
Ted Elcock, John K. Tsotsos
Dept. of Computer Science, Dept. of Computer Science,
Engineering and Mathematical 10 King's College Rd.,
Sciences Bldg., University of Toronto,
University of Western Ontario Toronto, Ontario, Canada,
London, Ontario, Canada M5S 1A4
N6A 5B9 (416)-978-3619
(519)-679-3567
* * * * * * *
!
* * * * * * *
CALL FOR PAPERS
COLING 84, TENTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL LINGUISTICS
COLING 84 is scheduled for 2-6 July 1984 at Stanford University, Stanford,
California. It will also constitute the 22nd Annual Meeting of the
Association for Computational Linguistics, which will host the conference.
Papers for the meeting are solicited on linguistically and computationally
significant topics, including but not limited to the following:
o Machine translation and machine-aided translation.
o Computational applications in syntax, semantics, anaphora, and
discourse.
o Knowledge representation.
o Speech analysis, synthesis, recognition, and understanding.
o Phonological and morpho-syntactic analysis.
o Algorithms.
o Computational models of linguistic theories.
o Parsing and generation.
o Lexicology and lexicography.
Authors wishing to present a paper should submit five copies of a summary
not more than eight double-spaced pages long, by 9 January 1984 to: Prof.
Yorick Wilks, Languages and Linguistics, University of Essex, Colchester,
Essex, CO4 3SQ, ENGLAND [phone: 44-(206)862 286; telex 98440 (UNILIB G)].
It is important that the summary contain sufficient information, including
references to relevant literature, to convey the new ideas and allow the
program committee to determine the scope of the work. Authors should clearly
indicate to what extent the work is complete and, if relevant, to what
extent it has been implemented. A summary exceeding eight double-spaced
pages in length may not receive the attention it deserves.
Authors will be notified of the acceptance of their papers by 2 April 1984.
Full length versions of accepted papers should be sent by 14 May 1984 to
Dr. Donald Walker, COLING 84, SRI International, Menlo Park, California,
94025, USA [phone: 1-(415)859-3071; arpanet: walker@sri-ai].
Other requests for information should be addressed to Dr. Martin Kay, Xerox
PARC, 3333 Coyote Hill Road, Palo Alto, California 94304, USA
[phone: 1-(415)494-4428; arpanet: kay@parc].
* * * * * * *
-------
∂05-Oct-83 2023 @SU-SCORE.ARPA:slndstrm@Shasta Re: meetings
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Oct 83 20:23:26 PDT
Received: from Shasta by SU-SCORE.ARPA with TCP; Wed 5 Oct 83 20:22:35-PDT
Date: Wednesday, 5 Oct 1983 20:21-PDT
To: Bruce Delagi <DELAGI@SUMEX-AIM>
Cc: ullman at Diablo <ullman@Diablo>, rfn at SAIL <rfn@SAIL>,
super at Score <super@Score>
Subject: Re: meetings
In-Reply-To: Your message of Wed 5 Oct 83 15:56:41-PDT.
From: Steve Lundstrom <slndstrm@Shasta>
Agree with Bruce, except that Wed. noon is a sometimes interesting
computer architecture seminar. (Why not have the meeting on Thurs
noon so that our afternoon speakers could visit with us informally
from time to time?)
Steve
∂06-Oct-83 0007 HST visit in januar 84
a paper of mine was accepted by a congress in hawaii ,4.-7.1.84.
to use the money most economic i could stay for some days at
stanford and peep a little bit in your stock of mit material.
to get it scientific i offer some themes for talks which i`m
able to present:
1. compilation understood as provable semantics-preserving
program transformation (example languages: lisp and fortran)
2. principles of object-oriented programming
3. machine-models, programming languages and programming styles
∂06-Oct-83 0829 RIGGS@SRI-AI.ARPA CSLI Education Committee Meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 6 Oct 83 08:29:16 PDT
Date: Thu 6 Oct 83 08:29:48-PDT
From: RIGGS@SRI-AI.ARPA
Subject: CSLI Education Committee Meeting
To: KJB@SRI-KL.ARPA, KAY@PARC-MAXC.ARPA, STAN@SRI-KL.ARPA, JMC@SU-AI.ARPA,
WASOW%psych@SUMEX-AIM.ARPA
cc: WUNDERMAN@SRI-KL.ARPA, JRP@SRI-KL.ARPA
The CSLI Education Committee meeting will be held on
Friday, October 7, 1983 at 1:15 p.m. in Room #20 at
Ventura Hall. I hope to see you then.
Sandy
-------
∂06-Oct-83 1001 BMACKEN@SRI-AI.ARPA meeting room
Received: from SRI-AI by SU-AI with TCP/SMTP; 6 Oct 83 10:01:24 PDT
Date: Thu 6 Oct 83 10:00:54-PDT
From: BMACKEN@SRI-AI.ARPA
Subject: meeting room
To: csli-people@SRI-AI.ARPA
Just a reminder that Room 6, downstairs in Ventura, is CSLI's and is
available all the time for conferences, discussions, reading, etc. We
are selecting and ordering comfortable furniture, but it will be a
while before we get it. We have ordered the Wall Street Journal; what
other newspapers would you like? Its a good place to leave tech reports,
papers, old journals, etc. ... anything you want to share and don't need
returned.
B.
-------
∂06-Oct-83 1150 YM computer facilities committee
To: JMC
CC: OP
Many students want to serve on this committee. How many would you like to have?
Two will suffice.
∂06-Oct-83 1330 DFH Kuo
According to correspondence in his file, we are to
pay him $575/mo. Is this still correct, and should
I go ahead with the appointment beginning Oct. 1?
If so, what account?
Yes, it is correct, Oct 1 is ok, and charge it to my unrestricted.
∂06-Oct-83 1342 DFH Livermore security form
The form seems to have been changed slightly, and
they are now asking for the date of birth (rather than
just ages) for your relatives, so I don't have this
from the previous form.
∂06-Oct-83 1345 DFH more travel
The only direct flight SFO/Austin on 10/20 is
at 3:25 in the afternoon arriving 9:50 pm. There
are a variety of connections that would get you
there somewhat earlier, though you would probably
still miss most of the talks that day. What is
your preference in this regard?
∂06-Oct-83 1623 DFH AI Qual for R. Vistnes
To: lenat@SU-SCORE, Buchanan@SUMEX-AIM, JMC@SU-AI, RV@SU-AI
CC: mullen@SUMEX-AIM
This is now scheduled for Mon., Nov. 7, 2 pm. I will
reserve a room and let you all know the location later.
∂06-Oct-83 2111 JMC
848-7610
∂07-Oct-83 0029 FEIGENBAUM@SUMEX-AIM Re: committee assignments
Received: from SUMEX-AIM by SU-AI with PUP; 07-Oct-83 00:29 PDT
Date: Fri 7 Oct 83 00:32:11-PDT
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM>
Subject: Re: committee assignments
To: GOLUB@SU-SCORE.ARPA
cc: jmc@SAIL
In-Reply-To: Your message of Wed 5 Oct 83 16:41:32-PDT
I would suggest that the Industrial Professorship committee be a one-person
committee, chaired and polulated by JMC. He agrees.
Ed
-------
Report on the Industrial Professorship:
There were exactly four volunteers for the four slots, so there
was no problem of selection. Two people from SRI jointly taught a course
on vision in the Spring of 83. The three courses for this academic
year are taught by Bob Moore of SRI, Stan Rosenschein of SRI and
John Greenstadt of IBM. The course descriptions are in the catalog.
The Department agreed to one year of Industrial Professorship, but
we won't have more than the Spring and part of the Fall to evaluate
before we have to decide whether it is a success. Therefore, I suggest
we continue for another year and make the main evaluation next Fall.
We need to get out the solicitationof applications again, say in early
November. The deadline for proposed catalog statements needs to be
February 1 so we can decide promptly and get the new statements
into the catalog. (We have to be sure that this year's statements
aren't continued next year). If there is a problem of selection
this year, I propose to put it to a Department meeting.
∂07-Oct-83 0828 AAAI-OFFICE@SUMEX-AIM Re: President's message
Received: from SUMEX-AIM by SU-AI with PUP; 07-Oct-83 08:26 PDT
Date: Fri 7 Oct 83 08:29:40-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Re: President's message
To: JMC@SAIL
cc: AAAI-OFFICE@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
In-Reply-To: Your message of Fri 7 Oct 83 01:40:00-PDT
John,
Bob is out of town so, please send the revised Presidential message to
AAAI@SCORE where we can begin to TEK it.
Thanks,
Claudia
-------
∂07-Oct-83 1150 AAAI-OFFICE@SUMEX-AIM Next meeting
Received: from SUMEX-AIM by SU-AI with PUP; 07-Oct-83 11:50 PDT
Date: Fri 7 Oct 83 11:53:03-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Next meeting
To: JMC@SAIL, Buchanan@SUMEX-AIM, Genesereth@SUMEX-AIM, Walker@SRI-AI.ARPA,
Amsler@SRI-AI.ARPA, Feinler@SRI-NIC.ARPA
cc: aaai-office@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I've made some arrangements for our next meeting regarding the on-line
abstract service/library proposal. It will be in Room 252 in SU's CS Dept
on Wednesday, November 2 at 10:00 am. If I don't hear from you, I can
assume you will be present at the meeting.
-- Claudia
-------
∂07-Oct-83 1217 AAAI-OFFICE@SUMEX-AIM Fredkin
Received: from SUMEX-AIM by SU-AI with PUP; 07-Oct-83 12:17 PDT
Date: Fri 7 Oct 83 12:20:33-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Fredkin
To: JMC@SAIL
cc: aaai-office@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
I talked to Nils about leading the Fredkin effort. He said he was too
busy with work, but felt we should have a meeting of about 10-15 AI people
prior to any grandeous meeting. He thought that CMU would be a good location
(given the fact that Allan Newell is having a difficult time traveling these
days).
CCM
-------
∂07-Oct-83 1303 PACK@SU-SCORE.ARPA blocks axioms
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Oct 83 13:03:39 PDT
Date: Fri 7 Oct 83 13:04:58-PDT
From: Leslie E. Pack <PACK@SU-SCORE.ARPA>
Subject: blocks axioms
To: jmc@SU-AI.ARPA
Using circumscription of abnormality to solve the frame problem.
The frame problem is simply defined as the question of what remains
the same in a situation that is the result of an action in some other
situation. This solution relies on the assumption that a property
persists across changing situations unless it in some aspect abnormal.
Using circumscription, we can infer that only the aspects that we can
directly infer are abnormal are abnormal.
These axioms state the persistence of properties where x is an object,
e an event, and s and result(e,s) situations:
~ab aspect1(x,e,s) -> location(x,result(e,s)) = location(x,s)
~ab aspect2(x,e,s) -> color(x,result(e,s)) = color(x,s)
The persistence of location is abnormal when the event is moving;
that of color, when the event is painting, as stated in the following
axioms:
~ab aspect3(x,position,s) -> ab aspect1(x,move(x,position),s)
~ab aspect4(x,color,s) -> ab aspect2(x,paint(x,color),s)
We must also include the positive inferences in these cases:
~ab aspect3(x,position,s) ->
location(x,result(move(x,position),s)) = position
~ab aspect4(x,color,s) ->
color(x,result(paint(x,color),s)) = color
The precedent in each of these axioms is necessary to take into
account the case in which the action is not fully or successfully
carried out. The block may not be movable or the position already
occupied; there may not be any paint of the correct color.
(location(z,s) = x or location(z,s) = position)
and ~ab aspect5(x,z,s,position) -> ab aspect3(x,position,s)
~(have(paint,s) and color(paint,s) = color)
and ~ab aspect6(paint,s) -> ab aspect4(x,color,s)
Including the ~ab clauses for the next level of abnormality in
antecedents of every axiom causes no trouble if they are not used, and
allows for anything to go wrong in any step of the way. It may be
possible to move a box with something on top of it if that thing is
tiny.
tiny z and ~ab aspect7(z) -> ab aspect5(x,z,s,position)
dense z and ~ab aspect8(z) -> ab aspect7(z)
-------
∂07-Oct-83 1332 oliger%SU-NAVAJO.ARPA@SU-SCORE.ARPA Re: meetings
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Oct 83 13:32:43 PDT
Received: from Navajo by Score with Pup; Fri 7 Oct 83 13:29:45-PDT
Date: Fri, 7 Oct 83 13:29:13 PDT
From: Joseph Oliger <oliger@Navajo>
Subject: Re: meetings
To: Jeff Ullman <ullman@Diablo>
Cc: rfn@Sail, super@Score
In-Reply-To: Your message of Wed, 5 Oct 83 09:54 PDT.
The Wed. noon idea that has been proposed would be fine with me. Is the
scheduled talk you refered to on Thurs 10/20 or Wed 10/19?
Joe
∂07-Oct-83 1335 ULLMAN@SU-SCORE.ARPA Next talk
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Oct 83 13:35:06 PDT
Date: Fri 7 Oct 83 13:29:49-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Next talk
To: CS440: ;
The Speaker for Thursday Oct., 13 is Vason Scrini of Berkeley.
We meet in 380-X (math building) at 4:15PM.
The abstract for the talk follows:
5-Oct-83 14:02:25-PDT,2380;000000000001
Return-Path: <TRATTNIG@SU-SIERRA.ARPA>
Received: from SU-SIERRA.ARPA by SU-SCORE.ARPA with TCP; Wed 5 Oct 83 14:02:21-PDT
Date: Wed 5 Oct 83 14:02:33-PDT
From: Werner Trattnig <TRATTNIG@SU-SIERRA.ARPA>
Subject: Scrini's talk
To: ullman@SU-SCORE.ARPA
Jeff,
here is an abstract of Scrini's seminar next week:
-----
Vason P. Scrini
Department of Computer Science
University of Alabama in Birmingham
A MESSAGE BASED PROCESSOR FOR SUPER COMPUTER SYSTEMS
The architecture of a message-based processor for a super computer
system based on a data driven model and comprising several processors
is described. The processor receives messages (tokens) containing
operations and/or data asynchronously. Transformations are performed
on the data and results forwarded to other processors. The processor
communicates with memory using packets to receive and send tokens
instead of accessing memory. The operations in a token are executed
correctly by a processor. If faults are detected in the execution of
operations in a token then there are mechanisms in the processor that
facilitate the reassignment of the token to another processor.
Nine classes of instructions have been identified for the
message-based processor. These instructions support message
communication, string operations, floating point operations, fixed
point operations, and manipulate structures such as tables and streams
of data. To achieve a good degree of performance in the computer
system, several features are included in the processor. One feature
is a feedback path from the outputs of functional units to the inputs
so that the token traffic in the dataflow system can be reduced.
Another feature is the availability of multiple functional units for
the concurrent execution of operations in compound instructions.
There is also a data memory (table memory) in the processor to do
string operations on a vector of data.
The key units of the processor have been designed using AMD 2900
bit-slice chips. A VLSI realization using NCR 32000 chips and custom
VLSI chips is in progress. A modified versionof Cray-1S with string
functional units, and datadriven control is also under design.
-------
-------
∂08-Oct-83 0925 CLT mud
I don't know who made the mess on the livingroom floor, but I don't
appreciate in the least. I've spent a fair ammount of time
since we returned trying to get the place clean.
Perhaps you could please see that the mess gets cleaned up.
I think it means vacuuming the rugs and going over the
exposed floor with a damp sponge. (There is the vacuum
in the linen closet, such as it is.)
Thanks,
∂08-Oct-83 1546 FEIGENBAUM@SUMEX-AIM
Received: from SUMEX-AIM by SU-AI with PUP; 08-Oct-83 15:46 PDT
Date: Sat 8 Oct 83 15:49:20-PDT
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM>
To: JMC@SAIL, golub@Score
In-Reply-To: Your message of Fri 7 Oct 83 00:55:00-PDT
Once again, I propose that JMC be a one-man committee to do the Industrial
Professoreships.
Ed
-------
∂08-Oct-83 2359 POURNE@MIT-MC
Received: from MIT-MC by SU-AI with TCP/SMTP; 8 Oct 83 23:59:47 PDT
Date: 9 October 1983 03:02 EDT
From: Jerry E. Pournelle <POURNE @ MIT-MC>
To: JMC @ SU-AI
In-reply-to: Msg of 27 Sep 83 2109 PDT from John McCarthy <JMC at SU-AI>
read report and make comments.
report went to Clark and President last week.
well receivred.
∂09-Oct-83 1342 ALMOG@SRI-AI.ARPA reminder on why context wont go away
Received: from SRI-AI by SU-AI with TCP/SMTP; 9 Oct 83 13:40:07 PDT
Date: 9 Oct 1983 1336-PDT
From: Almog at SRI-AI
Subject: reminder on why context wont go away
To: csli-friends at SRI-AI
Reminder : WHY CONTEXT WONT GO AWAY
Second meeting Tuesday 3.15, 10.11.83, Ventura Hall
Abstract
Impossibility results in context-based semantics:cognition
An analysis of the COGNITIVE impossiblity result will be given.
First, we shall distinguish the cognitive significance of atomic sentences
from the semantics of non atomic belief sentences. We shall go on to deal
with Frege's dispensability thesis ("indexical characterizations
of cognitive states are cognitively equivalent to purely descriptional
characterizations"). It was conceived in the realm of the eternal
and so was conceived in sin. We shall then see how and why Castaneda
Perry were able to refute it.
The new theory of cognitive states of Perry and Kaplan will
be discussed. The almost magical symmetry between the semantical theory
and cognitive notions will be analyzed. Perry's use of characters to
introduce the important notion of "reliable programs" will be mentioned.
Finally, I shall discuss problems in the Kaplan-Perry framework.
This includes allegations that it is too LIBERAL (in analyzing WHAT
we believe and think) and allegations that it is too CONSERVATIVE in
accepting the Fregean ideas that what determines reference is (i) what we
learn when we learn the language, and is (ii) what characterizes cognitive
significance.
-------
∂09-Oct-83 2023 ULLMAN@SU-SCORE.ARPA cs440
Received: from SU-SCORE by SU-AI with TCP/SMTP; 9 Oct 83 20:22:56 PDT
Delivery-Notice: While sending this message to SU-AI.ARPA, the
SU-SCORE.ARPA mailer was obliged to send this message in 50-byte
individually Pushed segments because normal TCP stream transmission
timed out. This probably indicates a problem with the receiving TCP
or SMTP server. See your site's software support if you have any questions.
Date: Sun 9 Oct 83 20:21:55-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: cs440
To: super@SU-SCORE.ARPA
The very sensible request has been made that we provide a list of
references that can be used to prepare for the talks.
If you are giving a talk in the series, or have some suggestions
for relevant reading, please forward them to me and I'll
compile the list and distribute it. Thanks.
The current schedule follows. Note that we have only one free slot.
This is the schedule of speakers for the CS440
Supercomputer Seminar.
It meets at 4:15 Thursdays in room 380-X.
Sept. 29: Rob Schreiber
Oct. 6: Ed Ashcroft "The SRI Dataflow Project"
Oct. 13: Vason Scrini "A message based processor for data-flow systems"
Oct. 20: Dick Gabriel "Queue-based multiprocessing LISP"
Oct. 27: Alain Hanover
Nov. 3: Bruce Delagi
Nov. 10: Steve Lundstrum
Nov. 17: Al Davis "Thoughts on Symbolic Supercomputers"
Dec. 1:
Dec. 8: Bob Keller
-------
∂10-Oct-83 0648 DOYLE@CMU-CS-C.ARPA
Received: from CMU-CS-C by SU-AI with TCP/SMTP; 10 Oct 83 06:48:14 PDT
Received: ID <DOYLE@CMU-CS-C.ARPA>; Mon 10 Oct 83 09:50:07-EDT
Date: Mon 10 Oct 83 09:50:05-EDT
From: DOYLE@CMU-CS-C.ARPA
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 10 Oct 83 01:36:00-EDT
Dear John,
Thank you for the preview of your AI Magazine article. I think it is
just the right thing to say, and am glad you are doing so. I will
post a copy on our reading wall so that our students can get the early
message.
Jon
-------
∂10-Oct-83 0817 DFH appointment
I gave Janet Coursey, a CS206 student, an appointment
to see you at 1 pm today. She said she needed about
15 minutes.
∂10-Oct-83 0900 CLT*
expenses
∂10-Oct-83 1018 DFH RPG talk
Jussi says he is talking to Feigenbaum's architecture
group today from 12 - 1:30 pm. He's not sure where.
Ellie says they usually just meet in Penny Nii's office,
but she is trying to check for me for sure. I'll send
you another message about it as soon as she has something.
Meanwhile, will tell the CS 206 student to come at 1:30
rather than 1.
∂10-Oct-83 1038 DFH appointment
I changed Janet Coursey's appt. time to 3:45 pm
this afternoon. If this is not OK, please let me
know.
∂10-Oct-83 1127 DFH RPG talk
Dick says he thinks it is in MJH 252
∂10-Oct-83 1509 RESTIVO@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 10 Oct 83 15:09:14 PDT
Date: Mon 10 Oct 83 15:10:20-PDT
From: Chuck Restivo <RESTIVO@SU-SCORE.ARPA>
To: jmc@SU-AI.ARPA
Again, thank you for your time, and the letter.
-------
∂10-Oct-83 1521 CLT
[CLT comments
Letter to Alice personally
Don't mention that you are offering money or she is
unlikely to get travel expenses from the Dutch
Send express or special delivery or ? There seems to be a dutch postal strike
She needs letter by oct 17
Content of letter something like the following -]
Adresses
(work)
(home)
Please decorate termeu.1[let,jmc]. It should go special deliver or express
mail and a copy should also be sent to the following home address.
St. Walbrugstraat 3
9712 HX Groningen
HOLLAND
∂10-Oct-83 1533 AMSLER@SRI-AI.ARPA Access to Stanford Libraries
Received: from SRI-AI by SU-AI with TCP/SMTP; 10 Oct 83 15:32:46 PDT
Date: Mon 10 Oct 83 12:50:57-PDT
From: Robert Amsler <AMSLER@SRI-AI.ARPA>
Subject: Access to Stanford Libraries
To: golub@SU-SCORE.ARPA
cc: sleeman@SUMEX-AIM.ARPA, misra%SU-Sierra@SU-SCORE.ARPA, jmc@SU-AI.ARPA,
walker@SRI-AI.ARPA
I'm Bob Amsler, a computational lexicologist working with
Don Walker at SRI in a new program within the Advanced
Computer Systems Department, called "Natural-Language and
Knowledge-Resource Systems". I have a problem that I would
greatly appreciate help with if possible.
Last year, when I was in the AI Center, I accepted the task
of writing a chapter for the upcoming 1984 issue of "Annual
Review of Information Science and Technology" on
Machine-Readable Dictionaries. These chapters are
accompanied by definitive bibliographic descriptions of the
literature in their areas, with 100-300 citations being
typical. Now to the problem. Last year, I had, through the
AI Center, a Visiting Scholar Card which permitted me entry
to the Green Library and which I found indispensible for
carrying out research on this project. However, access ended
in August and now I am afraid I won't be able to adequately
complete the chapter's background literature search. It is
both lack of access to the bibliographic indexes and the
actual journals that causes the problem--so, even if there
were funds to pay for the bibliographic searching, it would
still be a matter of obtaining the original articles to be
reviewed. There are, however, no funds for this work. Since
it really isn't an SRI project, but a scholarly one, I was
hoping I might be able to obtain a Visiting Scholar Card to
permit me convenient access to the library.
Any assistance you could provide in this matter would be
be greatly appreciated. The cc:'s on this message can be
contacted for "testimonials" as to my good, kind and
deserving nature.
Bob Amsler
859-3601
-------
∂10-Oct-83 1713 JRP@SRI-AI.ARPA This Thursday's approaches to natural language seminar
Received: from SRI-AI by SU-AI with TCP/SMTP; 10 Oct 83 17:13:43 PDT
Date: Mon 10 Oct 83 17:17:09-PDT
From: JRP@SRI-AI.ARPA
Subject: This Thursday's approaches to natural language seminar
To: csli-people@SRI-AI.ARPA
cc: peters@SRI-AI.ARPA
Situation Semantics
Barwise, Perry, Peters
-------
∂10-Oct-83 2242 @SU-SCORE.ARPA:HDG@SU-AI chess clocks
Received: from SU-SCORE by SU-AI with TCP/SMTP; 10 Oct 83 22:41:57 PDT
Received: from SU-AI.ARPA by SU-SCORE.ARPA with TCP; Mon 10 Oct 83 22:42:28-PDT
Date: 10 Oct 83 2220 PDT
From: David Goering <HDG@SU-AI>
Subject: chess clocks
To: "@CHESS.DIS[1,HDG]"@SU-AI
The new clocks have arrived, imported from Berkely. They aren't a matched
pair, but they take a licking and keep on ticking.
To avoid theftlike behavior, they will be kept in room 341; please return
them when finished.
So pay up. The total price was $80.83, divided by a total of fifteen
subscriptions -- $5.40 per full share, and $2.70 per half.
The response to our plea was gratifyingly large. I thank all of you.
Bughouse, anyone?
∂10-Oct-83 2353 @SRI-AI.ARPA:PULLUM%HP-HULK.HP-Labs@Rand-Relay Job available at HP Labs.
Received: from SRI-AI by SU-AI with TCP/SMTP; 10 Oct 83 23:53:42 PDT
Received: from rand-relay.ARPA by SRI-AI.ARPA with TCP; Mon 10 Oct 83 23:56:24-PDT
Date: 10 Oct 1983 1532-PDT
From: PULLUM.HP-HULK@Rand-Relay
Return-Path: <PULLUM%HP-HULK.HP-Labs@Rand-Relay>
Subject: Job available at HP Labs.
Received: by HP-VENUS via CHAOSNET; 10 Oct 1983 15:31:51-PDT
To: CSLI-FRIENDS@SRI-AI
Cc: pullum.HP-HULK@Rand-Relay, rosenberg.HP-HULK@Rand-Relay,
proudian.HP-HULK@Rand-Relay, goldstein.HP-HULK@Rand-Relay
Message-Id: <434673112.8357.hplabs@HP-VENUS>
Via: HP-Labs; 10 Oct 83 18:53-PDT
There is a full-time job opening in the Applications Technology Department of
Hewlett-Packard Laboratories in Palo Alto, for a person who works in areas
located at the interface between linguistics and computer science. The ideal
applicant would have
(1) an interest in the building of natural language processing systems;
(2) enough understanding of current syntactic and semantic theory to
grasp linguistic arguments and motivations readily;
(3) experience with programming in LISP;
(4) acquaintance with the theory and practice of parser design.
Exact details of qualifications and experience are of less importance than
interests and abilities that are represented in the above list. Anyone who
knows anyone who might be interested should tell them to send a resume to
Steven Rosenberg
Bldg 3U, HP Labs
1501 Page Mill Road
PALO ALTO
CA 94304
Net mail address: ROSENBERG@HP-LABS@RAND-RELAY.
[Please re-mail this to any promising locations on the net that you may
correspond with.]
-------
∂11-Oct-83 0905 JMC*
Mazda Pysch Today
∂11-Oct-83 0917 GROSZ@SRI-AI.ARPA use of CSLI mailing lists
Received: from SRI-AI by SU-AI with TCP/SMTP; 11 Oct 83 09:17:27 PDT
Date: Tue 11 Oct 83 07:48:08-PDT
From: GROSZ@SRI-AI.ARPA
Subject: use of CSLI mailing lists
To: csli-friends@SRI-AI.ARPA
Friends and folks:
The CSLI-FRIENDS and CSLI-FOLKS mailing list go out over the Arpanet
and should therefore not be used for advertising jobs r anything else)
in the private sector. The DOD frowns upon that sort of thing. So we
should police ourselves.
thanks
Barbara
-------
∂11-Oct-83 0944 100 (from: Doug Ferguson) Inforλ λλ$[D$[D¬¬λλλ$[D$[D$[D
This is a test to see if I can use AλSAIlλL vil tλλλ
This is a test to see if I can use SAIL via the 3081. Will be sending you
more infor mation on full text database developments Doug Ferguson
¬
!εl
?
?
∂11-Oct-83 1119 DFH office keys
RPG has asked that I get the three rooms that will have
LISP machines in them keyed to a master key. Betty would
prefer that we do the rooms for the whole section rather
than just those 3. Is this OK with you (including your
office)?
Changing the keys to one master including my office is ok.
∂11-Oct-83 1400 AAAI-OFFICE@SUMEX-AIM Confirmation of R. Fikes
Received: from SUMEX-AIM by SU-AI with PUP; 11-Oct-83 14:00 PDT
Date: Tue 11 Oct 83 14:02:01-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM>
Subject: Confirmation of R. Fikes
To: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BUCHANAN@SUMEX-AIM,
DAVIS@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA, LERMAN@SRI-KL.ARPA,
GENESERETH@SUMEX-AIM, GROSZ@SRI-AI.ARPA, HART@SRI-KL.ARPA, JMC@SAIL,
MINSKY@MIT-MC.ARPA, NILSSON@SRI-AI.ARPA, REDDY@CMU-CS-A.ARPA,
STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA, GJS@MIT-MC.ARPA,
TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA, BONNIE.UPENN@UDEL-RELAY.ARPA,
BLEDSOE@UTEXAS-20.ARPA, FIKES@PARC-MAXC.ARPA
cc: AAAI-OFFICE@SUMEX-AIM
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
It has been an unanimous approval of the nomination of Richard Fikes as the
the new Secretary/Treasurer of the AAAI. He will assume the responsibilities
of the Secretariat immediately.
I can assume we all wish him well in his new position and look forward
to working with him in the future.
Regards,
Claudia
-------
∂11-Oct-83 1438 PEREIRA@SRI-AI.ARPA Approaches to Computer Languages Seminar -- reminder
Received: from SRI-AI by SU-AI with TCP/SMTP; 11 Oct 83 14:37:52 PDT
Date: Tue 11 Oct 83 14:37:43-PDT
From: PEREIRA@SRI-AI.ARPA
Subject: Approaches to Computer Languages Seminar -- reminder
To: csli-friends@SRI-AI.ARPA
Thursday, October 13, 2:00 PM, Redwood Hall:
Terry Winograd will speak on "Specification Languages".
-------
∂11-Oct-83 1454 @SU-SCORE.ARPA:HDG@SU-AI chess clocks addendum
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Oct 83 14:53:57 PDT
Received: from SU-AI.ARPA by SU-SCORE.ARPA with TCP; Tue 11 Oct 83 14:55:07-PDT
Date: 11 Oct 83 1446 PDT
From: David Goering <HDG@SU-AI>
Subject: chess clocks addendum
To: "@CHESS.DIS[1,HDG]"@SU-AI
"And how shall we pay?" you may ask. Oh, any old way. By check (payable to
D. Goering) in my mailbox, or cash in my hand, or credit to my pony account (hdg),
or stamps, or whatever. Easy terms, receipts available on request.
David
∂11-Oct-83 1506 @SRI-AI.ARPA:SAG%HP-HULK.HP-Labs@Rand-Relay this week's colloquium
Received: from SRI-AI by SU-AI with TCP/SMTP; 11 Oct 83 15:06:19 PDT
Received: from rand-relay.ARPA by SRI-AI.ARPA with TCP; Tue 11 Oct 83 15:08:46-PDT
Date: 11 Oct 1983 1052-PDT
From: SAG.HP-HULK@Rand-Relay
Return-Path: <SAG%HP-HULK.HP-Labs@Rand-Relay>
Subject: this week's colloquium
Received: by HP-VENUS via CHAOSNET; 11 Oct 1983 10:52:31-PDT
To: csli-friends@SRI-AI
Message-Id: <434742752.903.hplabs@HP-VENUS>
Via: HP-Labs; 11 Oct 83 14:33-PDT
Slight Modification:
This week's [Oct. 13] colloquium will be:
"Theoretical" vs. "Computational" LInguistics: Three Perspectives
Lauri Karttunen Jerry Hobbs Geoffrey K. Pullum
Univ. of Texas SRI-CSLI UC Santa Cruz
Redwood Hall Classroom
4:15 PM
-------
∂11-Oct-83 2225 JMC*
coffee
∂11-Oct-83 2326 GOLUB@SU-SCORE.ARPA Industrial Lectureships
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Oct 83 23:26:25 PDT
Date: Tue 11 Oct 83 23:27:43-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: Industrial Lectureships
To: JMC@SU-AI.ARPA
cc: BSCOTT@SU-SCORE.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Fri 7 Oct 83 00:55:00-PDT
John, The Department is having a financial crisis. We have an estimated bill
of $60,000 for lecturers and the Dean allows us $10,000. We really must
control our finances much more carefully. We cannot afford the industrial
lectureships in our present situation. Is it possible some persons might
lecture gratus? I believe the rule of the University is that we cannot
offer a salary if one is being paid by one's institution. That is, no
double pay. At any rate, I must insist that you hold off until we get our
finances in better shape. It's a great idea! Let's see if we can find a means
for implementing it. GENE
-------
Are you talking about holding off for 84-85 or reneging on the lecturers
already recruited for this year whose courses are already announced in the
catalog? As for 84-85, we can hold off any attempt to recruit lecturers
until January. If we hold off longer, we should skip 84-85.
Bob Moore is already lecturing this Fall, Stan Rosenschein is scheduled
for this Winter and John Greenstadt for this Spring. I have made no
specific financial arrangement, but the announcement and oral communications
have said that we would pay what the Department usually pays lecturers.
∂11-Oct-83 2336 GOLUB@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Oct 83 23:36:09 PDT
Date: Tue 11 Oct 83 23:37:29-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 11 Oct 83 23:31:00-PDT
Obviously we have a commitment for this year and I shall try to honor it
but if worse comes to worse.....
GENE
-------
If worst comes to worst, you might try to get IBM to donate the services
of John Greenstadt, but I don't know what arrangements the individuals
have made with IBM and SRI respectively about whether they are docked
for the time off.
∂11-Oct-83 2347 GOLUB@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Oct 83 23:47:00 PDT
Date: Tue 11 Oct 83 23:48:20-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 11 Oct 83 23:39:00-PDT
I think we might get IBM to pay. GENE
-------
I leave IBM and SRI to you, since my commitment involved only finding
reasonable lecturers and getting reasonable course descriptions from
them and explicitly did not include settling financial arrangements.
∂11-Oct-83 2354 GOLUB@SU-SCORE.ARPA teaching schedule
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Oct 83 23:54:15 PDT
Date: Tue 11 Oct 83 23:55:28-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: teaching schedule
To: JMC@SU-AI.ARPA
cc: genesereth@SU-SCORE.ARPA, floyd@SU-SCORE.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 10 Oct 83 15:51:00-PDT
John, It seems to me that you ought to clear this with the curriculum
committee headed by Bob Floyd. Also if you teach this jointly with Genesereth
each of yu will have to teach another 1/2 course.
GENE
-------
There are 120 students in CS 223 as opposed to about 10 usually in
CS258. However, it seems to me better, in view of the complications, to give
up the idea. Therefore, I'll teach CS258 as originally scheduled.
∂12-Oct-83 0900 JMC*
cleaning and ultrasuede
∂12-Oct-83 1055 DFH Brazil ticket
This should be here this afternoon. There was one last
problem, in that the exchange rate had changed and an
additional charge was due. I put in on your American
Express --- $71. Renteria will give you this amount on
your arrival in Rio. I am putting flight info. on your calendar.
∂12-Oct-83 1200 @SRI-AI.ARPA:TW@SU-AI Talkware seminar schedule (and current abstract)
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Oct 83 11:57:05 PDT
Received: from SU-AI.ARPA by SRI-AI.ARPA with TCP; Wed 12 Oct 83 11:59:28-PDT
Date: 12 Oct 83 1152 PDT
From: Terry Winograd <TW@SU-AI>
Subject: Talkware seminar schedule (and current abstract)
To: "@377.DIS[1,TW]"@SU-AI
Talkware Seminar - CS 377
Date: October 12
Speaker: Harold Ossher
Topic: FABLE, a language for IC process-automation
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract:
The Stanford University Center for Integrated Systems is embarking on an
ambitious project to formally characterize integrated circuit fabrication
processes, and to provide a degree of automation of research and
prototyping activities in the IC fabrication facility. A crucial
component of this project is the ability to represent an IC fabrication
"recipe" in a repeatable, transportable, device-independent fashion. We
have designed the language Fable for this purpose; it offers some novel
approaches to abstraction and modularity.
Date: October 19
Speaker: Brian Reid (Stanford CS/EE)
Topic: Printing and formatting languages
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract: to come later
Date: October 26
Speaker: Greg Nelson (Xerox PARC)
Topic: JUNO: a constraint based language for graphics
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract: to come later
∂12-Oct-83 1211 GOLUB@SU-SCORE.ARPA Visiting Scholar
Received: from SU-SCORE by SU-AI with TCP/SMTP; 12 Oct 83 12:11:06 PDT
Date: Wed 12 Oct 83 12:12:50-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: Visiting Scholar
To: jmc@SU-SCORE.ARPA
Do you think it is a good idea we give Amsler a visiting Scholar card?
GENE
---------------
Return-Path: <AMSLER@SRI-AI.ARPA>
Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Mon 10 Oct 83 12:57:55-PDT
Date: Mon 10 Oct 83 12:50:57-PDT
From: Robert Amsler <AMSLER@SRI-AI.ARPA>
Subject: Access to Stanford Libraries
To: golub@SU-SCORE.ARPA
cc: sleeman@SUMEX-AIM.ARPA, misra%SU-Sierra@SU-SCORE.ARPA, jmc@SU-AI.ARPA,
walker@SRI-AI.ARPA
I'm Bob Amsler, a computational lexicologist working with
Don Walker at SRI in a new program within the Advanced
Computer Systems Department, called "Natural-Language and
Knowledge-Resource Systems". I have a problem that I would
greatly appreciate help with if possible.
Last year, when I was in the AI Center, I accepted the task
of writing a chapter for the upcoming 1984 issue of "Annual
Review of Information Science and Technology" on
Machine-Readable Dictionaries. These chapters are
accompanied by definitive bibliographic descriptions of the
literature in their areas, with 100-300 citations being
typical. Now to the problem. Last year, I had, through the
AI Center, a Visiting Scholar Card which permitted me entry
to the Green Library and which I found indispensible for
carrying out research on this project. However, access ended
in August and now I am afraid I won't be able to adequately
complete the chapter's background literature search. It is
both lack of access to the bibliographic indexes and the
actual journals that causes the problem--so, even if there
were funds to pay for the bibliographic searching, it would
still be a matter of obtaining the original articles to be
reviewed. There are, however, no funds for this work. Since
it really isn't an SRI project, but a scholarly one, I was
hoping I might be able to obtain a Visiting Scholar Card to
permit me convenient access to the library.
Any assistance you could provide in this matter would be
be greatly appreciated. The cc:'s on this message can be
contacted for "testimonials" as to my good, kind and
deserving nature.
Bob Amsler
859-3601
-------
-------
∂12-Oct-83 1212 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 13
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Oct 83 12:12:33 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Wed 12 Oct 83 12:15:05-PDT
Date: Wed, 12 Oct 83 12:11 PDT
From: desRivieres.PA@PARC-MAXC.ARPA
Subject: CSLI Activities for Thursday Oct. 13
To: csli-friends@SRI-AI.ARPA
Reply-to: desRivieres.PA@PARC-MAXC.ARPA
CSLI SCHEDULE FOR THURSDAY, OCTOBER 13, 1983
10:00 Research Seminar on Natural Language
Speakers: Barwise, Perry, Peters (CSLI)
Title: "Situation Semantics"
Place: Redwood Hall, room G-19
12:00 TINLunch
Discussion leader: Kurt Konolige
Paper for discussion: "On Inheritance Hierarchies with Exceptions"
by David W. Etherington and Raymond Reiter,
AAAI '83, pp. 104-108.
Place: Ventura Hall
2:00 Research Seminar on Computer Languages
Speaker: Terry Winograd (CSLI)
Title: "Specification Languages"
Place: Redwood Hall, room G-19
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speakers: Lauri Karttunen (U. Texas), Jerry Hobbs (SRI-CSLI),
Geoffrey K. Pullum (UC Santa Cruz)
Title: ""Theoretical" vs. "Computational" Linguistics: Three Perspectives"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It
can be reached from Campus Drive or Panama Street. From Campus Drive
follow the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
-------
∂12-Oct-83 1227 GOLUB@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 12 Oct 83 12:25:29 PDT
Date: Wed 12 Oct 83 12:27:09-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 12 Oct 83 00:25:00-PDT
Basically I am in favor of the Industrial Lectureships. My only qualm
these days is with the finances of the department. We are being systematically
under-budgeted by the Dean and hence we cannot have as rich a program as we
think we should. I think it is likely the lecturers will be hired.
GENE
-------
∂12-Oct-83 1401 YOM Manuals
To: JMC, JJW
I originally counted about 35 students who didn't have the manual
when we checked in class. The next day, I counted 40 manuals on the
bookstore shelf. I don't expect any problems, although obviously
there are some students that didn't rush to buy the manuals as I
suggested. As the manuals are more complete than the summaries, the
students should be encouraged to purchase those.
-Yoram.
One student told me he was unable to find a manual. Please check tomorrow.
∂12-Oct-83 1523 JJW EKL at LOTS
To: JMC, JK, YOM
EKL is now running at LOTSA. EKL.EXE is on the directory PS:<EKL>, so you
can either say
@<ekl>ekl
or put something like
@define sys: sys:,ps:<ekl>
in your LOGIN.CMD file and then just type
@ekl
(I haven't tried that, but it should work.) The Lisp axioms are on
<EKL>LISPAX.LSP and in internal form as <EKL>LISPAX.PRF.
∂12-Oct-83 1539 MRC@SU-SCORE.ARPA primes
Received: from SU-SCORE by SU-AI with TCP/SMTP; 12 Oct 83 15:39:31 PDT
Date: Wed 12 Oct 83 15:41:12-PDT
From: Mark Crispin <MRC@SU-SCORE.ARPA>
Subject: primes
To: JMC@SU-AI.ARPA
Postal-Address: 725 Mariposa Ave. #103; Mountain View, CA 94041
Phone: (415) 497-1407 (Stanford); (415) 968-1052 (residence)
Sigh. All I'm trying to do is get the next several primes after 1201.
I have to increase a monitor table which must be prime; I think a prime
around 2000 would be nice... I didn't want to have to write a sieve
program but guess I must.
-------
How about 1999 which is prime?
∂12-Oct-83 1546 MRC@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 12 Oct 83 15:46:42 PDT
Date: Wed 12 Oct 83 15:48:28-PDT
From: Mark Crispin <MRC@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 12 Oct 83 15:44:00-PDT
Postal-Address: 725 Mariposa Ave. #103; Mountain View, CA 94041
Phone: (415) 497-1407 (Stanford); (415) 968-1052 (residence)
That's fine! I was just typing in the canonical prime generator in Basic.
I only have memorized the primes up to about 100 or so. 1999 will do
quite nicely.
-------
I just guessed and divided by the primes to 43 using pocket calculator.
I thought that would be faster.
(do
((n 3.0 (plus 2. n))
(l nil (cons (quotient 1999. n) l)))
((greaterp n 45.) l))
∂12-Oct-83 1614 BSCOTT@SU-SCORE.ARPA New ARPA Contract
Received: from SU-SCORE by SU-AI with TCP/SMTP; 12 Oct 83 16:14:08 PDT
Date: Wed 12 Oct 83 16:15:23-PDT
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: New ARPA Contract
To: JMC@SU-AI.ARPA, DCL@SU-AI.ARPA, ZM@SU-AI.ARPA, TOB@SU-AI.ARPA,
Wiederhold@SUMEX-AIM.ARPA
cc: DFH@SU-AI.ARPA, MAS@SU-AI.ARPA, Atkinson@SU-SCORE.ARPA, Brock@SU-AI.ARPA,
BScott@SU-SCORE.ARPA
There will be a two-month delay in the funding of the new ARPA contract;
Ron Ohlander tells me that the planned effective date is 12/1, and that
we must make the funds we have tide us over until that time. There are funds
remaining in the current contract allocations for both Binford and Luckham.
These funds will be redistributed in order to cover salary payments for
the two months for all Principal Investigators.
John Machado is supposed to send you a message explaining what is being
done, and why, and assuring both Tom and David that they will not lose
their equipment money. The equipment purchases will have to be delayed
until we have the new contract.
I will pass further information along to you as I receive it. I am now
identifying salary needs for all of you for the two months.
Betty
-------
∂12-Oct-83 1619 PEREIRA@SRI-AI.ARPA Approaches to Comp. Lang., Oct 27
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Oct 83 16:19:43 PDT
Date: Wed 12 Oct 83 16:20:33-PDT
From: PEREIRA@SRI-AI.ARPA
Subject: Approaches to Comp. Lang., Oct 27
To: csli-friends@SRI-AI.ARPA, dkanerva@SRI-AI.ARPA
CSLI
Approaches to Computer Languages Seminar
Thursday, October 27, 2:00 PM, Redwood Hall
Peter Deutsch
Xerox PARC
SMALLTALK-80: LANGUAGE AND STYLE IN A
REAL OBJECT-ORIENTED PROGRAMMING SYSTEM
Smalltalk-80 is an object-oriented programming language and system that
has been in daily use by several dozen people for several years. This
talk will cover the more interesting aspects of the language, and some
equally interesting observations on how the language features are
actually used to create program structures. I will mention some other
object-oriented programming languages (Actors, LOOPS, Concurrent
Prolog), but not discuss them in depth.
-------
∂12-Oct-83 1702 KJB@SRI-AI.ARPA Executive committee meetings
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Oct 83 17:02:01 PDT
Date: Wed 12 Oct 83 17:00:21-PDT
From: KJB@SRI-AI.ARPA
Subject: Executive committee meetings
To: csli-folks@SRI-AI.ARPA
If you ever have anything that you feel needs to be changed or
thought about, please feel free to tell me or anyone else on
the executive committee, or Betsy.
-------
∂12-Oct-83 1759 DKANERVA@SRI-AI.ARPA October 13 Newsletter (No. 4)
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Oct 83 17:58:58 PDT
Date: Wed 12 Oct 83 17:56:37-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: October 13 Newsletter (No. 4)
To: csli-friends@SRI-AI.ARPA
CSLI Newsletter
October 13, 1983 * * * Number 4
THE CHANGING FACE OF LOGIC
CSLI is not the only thing to come out of the interaction with
computer science. The vernerable "Annals of Mathematical Logic,"
published by North-Holland, has now changed its name to "Annals of
Pure and Applied Logic" and "will accept papers in the field of
computer science which have substantial logical content, or provide
interesting applications." Surely some of the work of CSLI will fall
in that class.
* * * * * * *
CONSULTANT IN COMPUTER LANGUAGE SEMANTICS
Professor Yiannis Moschovakis, who is one of the world's experts
in recursion theory and who has been working on the semantics of
computer languages for the past couple of years, has agreed to serve
as a consultant this quarter. He will be coming up from UCLA most
Tuesdays in conjunction with the Working Group on Computer Language
Semantics, which meets at 9:30 a.m. Tuesdays at PARC. Next week's
meeting, by way of reminder, will feature Henson Graves discussing the
relation between Category Theory and the Programming Languages.
* * * * * * *
COMPUTING COMMITTEE
As of this week, Raymond Perrault of SRI International is the new
chairman of the CSLI Computing Committee, replacing Brian Smith of
Xerox PARC who continues to serve on the committee.
* * * * * * *
CSLI READING ROOM
Just a reminder that Room 6, downstairs in Ventura Hall, is one
of CSLI's rooms and is available all the time for conferences,
discussions, reading, and other similar activities. Comfortable
furniture is being ordered, but it will be a while before it arrives.
Some newspapers have also been ordered. It is a good place to leave
technical reports, old journals, and the like that you want to share
but do not need returned.
* * * * * * *
! * * * * * * *
CSLI SCHEDULE FOR THURSDAY, OCTOBER 13, 1983
10:00 Research Seminar on Natural Language
Speakers: Jon Barwise, John Perry, Stanley Peters (CSLI)
Title: "Situation Semantics"
Place: Redwood Hall, room G-19
12:00 TINLunch
Discussion leader: Kurt Konolige
Paper for discussion: "On Inheritance Hierarchies with Exceptions"
by David W. Etherington and Raymond Reiter,
AAAI '83, pp. 104-108.
Place: Ventura Hall
2:00 Research Seminar on Computer Languages
Speaker: Terry Winograd (CSLI)
Title: "Specification Languages"
Place: Redwood Hall, room G-19
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speakers: Lauri Karttunen (U. Texas), Jerry Hobbs (SRI-CSLI),
Geoffrey K. Pullum (UC Santa Cruz)
Title: "`Theoretical' vs. `Computational' Linguistics:
Three Perspectives"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It
can be reached from Campus Drive or Panama Street. From Campus Drive
follow the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
* * * * * * *
! * * * * * * *
TINLUNCH SCHEDULE
TINLunch will be held at 12:00 on Thursday, October 13, 1983, at
Ventura Hall, Stanford University. Kurt Konolige will lead the
discussion. The paper for discussion will be
ON INHERITANCE HIERARCHIES WITH EXCEPTIONS
by
David W. Etherington and Raymond Reiter
TINLunch will be held each Thursday at Ventura Hall on the
Stanford University campus as a part of CSLI activities. Copies of
TINLunch papers will be at SRI in EJ251 and at Stanford University in
Ventura Hall.
NEXT WEEK: BELIEF-SENTENCES AND THE LIMITS OF SEMANTICS
Barbara Hall Partee
October 13: Kurt Konolige
October 20: Per-Christian Halvorsen
October 27: Michael Georgeff
November 3: Ron Kaplan
November 10: Martin Kay
* * * * * * *
APPROACHES TO COMPUTER LANGUAGES SEMINAR, FALL 1983
October 13: T. Winograd Stanford Specification Languages
October 20: H. Levesque Fairchild Knowledge Representation
October 27: P. Deutsch XEROX Object-Oriented Languages
November 3: C. Talcott Stanford Logic of Programs
November 10: G. Winskel CMU Semantics of Programming Languages
November 17: to be announced
December 1: to be announced
December 8: J. Goguen SRI Data Abstraction
* * * * * * *
! * * * * * * *
TALKWARE SEMINARS - CS 377
Date: October 12
Speaker: Harold Ossher
Topic: FABLE, a language for IC process-automation
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract:
The Stanford University Center for Integrated Systems is
embarking on an ambitious project to formally characterize integrated
circuit fabrication processes, and to provide a degree of automation
of research and prototyping activities in the IC fabrication facility.
A crucial component of this project is the ability to represent an IC
fabrication "recipe" in a repeatable, transportable,
device-independent fashion. We have designed the language Fable for
this purpose; it offers some novel approaches to abstraction and
modularity.
Date: October 19
Speaker: Brian Reid (Stanford CS/EE)
Topic: Printing and formatting languages
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract: to come later
Date: October 26
Speaker: Greg Nelson (Xerox PARC)
Topic: JUNO: a constraint based language for graphics
Time: 2:15 - 4
Place: 380D (Math corner)
Abstract: to come later
* * * * * * *
-------
∂13-Oct-83 1049 DELAGI%SUMEX-AIM.ARPA@SU-SCORE.ARPA SuperLunch - PreSeminar Meeting
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 10:49:01 PDT
Received: from SUMEX-AIM by Score with Pup; Thu 13 Oct 83 10:47:07-PDT
Date: Thu 13 Oct 83 10:48:16-PDT
From: Bruce Delagi <DELAGI@SUMEX-AIM>
Subject: SuperLunch - PreSeminar Meeting
To: super@Score
cc: decwrl!rhea::vaxwrk::sclara@Shasta
The November 3 CS440 seminar will concern itself with some current practice in
parallel computation for scientific applications (at NASA Ames). The speaker,
Tony Fleig, will talk on the work he did in formulating the problems for
execution in a shared memory environment complemented with a high speed local
area net. He will outline the alternatives considered (from a software
engineering perspective) in the degree of application customization employed.
Tony has agreed to meet with a smaller group (the old "super" group?) to get
into a bit more of exactly what the performance and effort tradeoffs were/are,
how much of the supplied VMS stuff was useful in the parallel computation
environment he set up, and what the performance tradeoffs were (as he saw them).
It seems useful to me for at least some of us to understand this in some detail.
There is also the possibility of supplementing the lunch and seminar with one
or two one-on-one meetings with Tony.
Please let me know of your interest so I can set things up.
/bruce
-------
∂13-Oct-83 1227 oliger%SU-NAVAJO.ARPA@SU-SCORE.ARPA Re: SuperLunch - PreSeminar Meeting
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 11:38:37 PDT
Received: from Navajo by Score with Pup; Thu 13 Oct 83 11:33:32-PDT
Date: Thu, 13 Oct 83 11:32:16 PDT
From: Joseph Oliger <oliger@Navajo>
Subject: Re: SuperLunch - PreSeminar Meeting
To: Bruce Delagi <DELAGI@SUMEX-AIM>
Cc: super@Score, decwrl!rhea::vaxwrk::sclara@Shasta
In-Reply-To: Your message of Thu 13 Oct 83 10:48:16-PDT.
Sounds interesting. How about Thursday noon?
Joe
∂13-Oct-83 1229 ULLMAN@SU-SCORE.ARPA future talks
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 12:28:03 PDT
Date: Thu 13 Oct 83 12:28:10-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: future talks
To: CS440: ;
The following is the list of talks for the balance of the quarter,
with a few words about the topic in some cases.
This is the schedule of speakers for the CS440
Supercomputer Seminar.
It meets at 4:15 Thursdays in room 380-X.
Oct. 13: Vason Scrini (Berkeley) "A message based processor for data-flow systems"
Oct. 20: Dick Gabriel "Queue-based multiprocessing LISP"
Oct. 27: Alain Hanover (DEC)
Nov. 3: Tony Fleig (NASA) "A shared memory/local area net environment
for parallel computation"
Nov. 10: Steve Lundstrum
Nov. 17: Al Davis (Fairchild) "Thoughts on Symbolic Supercomputers"
Dec. 1: Jay Misra (U. Texas) "Simulation and multiprocessors"
Dec. 8: Bob Keller (Utah/Livermore)
-------
∂13-Oct-83 1433 DFH
To: "@FACCO2.[1,DFH]"@SU-AI
Facilities & Space Committee Meeting
There will be a joint meeting of the Facilities Committee and the Space
Committee with Len Bosack Tue, Oct. 18, 4 pm room 252. Please let
Marlie Yearwood (Yearwood@score) know if you can attend. Thank you.
∂13-Oct-83 1458 GOLUB@SU-SCORE.ARPA appointments
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 14:58:30 PDT
Date: Thu 13 Oct 83 14:59:53-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: appointments
To: full-professors@SU-SCORE.ARPA
I had a discussion with Gordon Bower about the direction of the department
with respect to appointments. He says the department will get two new
appointments in the next five years. We have already voted for one, Leo
Guibas and several of you are enthusistic about the appointment of
Nilsson as chairman. But if Nilsson comes in as chairman, he will undoubtedly
want some freedom in making appointments. (One new appointment is not very
many!) Bower thinks we should think of appointing Nillson first and then see
what happens.
These are pretty important issues that we cannot settle by electronic mail.
Therefore I am calling a meeting this Tuesday ---Oct 18 at 2:30 in my
Conference room. I hope you can come. GENE
-------
∂13-Oct-83 1617 CASLEY@SU-SCORE.ARPA Research Assistantship
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 16:17:18 PDT
Date: Thu 13 Oct 83 16:18:59-PDT
From: Ross T. Casley <CASLEY@SU-SCORE.ARPA>
Subject: Research Assistantship
To: jmc@SU-AI.ARPA
Prof. McCarthy,
I spoke with you about getting support from you, but I have
decided to work with Richard Weyrauch. Thank you again.
Ross Casley.
-------
∂13-Oct-83 1644 ROODE@SRI-NIC MCI Mail and disk storage
Received: from SRI-NIC by SU-AI with TCP/SMTP; 13 Oct 83 16:44:06 PDT
Date: Thu 13 Oct 83 16:46:29-PDT
From: David Roode <ROODE@SRI-NIC>
Subject: MCI Mail and disk storage
To: JMC@SU-AI
Location: EJ286 Phone: (415) 859-2774
Have you seen the announcements for MCI mail? They are charging a flat
$1 per message fee for online delivery or $2 for next day U.S. mail delivery,
with other options. They are accepting account requests in the
LOTS style at 800-323-7751 with 110-1200 baud modems. You log on with
username REGISTER and password REGISTER, and then it leads you by the hand.
The suggest a username with your first initial and your last name, but
I wonder what they expect to do with a whole country full of users? It should
be interesting.
I missed a meeting with you AAAI folks that I was not given much
notice of. Realizing that there are many other variables, and even
thinking of the need as unlikely, I still think you will be interested
to know that DEC currently offers a 500 megabyte "mini module" style
drive capability for the DEC-20 at an incremental cost of $21,000
each. Three fit in one cabinet, i.e. unit of floor space. This
requires their CI bus and HSC-50 controller, which has a start up cost
of $50,000. Something which is not clear to me but which Ralph might
know is whether you could buy brand X disk drives and connect them to
your HSC-50, i.e. whether it is an industry standard interface.
-------
∂13-Oct-83 1806 pratt%SU-NAVAJO.ARPA@SU-SCORE.ARPA appointments
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Oct 83 18:06:06 PDT
Received: from Navajo by Score with Pup; Thu 13 Oct 83 18:07:42-PDT
Date: Thu, 13 Oct 83 18:07 PDT
From: Vaughan Pratt <pratt@Navajo>
Subject: appointments
To: full-professors@score
I would approach this two-professors-in-five-years limit by hiring
two excellent professors in the first year and then making sure we are
in the same state the next year, namely a limit of two professors for
the next five years.
-v
∂14-Oct-83 0948 GOLUB@SU-SCORE.ARPA Meeting on Tuesday
Received: from SU-SCORE by SU-AI with TCP/SMTP; 14 Oct 83 09:47:52 PDT
Date: Fri 14 Oct 83 09:49:18-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: Meeting on Tuesday
To: full-professors@SU-SCORE.ARPA
It seems many people are out of town on Tuesday and there are conflicting
events. Therefore I am cancelling the meeting on Tuesday.
Remember we have a Senior Faculty meeting on Tuesday Nov 1 at 2:30.
GENE
-------
∂14-Oct-83 1123 KJB@SRI-AI.ARPA Seminar for next quarter
Received: from SRI-AI by SU-AI with TCP/SMTP; 14 Oct 83 11:22:59 PDT
Date: Fri 14 Oct 83 11:22:09-PDT
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Seminar for next quarter
To: briansmith@PARC-MAXC.ARPA, tw@SU-AI.ARPA, jmc@SU-AI.ARPA
cc: bmacken@SRI-AI.ARPA
Brian, Your lisp as language course will need a room. Will
the redwood room work? It has to be reserved, and will have
low priority unless it is an official dept course. I wonder if
it could be listed as a c.s. course. Jon
-------
∂14-Oct-83 1319 DFH schedule conflict
It looks to me like the CIS panel on Nov. 9 will conflict with
your 11 am appt.
∂15-Oct-83 1533 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS
Received: from SRI-AI by SU-AI with TCP/SMTP; 15 Oct 83 15:32:57 PDT
Received: from SU-AI.ARPA by SRI-AI.ARPA with TCP; Sat 15 Oct 83 15:35:06-PDT
Date: 15 Oct 83 1524 PDT
From: Carolyn Talcott <CLT@SU-AI>
Subject: SEMINAR IN LOGIC AND FOUNDATIONS
To: "@TMP.DIS[1,CLT]"@SU-AI
SPEAKER: Prof. J. E. Fenstad, University of Oslo
TITLE: Hyperfinite probability theory; basic ideas and applications
in natural sciences
TIME: Wednesday, Oct. 19, 4:15-5:30 PM
PLACE: Stanford Mathematics Dept. Faculty Lounge, 383N
The talk will assume some acquaintance with non-standard analysis
(existence of the extensions, transfer). But the ideas of hyperfinite
probability theory (e.g. Loeb construction) will be explained before turning
to applications, which will mainly be to hyperfinite spin systems
(statistical mechanics, polymer models, field theory). The models
will be fully explained, so no knowledge of "advanced" physics is presupposed.
S. Feferman
[ps - I have had some trouble with the mailer, so if you get
two copies of this that may be why - CLT]
∂15-Oct-83 1919 BrianSmith.pa@PARC-MAXC.ARPA Lisp as Language
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 15 Oct 83 19:19:12 PDT
Date: 15 Oct 83 19:19 PDT
From: BrianSmith.pa@PARC-MAXC.ARPA
Subject: Lisp as Language
To: CSLI-Executives@SRI-AI.ARPA, Winograd.pa@PARC-MAXC.ARPA,
JMC@SU-AI.ARPA, Pereira@SRI-AI.ARPA, RPerrault@SRI-AI.ARPA,
desRivieres.pa@PARC-MAXC.ARPA, Levesque@SRI-KL.ARPA
cc: BrianSmith.pa@PARC-MAXC.ARPA
I am appending a first draft description of the "Lisp as Language"
course, to be given next quarter. I want to put a copy in the CSLI
newsletter, so that we can begin to get a sense of how many people are
going to take it, figure out how to provide enough machines, draft
people to help TA it (particularly to help with the programming
tutorials), etc. My current request to you all is to let me know if it
seems a reasonable overview -- reasonable content, appealing to the
right audience, etc.
One worry I have is that I am used to semesters, not quarters; I will
try to develop a more detailed outline, to see what fraction of this I
can reasonably expect to fit into 10 weeks.
Many thanks.
Brian
-----------------------------
LISP as Language
An unassuming but rigorous introduction to the practices and concepts of
programming, based on a simple reconstructed dialect of LISP. The aim
is to present (to linguists, philosophers, mathematicians, and anyone
else who is interested) an explicit introduction to the knowledge that
programmers typically acquire tacitly, after months of late-night
debugging. The course will involve a combination of conceptual analysis
and hands-on programming. We will try to formulate the analysis using
vocabulary that would be useful in studying any linguistic system.
No previous exposure to computing is required (although familiarity with
some formal systems will help). Participants will be provided with
computer access and programming instruction. There will probably be one
two-hour class presentation and one three-hour tutorial section per
week.
Topics to be covered will include:
-- Procedural and data abstraction;
-- Objects, modularity, state, and encapsulation;
-- Meta-linguistic abstraction, and problems of intensional grain;
-- Contexts, scoping mechanisms, and locality;
-- Procedural and declarative notions of semantics;
-- Architecture, implementation, and abstract machines;
-- Self-reference, meta-circular interpreters, and reflection;
-- Communication protocols, and models of input/output;
-- Implicit and explicit information -- models of representations;
-- Compilation, interpretation, and execution -- models of
processing;
The course will be based in part on the "Structure and Interpretation of
Computer Programs" textbook, by Abelson and Sussman, that has been used
at M.I.T., although we will present much of that material under a
linguistic reconstruction.
-----------------------------
∂15-Oct-83 1934 BrianSmith.pa@PARC-MAXC.ARPA [BrianSmith.pa: Lisp as Language]
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 15 Oct 83 19:32:38 PDT
Date: 15 Oct 83 19:33 PDT
From: BrianSmith.pa@PARC-MAXC.ARPA
Subject: [BrianSmith.pa: Lisp as Language]
To: CSLI-Executives@SRI-AI.ARPA, Winograd.pa@PARC-MAXC.ARPA,
JMC@SU-AI.ARPA, Pereira@SRI-AI.ARPA, RPerrault@SRI-AI.ARPA,
desRivieres.pa@PARC-MAXC.ARPA, Levesque@SRI-KL.ARPA
cc: BrianSmith.pa@PARC-MAXC.ARPA
Second try (some mailer on some system took a big chunk out of this
message the first time I sent it).
----- Fowarded Messages -----
Date: 15 Oct 83 19:19 PDT
From: BrianSmith.pa
Subject: Lisp as Language
To: CSLI-Executives@SRI-AI, Winograd, JMC@SU-AI, Pereira@SRI-AI,
RPerrault@SRI-AI, desRivieres, Levesque@SRI-KL
cc: BrianSmith.pa
I am appending a first draft description of the "Lisp as Language"
course, to be given next quarter. I want to put a copy in the CSLI
newsletter, so that we can begin to get a sense of how many people are
going to take it, figure out how to provide enough machines, draft
people to help TA it (particularly to help with the programming
tutorials), etc. My current request to you all is to let me know if it
seems a reasonable overview -- reasonable content, appealing to the
right audience, etc.
One worry I have is that I am used to semesters, not quarters; I will
try to develop a more detailed outline, to see what fraction of this I
can reasonably expect to fit into 10 weeks.
Many thanks.
Brian
-----------------------------
LISP as Language
An unassuming but rigorous introduction to the practices and concepts of
programming, based on a simple reconstructed dialect of LISP. The aim
is to present (to linguists, philosophers, mathematicians, and anyone
else who is interested) an explicit introduction to the knowledge that
programmers typically acquire tacitly, after months of late-night
debugging. The course will involve a combination of conceptual analysis
and hands-on programming. We will try to formulate the analysis using
vocabulary that would be useful in studying any linguistic system.
No previous exposure to computing is required (although familiarity with
some formal systems will help). Participants will be provided with
computer access and programming instruction. There will probably be one
two-hour class presentation and one three-hour tutorial section per
week.
Topics to be covered will include:
-- Procedural and data abstraction;
-- Objects, modularity, state, and encapsulation;
-- Meta-linguistic abstraction, and problems of intensional grain;
-- Contexts, scoping mechanisms, and locality;
-- Procedural and declarative notions of semantics;
-- Architecture, implementation, and abstract machines;
-- Self-reference, meta-circular interpreters, and reflection;
-- Communication protocols, and models of input/output;
-- Implicit and explicit information -- models of representations;
-- Compilation, interpretation, and execution -- models of
processing;
The course will be based in part on the "Structure and Interpretation of
Computer Programs" textbook, by Abelson and Sussman, that has been used
at M.I.T., although we will present much of that material under a
linguistic reconstruction.
-----------------------------
----- End of Forwarded Messages -----
∂17-Oct-83 0937 ALMOG@SRI-AI.ARPA reminder on why-context-wont-go-away
Received: from SRI-AI by SU-AI with TCP/SMTP; 17 Oct 83 09:37:22 PDT
Date: 17 Oct 1983 0937-PDT
From: Almog at SRI-AI
Subject: reminder on why-context-wont-go-away
To: csli-friends at SRI-AI
cc: almog, kjb
Tomorrow, Tuesday 10.18.83 at 3.15 in Ventura Hall, we have our
third meeting on WHY CONTEXT WONT GO AWAY.
The speaker is B.Grosz from SRI. Attached is an abstract of her talk.
OVERVIEW OF AI RESEARCH ON CONTEXT
The context-dependence of language use has been a concern of research
on (natural) language processing in Artificial Intelligence almost
from its beginning. For a system to carry on an extended dialogue with
a user, (or to interpret a text), it must be able to use information
about the context in which utterances occur both for interpreting what
is said and for generating appropriate responses. As a result, AI
researchers have been forced to provide some treatment of many kinds
of linguistic expressions that require taking context into account
(e.g., definite descriptions, pronouns).
Like "pragmatics", "context" is a term that includes many different
phenomena. Within AI, at least three different kinds of contextual
information have been given some treatment: (1) the context
provided by the discourse itself (what has been said, and in
some cases how it has been said); (2) the context provided by the
(shared) common-sense knowledge of the discourse participants; (3) the
context of the ongoing activity and/or intentions of the discourse
participants.
We will provide an overview of research in AI, pointing out flaws
in some of the earlier attempts (e.g., editors' rules for pronoun
use aren't always followed), describing initial formulations of
computational theories, and discussing problems currently being
investigated.
-------
-------
∂17-Oct-83 1113 HANS@SRI-AI.ARPA Mailing addresses for the individual research projects
Received: from SRI-AI by SU-AI with TCP/SMTP; 17 Oct 83 11:13:17 PDT
Date: Mon 17 Oct 83 11:13:51-PDT
From: Hans Uszkoreit <Hans@SRI-AI.ARPA>
Subject: Mailing addresses for the individual research projects
To: csli-folks@SRI-AI.ARPA
cc: CSLI-REQUESTS@SRI-AI.ARPA
There exist now 16 mail addresses for the individual project groups
A1,...,D4. The addresses are CSLI-XN@SRI-AI, where XN stands for the
group name. These addresses take files in our directory <CSLI> as
distribution list. The names of the files are <CSLI>CSLI.-XN.
So far, each distribution list contains only the name of the project
leader. So, if you send a letter to CSLI-A1@SRI-AI, the mailer will
use the mailing list <CSLI>CSLI.-A1 for distributing. It will find there
the address Kaplan@parc and forward the message to Ron.
If you want to use these addresses to send mail to all members of your
group, then you have to add their addresses to the corresponding
distribution list. The addresses have to be separated by commas. (Look
at <CSLI>CSLI-FOLKS for the right format of list and addresses.) In this
case you also have to update your mailing list. If no group member has
an account on the SRI-AI machine, please ask a friend or a secretary for
help.
Please report problems to CSLI-REQUESTS@SRI-AI.
-------
∂17-Oct-83 1501 @SU-SCORE.ARPA:slndstrm@Shasta Visit Schedule - Burton Smith, Denelcor
Received: from SU-SCORE by SU-AI with TCP/SMTP; 17 Oct 83 15:01:21 PDT
Received: from Shasta by SU-SCORE.ARPA with TCP; Mon 17 Oct 83 14:57:46-PDT
Date: Monday, 17 Oct 1983 14:54-PDT
To: super at Score <super@Score>, tajnai at Score <tajnai@Score>
Cc: slndstrm at Shasta <slndstrm@Shasta>
Subject: Visit Schedule - Burton Smith, Denelcor
From: Steve Lundstrom <slndstrm@Shasta>
Burton Smith DENELCOR
Visit to Stanford University
Tentative Agenda
Wednesday, October 19, 1983
3:00pm Attend CLASSIC meeting (Lundstrom speaker)
4:15pm Present EE380 Seminar (Skilling Aud)
Thursday, October 20, 1983
9:00am Lundstrom
10:00am Trattnig
11:00am Tour of Facilities
12:00noon SUPER LUNCH AEL102
Informal discussions with Burton Smith
concerning future of supercomputing, the
recent Frontiers of Supercomputing meetings,
etc. (Bring your own lunch)
2:00pm Not Scheduled (Send mail to Lundstrom if you want
to be scheduled)
4:15pm Attend CS440 Seminar (Gabriel speaker)
Friday, October 21, 1983
8:45am Tajnai (FORUM)
9:30am Gabriel, et al (AEL102)
Informal Discussions re LISP on multiprocessors
such as the Denelcor HEP.
11:00am Not Scheduled Yet (Send mail to Lundstrom to get scheduled)
12:00noon Lunch
∂17-Oct-83 1506 @SU-SCORE.ARPA:RPG@SU-AI Queue-based Multiprocessing Lisp
Received: from SU-SCORE by SU-AI with TCP/SMTP; 17 Oct 83 15:05:49 PDT
Received: from SU-AI.ARPA by SU-SCORE.ARPA with TCP; Mon 17 Oct 83 15:00:19-PDT
Date: 17 Oct 83 1457 PDT
From: Dick Gabriel <RPG@SU-AI>
Subject: Queue-based Multiprocessing Lisp
To: super@SU-SCORE
If you would like to be decently-prepared for my talk, you ought to be
familiar with Lisp, even at a high level, and you ought to know a little
bit about closures. The best way to do this is to read ``Lambda: the
Ultimate Imperative'' by Steele and Sussman, an MIT AI Memo. I have a copy
in my office (360 MJH) that can be copied.
-rpg-
∂17-Oct-83 1510 ULLMAN@SU-SCORE.ARPA [Dick Gabriel <RPG@SU-AI>: Queue-based Multiprocessing Lisp ]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 17 Oct 83 15:10:20 PDT
Date: Mon 17 Oct 83 15:02:11-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: [Dick Gabriel <RPG@SU-AI>: Queue-based Multiprocessing Lisp ]
To: CS440: ;
The following message may be of use to CS440 people who are planning
to attend this Thursday's Seminar.
Remember that we meet in 380-Y, not "X".
---------------
Return-Path: <RPG@SU-AI>
Received: from SU-AI.ARPA by SU-SCORE.ARPA with TCP; Mon 17 Oct 83 15:00:19-PDT
Date: 17 Oct 83 1457 PDT
From: Dick Gabriel <RPG@SU-AI>
Subject: Queue-based Multiprocessing Lisp
To: super@SU-SCORE
If you would like to be decently-prepared for my talk, you ought to be
familiar with Lisp, even at a high level, and you ought to know a little
bit about closures. The best way to do this is to read ``Lambda: the
Ultimate Imperative'' by Steele and Sussman, an MIT AI Memo. I have a copy
in my office (360 MJH) that can be copied.
-rpg-
-------
∂17-Oct-83 1521 KONOLIGE@SRI-AI.ARPA thesis defense
Received: from SRI-AI by SU-AI with TCP/SMTP; 17 Oct 83 15:21:05 PDT
Date: 17 Oct 1983 1522-PDT
From: Konolige at SRI-AI
Subject: thesis defense
To: nilsson, genesereth at SCORE, jmc at SAIL, sgf at SAIL
cc: konolige at SRI-AI
I just talked with Solomon, and he has Tuesday (15th Nov) open in
the afternoon. I am going to schedule my thesis orals for that day at
4pm. If there are any objections, please let me know ASAP. The alternate
day would be the afternoon of the 14th. --kk
-------
∂17-Oct-83 1701 ULLMAN@SU-SCORE.ARPA New Room
Received: from SU-SCORE by SU-AI with TCP/SMTP; 17 Oct 83 17:01:27 PDT
Date: Mon 17 Oct 83 16:59:43-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: New Room
To: CS440: ;
cc: reges@SU-SCORE.ARPA
It turns out that 380-Y is not available regularly.
We have therefore gotten as our permanent room 200-034 in the History
Corner.
This change includes the talk by Dick Gabriel this Thursday 10/20.
-------
∂17-Oct-83 1936 CLT car
in case you login before finding out otherwise,
your car is at school (far lot). No juice (electrical).
Keys are in your desk drawer (at school).
Not much I can do about it at this time.
∂18-Oct-83 0908 DFH phone messages
1. Harry Lull, CS Library, please call 7-0864
2. Darlene Vian of Dr. Shortliffe's office, 7-6979. Will AAAI sponsor
AMSI meeting?
∂18-Oct-83 1542 RINDFLEISCH@SUMEX-AIM.ARPA More Symbolics Machines Coming
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 18 Oct 83 15:42:13 PDT
Date: Tue 18 Oct 83 15:44:07-PDT
From: T. C. Rindfleisch <Rindfleisch@SUMEX-AIM.ARPA>
Subject: More Symbolics Machines Coming
To: Pattermann@SUMEX-AIM.ARPA, Veizades@SUMEX-AIM.ARPA,
Spurgeon@SUMEX-AIM.ARPA, Schmidt@SUMEX-AIM.ARPA, Yeager@SUMEX-AIM.ARPA,
Bosack@SU-SCORE.ARPA, JMC@SU-AI.ARPA, RPG@SU-AI.ARPA
cc: Rindfleisch@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA
I just got a call from Mayflower Van Lines that the next three 3600's
are coming next Monday (10/24) -- I don't know what time of day they
will arrive. Two will be delivered at Welch Road and one more here in
MJH. The WR machines will also include a color display. I'd like to
ask Nick Veizades and Len Bosack to make necessary arrangements at WR
and MJH respectively.
Thanks, Tom R.
-------
∂18-Oct-83 1543 DFH Security form
I got addresses for the organizations listed except for
the L-5 Society and the American Go Assn. If you have L-5
address at home, could you mail it to me.
Birthdates. Have so far been unable to reach your ex-wife or
brother by phone. Both the home and work phone listed for
Martha Coyote were out of date. The new number given for the
home phone is 848-7610. I finally sent them both a note
asking for the information, but perhaps you would want to try
in the evening.
Martha's birthday is April 24, 1933.
Patrick McCarthy's is July 6, 1929.
∂18-Oct-83 1726 PETERS@SRI-AI.ARPA Thursday's Approaches to Natural Languages seminar
Received: from SRI-AI by SU-AI with TCP/SMTP; 18 Oct 83 17:26:12 PDT
Date: Tue 18 Oct 83 17:28:12-PDT
From: Stanley Peters <PETERS@SRI-AI.ARPA>
Subject: Thursday's Approaches to Natural Languages seminar
To: csli-friends@SRI-AI.ARPA
Stanley Peters will speak in the Approaches to Natural Language
seminar this Thursday. The topic is situation semantics for
referring noun phrases and relative clauses. The time 10:00 and
place Redwood G-19 are as usual.
-------
∂18-Oct-83 1741 @SRI-AI.ARPA:desRivieres.PA@PARC-MAXC.ARPA CSLI Activities for Thursday Oct. 20th
Received: from SRI-AI by SU-AI with TCP/SMTP; 18 Oct 83 17:41:24 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Tue 18 Oct 83 17:43:18-PDT
Date: Tue, 18 Oct 83 17:34 PDT
From: desRivieres.PA@PARC-MAXC.ARPA
Subject: CSLI Activities for Thursday Oct. 20th
To: csli-friends@SRI-AI.ARPA
Reply-to: desRivieres.PA@PARC-MAXC.ARPA
CSLI SCHEDULE FOR THURSDAY, OCTOBER 20, 1983
10:00 Research Seminar on Natural Language
Speaker: Stanley Peters (CSLI-Stanford)
Title: More about situation semantics.
Place: Redwood Hall, room G-19
12:00 TINLunch
Discussion leader: Per-Kristian Halvorsen (Xerox-CSLI)
Paper for discussion: "Belief-sentences and the Limits of Semantics"
by Barbara Hall Partee,
Place: Ventura Hall
2:00 Research Seminar on Computer Languages
Speaker: Hector Levesque (Fairchild AI Lab)
Title: "Knowledge Representation"
Place: Redwood Hall, room G-19
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speaker: Jerry Hobbs (SRI AI Lab)
Title: "Discourse Coherence and Discourse Structure"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It
can be reached from Campus Drive or Panama Street. From Campus Drive
follow the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
∂18-Oct-83 1831 CHANDRASEKARAN@RUTGERS.ARPA Simulation and Reasoning
Received: from RUTGERS by SU-AI with TCP/SMTP; 18 Oct 83 18:31:20 PDT
Date: 18 Oct 83 21:33:13 EDT
From: Chandra-at-OhioState <Chandrasekaran@RUTGERS.ARPA>
Subject: Simulation and Reasoning
To: JMC@SU-AI.ARPA
cc: Chandrasekaran@RUTGERS.ARPA
You might remember that a couple of years ago you commented
on my proposal for Prof. EDward Teller. After that we engaged in a
conversation about how simulation programs and AI programs which
do qualitative reasoning can interface with each other.
Your position was that in the nuclear industry there must obviously
be a number of programs which actually simulate
a reactor (or parts of it), and it would be important for
this information to be used by a reasoning system. I was
arguing for the dominance of the qualitative reasoning
component
and selcetive numerical simulation as determined
by the qualitative reasoning system. I have done more
thinking on that topic and I believe have gone closer
to understanding how the qualitative system may
actually invoke the simulation routines for specific
purposes. I know you were quite interested in the issue
at that time , and I was wondering if you would be
interested in taking a look at the approach that I am
converging on for this purpose. On hearing from you,
I'll be happy tomail a short description of the
approach to you.
I am sure you have taken a look at the recent Weizenbaum
broadside on AI in the New York Review as part of his review
of Feigenbaum and McCorduck's Fifth Generation book.
Weizenbaum is as irrational about AI as before, but I
think he scores a few points off the authors.
The problem is that within AI there is a lot of disquiet
about the solutions for long-standing AI problems claimed
in the book, but it is going to be trick how to respond
to reviews such as Weizenbaum's without appearing
to embrace wholeheartedly the "line" of the book.
I wonder if you have any suggestions, since I was thinking
of taking a crack at replying to the review.
Thanks for yyour attention.
-------
∂19-Oct-83 0818 LIBRARY@SU-SCORE.ARPA Reply to McCarthy and Keller concerning Library Services
Received: from SU-SCORE by SU-AI with TCP/SMTP; 19 Oct 83 08:18:36 PDT
Date: Wed 19 Oct 83 08:19:43-PDT
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: Reply to McCarthy and Keller concerning Library Services
To: su-bboards@SU-SCORE.ARPA
cc: jmc@SU-AI.ARPA, ark@SU-AI.ARPA, cottle@SU-SIERRA.ARPA,
herriot@SU-SCORE.ARPA, faculty@SU-SCORE.ARPA
I felt I should address the issues Prof. McCarthy and Prof. Keller have
raised and if the department as a whole feels the way the do I will
difintely want to address this issue of library services. The key issue
will be addressed with the library committee.
In regards to library services, one of my main goals when I came was to
expand the information services of the library. During those four
years I have done the following: videotapes of computer science classes
are offered for viewing in the library; I offer bibliographic instruction
and orientations to new students, I have had 50 new students attending
each year; computerized literature searching and demonstrations;
communicating with the department through score, sierra etc for reference
questions, overdues, recalls etc; increased reference services in the
library; new technical reports list online (and we hope to have the whole
file online). Some very specific decisions I made with computer science
in mind because the former math library had a different policy was to
allow graduate reserves out over night and not to send overdues or fines
on technical reports and reserves unless a user refused to bring material
back for another patron. These are the types of decisions I have implemented
that users often take for granted. If you have been positively impacted
by these decisions let me know.
In reference to our losses, to lessen the impact I have borrowed books
from Berkeley on my name in order to get the material to the patron quickly.
Within the past to months I have had to request over 30 items because we
were not able to find them in the library. It would have been much easier
just to tell the patron the material is not here or it is lost and turn
my back. However that is not how I operate. When I or my staff find
that something a patron needs is lost, we drop everything and go to any
lengths to get it quickly. From the reaction of some people in the
department it might have been better if I had taken that approach. However
the person who would have suffered would have been the graduate student
or researcher who needed the information then.
I have talked with some of you concerning the lack of staffing and increase
of use in the Math/CS Library. I try not to overdo on this because I am
very aware that the department is having its own struggles for space, staff,
and money. But you need to be aware that as the deparment grows the use
of the library also increases. In addition, as the impact of computers
impacts all of society we are having more and more people needing help
in the area of their information needs. We are the Math/CS Library but
we are required to serve the information needs of all the Stanford
community in the areas of computer science, mathematics, and statistics.
(By the way, all of Silicon Valley thinks we should serve them also but
we have to stop somewhere, I guess that is another example of bureaucracy?)
Prof. McCarthy suggests that we should turn to computers to solve this problem.
On this issue, I agree with him. However computers will help libraries give
better service, not put them out of business. I will like to encourage all
users to communicate with us through SCORE, SAIL, etc. I have tried to
announce new books, call for papers and conference dates, technical reports,
new journals etc. on the bulletin board. We can save you time if you
use the electronic mail for various library questions you have. It will
also help us in structuring our day when working on questions that come
through the electronic mail.
If you honestly feel that services have diminished, I want to hear from you
and I want to attempt to make our services more effective. However, I
would like to address this question with specific examples. In the past,
a few people often got very personalized service. My goal is to try to
keep as much of the personalized service as possible but not to waste
staff time on overdues, deliverying books personally etc. Instead I have
staff working on new technical reports exchange agreements, monitoring
the journals and conferences and making sure we get them as fast as possible,
ordering dissertations, books, and technical reports on demand, using
computerized databases for answering reference questions, videotapes.
Thank you in advance for your input.
Harry Llull
-------
As far as I can see, the latest message
from the Harry Llull Library is a mere advertisement
proposing an expansion of the empire
and an obfuscation. It is what we usually get
when bureaucracies feel nervous. It in no way addresses
the matter of keys which is the only issue I raised and
about which there is probably little new that can be said.
I still think we should plan a Computer Science Library in
our new building whenever that becomes a real possibility.
∂19-Oct-83 0924 @SRI-AI.ARPA:TW@SU-AI Abstract for Brian Reid talkware seminar today --NEW ROOM 380Y
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 09:23:56 PDT
Received: from SU-AI.ARPA by SRI-AI.ARPA with TCP; Wed 19 Oct 83 09:26:20-PDT
Date: 19 Oct 83 0919 PDT
From: Terry Winograd <TW@SU-AI>
Subject: Abstract for Brian Reid talkware seminar today --NEW ROOM 380Y
To: "@377.DIS[1,TW]"@SU-AI
Talkware Seminar
Date: October 19
Speaker: Brian Reid (Stanford CS/EE)
Topic: The Representation and Specification of Documents
Time: 2:15 - 4
Place: 380Y (Math corner) NOTE NEW (MAYBE STABLE) PLACE!
The way that a document is represented determines the kinds of
operations that we can perform on it. The taxonomy of representations
includes @i[pictorial] vs. @i[synthetic], @i[textual] vs.
@i[structural], @i[abstract] vs. @i[concrete], and @i[mutable] vs.
@i[immutable]. Representations are important in both the communication and
storage of a document.
A piece of paper, or a facsimile image, is a pictorial representation
of a document. It is completely concrete, unstructured, and relatively
immutable. A character file is a @i[textual] representation of a
document. It can be relatively abstract or relatively concrete. A
source file for a text formatter like Scribe is synthetic, primarily
textual, relatively abstract, and completely mutable. When that
document is run through Scribe for typesetting, the resulting output
file is synthetic, completely textual, completely concrete, and nearly
immutable.
I will discuss many schemes for the representation of documents, and the
issues associated with each.
∂19-Oct-83 1004 cheriton@Diablo Re: Reply to McCarthy and Keller concerning Library Services
Received: from SU-HNV by SU-AI with PUP; 19-Oct-83 10:04 PDT
Date: Wed, 19 Oct 83 10:04 PDT
From: David Cheriton <cheriton@Diablo>
Subject: Re: Reply to McCarthy and Keller concerning Library Services
To: LIBRARY@SU-Score, su-bboards@SU-Score
Cc: ark@Sail, cottle@SU-SIERRA.ARPA, faculty@SU-Score, herriot@SU-Score,
jmc@Sail
I have generally quite pleased with the library and feel some comments
being made are unfair. In my experience, computer science libraries
have an especially difficult time with theft of usually the most
valuable (from a reference standpoint) material. That is, the latest
conference publkications are stolen before the 1964 OS/360 JCL manual.
I dont care whether the lost rate is 5 percent or 15 percent if
everything I am interested in has been stolen.
Surely, we are collectively concerned about access to materials.
This is reduced by theft as well as tighter security. I would hope that
CSD and the library admin. can agree on a solution that is optimal
in terms of access.
P.S. How about appointing JMC and ARK as "honorary librarians" so they
can have keys?
∂19-Oct-83 1211 KONOLIGE@SRI-AI.ARPA Thesis Orals
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 12:11:42 PDT
Date: 19 Oct 1983 1211-PDT
From: Konolige at SRI-AI
Subject: Thesis Orals
To: nilsson, genesereth at SCORE, jmc at SAIL, sgf at SAIL
cc: konolige
They have been scheduled for Tuesday Nov. 15th at 2:30pm (note
the earlier time). --kk
-------
∂19-Oct-83 1318 TW
To: JMC, TW
This is a reminder that I need an abstract for your talk at 2:15 on Nov 9. --t
∂19-Oct-83 1320 DFH Travel (Austin/Urbana)
You should take a cab or the courtesy car to
the Hyatt Regency when you arrive in Austin. They
can't pick you up because your flight arrives during
a scheduled dinner. The Univ. of Il. will meet you
at the airport. I have given them the flight info.
∂19-Oct-83 1536 KJB@SRI-AI.ARPA visits
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 15:36:35 PDT
Date: Wed 19 Oct 83 15:33:16-PDT
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: visits
To: csli-folks@SRI-AI.ARPA
We will have a column in the newsletter called "Schedule of Visitors".
Please let Dianne know as you hear when people are coming to visit, so
that we can keep this up to date. Thanks.
-------
∂19-Oct-83 1750 BMACKEN@SRI-AI.ARPA budget
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 17:50:07 PDT
Date: Wed 19 Oct 83 17:53:04-PDT
From: BMACKEN@SRI-AI.ARPA
Subject: budget
To: csli-folks@SRI-AI.ARPA
The executive committee and I have completed our target allocations of
the first year budget. We made allocations to Areas A, B, C, and D, but
not to the projects within each area. The area managers in consultation
with the Executive Committee will make decisions regarding the further
allocation to individual projects.
I will be sending copies of CSLI expense request forms to each of you to
use as needed. The forms include space for the approval of the project
leader, the area manager, and me. There is also space for you to explain
the need for the funds in relation to a particular project. I will need
this information at the end of the year in reporting to SDF how the money
has been spent and in making decisions about next year's budget, so please
be as specific as possible.
Thanks,
B.
-------
∂19-Oct-83 1903 @SRI-AI.ARPA:BRESNAN.PA@PARC-MAXC.ARPA A2 meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 19:03:36 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Wed 19 Oct 83 19:05:07-PDT
Date: Wed, 19 Oct 83 19:00 PDT
From: BRESNAN.PA@PARC-MAXC.ARPA
Subject: A2 meeting
To: csli-people@sri-ai.ARPA
cc: BRESNAN.PA@PARC-MAXC.ARPA
Project A2 members met today at 2:00 in Ventura Hall. Martin Kay talked
about "A New Proposal for Patterns in Functional Unification Grammar".
Unification Grammar (UG) is very similar to LFG, except that constraints
on categories are not factored from constraints on functional relations.
A constituent in UG is a functional unit (e.g. SUBJ, PRED, OBJ) which
has a category attribute and is the domain of ordering relations on
subconstituents. Nonconfigurational languages like Warlpiri have posed
a problem for UG: the order patterns of UG are local to constituents,
but these languages have essentially free word order among both
constituents and their immediate subconstituents. Thus, in Warlpiri, a
subject may consist of several discontinuous parts spread out in the
sentence.
In "A New Proposal. . .", Martin proposed defining the order patterns
not on functional units, but on string elements, following a suggestion
of Henry Thompson. Order among string elements of the entire sentence
is assumed to be completely free, and patterns state constraints on them
in terms of the first and last elements of the functional units and
primitive relations such as adjacency and precedence. Thus constituents
may have "tangled branches" in relation to the terminal string.
In the discussion of this idea several problems were put forth. First,
can one describe in this new pattern language a generalization such as
"bounded scrambling", in which free word order is permitted across a
finite number of constituent boundaries? Second, how can one describe
the "AUX-second" constraint in languages like Warlpiri (Simpson 1983)
and Ngiyambaa (Klavans 1982)? According to this constraint, a
constituent such as SUBJ or OBJ may precede the AUX and any "piece" of a
constituent may precede the AUX, but nonconstituent sequences may not;
elsewhere word order is free as described above. It appears that the
pattern language is too impoverished to describe ordering relations of
this type, because there is no way to refer to "pieces" of structure
that consist solely of adjacent string elements.
Our next meeting will be on November 2, when Stusan Stucky will discuss
the syntax of the topic function in Makua.
Joan Bresnan
∂19-Oct-83 2051 DKANERVA@SRI-AI.ARPA Newsletter No. 5, October 20, 1983
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Oct 83 20:50:34 PDT
Date: Wed 19 Oct 83 20:50:15-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: Newsletter No. 5, October 20, 1983
To: csli-friends@SRI-AI.ARPA
.............
! CSLI Newsletter
October 20, 1983 * * * Number 5
MEETINGS OF PRINCIPALS AND ASSOCIATES
To get a better feeling for what is happening and to let people
know some things that are developing, Betsy and I would like to meet
with all the principals and associates next week. Because we want to
keep the meetings down to a size where discussion can take place, we
will have two meetings.
The principals will meet at Ventura Monday, October 17, at 4 p.m.
after tea. It shouldn't last much past 5:30. We will break at 6 at
the very latest.
The associates will meet at Ventura on Wednesday, October 26,
at 4 p.m. after tea, with the same time constraints as above.
- Jon Barwise
* * * * * * *
ADVISORY PANEL
The Advisory Panel has now been formed, at least for this fall's
review. It consists of Rod Burstall, Jerry Fodor, George Miller, Nils
Nilsson, Barbara Partee, and Bob Ritchie. The meeting with them will
be Thursday through Saturday, November 17-19. I hope that everyone
will be here then. They will attend that Thursday's activities. I
also hope that everyone, but especially the project managers, will be
available to talk with members of the Panel in their area on Friday
afternoon, November 18. Only the Executive Committee needs to keep
Saturday (morning) free.
- Jon Barwise
* * * * * * *
MEETING OF THE CSLI EDUCATION COMMITTEE
On October 7, the CSLI Education Committee met to establish the
requirements of the Graduate Program in Language in Information. The
meeting was chaired by John Perry and attended by Jon Barwise (ex
officio), Martin Kay, Stanley Rosenschein, and Thomas Wasow.
After discussion of the program as described in the proposal, the
prerequisites for the program and the courses required during the
program were decided. The prerequisites for the program, to be
completed during the first or second year of graduate study (if not
completed prior to matriculation), are Philosophy 160A, Syntax 230,
Computer Science/Linguistics 275 or 276, and Computer Science 102 (or,
in each case, something equivalent or more advanced). The program
itself will require that students (1) participate in the Approaches to
Natural Languages Seminar or the Approaches to Computer Languages
Seminar for at least one year, (2) take both the Situated Language in
Action course and the Foundations of Situated Language course (which
are each one-year sequences we plan to develop and offer), and (3)
have a CSLI advisor as a member of their dissertation committee.
! Page 2
Rosenschein and Kay are the Committee on Graduate Course
Development. They will begin developing the Situated Language courses
described above, rethinking the basic organizational ideas as
necessary; those with ideas should contact them. Perry and Wasow will
discuss with departments the compatibility of our proposed program
with their Ph.D. programs.
As the program does not yet exist, students supported this year,
and unsupported students we are encouraging to participate in our
activities, have a somewhat special status. They are not committed to
participate in the program when it is developed; we are not committed
to admit them. They should be of help to us in developing the
program, and the activities of this year should be valuable for them.
* * * * * * *
PROJECT A2 MEETING
Project A2 members met this Wednesday, October 19, at 2:00 in
Ventura Hall. Martin Kay talked about "A New Proposal for Patterns in
Functional Unification Grammar." In his talk, Martin proposed
defining the order patterns not on functional units, but on string
elements, following a suggestion of Henry Thompson. Our next meeting
will be on November 2, when Stusan Stucky will discuss the syntax of
the topic function in Makua.
- Joan Bresnan
* * * * * * *
WEEKLY STAFF MEETINGS
The CSLI staff at Ventura Hall will meet at 9 a.m. every Friday
in the Seminar Room to keep up to date on CSLI developments and to
coordinate their work. For that hour or so on Friday mornings, none
of the staff will be available to answer phones at Ventura, so it may
be necessary to call again after 10 a.m.
* * * * * * *
OFFICE AND PHONE CHANGES
This past week, several additional CSLI telephones were installed
in Ventura and Casita Halls. Some of the staff offices were changed
at the same time. Office and phone lists reflecting these changes
will be available soon.
* * * * * * *
VENTURA LANDSCAPING
There have been some inquiries into the reasons for the unique
landscaping around Ventura Hall. As most of you probably know,
Ventura and Casita were relocated last year. By the time the project
was finished, it was considerably over budget. One of the first items
to be cut was extensive landscaping. Instead of costly grass and
sprinker systems, drought-resistant trees and shrubs were selected.
Smaller trees rather than larger ones were planted. Someday the stand
of redwood trees on the corner will look impressive, and we do have a
superb (according to the landscape architect) specimen of madrona
between Ventura and Casita. The University does plan to seed the area
with wildflowers when the rains begin, so perhaps we'll look like
we're in a meadow for a while. But, for the foreseeable future, we'll
just have to wait for the trees to grow.
- Joyce Firstenberger
* * * * * * *
! Page 3
* * * * * * *
CSLI SCHEDULE FOR *THIS* THURSDAY, OCTOBER 20, 1983
10:00 Research Seminar on Natural Language
Speaker: Stanley Peters (CSLI-Stanford)
Title: More about situation semantics.
Place: Redwood Hall, room G-19
12:00 TINLunch
Discussion leader: Per-Kristian Halvorsen (Xerox-CSLI)
Paper for discussion: "Belief-sentences and the Limits of Semantics"
by Barbara Hall Partee,
Place: Ventura Hall
2:00 Research Seminar on Computer Languages
Speaker: Hector Levesque (Fairchild AI Lab)
Title: "Knowledge Representation"
Place: Redwood Hall, room G-19
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speaker: Jerry Hobbs (SRI AI Lab)
Title: "Discourse Coherence and Discourse Structure"
Place: Redwood Hall, room G-19
Note to visitors:
Redwood Hall is close to Ventura Hall on the Stanford Campus. It
can be reached from Campus Drive or Panama Street. From Campus Drive
follow the sign for Jordan Quad. Parking is in the C-lot between
Ventura and Jordan Quad.
* * * * * * *
MAIL SLOTS FOR OTHER SITES
Along with the new mail slots for individuals at Ventura Hall,
there are slots provided (bottom row) for the Computer Science,
Linguistics, and Philosophy Departments at Stanford and for Fairchild,
SRI, and Xerox-PARC. These mail slots are in the passageway between
the front hall and the kitchen.
If you are at Ventura, you might check the slot for your site and
take any items that might be there back with you to be distributed.
We will still try to get things through the mail and by other means to
the sites involved in the Center, but this might be a way of speeding
up the delivery of announcements and the like.
* * * * * * *
! Page 4
* * * * * * *
CSLI SCHEDULE FOR *NEXT* THURSDAY, October 27, 1983
10:00 Research Seminar on Natural Language
Speaker: Ray Perrault (CSLI-SRI)
Title: "Speech Acts and Plans"
Place: Redwood Hall, room G-19
We will examine some of the benefits and problems involved in
attempting to integrate elements of the speech act theory developed by
philosophers of language with accounts of action, planning, and
attitudes coming from the AI community. Special attention will be
given to indirect speech acts and to the integration of syntactic and
pragmatic information.
12:00 TINLunch
Discussion leader: Jeffrey S. Rosenschein
Paper for discussion: "Synchronization of Multi-Agent Plans"
by Jeffrey S. Rosenschein.
Place: Ventura Hall
2:00 Research Seminar on Computer Languages
Speaker: Peter Deutsch (Xerox PARC)
Title: "Smalltalk-80: Language and Style in Real
Object-oriented Programming System"
Place: Redwood Hall, room G-19
Smalltalk-80 is an object-oriented programming language and system
that has been in daily use by several dozen people for several years.
This talk will cover the more interesting aspects of the language, and
some equally interesting observations on how the language features are
actually used to create program structures. I will mention some other
object-oriented programming languages (Actors, LOOPS, Concurrent
Prolog), but not discuss them in depth.
3:30 Tea
Place: Ventura Hall
4:15 Colloquium
Speaker: Jay M. Tenenbaum (Fairchild AI Lab)
Title: "A.I. Research at Fairchild"
Place: Redwood Hall, room G-19
* * * * * * *
! Page 5
* * * * * * *
TINLUNCH SCHEDULE
TINLunch will be held on each Thursday at Ventura Hall on the
Stanford University campus as a part of CSLI activities. Copies of
TINLunch papers will be at SRI in EJ251 and at Stanford University in
Ventura Hall.
NEXT WEEK: SYNCHRONIZATION OF MULTI-AGENT PLANS
Jeffrey S. ROSENSCHEIN
NOTE: The author will be present on 10/27/83 for TINLUNCH.
SCHEDULE
October 20 Per-Kristian Halvorsen
October 27 Michael Georgeff
November 3 Ron Kaplan
November 10 Martin Kay
November 17 Jerry Hobbs
* * * * * * *
WHY CONTEXT WONT GO AWAY - Fourth meeting
Tuesday, Oct. 25, Ventura Hall, 3:15
"Computational Models of Mind and Meaning"
Speaker: B. Moore, SRI
We will give an overview of the sort of computational model of
mind that underlies much of the research in artificial intelligence,
and consider its implications for a theory of meaning. We will argue
that given such a model, a speaker's semantic competence does not
consist in knowing definitions of words or even knowing (except in an
almost trivial sense) the truth conditions of the sentences of his
language, but consists instead of knowing a commonsense theory of
those aspects of the world to which the terms of his language refer.
Finally, we will explain how this view of meaning undermines the
assumptions that lead to attempts to eliminate context from the theory
of meaning.
* * * * * * *
STANFORD PHILOSOPHY DEPARTMENT:
"Semantical Theory of Anaphora"
by
Jaakko Hintikka
Philosophy Department
Florida State University
Friday, October 21, 3:15 p.m.
Bldg. 90, Room 92Q (seminar room, 2nd floor)
* * * * * * *
! Page 7
* * * * * * *
TALKWARE SEMINAR
Date: October 19
Speaker: Brian Reid (Stanford CS/EE)
Topic: The Representation and Specification of Documents
Time: 2:15 - 4
Place: 380Y (Math corner) NOTE NEW (MAYBE STABLE) PLACE!
The way that a document is represented determines the kinds of
operations that we can perform on it. The taxonomy of representations
includes pictorial vs. synthetic, textual vs. structural, abstract
vs. concrete, and mutable vs. immutable. Representations are
important in both the communication and storage of a document. A
piece of paper, or a facsimile image, is a pictorial representation of
a document. It is completely concrete, unstructured, and relatively
immutable. A character file is a textual representation of a
document. It can be relatively abstract or relatively concrete. A
source file for a text formatter like Scribe is synthetic, primarily
textual, relatively abstract, and completely mutable. When that
document is run through Scribe for typesetting, the resulting output
file is synthetic, completely textual, completely concrete, and nearly
immutable. I will discuss many schemes for the representation of
documents, and the issues associated with each.
Date: October 26
Speaker: Greg Nelson (Xerox PARC)
Topic: JUNO: a constraint based language for graphics
Time: 2:15 - 4
Place: 380Y (Math corner)
Abstract: to come later
Visitor on Nov 1: Kristen Nygaard, Norwegian computing Center and
University of Oslo, will be speaking at Talkware seminar (CS377) at a
nonstandard time Tuesday at 1:15 (place not yet determined).
* * * * * * *
SEMINAR IN LOGIC AND FOUNDATIONS
SPEAKER: Prof. J. E. Fenstad, University of Oslo
TITLE: "Hyperfinite Probability Theory: Basic Ideas and Applications
in Natural Sciences"
TIME: Wednesday, Oct. 19, 4:15-5:30 PM
PLACE: Stanford Mathematics Dept. Faculty Lounge, 383N
The talk will assume some acquaintance with nonstandard analysis
(existence of the extensions, transfer). But the ideas of hyperfinite
probability theory (e.g., Loeb construction) will be explained before
turning to applications, which will mainly be to hyperfinite spin
systems (statistical mechanics, polymer models, field theory). The
models will be fully explained, so no knowledge of "advanced" physics
is presupposed.
- S. Feferman
* * * * * * *
TWO JOURNALS: L&P AND JSL
The journal LINGUISTICS AND PHILOSOPHY is currently short of
first-rate papers. It only has about half of what it needs for the
next issue. The JOURNAL OF SYMBOLIC LOGIC is now interested in
publishing technically sophisticated papers applying ideas from logic
to computer science, linguistics, and philosophy. Jon Barwise is an
editor for both these journals. If you have a paper that you think
might be appropriate for one of the journals, see him.
* * * * * * *
-------
∂20-Oct-83 0843 BSCOTT@SU-SCORE.ARPA Re: Martin Brooks expenses
Received: from SU-SCORE by SU-AI with TCP/SMTP; 20 Oct 83 08:43:31 PDT
Date: Thu 20 Oct 83 08:45:25-PDT
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: Re: Martin Brooks expenses
To: JMC@SU-AI.ARPA
cc: BSCOTT@SU-SCORE.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 18 Oct 83 16:09:00-PDT
Diana, I assume you're doing something about this?
Betty
-------
∂20-Oct-83 0901 KJB@SRI-AI.ARPA Putting up the advisory panel
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Oct 83 09:00:14 PDT
Date: Thu 20 Oct 83 08:59:05-PDT
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Putting up the advisory panel
To: csli-principals@SRI-AI.ARPA
cc: bmacken@SRI-AI.ARPA
If any of you are friends of our out of town advisory panel members,
you might like to invite them to stay with you. It would both be more
pleasant for them, and be a more accurate reflection of our financial
situation, than if we were to put them up at a nice motel. Let Betsy
know if you make such plans, so she can cancel reservations for them.
Jon
(The list of panel members is in todays newsletter.)
-------
∂20-Oct-83 0944 DKANERVA@SRI-AI.ARPA Correction for Newsletter
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Oct 83 09:44:49 PDT
Date: Thu 20 Oct 83 09:42:55-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: Correction for Newsletter
To: csli-friends@SRI-AI.ARPA
The meeting of the CSLI Principals will be held Monday, October 24,
at 4 p.m., not October 17, as erroneously given on the first page
of this week's newsletter. - Dianne Kanerva
-------
∂20-Oct-83 0950 BMACKEN@SRI-AI.ARPA Associations
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Oct 83 09:50:05 PDT
Date: Thu 20 Oct 83 09:33:13-PDT
From: BMACKEN@SRI-AI.ARPA
Subject: Associations
To: csli-folks@SRI-AI.ARPA
I have updated <CSLI>associations again. Let me know if there are further
changes.
A further note about the allocation percentages: one use of the pecentages
was to help in making the budget allocations; another will be in reporting
to SDF about the relative effort in different areas. While we did not need
the percentage information from those of you not getting paid by CSLI funds
for the first use, we will for the second. Thus, I do need estimated
percentages from all of you eventually. Let me know as you know them and
update them as needed.
B.
-------
∂20-Oct-83 1013 DKANERVA@SRI-AI.ARPA Re: mailing lists
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Oct 83 10:13:03 PDT
Date: Thu 20 Oct 83 10:17:03-PDT
From: DKANERVA@SRI-AI.ARPA
Subject: Re: mailing lists
To: JMC@SU-AI.ARPA
cc: Wunderman@SRI-AI.ARPA, DKanerva@SRI-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Thu 20 Oct 83 09:46:00-PDT
I'll see that the CSLI lists all get changed to show your net
address as JMC-LISTS@SU-AI, as you requested. -- Dianne Kanerva
-------
∂20-Oct-83 1126 @SU-SCORE.ARPA:TRATTNIG@SU-SIERRA.ARPA Today's lunch with Burton Smith
Received: from SU-SCORE by SU-AI with TCP/SMTP; 20 Oct 83 11:26:44 PDT
Received: from SU-SIERRA.ARPA by SU-SCORE.ARPA with TCP; Thu 20 Oct 83 11:26:14-PDT
Date: Thu 20 Oct 83 11:26:09-PDT
From: Werner Trattnig <TRATTNIG@SU-SIERRA.ARPA>
Subject: Today's lunch with Burton Smith
To: super@SU-SCORE.ARPA
This is a remainder:
12:00noon SUPER LUNCH AEL109
---------------------------------
Informal discussions with Burton Smith
concerning future of supercomputing, the
recent Frontiers of Supercomputing meetings,
etc. (Bring your own lunch)
-------
∂20-Oct-83 1605 @SRI-AI.ARPA:CLT@SU-AI SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Oct 83 16:03:54 PDT
Received: from SU-AI.ARPA by SRI-AI.ARPA with TCP; Thu 20 Oct 83 16:05:47-PDT
Date: 20 Oct 83 1555 PDT
From: Carolyn Talcott <CLT@SU-AI>
Subject: SEMINAR IN LOGIC AND FOUNDATIONS OF MATHEMATICS
To: "@DIS.DIS[1,CLT]"@SU-AI
SPEAKER: S. Feferman
TITLE An introduction to "Reverse Mathematics"
TIME: Wednesday, Oct. 26, 4:15-5:30 PM
PLACE: Stanford Mathematics Dept. Faculty Lounge (383-N)
The talk will introduce and survey work by Friedman, Simpson and others,
providing sharp information in the form of equivalences as to which set-
existence axioms are needed to prove various statements in analysis and
algebra.
S. Feferman
∂21-Oct-83 0011 ARK A New Income Source for CSD-CF?
To: Bosack@SU-SCORE
CC: ME@SU-AI, JMC@SU-AI, ARK@SU-AI
I wonder if the NS program and files could be used to get some income for
CSD-CF. Perhaps, we could get university officials to get accounts on
SAIL to supplement their clipping services. Suppose we set up NS
notification requests, printed the files every so often, and ID mailed the
results out. What could we charge for such a service, how much would it
cost us? It seems like something that would bring in income without
adverse effect on users. (The problem I see with this is the labor in
distributing the printouts, which could be eliminated with electronic
mail, but we'd need to be able to do it to CIT.)
Arthur
To use NS for income would violate our agreements with AP and NYT.
∂21-Oct-83 0950 @SU-SCORE.ARPA:GEORGEFF@SRI-AI.ARPA LISP 1980 Conference
Received: from SU-SCORE by SU-AI with TCP/SMTP; 21 Oct 83 09:50:25 PDT
Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Fri 21 Oct 83 09:52:17-PDT
Date: Fri 21 Oct 83 09:54:06-PDT
From: Michael Georgeff <georgeff@SRI-AI.ARPA>
Subject: LISP 1980 Conference
To: MCCarthy@SU-SCORE.ARPA
Dear John,
I am trying to trace a paper by McDermott entitled "An Efficient Environment
Allocation Scheme ..." given at the 1980 LISP conference. Unfortunately,
no one around here has a copy. Do you have a copy of the proceedings, or
of Drew's paper?
Thanks,
Michael Georgeff.
-------
∂21-Oct-83 1033 ME selling the news
To: ARK@SU-AI, LB@SU-AI, JMC@SU-AI
∂21-Oct-83 0011 ARK A New Income Source for CSD-CF?
To: Bosack@SU-SCORE
CC: ME@SU-AI, JMC@SU-AI, ARK@SU-AI
I wonder if the NS program and files could be used to get some income for
CSD-CF. Perhaps, we could get university officials to get accounts on
SAIL to supplement their clipping services. Suppose we set up NS
notification requests, printed the files every so often, and ID mailed the
results out. What could we charge for such a service, how much would it
cost us? It seems like something that would bring in income without
adverse effect on users. (The problem I see with this is the labor in
distributing the printouts, which could be eliminated with electronic
mail, but we'd need to be able to do it to CIT.)
Arthur
ME - I believe this would violate the spirit and letter of our agreements
with the AP and NYT.
∂21-Oct-83 1227 Winograd.PA@PARC-MAXC.ARPA Re: Lisp as Language
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 21 Oct 83 12:27:42 PDT
Date: 21 Oct 83 12:13 PDT
From: Winograd.PA@PARC-MAXC.ARPA
Subject: Re: Lisp as Language
In-reply-to: BrianSmith's message of 15 Oct 83 19:19 PDT
To: BrianSmith.PA@PARC-MAXC.ARPA
cc: CSLI-Executives@SRI-AI.ARPA, Winograd.PA@PARC-MAXC.ARPA,
JMC@SU-AI.ARPA, Pereira@SRI-AI.ARPA, RPerrault@SRI-AI.ARPA,
desRivieres.PA@PARC-MAXC.ARPA, Levesque@SRI-KL.ARPA
The basic structure looks fine, but there is a prima facia inconsistency in
combining the following phrases (or their paraphrases)...
"No previous exposure to computing"
"rigourous introduction"
"topics will include..........................."
"one quarter course"
"one two-hour class presentation and one three-hour tutorial section per week
Consistency will require inversion (to the other pole) of the first, drastic
shortening of one of the next two or lengthening of one of the last two. --t
∂22-Oct-83 1143 AAAI-OFFICE@SUMEX-AIM.ARPA workshop
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 22 Oct 83 11:43:15 PDT
Date: Sat 22 Oct 83 11:46:19-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: workshop
To: JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I talked to Pat Hayes on Friday and described our intent to assist
him financially with this "commmonsense physics" workshop> He was delighted
with the offer and said he would be getting back to me later.
Re: Shortliffe's request. Nils said he can't remember what Ted wanted
exactly. So, it looks like we will have to wait until he calls in to
determine exactly what he wants. Concurrently, I got a phone message
from a Morris Collin, head of the Amer Assoc for Medical Systems and
Information Congress which will be sponsoring Ted's workshop at their
national conference, asking for a letter of AAAI's approval so that
AAAI can be listed as a co-sponsor of this conference. As soon as
Collin calls me back, I'll find out exactly what are the responsibilities
associated with co-sponsorship. I'll keep you informed as things progress.
Claudia
-------
∂23-Oct-83 2117 ALMOG@SRI-AI.ARPA reminder on why context wont go away
Received: from SRI-AI by SU-AI with TCP/SMTP; 23 Oct 83 21:17:38 PDT
Date: 23 Oct 1983 2114-PDT
From: Almog at SRI-AI
Subject: reminder on why context wont go away
To: csli-friends at SRI-AI
On Tuesday 25.10.83 we have our fourth meeting. The speaker will be
Bob Moore. Here is an abstract of his talk.
WHY CONTEXT WONT GO AWAY- Fourth meeting
Tuesday,10.25.83,Ventura Hall, 3.15
Speaker: B.Moore , SRI
Computational Models of Mind and Meaning
We will give an overview of the sort of computational model of
mind that underlies much of the research in artificial intelligence,
and consider its implications for a theory of meaning. We will argue
that given such a model, a speaker's semantic competence does not
consist in knowing definitions of words or even knowing (except in an
almost trivial sense) the truth conditions of the sentences of his
language, but consists instead of knowing a commonsense theory of
those aspects of the world to which the terms of his language refer.
Finally, we will explain how this view of meaning undermines the
assumptions that lead to attempts to eliminate context from the theory
of meaning.
-------
∂24-Oct-83 1354 @SRI-AI.ARPA:BrianSmith.pa@PARC-MAXC.ARPA LISP as Language Course, Winter Quarter
Received: from SRI-AI by SU-AI with TCP/SMTP; 24 Oct 83 13:54:20 PDT
Received: from PARC-MAXC.ARPA by SRI-AI.ARPA with TCP; Mon 24 Oct 83 13:49:49-PDT
Date: 24 Oct 83 13:45 PDT
From: BrianSmith.pa@PARC-MAXC.ARPA
Subject: LISP as Language Course, Winter Quarter
To: CSLI-Friends@SRI-AI.ARPA
cc: BrianSmith.pa@PARC-MAXC.ARPA
Appended is an announcement of the Lisp as Language seminar that I will
be teaching next quarter. We will probably aim for two
one-and-a-quarter hour class sessions per week, plus a three-hour
session once a week for programming help. Time, place, etc., are all to
be determined.
If you think you will be interested in taking the course, please let me
know, so that I can plan for a classroom, machines, TA's, etc.
Many thanks.
Brian
-----------------------------------------------------------------------
LISP as Language
A systematic introduction to the concepts and practices of programming,
based on a simple reconstructed dialect of LISP. The aim is to make
explicit the knowledge that nights and weekends of programming make
implicit. The material will be presented under a "linguistic
reconstruction", using vocabulary that should be of use in studying any
linguistic system. Although intended primarily for linguists,
philosophers, and mathematicians, anyone interested in computation is
welcome.
Although no previous exposure to computation is required, we will aim
for rigorous analyses. Familiarity with at least some formal system is
therefore essential. Participants will be provided with tutorial
programming instruction.
Topics to be covered include:
-- Procedural and data abstraction;
-- Objects, modularity, state, and encapsulation;
-- Input/output, notation, and communication protocols;
-- Meta-linguistic abstraction, and problems of intensional grain;
-- Self-reference, meta-circular interpreters, and reflection.
Throughout the course, we will pay particular attention to the following
themes:
-- Procedural and declarative notions of semantics;
-- Interpretation, compilation, and other models of processing;
-- Architecture, implementation, and abstract machines;
-- Implicit vs. explicit representation of information;
-- Contextual relativity, scoping mechanisms, and locality.
The course will be based in part on the "Structure and Interpretation of
Computer Programs" textbook, by Abelson and Sussman, that has been used
at M.I.T., although the linguistic orientation will affect our dialects
and terminology.
-----------------------------------------------------------------------
∂24-Oct-83 1633 DFH
Messages
10/20 Alphonse Juilland. ofc. 7-4460, home 321-7819. He says it's
important. (called again 10/24)
10/20 Walter Willis. Wants to know if there is anything new about the
"Rolling Skyhook". 415-368-4069.
10/21 Carole Tretkoff, Brooklyn College. Would like you to speak at their
conference on AI Dec. 16 & 17.. If you phone her Tues, the no. is
609-924-5185.
10/21 Judson Hewitt. referred by Peter Vajk re self-reproducing machinery.
346-5421 or 845-7677. There is also a letter from him.
10/21 Prof. Gil Kim of the Korean Inst. called, I gave him your flight info.
They will meet you at the airport, and he wants you to use the left
customs exit. You will be at the Shilla Hotel.
∂24-Oct-83 1706 ULLMAN@SU-SCORE.ARPA Next Seminar
Received: from SU-SCORE by SU-AI with TCP/SMTP; 24 Oct 83 17:06:20 PDT
Date: Mon 24 Oct 83 17:04:23-PDT
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Next Seminar
To: CS440: ;
Remember we meet in Room 034 of the History Corner (Bldg. 200).
Here is the next talk:
"DATAFLOW MACHINE AND SISAL, A NEW LANGUAGE FOR MULTIPROCESSING"
by Alain Hanover
Manager
VLSI Architecture Advanced Development
Digital Equipment Corporation
Digital has been involved in a cooperative dataflow research project
with the University of Manchester for over three years. Our work on
the dataflow machine has led us to try to develop a new language,
SISAL-- Streams and Iterations in a Single-Assignment Language.
The SISAL language has been developed jointly with the University of
Manchester, Lawrence Livermore National Laboratories, and Colorado State
University researchers who also wish to program supercomputers such as
the CRAY and HEP. Currently we are jointly developing a SISAL front-
end compiler and several back-end code generators for different target
machines as well as extensive benchmarks to gain performance results.
Digital's Dataflow research effort, SISAL, and our strategy for language
and applications software will be discussed in this talk.
-------
∂25-Oct-83 0918 BOSACK@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 25 Oct 83 09:18:33 PDT
Date: Tue 25 Oct 83 09:19:57-PDT
From: Len Bosack <BOSACK@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 25 Oct 83 00:28:00-PDT
I have not seriously considered offering NS as a 'product'. I think we
agree not to publish the text and so qualify for a lower rate.
Len
-------
We don't "qualify for a lower rate" according to some algorithm. We have
a specific agreement for the experimental use of the news wires.
Incidentally, the lower rate is lower by a factor of tens.
∂25-Oct-83 1125 AAAI-OFFICE@SUMEX-AIM.ARPA fellowships
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 25 Oct 83 11:25:15 PDT
Date: Tue 25 Oct 83 11:27:49-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: fellowships
To: JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
I'm planning to begin an investigation of fellowship programs
administered by private foundations. I thought I'd begin by
talking to the MacArthur, Ford and Rockerfeller Foundations.
Do you know of any private foundatioon that currently provides
for fellowships in computer science?
Claudia
-------
∂25-Oct-83 1157 @SU-SCORE.ARPA:DELAGI@SUMEX-AIM.ARPA Visit by PPA Developers
Received: from SU-SCORE by SU-AI with TCP/SMTP; 25 Oct 83 11:57:08 PDT
Received: from SUMEX-AIM.ARPA by SU-SCORE.ARPA with TCP; Tue 25 Oct 83 11:52:33-PDT
Date: Tue 25 Oct 83 11:52:35-PDT
From: Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>
Subject: Visit by PPA Developers
To: super@SU-SCORE.ARPA,
Adv.Architectures: ;
cc: "decwrl!rhea::delagi"@SU-SHASTA.ARPA
On November 3, a few of the lead designers of the PPA hardware and software
will be at Stanford to talk about what they're doing and learn about the
projects here that are pertinent to multiprocessing. In the afternoon, they'll
be joined by Tony Fleig who (on assignment from DEC) has put together the
application environment for a shared memory supplemented computer cluster
at NASA Ames.
The afternoon will be dedicated to presentations by DEC people (beginning with
a lunch at 12:30) ending with a public seminar by Tony Fleig (with a few words
added by the PPA developers) at 4:15 as part of the CS440 seminar series. The
detailed discussion of the PPA after lunch will be for those people who have
signed the non-disclosure agreement (as requested by the PPA project management)
In the morning of November 3, the design folks would like to meet with people
here to learn what is happening at Stanford in multiprocessing research. Please
drop me a note if you'd like to talk with them then.
If you plan to come to the lunch, please let me know so I can let the sandwich
supplier know the head count.
/bruce
-------
∂25-Oct-83 1410 TW
John, I would like an abstract for your talk on the 9th. Thanks --t
∂26-Oct-83 0018 GOLUB@SU-SCORE.ARPA Re: library keys
Received: from SU-SCORE by SU-AI with TCP/SMTP; 26 Oct 83 00:18:46 PDT
Date: Wed 26 Oct 83 00:17:45-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
Subject: Re: library keys
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 25 Oct 83 13:04:00-PDT
John! I certainly will postpone my decision until you return. GENE
-------
∂26-Oct-83 0505 CL.BOYER@UTEXAS-20.ARPA Re: visit
Received: from UTEXAS-20 by SU-AI with TCP/SMTP; 26 Oct 83 05:04:58 PDT
Date: Wed 26 Oct 83 07:04:30-CDT
From: Bob Boyer <CL.BOYER@UTEXAS-20.ARPA>
Subject: Re: visit
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 25 Oct 83 16:54:00-CDT
I am delighted.
Ps. to Carolyn. Congratulations to you too on your marriage.
-------
∂26-Oct-83 1138 AAAI-OFFICE@SUMEX-AIM.ARPA ACS
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 26 Oct 83 11:36:40 PDT
Date: Wed 26 Oct 83 11:37:04-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: ACS
To: JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
Finally, ACS returned my calls.
They are sending me a history of the ACS that describes evolution
of the association. Re: advice with conference organization, they
said they use the resources of the ASAE (Amer Society for Assoc
Executives) for conference planning. I'll call ASAE today.
ACS wants the book back immediately. So, I'll copy relevant
chapters and send you a copy.
Claudia
-------
∂26-Oct-83 1545 GOLUB@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 26 Oct 83 15:45:38 PDT
Date: Wed 26 Oct 83 15:45:14-PDT
From: Gene Golub <GOLUB@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 18 Oct 83 23:59:00-PDT
There was no vote. I have told Jack any faculty member who wants
a key should get one. GENE
-------
∂26-Oct-83 1748 PACK@SU-SCORE.ARPA Circumscription etc
Received: from SU-SCORE by SU-AI with TCP/SMTP; 26 Oct 83 17:47:58 PDT
Date: Wed 26 Oct 83 17:43:17-PDT
From: Leslie E. Pack <PACK@SU-SCORE.ARPA>
Subject: Circumscription etc
To: jmc@SU-AI.ARPA
I have worked up axioms for dealing with piling the blocks on top of
each other, and discovered a problem. In the case where there is a
block in the position to which we want to move or on top of the block
we want to move, we would like to say that unless it is abnormal, we
can't move our original block the way we intended. It is abnormal if
we can move the offending block (on top, or at the destination). It
would seem cleaner if we could treat both of these cases as having
the same abnormality, just with different arguments. (In one case it's
aspect9(z,x,s) in the other case it's aspect9(z,p,s). The difficulty
arises when trying to generate the fopositive inference because we no
longer have all of the information we need.
The only solution to this that I see is putting the ~ab clause in
front of all of the axioms, including those that deal with the next
higher level abnormality. This would allow us to leave out the
positive axioms after their highest level mention. You mentioned a
problem this caused when doing the circumscription, but I didn't see
what is was. If this is in fact the case, perhaps you should amend
your paper. In the most recent version, you retained
~ab aspect3(x,p,s) -> ab aspect1(x,move(x,p),s).
Here is my file, so far...
Just taking position into account for now:
can←do(move(b1,p),s) & can←do(move(b2,b1),result(move(b1,p,s))) &
can←do(move(b3,b2),result(move(b2,b1),result(move(b1,p),2))) ->
can←do(make←tower(p,b1,b2,b3),s)
(Es1.location(x,p,s1) & temporally←greater(s1,s)) -> can←do(move(x,p),s)
~ab aspect1(x,e,s) -> location (x,result(e,s)) = location (x,s)
ab aspect1(x,move(x,p),s)
~ab aspect3(x,p,s) -> location(x,result(move(x,p),s)) = p
Ez.location(z,s) = x or location(z,s) = p or ~fits←on(x,p,s) ->
ab aspect3(x,p,s)
Ez.location(z,s) = x & ~ab aspect8(x,z,p,s) ->
location(x,result(move(x,p),s) = location(x,s)
Ez.location(z,s) = p & ~ab aspect9(x,z,p,s) ->
location(x,result(move(x,p),s) = location(x,s)
~fits←on(x,p,s) -> location(x,result(move(x,p),s) = location(x,s)
Ep1.can←do(move(z,p1),s) & ~p1=x -> ab aspect8(x,z,p,s)
Ep1.can←do(move(z,p1),s) & ~p1=x -> location(x,result(move(x,p),s) = p
Ep1.can←do(move(z,p1),s) & ~p1=p -> ab aspect9(x,z,p,s)
Ep1.can←do(move(z,p1),s) & ~p1=p -> locaiton(x,result(move(x,p),s) = pp
If the blocks don't fit on top of one another or the obstacles cannot
be removed, then we fail. Other abnormalities can be introduced if we
think of ways to fix such things.
It would be nice if the axioms dealing with aspect8 could be sufficiently
general to deal with any situation that requires a block to be moved. In
this case, there are two cases in which the situation can be abnormal in
this way.
-------
∂27-Oct-83 1020 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doctoral fellowships
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 27 Oct 83 10:20:32 PDT
Date: Thu 27 Oct 83 10:21:03-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Post-Doctoral fellowships
To: JMC@SU-AI.ARPA
cc: Buchanan@SUMEX-AIM.ARPA, aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
I made a phone call to Dr Wilson Tally, President of the Hertz
Foundation regarding our interest in establishing a post-doctoral
fellowship program and having an outside foundation administer it. He
said that the Hertz Foundation does this all the time, and he thought
his organization would be interested in administering our program.
Before I go any further, let me give you some information
about the Hertz Foundation. This year's endowment is $1.5 million
for undergraduate scholarships and graduate fellowships. 20-25% went
to computer science and the rest went to the physical sciences.
The flow of monies begins with their organization and flows into
an academic institution directly. Their selection process is
similar to the Rhodes and Fullbright procedures; that is, written
materials, references, and personal interviews are required in
the selection process. They require the recipients to prepare
status reports at the end of each semester or quarter.
If we are interested in working with the Hertz Foundation,
we need to prepare a letter (with the names of our executive council members)
describing our intent, noting that we would be responsible for the
selection process, and acknowledging that the Hertz Foundation
will not access an administrative fee for its efforts by the
end of week (Nov 4). Tally would like this letter so that
he can take it with him to their board meeting in Boston on
November 14.
I'd like to hear your comments on this development.
Regards,
Claudia
-------
∂27-Oct-83 1046 BUCHANAN@SUMEX-AIM.ARPA Re: Post-Doctoral fellowships
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 27 Oct 83 10:46:02 PDT
Date: Thu 27 Oct 83 10:46:34-PDT
From: Bruce Buchanan <BUCHANAN@SUMEX-AIM.ARPA>
Subject: Re: Post-Doctoral fellowships
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: JMC@SU-AI.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of Thu 27 Oct 83 10:21:08-PDT
I think it is a good idea to have the AAAI fellowships professionally
administered. We should make sure that the money is separately
identified, however.
bgb
-------
∂28-Oct-83 0929 AAAI-OFFICE@SUMEX-AIM.ARPA vacation
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 28 Oct 83 09:29:21 PDT
Date: Fri 28 Oct 83 09:29:41-PDT
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: vacation
To: JMC@SU-AI.ARPA, FIKES@PARC-MAXC.ARPA, ENGELMORE@SRI-KL.ARPA
cc: AAAI-OFFICE@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I'm planning to take a vacation between November 19-28, 1983.
If you really need me, I can be reached at home, 328-8676.
Regards,
Claudia
-------
∂29-Oct-83 0719 ATP.BLEDSOE@UTEXAS-20.ARPA Re: visit
Received: from UTEXAS-20 by SU-AI with TCP/SMTP; 29 Oct 83 07:19:30 PDT
Date: Sat 29 Oct 83 09:18:57-CDT
From: Woody Bledsoe <ATP.Bledsoe@UTEXAS-20.ARPA>
Subject: Re: visit
To: JMC@SU-AI.ARPA
cc: ATP.Bledsoe@UTEXAS-20.ARPA, clt@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 25 Oct 83 16:54:00-CDT
John and Carolyn,
We would be happy to have you visit Austin when you can. It would
be better while school is in session so that many people won't be gone
(i.e., not Dec 12 - Jan 15), but even then is possible. Why don't you
suggest some times, and let me know.
Woody
-------
∂30-Oct-83 1732 CLT Sarah
called. Wants you to send a letter saying you have own the
car and authorize S to dispose of it.
She has someone who wants to take it away.
∂31-Oct-83 0916 DFH K. Clark Parlog paper
A. Van Gelder was by wanting to borrow this and
a couple of other papers, which I gave to him. I
could not find the Parlog paper, however--it's never
been filed in the library. If you happen to have it
at home, he would still like to borrow it. (I looked on
your desk too).
∂31-Oct-83 0955 DFH Claudia Mazzetti
would like you to call her.
∂31-Oct-83 1058 DFH D. Chudnovsky called
He will try to call you again tomorrow.
∂31-Oct-83 1250 ULLMAN@SU-SCORE.ARPA Next Talk
Received: from SU-SCORE by SU-AI with TCP/SMTP; 31 Oct 83 12:50:08 PST
Date: Mon 31 Oct 83 12:46:46-PST
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Next Talk
To: CS440: ;
The speaker for Thursday, Nov. 3 is Tony Fleig of NASA/Ames talking on
"A shared memory/local area network environment for parallel computation."
As usual, we meet in the History Corner, rm. 200-034, 4:15PM.
-------
∂31-Oct-83 1545 @SU-SCORE.ARPA:DELAGI@SUMEX-AIM.ARPA PPA Lunch Location
Received: from SU-SCORE by SU-AI with TCP/SMTP; 31 Oct 83 15:45:19 PST
Received: from SUMEX-AIM.ARPA by SU-SCORE.ARPA with TCP; Mon 31 Oct 83 15:36:53-PST
Date: Mon 31 Oct 83 14:57:40-PST
From: Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>
Subject: PPA Lunch Location
To: super@SU-SCORE.ARPA,
Adv.Architectures: ;
cc: ejg@SU-AI.ARPA
This Thursday (3 November) there'll be lunch with the PPA developers
in MJH 301.
/bruce
-------
∂31-Oct-83 2020 PACK@SU-SCORE.ARPA towers
Received: from SU-SCORE by SU-AI with TCP/SMTP; 31 Oct 83 20:20:41 PST
Date: Mon 31 Oct 83 20:20:17-PST
From: Leslie E. Pack <PACK@SU-SCORE.ARPA>
Subject: towers
To: jmc@SU-AI.ARPA
I think I have a workable set of axioms for a fairly simple world. A
big problem I came across has to do with making aspects fairly
general. The problem manifests itself in aspects 8 and 9. It seems
to me that it really should only be one aspect, depending on the thing
that's in the way, what it's in the way of, and the situation. That
won't work and allow use of the positive inferences.
In the most recent draft of your paper, you still have ab in the
antecedent of a sentence, the consequent of which is ab (something).
I thought this caused a problem of getting a disjuction when
circumscribed?
I will be around most afternoons this week. Send mail telling me when
it's convenient.
Axioms follow...
(Es1.location(b1,s1) = p and
location(b2,s1) = b1 and
location(b3,s1) = b2 and
color(b1,s1) = c1 and
color(b2,s1) = c2 and
color(b3,s1) = c3 and
temporally←greater(s1,s))
=> can←do (make←tower(p,c1,c2,c3),s)
~ab aspect1(x,e,s) -> location (x,result(e,s)) = location (x,s)
~ab aspect2(x,e,s) -> color(x,result(e,s)) = color(x,s)
ab aspect1(x,move(x,p),s)
ab aspect2(x,paint(x,c),s)
~ab aspect3(x,p,s) -> location(x,result(move(x,p),s)) = p
~ab aspect4(x,c,s) -> color(x,result(paint(x,c),s)) = c
~(Ep.color(p,s) = c & paint(p,s))) -> ab aspect4(x,c,s)
~(Ep.color(p,s) = c & paint(p,s)) & ~ab aspect7(p,c,s) ->
color(x,result(paint(x,c),s)) = color(x,s)
Ez.location(z,s) = x or location(z,s) = p or ~fits←on(x,p,s) ->
ab aspect3(x,p,s)
Ez.location(z,s) = x & ~ab aspect8(x,z,p,s) ->
location(x,result(move(x,p),s) = location(x,s)
Ez.location(z,s) = p & ~ab aspect9(x,z,p,s) ->
location(x,result(move(x,p),s) = location(x,s)
~fits←on(x,p,s) -> location(x,result(move(x,p),s) = location(x,s)
Ep1.can←do(move(z,p1),s) & ~p1=x -> ab aspect8(x,z,p,s)
Ep1.can←do(move(z,p1),s) & ~p1=x -> location(x,result(move(x,p),s) = p
Ep1.can←do(move(z,p1),s) & ~p1=p -> ab aspect9(x,z,p,s)
Ep1.can←do(move(z,p1),s) & ~p1=p -> locaiton(x,result(move(x,p),s) = pp
If the blocks don't fit on top of one another or the obstacles cannot
be removed, then we fail. Other abnormalities can be introduced if we
think of ways to fix such things.
-------
∂01-Nov-83 1009 LGC H-P recommendation
John, I've been pursuing the possibility of a job with Ira Goldstein's
group at the Hewlett-Packard computer research center. Since you are one of
my references, you should expect a call very soon from someone in that group
(probably Steven Rosenberg, a manager who works for Ira). In this connection,
it's a good thing that I invested heavily in a programming effort over the
past two years and have substantially improved programming skills to show for
it, since they need someone who can help them with implementation as well as
evolutionary design of their "Heuristic Programming and Representation
Language". They also have a small natural language group to whose work I
would expect to contribute.
I am seriously interested in this job, and will probably accept it if it's
offered to me. -- Lew
∂01-Nov-83 1932 TREITEL@SUMEX-AIM.ARPA Security
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 1 Nov 83 19:32:08 PST
Date: Tue 1 Nov 83 19:23:18-PST
From: Richard Treitel <TREITEL@SUMEX-AIM.ARPA>
Subject: Security
To: jmc@SU-AI.ARPA, su-bboards@SUMEX-AIM.ARPA
cc: treitel@SUMEX-AIM.ARPA
There probably is no need for a renewed flap, but there could be some need for
a renewed emphasis of the policy, since the set difference between the people
who have some reason to use our machinery and those entitled to get in after
hours is now very large.
- Richard
-------
My opinion is that all those with SCORE accounts should be allowed in
after hours to pick up output. If it is necessary to prevent them from
crowding local terminals after hours, this should be done some other way.
∂01-Nov-83 2004 KJB@SRI-AI.ARPA Lunch?
Received: from SRI-AI by SU-AI with TCP/SMTP; 1 Nov 83 20:04:38 PST
Date: Tue 1 Nov 83 20:05:40-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Lunch?
To: GHG@SU-AI.ARPA
cc: jrp@SRI-AI.ARPA, tw@SU-AI.ARPA, jmc@SU-AI.ARPA
Nov 1, 1983
Dear Professor Golub,
I understand from Terry Winograd that the computer science
department has a few concerns regarding our request for consulting
professorships for some members of CSLI. I wonder if you, John Perry
and I might have lunch one day soon. Perhaps we can answer some of
these concerns, and explain more fully just how we envision ways that
CSLI can cooperate with the computer science department in the coming
years.
Terry said you would be leaving town soon, so maybe we could
make a date for soon after your return. At this time, any day after
November 10 is fine with me except for the 17th and 18th. Hope to
meet you soon.
Sincerely,
Jon Barwise
-------
∂01-Nov-83 2045 @MIT-MULTICS.ARPA:"Roy Goodman Amdahl(EX)."@ODEN.MAILNET Dear John,
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 1 Nov 83 20:45:01 PST
Received: from ODEN.MAILNET by MIT-MULTICS.ARPA with Mailnet id <2614033403732635@MIT-MULTICS.ARPA>; 01 Nov 1983 19:23:23 est
Date: 01-Nov-83 17:56-+0100
From: "Roy Goodman Amdahl(EX)."@ODEN.MAILNET
Reply-to: "Roy Goodman Amdahl(EX)."@ODEN.MAILNET
To: JMC@SU-AI.ARPA
Subject: Dear John,
Message-ID: <28641@QZCOM>
Last time we met, you were ceramically but soberly frying a Datamation
egg, pre-dawn, in the kitchen of Houston's Astroworld Hotel Penthouse
(corridors lined with plastic tree-ranks, walls COVERED with "paintings").
I didn't stay to see the egg's fate, but read something of yours
recently in a yellowing OMNI.
How would you like to dust off your thoughts on your computer mail
terminal?
I'm just starting a self-propelling 5-to-7-year project that, amongst
it's skeptically-spritely activities WILL produce a rather special
(as we say here) communication device for its internationally-conferring
Members during 1986. And annual "throwaway" versions thereafter.
This message, I suspect, is being relayed by a modern analog of The
Pony Express. I hope it (and the pony and rider) survive the journey.
And therefore, I hope to hear from you.
If you still use a telephone, dialling 44-628 32302 reaches the same
home telephone through which I'm now typing this message. I'm
definitely contactable via letters to COM at QZ, Sweden and to
3 Windsor Road, Maidenjead, Berkshire SL6 1UZ, England.
Kindest regards,
(no facsimile yet)
Roy Goodman.
------------
(Text 28641)------------------------------
∂01-Nov-83 2047 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET Testing connection from University of Stockholm
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 1 Nov 83 20:47:31 PST
Received: from ODEN.MAILNET by MIT-MULTICS.ARPA with Mailnet id <2614034561192914@MIT-MULTICS.ARPA>; 01 Nov 1983 19:42:41 est
Date: 29-Oct-83 00:08-+0100
From: "Tommy Ericson QZ"@ODEN.MAILNET
Reply-to: "Tommy Ericson QZ"@ODEN.MAILNET
To: JMC@SU-AI.ARPA
Subject: Testing connection from University of Stockholm
Message-ID: <28377 @ QZCOM>
Hello, I hope that this message reaches you. If you would kindly
reply if this is received to confirm that our gateway is working.
Thank you in advance!
Tommy Ericson,
Stockholm University Computing Center.
(Text 28377)------------------------------
I was unsuccessful in replying to this message. Can you tell me how?
∂01-Nov-83 2258 CLT Calendar items
Wed. 16 Nov 20:00 guarneri (tresidder)
Thu. 17 Nov ??:?? Dinner Dantzig
∂01-Nov-83 2313 Mailer failed mail returned
In processing the following command:
MAIL "Tommy Ericson QZ"%oden Got your message.
The following message was unsent because of a command error:
------- Begin undelivered message: -------
∂01-Nov-83 2313 JMC
Got your message.
------- End undelivered message -------
∂01-Nov-83 2343 ARK
Welcome Party
Come help us celebrate the long awaited, much delayed Welcome Party.
We have a complete set of hosts and hostesses for this event, namely
Arthur Keller, Doug Appelt, Joe Weening, Maureen Tjan, and Noriko
Minaka. This gala event will be on Saturday, 5 November from 8 p.m.
until dawn at 3400 Kenneth Drive, Palo Alto (856-3550), off Greer near
Loma Verde. Maps are posted on most CSD bulletin boards, and extras
are by the receptionist's desk in MJH.
Come prepared for some serious carousing. We will furnish some above
average beer and goodies. Please don't feel inhibited about bringing
your favorite consumables.
Arthur Keller
∂02-Nov-83 0645 OP things being stolen
although I don't necessarily think a change is in order, I thought I'd let
you know that things do get stolen off-hours. I've had three things taken
from my desk top (in 450, an unlockable office)---the one I was most
pissed-off about was taken between 6:15 pm on a Friday, and 6:50am on the
next Saturday morning---it was an irreplaceable picture of my wife when
she was six; presumably whoever took it wanted the cute (but cheap)
plastic frame.
--Oren
∂02-Nov-83 0813 DFH Vistnes AI Qual
To: JMC@SU-AI, buchanan@SUMEX-AIM, lenat@SU-SCORE
Rick asked me to remind you of this. I reserved
room 352 Mon., Nov. 7 from 2 - 4 pm.
∂02-Nov-83 0820 DFH Personnel Meeting
I am scheduled to go to a one-year-after orientation
meeting from 1 - 3:30 this afternoon. I can reschedule
this for a later date if this is inconvenient.
∂02-Nov-83 0932 TREITEL@SUMEX-AIM.ARPA
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 2 Nov 83 09:32:02 PST
Date: Wed 2 Nov 83 09:32:33-PST
From: Richard Treitel <TREITEL@SUMEX-AIM.ARPA>
To: JMC@SU-AI.ARPA
cc: treitel@SUMEX-AIM.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 1 Nov 83 21:29:00-PST
As a matter of practice, I have always done that (and given them a lecture on
CSD policy about keys). But if enough of the faculty agree with you, the CSD
should presumably change its policy.
-------
∂02-Nov-83 1016 LGC Where's Keith Clark?
Do you know Keith Clark's current computer and/or mailing address, or failing
that, a higher-level description of where he's working now?
∂02-Nov-83 1103 NILSSON@SRI-AI.ARPA interview
Received: from SRI-AI by SU-AI with TCP/SMTP; 2 Nov 83 11:02:41 PST
Date: 2 Nov 1983 1102-PST
From: Nilsson at SRI-AI
Subject: interview
To: jmc at SU-AI
cc: Nilsson
John, A Mr. Chris Roper called me yesterday. He is writing a book
on the history of computing languages and has talked with people like
Zuse, Backus, Algol people and others involved in the development of
languages. It will be written for intelligent lay people, and he
wants to include historical anecdotes as well as understandable
descriptions of the languages. I recommended he talk to you about
LISP and gave him your phone number. He sounded like a pretty
competent person. -Nils
-------
∂03-Nov-83 1537 PACK@SU-SCORE.ARPA Towers
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Nov 83 15:37:13 PST
Date: Thu 3 Nov 83 15:35:07-PST
From: Leslie E. Pack <PACK@SU-SCORE.ARPA>
Subject: Towers
To: jmc@SU-SCORE.ARPA
Are you still interested in discussing them (or your new ideas)?
Leslie
-------
∂04-Nov-83 0900 JMC*
Ask Bark whether he has anti-Hoover petition.
∂04-Nov-83 1026 PACK@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Nov 83 10:26:25 PST
Date: Fri 4 Nov 83 10:25:11-PST
From: Leslie E. Pack <PACK@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Fri 4 Nov 83 01:06:00-PST
I'll come by late this afternoon.
-------
∂04-Nov-83 1059 AAAI-OFFICE@SUMEX-AIM.ARPA '86 site selection
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 4 Nov 83 10:58:43 PST
Date: Fri 4 Nov 83 10:54:15-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: '86 site selection
To: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BLEDSOE@UTEXAS-20.ARPA,
BUCHANAN@SUMEX-AIM.ARPA, KRD@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA,
LERMAN@SRI-KL.ARPA, GENESERETH@SUMEX-AIM.ARPA, GROSZ@SRI-AI.ARPA,
HART@SRI-KL.ARPA, JMC@SU-AI.ARPA, MINSKY@MIT-MC.ARPA, NILSSON@SRI-AI.ARPA,
REDDY@CMU-CS-A.ARPA, STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA,
GJS@MIT-MC.ARPA, TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA,
BONNIE.UPENN@UDEL-RELAY.ARPA, FIKES@PARC-MAXC.ARPA,
AAAI-OFFICE@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
At this year's Executive Council meeting, Bonnie Webber (via Aravind
Joshi) submitted a proposal to have AAAI-86 held in Philadelphia.
Prior to the submission of this proposal, it had been suggested that
AAAI-86 be held at UMASS. Marty Tenenbaum was directed to investigate
both sites and then present his recommendations to the EXECCOM. Below
is a comparison of both sites.
Philadelphia UMASS
------------------------------------------------------------------
Facilities
*1-Ballroom for 2500 people; *Concert Hall for 2053 people
this room can be divided into
4 smaller rooms (600)
* 1 ballroom for 700 * Theater for 666
* 1 hall for 55-8'x10' booths * Recital Hall for 200
* 1 ballroom for 425 in the Fine Arts Complex
* lots of small rooms * 1 auditorium for 900
ranging from 30 to 125 * 2 auditoriums 650 each
all in the Franklin * lots of small meeting
Plaza Hotel rooms ranging from 18-100
in the Lincoln Conference Ctr
* We can reserve meeting
rooms at the Philadelphia
Center (formerly the Shearton)
for tutorials
Housing
* 650 rooms at the Franklin Plaza * 116 hotel rooms at UMASS's
Lincoln Conference Ctr
* 150 rooms at the Four Seasons * lots of dorm rooms
* 400 rooms at the Philadelphia * ? local hotel rooms
Center
* 400-500 dorm rooms at UPENN
Transportation
* major metropolitan airport; * no direct flights; Amherst is a 45
10-15 minutes to the hotels minute drive from Bradley Airport in
and dorms Windsor Locks, CT; from Logan, it is
a two hour drive to Amherst
We recommend that Philadelphia be the site for AAAI-86 National
Conference for the following reasons:
1. The Franklin Plaza has a general meeting room that can hold
2500 people. UMASS does not have space that can seat 2000
people;
2. The Franklin Plaza has space for an exhibit program; UMASS has
no designated space for exhibits;
3. Easy access in and out of Philadelphia;
4. Philadelphia site can provide a variety of housing choices -
inexpensive dorm rooms as well as adequate hotel rooms; and
5. The conference can be held in one location (except perhaps for
the tutorials).
We'd like to hear your comments regarding this matter. If we do not
hear from you within the next 7 days, we can assume you concur with
the recommendation.
Regards,
Marty Tenenbaum Claudia Mazzetti
Conference Chair
-------
∂04-Nov-83 2355 long@csnet-cic Mail problem at CSNET-RELAY
Received: from CSNET-CIC by SU-AI with TCP/SMTP; 4 Nov 83 23:55:02 PST
Date: Sat, 5 Nov 83 2:45:32 EST
From: Daniel Long <long@BBN-UNIX>
Subject: Mail problem at CSNET-RELAY
To: JMC@SU-AI
Cc: long@BBN-UNIX
Due to an operations error at csnet-relay on Thursday, November 3, attempts to
send mail to csnet-relay between 18:20 and 23:30 returned the "User Unknown"
error. A log file was saved covering the period 20:30 to 23:30. This log
indicates that at 21:27:21 a message from you to the following recipients(s) was
rejected:
jsv.brown@CSNET-CIC
The error has been corrected and procedures have been established to insure
that it doesn't happen again. We regret any inconvenience this problem may
have caused. We would appreciate your re-sending any messages returned to you
during this period.
Thanks,
Dan Long
CSNET CIC Technical Liaison
∂05-Nov-83 1129 ALMOG@SRI-AI.ARPA talk
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Nov 83 11:29:20 PST
Date: 5 Nov 1983 1126-PST
From: Almog at SRI-AI
Subject: talk
To: jmc at SU-AI
Here is a suggestion of a date: Tuesday 12.6.83 (3.15pm).
I hope it suits you. Independently, I would love to talk
to you on some of your ideas on a common-sense theory. Naively perhaps,
I believe some ideas in the phil.of language may be applicable here.
I didnt want to push myself on you, but plewase tell me if we could talk.
Thanks a lot. Joseph Almog.
-------
∂05-Nov-83 1459 AMAREL@RUTGERS.ARPA Re: '86 site selection
Received: from RUTGERS by SU-AI with TCP/SMTP; 5 Nov 83 14:59:39 PST
Date: 5 Nov 83 17:54:10 EST
From: AMAREL@RUTGERS.ARPA
Subject: Re: '86 site selection
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: BRACHMAN@SRI-KL.ARPA, BLEDSOE@UTEXAS-20.ARPA, BUCHANAN@SUMEX-AIM.ARPA,
KRD@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA, LERMAN@SRI-KL.ARPA,
GENESERETH@SUMEX-AIM.ARPA, GROSZ@SRI-AI.ARPA, HART@SRI-KL.ARPA,
JMC@SU-AI.ARPA, MINSKY@MIT-MC.ARPA, NILSSON@SRI-AI.ARPA,
REDDY@CMU-CS-A.ARPA, STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA,
GJS@MIT-MC.ARPA, TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA,
BONNIE.UPENN@UDEL-RELAY.ARPA, FIKES@PARC-MAXC.ARPA, AMAREL@RUTGERS.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of 4 Nov 83 13:54:15 EST
I agree with the choice of Philadelphia for '86.
Saul
-------
∂06-Nov-83 0032 RPG
∂05-Nov-83 2357 JMC
The files here in ess,jmc now have different names from any at s1.
Transfer to sail:[ess,jmc] complete.
Thanks a lot.
∂06-Nov-83 1340 Fikes.pa@PARC-MAXC.ARPA Re: '86 site selection
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 6 Nov 83 13:40:28 PST
Date: Sun, 6 Nov 83 13:38 PST
From: Fikes.pa@PARC-MAXC.ARPA
Subject: Re: '86 site selection
In-reply-to: "AAAI-OFFICE@SUMEX-AIM.ARPA's message of Fri, 4 Nov 83
10:54:15 PST"
To: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
cc: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BLEDSOE@UTEXAS-20.ARPA,
BUCHANAN@SUMEX-AIM.ARPA, KRD@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA,
LERMAN@SRI-KL.ARPA, GENESERETH@SUMEX-AIM.ARPA, GROSZ@SRI-AI.ARPA,
HART@SRI-KL.ARPA, JMC@SU-AI.ARPA, MINSKY@MIT-MC.ARPA,
NILSSON@SRI-AI.ARPA, REDDY@CMU-CS-A.ARPA, STAN@SRI-AI.ARPA,
STEFIK.pa@PARC-MAXC.ARPA, GJS@MIT-MC.ARPA, TENENBAUM@SRI-KL.ARPA,
WALKER@SRI-AI.ARPA, BONNIE.UPENN@UDEL-RELAY.ARPA,
FIKES.pa@PARC-MAXC.ARPA, AAAI-OFFICE@SUMEX-AIM.ARPA
I agree with the recommendation to have the '86 conference in Philadelphia.
richard
∂07-Nov-83 0852 Hans.Berliner@CMU-CS-A your presidential msg
Received: from CMU-CS-A by SU-AI with TCP/SMTP; 7 Nov 83 08:51:17 PST
Received: from [128.2.254.192] by CMU-CS-PT with CMUFTP; 7 Nov 83 11:38:47 EST
Date: 7 Nov 83 1123 EST (Monday)
From: Hans.Berliner@CMU-CS-A (C350HB03)
To: mccarthy@su-ai
Subject: your presidential msg
CC: Hans.Berliner@CMU-CS-A
Message-Id: <07Nov83.112317.HB03@CMU-CS-A>
John,
In general I am in fairly close agreement with the things you said,
with the exception that I believe you are expecting more from the
world of formal mechanisms that it can ever hope to deliver. However,
I did want to make some comments about computer chess (section 4).
Anyone who belives that the most successful people working on
computer chess are just doing it for sport is very naive. I know
that is what they all say, but all one has to do (as I just did) is
attend a ACM World championship and watch Belle being wheeled in by a
team of technicians that check the voltage on the line, install loads
of whatzits, etc, and consider that Belle's opening book now contains
over 300,000 positions to know that this is not just "sport". This
is far from the only team like this, it was much more the rule than
the exception. I was one of 3 or 4 "teams" (out of the 22 competing)
that was represented by a single person. I felt more as if I was at
the Indianapolis speedway than at a chess tournament. Computer chess
is big business (not commercial but certainly effort) to lots of
people. The CRAY BLITZ people that won were there with three people
who had worked on the program and two students and had open lines to
Cray Research for help as needed. In short a lot of people feel
there is a lot to be achieved in computer chess but they may not be
achieving it the way the founding fathers of AI originally conceived
it would be done. However, what they are doing is only a little
short of amazing. Just as one example, you would be very, very
surprised at what has happened to alpha-beta since you first
formulated the recursive version. It now has iterative deepening,
hash tables, zero-windows, and other things that have made it into a
super-efficient tree searcher. This has been the result of a lot of
people working on the problem (mostly on their own time) and
exchanging results. It is true that not too much has been published
(although all the above have but not in the standard AI literature),
but I know of a recent case where a very fine article on tree
searching in computer chess was rejected by a major AI conference
without even giving a reason. All you have to do is look at the call
for papers for AAAI-84 to see that game playing is not even mentioned
as a topic to see that maybe something is wrong with the AI
establishment rather than chess as a problem.
I for one feel that a great deal of progress has been made. It is
true that it has been made principally by superb programmers rather
than the AI researcher that we all yearn to be. However, if one goes
down the brute-force road, that is what is required. That is not to
say that brute-force has no knowledge; far from it: some of these
programs know a lot (although there is always more to be put in).
In any case, I believe that chess will still be the first place where
a clear breakthru to the top levels of human competence will occur.
It may not come by brute-force (and I am working on methods that are
very different from that ), but it seems that programs of the present
ilk, with about twice the knowledge they have at present and another
factor of 25 in speed would probably be able to compete with the best
players in the world. How long that would take, and whether some
shorter road (such as what I am looking for) will be found is
difficult to predict, but I feel that having a computer among the top
ten in the world by the end of 1990 is an even money shot. I fault
the AI community for having contributed so little to actually making
a good program. It is a lot of hard work, and even though an
infusion of good ideas is necessary for such an endeavor (and I get
the distinct impression that people in AI are more interested in
doing the inventing that in applying the inventions to a development
of a vehicle whose performance can be measured) the people who have
been building the good programs have produced a lot of meaningful
innovations.
Now it is my turn to apologize for the garbling of a lot of ideas.
However, I did want to say that I feel that computer chess is still out
there. It is a VERY difficult problem (a worthy challenge to any
researcher) and we will see that when a program is world champion (or
equivalent) that we will have learned a lot about intelligence. I
wonder if any of our prominent AI institutions or funding agencies is
willing to devote the manpower/resourses to be among the high place
finishers when it is all over.
Hans
Hans,
I agree in the main with what you said in your message and suggest
you put it in a letter to the AI Magazine. Part of what you describe,
however, amounts to computer chess becoming a professional sport rather than
an amateur sport. The criticism you about being more interested in
inventing than applying is applicable to me, but I can't help it.
As you know I have supported chess as an AI domain, although I wasn't
as much help to Dave Wilkins as I had hoped to be.
What would you think of a one class computer chess tournament, e.g. IBM PC
or VAX/750? The idea would be to pick the best widely available computer
and force the competitors to concentrate on programming. I would put no
restriction on amount of memory, because progress will come from
elaboration.
Perhaps even better than a letter would be an article on chess as an
AI domain, spelling out in detail your contention that chess is a good
bet for reaching human levels of intellectual performance first.
∂07-Nov-83 0900 JMC*
Feigenbaum review
∂07-Nov-83 1031 DFH
Room 302
Marlie says she has Prof. Tucker in there through Autumn quarter. She says
Ginsberg could use the third desk that is in there now. I don't know how
practical this is as Paul says he will have a LISP console in there and that
a variety of people will be using it. She says she won't put anyone in there
after Tucker leaves.
∂07-Nov-83 1037 AAAI-OFFICE@SUMEX-AIM.ARPA Next meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 7 Nov 83 10:37:20 PST
Date: Mon 7 Nov 83 10:37:20-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Next meeting
To: Feinler@SRI-NIC.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA,
JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
This is a reminder that the next meeting regarding the On-line abstract/
full text library will be on Friday, Nov 18, 10 am in Rm 252, CS Dept, SU.
Thanks,
Claudia
-------
∂07-Nov-83 1100 JMC*
meeting in Brooklyn
∂07-Nov-83 1118 BUCHANAN@SUMEX-AIM.ARPA Re: Next meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 7 Nov 83 11:16:10 PST
Date: Mon 7 Nov 83 11:13:46-PST
From: Bruce Buchanan <BUCHANAN@SUMEX-AIM.ARPA>
Subject: Re: Next meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: Feinler@SRI-NIC.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA,
JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of Mon 7 Nov 83 10:37:26-PST
I'll be a few minutes late, but I'll be there.
bgb
-------
∂07-Nov-83 1141 DFH
To: "@GROUP.[1,DFH]"
Office keys
For those who weren't here when this occurred, the locks on our offices have
been changed and are now keyed to a master. The old keys will not work.
Unfortunately, they did not make enough master keys for everyone at this point,
so until I can get more made, some of you will still have to make do with
individual office keys. I should have more masters shortly. The new keys are
in my office.
∂07-Nov-83 1214 ULLMAN@SU-SCORE.ARPA Next Seminar
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Nov 83 12:14:09 PST
Date: Mon 7 Nov 83 12:02:22-PST
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Next Seminar
To: CS440: ;
The speaker is our own Steve Lundstrom, 4:15 Thursday, rm. 200-034
Here is the abstract.
PERFORMANCE MODELING AND PROJECTION
IN
HIGH-PERFORMANCE MIMD SYSTEMS
Dr. Stephen F. Lundstrom
ABSTRACT
The ability to project (with reasonable accuracy) the likely performance of a
planned multiple-processor supercomputer is an exceptionally difficult, but
important problem. The talk will discuss this problem from the point of view
of a specific case, the Numerical Aerodynamic Simulator (a system designed to
sustain execution rates in excess of 1 billion floating point operations per
second - 1 GFLOP). The talk will include a brief overview of the system
design. Discussions will center on the performance modelling and analysis
process, the tools and languages developed to support such a process, and the
state of progress to date.
-------
∂07-Nov-83 1232 pratt@Navajo
Received: from SU-NAVAJO by SU-AI with PUP; 07-Nov-83 12:32 PST
Date: Mon, 7 Nov 83 12:31 PST
From: Vaughan Pratt <pratt@Navajo>
To: clt@sail, jmc@sail, owicki@sierra, zm@sail
(Message inbox:277)
Received: from SU-SCORE.ARPA by Navajo with TCP; Wed, 2 Nov 83 09:55:03 pst
Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Wed 2 Nov 83 09:54:40-PST
Date: Wed 2 Nov 83 09:56:11-PST
From: DKANERVA@SRI-AI.ARPA
Subject: CSLI Lectures by Glynn Winskel, Nov. 3-11
To: Pratt@SU-SCORE.ARPA
SPECIAL CSLI LECTURE SERIES BY GLYNN WINSKEL
CSLI announces a special series of lectures by Glynn Winskel,
of the CMU Computer Sciences Department, who will be visiting CSLI
from November 3 through 11. During his stay, he will be using room 27
in Ventura Hall (497-1710). His lectures will be as follows:
1. The CSLI Colloquium, 4:15 p.m., Thursday, November 3, Redwood Hall
"The Semantics of a Simple Programming Language"
The operational and denotational semantics of a simple
programming language are presented and used to illustrate
some basic issues in the semantics of programming languages.
I will try to show how the more abstract concepts of denotational
semantics connect with more basic operational ideas. Specifically,
I will define what it means for the semantics to be equivalent
and indicate briefly how to prove them equivalent. I'll explain
what it means for a denotational semantics to be fully abstract
with respect to an operational semantics. Full abstraction
is a useful criterion for the agreement of denotational and
operational semantics; it has been used particularly in murky
areas like the semantics of concurrency where at present there
is no generally accepted model. I'll motivate the basic concepts
of denotational semantics like complete partial orders (cpo's)
and the continuous functions on them.
2. Working Group in Semantics of Computer Languages,
9:30 a.m., Tuesday, November 8, at Xerox PARC.
Come to lower entrance at 9:25.
"The Semantics of Nondeterminism"
The programming language of the first talk will be extended
to admit nondeterminism. Starting from an operational semantics
there will be three natural equivalence relations between programs
based on their possible and inevitable behaviour. Accordingly
when we move over to the denotational semantics there will be
three different power domains with which to give the semantics.
(Power domains are the cpo analogue of powerset and they capture
information about nondeterministic behaviour of a computation,
roughly the set of values it produces.) With the intuitions
secure (hopefully), we'll turn to a more abstract treatment of
power domains and show how they are used to give denotational
semantics to parallelism. In this talk both the operational
and denotational semantics will use the nondeterministic
interleaving (shuffling) of atomic actions to handle parallelism.
3. Approaches to Computer Languages Seminar, 2 p.m., Thursday,
November 10, Redwood Hall.
"The Semantics of Communicating Processes"
This talk is intended as an introduction to the work of Milner
and associates in Edinburgh and Hoare and associates in Oxford
on programming languages and semantics for communicating processes.
Milner's language Calculus of Communicating Systems (CCS) and
Hoare's Communicating Sequential Processes (CSP) are fairly similar.
Both are based on synchronized communication between processes.
4. Special meeting of C1 group, 3:30 p.m., Friday, November 11,
at SRI, conference room EL369. Visitors whould come to the
entrance of Building E at 3:25 p.m.
"Event Structure Semantics of Communicating Processes"
An event structures consists of a set of events related by
causality relations specifying how an event depends for its
occurrence on the previous occurrence of events and how the
occurrence of some events excludes others. Here we focus on
their use to give a semantics to languages like CCS and CSP.
Event structures capture concurrency as causal independency
and so give a noninterleaving model of concurrent (or parallel)
computations. Adopting a notion of morphism appropriate to
synchronizing processes we obtain a category of event structures
with categorical constructions closely related to those
constructions used by Milner and Hoare. We show how relations
between event structures and other models like Petri nets and
some of the interleaving models of Milner and Hoare, are
expressed as adjunctions.
-------
∂07-Nov-83 1731 YOM Midterm
When and how do you want to give Helene Taran, the student that was
granted an extension on the Midterm, her exam?
I mt her today, and she could take it tommorow either before or during
class (she prefers not to miss class, if possible).
Her net address is I.ikabod9@LOTSA.
Yoram.
Please arrange to give Helene Taran the midterm before class then, since
I want to talk about it in class. Also please be prepared to talk about
the problems you graded. I expect to be done with my grading in time to
leave them for you late tonight or tomorrow morning.
∂07-Nov-83 1905 YOM CS206
The Comp Committee is having an important meeting tomorrow from
12:30 to 2:30. I therefore will only be able to make it for the
last ten minutes of class. Can we postpone the presentation of
the solutions to Thursday, by which time I hope to have solution
sheets ready to hand out?
-yom
∂07-Nov-83 2227 YOM
∂07-Nov-83 2059 JMC
i.ikabod9 doesn't exist at lotsa and neither does i.ikabod
------------------
Sorry about that one, it exists on lotsb (although the class is on
lotsa), and I sent her a message regarding tommorow.
No reply yet.
∂08-Nov-83 1010 ullman%SU-HNV.ARPA@SU-SCORE.ARPA Browne workshop
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Nov 83 10:08:56 PST
Received: from Diablo by Score with Pup; Tue 8 Nov 83 10:04:37-PST
Date: Tue, 8 Nov 83 10:04 PST
From: Jeff Ullman <ullman@Diablo>
Subject: Browne workshop
To: super@score
I have a collection of "position papers" that people have submitted to
the workshop on university/industry/govt. cooperation in the supercomputer
field on the 15th of this month, and I'll mail copies to anyone interested
in seeing them. If you are actually interested in *attending* the
meeting, I'll mail you *two* copies, free of charge.
∂08-Nov-83 1546 RESTIVO@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Nov 83 15:46:18 PST
Date: Tue 8 Nov 83 15:45:42-PST
From: Chuck Restivo <RESTIVO@SU-SCORE.ARPA>
To: jmc@SU-AI.ARPA
would The Formal Reasoning Group be interested in using
higher-performance Prolog systems in it's research ?
-------
What do you have mind?
I'm always interested in what hardware and software are available,
but the Formal Reasoning Group isn't using Prolog at all at
present. Some of us, including me, have written Prolog programs
in order to understand it better.
∂08-Nov-83 1619 RESTIVO@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Nov 83 16:19:41 PST
Date: Tue 8 Nov 83 16:19:16-PST
From: Chuck Restivo <RESTIVO@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 8 Nov 83 15:55:00-PST
I have been talking with people at Silogic about the possibility of
using the evolving family of Prolog and Prolog environments. If
there was sufficient interest at Stanford, and the terms of the
relationship were acceptable from Stanford'perspective I would
consider acting as some sort of liason between us and them. I just
wanted to poll in and see what your altitude might be about this
possibility.
JJW mentioned that LGC was interested in using Prolog for some of
his work. Thought there might be a growing interest.
There is an unconfirmed assertion that DWarren's section of things
is seeking venture finance to build a 1G-LIP Prolog Machine.
-------
I guess you better include me out for the time being.
∂08-Nov-83 1719 CLT
Fri. 18 Nov 20:00 Kronos
∂09-Nov-83 0943 STAN@SRI-AI.ARPA Mind & Action meeting Friday afternoon
Received: from SRI-AI by SU-AI with TCP/SMTP; 9 Nov 83 09:43:37 PST
Date: 9 Nov 1983 0945-PST
From: Stan at SRI-AI
Subject: Mind & Action meeting Friday afternoon
To: jmc at SAIL
cc: stan
John,
Bob Moore suggested you might be interested in attending all or part
of an all-afternoon working meeting of the Mind & Action group of CSLI.
It includes some philosophers (Perry, Bratman, Etchemendy, Almog)
and some computer scientists (Moore, Rosenschein) and will deal with
models of mind (computational and otherwise). An important issue
for me in all this is whether there will be enough technical content
to make it interesting & worthwhile. I know you feel the same way
(because you once decided to skip Bratman's tinlunch because you thought
the paper was technically thin). Even if you don't plan to come,
I would welcome your thoughts on what the main technical issues OUGHT to
be. I am including part of a message I sent out about the meeting.
Some additional focus issues might be: syntactic vs. semantic characterization
of belief states (I'm thinking of Konolige vs. Barwise & Perry situation
semantics); identifying what constraints computation puts on an
epistomological theory. Please let me know if you plan to attend.
Mind & action in retreat
We plan to have a "retreat" next Friday, Nov. 11, from about 12:30
till whenever (e.g. 5 pm). The place will be the Ventura conference
room. The format will be informal, and the participants will include
the following: John Perry, Michael Bratman, Joseph Almog, Bob Moore,
John Etchemendy, Helen Nissenbaum, and Stan Rosenschein.
We can have sandwiches ordered & get down to work. Since we're still
in the stage of discovering one another's interests, we should give
people with something to say the opportunity to say it (e.g. a 15-min
time slice). But so that there will be some focus, we thought it
would be good for people to address themselves to some issues that
were formulated in advance. Off the top of our heads we came up with:
1. What does your research have to say about the causation of action?
2. What is your conception of a mental state? Is that state
viewed as having "components"? What sorts of criteria (e.g.
evidence from NL semantics of prop. attitudes, empirical evidence
from "real" (non-folk) psychology, computational models of knowledge
and action) apply to the decision to include/not-include a
component (e.g. intentions or emotions)?
We need to revise and/or add to this list so that people will be able
to approach the meeting in the right "mental state" and make it
productive.
Helen Nissenbaum & I have been doing such planning as has been done on
this. We welcome suggestions.
--Stan Rosenschein
-------
∂09-Nov-83 1120 TW
TALK TODAY at 2:15 in 380Y
I'll be coming late so go ahead and start it yourself.
Also, there is no transparency projector there, so if you
need one, bbring it from the dept. There is a screen.
--t
∂09-Nov-83 1048 Raj.Reddy@CMU-CS-A Paris Trip
Received: from CMU-CS-A by SU-AI with TCP/SMTP; 9 Nov 83 10:47:58 PST
Received: from [128.2.254.192] by CMU-CS-PT with CMUFTP; 9 Nov 83 13:33:43 EST
Date: 9 Nov 83 1332 EST (Wednesday)
From: Raj.Reddy@CMU-CS-A
To: JMC@SU-AI
Subject: Paris Trip
CC: Raj.Reddy@CMU-CS-A
Message-Id: <09Nov83.133202.RR29@CMU-CS-A>
Dear John,
Sorry I couldn't get back to you at 8:30 your time. I will be going back
to Paris for a few days next Monday. I wanted to know if you have any
specific suggestions about what we should try to accomplish over the next
few months.
Raj
My only suggestion is that they get a local manager for the project. I can
possibly visit in December or January.
∂09-Nov-83 0958 AAAI-OFFICE@SUMEX-AIM.ARPA Future Conference sites
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 9 Nov 83 09:58:22 PST
Date: Wed 9 Nov 83 09:57:57-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Future Conference sites
To: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BLEDSOE@UTEXAS-20.ARPA,
BUCHANAN@SUMEX-AIM.ARPA, KRD@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA,
LERMAN@SRI-KL.ARPA, GENESERETH@SUMEX-AIM.ARPA, GROSZ@SRI-AI.ARPA,
HART@SRI-KL.ARPA, JMC@SU-AI.ARPA, MINSKY@MIT-MC.ARPA, NILSSON@SRI-AI.ARPA,
REDDY@CMU-CS-A.ARPA, STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA,
GFS@MIT-MC.ARPA, TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA,
BONNIE.UPENN@UDEL-RELAY.ARPA, FIKES@PARC-MAXC.ARPA
cc: AAAI-OFFICE@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
We like to solicit your suggestions for conference site locations beyond
1986. A preliminary list might include the San Francisco Bay Area (SU),
Boston (MIT), Chicago-Urbana (Illinois) region. As you prepare your
suggestions, please remember that any facilities should be able to
accomodate atleast 2000 people and inexpensive housing should be available
in the area.
We look forward to hearing your suggestions.
Regards,
Claudia Mazzetti Marty Tenenbaum
Conference Chair
-------
∂09-Nov-83 1705 ME Testing connection from University of Stockholm
∂09-Nov-83 1636 JMC Testing connection from University of Stockholm
I was unsuccessful in replying to this message. Can you tell me how?
∂01-Nov-83 2047 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET Testing connection from University of Stockholm
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 1 Nov 83 20:47:31 PST
Received: from ODEN.MAILNET by MIT-MULTICS.ARPA with Mailnet id <2614034561192914@MIT-MULTICS.ARPA>; 01 Nov 1983 19:42:41 est
Date: 29-Oct-83 00:08-+0100
From: "Tommy Ericson QZ"@ODEN.MAILNET
Reply-to: "Tommy Ericson QZ"@ODEN.MAILNET
To: JMC@SU-AI.ARPA
Subject: Testing connection from University of Stockholm
Message-ID: <28377 @ QZCOM>
Hello, I hope that this message reaches you. If you would kindly
reply if this is received to confirm that our gateway is working.
Thank you in advance!
Tommy Ericson,
Stockholm University Computing Center.
(Text 28377)------------------------------
ME - Try mailing to "Tommy Ericson QZ@ODEN.MAILNET"%mit-multics.
mail "Tommy Ericson QZ@ODEN.MAILNET"%mit-multics/su
Acknowledgement
Here's another try at acknowledging your Oct 29 message received Nov 1.
My first tries were rejected by our mailer.
Include the quotes in what you give MAIL.
Another possibility would be:
MAIL ↓"Tommy Ericson QZ"@ODEN.MAILNET↓%mit-multics
which has two types of quotes in it, namely downarrows and double quotes.
∂09-Nov-83 2034 CLT project j
To: JMC, RWW, SGF, JK, CG, CLT
Event: Meeting to discuss proposal etc.
Time: 11am Friday November 11
Place: CLT office or nearby
∂10-Nov-83 0955 KONOLIGE@SRI-AI.ARPA Oral exam
Received: from SRI-AI by SU-AI with TCP/SMTP; 10 Nov 83 09:55:02 PST
Date: Thu 10 Nov 83 09:56:46-PST
From: Kurt Konolige <Konolige@SRI-AI.ARPA>
Subject: Oral exam
To: nilsson@SRI-AI.ARPA, sgf@SU-AI.ARPA, jmc@SU-AI.ARPA,
genesereth@SU-SCORE.ARPA
Marilyn Walker has informed me that the oral exam is scheduled for
the Building 170 conference room (that's on the History corner of the quad).
The time and date (Tues. Nov. 15 at 2:30pm) remain the same. Please confirm
that you have received this information about the room change. --kk
-------
170 it is.
∂10-Nov-83 1500 AAAI-OFFICE@SUMEX-AIM.ARPA revised date for on-line abstract meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 10 Nov 83 15:00:28 PST
Date: Thu 10 Nov 83 15:00:59-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: revised date for on-line abstract meeting
To: Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA, Feinler@SRI-NIC.ARPA,
Buchanan@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA, JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
The on-line abstract/full text meeting has been changed to Monday,
November 21 at 10:30am in Rm252, CS Dept. If you have problems
with that time or date, pls send me a message.
Thanks,
Claudia
-------
∂10-Nov-83 2117 ullman@Shasta panel
Received: from SU-SHASTA by SU-AI with TCP/SMTP; 10 Nov 83 21:17:52 PST
Date: Thu, 10 Nov 83 21:16 PST
From: Jeff Ullman <ullman@Shasta>
Subject: panel
To: genesereth@sumex, jmc@sail, owicki@sierra, pratt@navajo,
reid@glacier, trattnig@sierra
I'd like to thank you all for teaming up to produce an outstanding
panel discussion. I've received a number of very nice comments, as
I'm sure you have.
∂11-Nov-83 0132 HST VISIT
HALLO JOHN.YOU DID NOT ANSWER ON MY LAST MESSAGE.CAN I INFER YOU
ARE NOT INTERESTED?WHAT ABOUT YOUR INTERVENTION AT DOCUMENT DISTRIBUTION?
HERBERT
Sorry, I have too many messages to answer, and this one evidently
got lost. Could you say again what you have in mind? I kept
forgetting to check with the document person, but I'll do it
today. I am trying to make sure that you are invited to give
a historical paper at the LISP conference in Austin, Texas
next August. Would you need transportatin money or would your
university or some other German source pay for it?
∂11-Nov-83 0824 ALMOG@SRI-AI.ARPA thanks
Received: from SRI-AI by SU-AI with TCP/SMTP; 11 Nov 83 08:24:32 PST
Date: 11 Nov 1983 0822-PST
From: Almog at SRI-AI
Subject: thanks
To: jmc at SU-AI
Thanks very much for your msg. There is no meeting this quarter after
dec.6 and the three before that are booked. Why wont you open next
quarter? this would be Jan.10 or if you cannot other tuesdays in jan.
will do just as well. I look forward for that many thanks for
agreeing to appear.
Next Tuesday at 11 will be just fine. Where should I come to?
and,besides, thanks a lot for having a time to chat with me,Joseph
-------
∂11-Nov-83 1153 sap@Shasta friday testing
Received: from SU-SHASTA by SU-AI with TCP/SMTP; 11 Nov 83 11:50:39 PST
Date: Friday, 11 Nov 1983 11:48-PST
To: shott at Shasta <shott@Shasta>, shott at Glacier <shott@Glacier>,
mccarthy at Shasta <mccarthy@Shasta>
Cc: sap at Shasta <sap@Shasta>, hou at Shasta <hou@Shasta>
Subject: friday testing
From: Steve Przybylski <sap@Shasta>
I had forgotten about he weekly MIPS meeting. It starts at 1:00 and
usually lasts till 2:30 or 3:00. It would be best if we could start
after that. How's that sound?
Steven
∂11-Nov-83 1218 sap@Shasta
Received: from SU-SHASTA by SU-AI with TCP/SMTP; 11 Nov 83 12:18:00 PST
Date: Friday, 11 Nov 1983 12:17-PST
To: John McCarthy <JMC@SU-AI>
Cc: sap at SU-SHASTA <sap@SU-SHASTA>
In-Reply-To: Your message of 11 Nov 83 1213 PST.
From: Steve Przybylski <sap@Shasta>
My apologies. Right name, wrong machine.
∂11-Nov-83 1440 LENAT@SU-SCORE.ARPA ...if you haven't heard it before
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Nov 83 14:40:42 PST
Date: Fri 11 Nov 83 14:36:46-PST
From: Doug Lenat <LENAT@SU-SCORE.ARPA>
Subject: ...if you haven't heard it before
To: winograD@SU-SCORE.ARPA, mccarthY@SU-SCORE.ARPA, buchanaN@SU-SCORE.ARPA,
feigenbauM@SU-SCORE.ARPA, tob@SU-AI.ARPA
cc: lenat@SU-SCORE.ARPA
To qualify my last "invitation" message for you, the AI faculty,
this talk will overlap the invited talks I gave atat IJCAI and AAAI
in August, so you may want to take that into consideration when
deciding whether to come or not.
--Doug
-------
∂11-Nov-83 1622 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET Acknowledgement
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 11 Nov 83 16:21:54 PST
Received: from ODEN.MAILNET by MIT-MULTICS.ARPA with Mailnet id <2614897268941886@MIT-MULTICS.ARPA>; 11 Nov 1983 19:21:08 est
Date: 11-Nov-83 12:37-+0100
From: "Tommy Ericson QZ"@ODEN.MAILNET
Reply-to: "Tommy Ericson QZ"@ODEN.MAILNET
To: "John McCarthy" <JMC@SU-AI.ARPA>
Subject: Acknowledgement
Message-ID: <29666@QZCOM>
In-Reply-To: <29562@QZCOM>
Fine, it got through, your message arrived Nov 11 19.46.
Are there some people at the university of Stockholm or Uppsala
for which you would like to know the mailbox name of? Then just
send a question to our local Postmaster function.
(Text 29666)------------------------------
"Tommy Ericson QZ@ODEN.MAILNET"%mit-multics/su addresses
I don't know the address of your local Postmaster. Would it be
"Postmaster QZ@ODEN.MAILNET"%mit-multics? Anyway I would like the addresses
of Kenneth Kahn and Sten-Ake Tarnlund.
∂13-Nov-83 0005 ALMOG@SRI-AI.ARPA baby's comment on friday
Received: from SRI-AI by SU-AI with TCP/SMTP; 13 Nov 83 00:04:13 PST
Date: 12 Nov 1983 2359-PST
From: Almog at SRI-AI
Subject: baby's comment on friday
To: stan, bmoore, jrp, jmc at SU-AI
cc: almog
In view of Prof. McCarthy's comments on Friday discussion, allow me to
make the following commment on something that personally disturbed me. The
point concerns our discussion of logic and reasoning(Prof.McCarthy had left
by that time). I may feel this way because of my lack of sophistication and
naivete concerning the topic on which I will comment, but still take
it as the baby's comment on some questions that are puzzling to him.
With all due respect to our excitement from T-machine simulations of
reasoning, I feel that at least some, rather elementary lessons, already
drawn by none other than Turing himself in his 1938 Ph.D.
should have been respected in our discussion.
I firmly believe that if a very sloppy amateur in logic like
myself is aware of this,for you, who use daily the jargon of T-machines,
all this is a rather boring old hat. Do correct me if I am wrong.
My point, I should note, has nothing to do with the "big" issues in
the phil. of math. and mind that some of us were quick to jump into( for
a little note on these issues see digression below). Rather, this
has to do with FACTS, facts first pointed out by Turing himself a year
and a half after he himself thought he mechanised all reasoning(in the 37
paper on computable numbers and the Entscheidungsproblem).
OK. The point has to do with simple facts about Turing's ordinal
logics. So,indeed we can add axioms to the original arithmetical system
in such a way that the "true but unprovable" sentences could be
proved. But to stop the cycle we need,ultimately,infinitely many axioms.
Turing's ordinal logics studied such extensions of Arithmetic. In one sense
it was obvious that all ordinal logics werent complete since either they
had infinintely many axioms or else if a finite rule generated those axioms,
Godel's result about unprovability would still hold.
But if you look at his ordinal logics you see that you generate an
axiom by putting in an ordinal formula for another formula.
Now the question becomes: which formulas are ordinal formulas? So the
trick would be here to say: You can prove everything from the axioms, but
you cannot decide mechanically what's an axiom(which formula is
an ordinal one).
You all know Turing's answer(Proc. London Math.soc. 45,1939):
There are complete ordinal logics, but not in the INTENDED sense of
knowing how many applications of "intuitions" would get you to a specific
theorem.
Moreover, what were these "intuitions"?
Well,... again I assume this is no news to you, but still to use
Turing's own words:
"Mathematical reasoning may be regarded rather schematically as the
exercise of a combination of two faculties which we may call "intuition"
and "ingenuity". The activity of the intuition consists
in making spontaneous judgements which are not the result of any conscious
train of reasoning."
Of course, this is not meant to settle any issue. But I have
the rather personal feeling that we arent really helped by quick strong
statements.The same applies to ,what seem to me,rather fast "ideological"
uses of "Church's thesis".
I mean, how could we do that without being very careful in light
of (at least):
(1)Turing's own study of ordinal logics, (2)Godel's own differentation
between a "Church's thesis" on what is M-comptuable from another thesis
on what is HUMANLY-computable( where G. is questioning the
finite mental state hypothesis for HUMAN computations)
and (3) Kreisel's careful and deep analyses of the issue in many varied
ways.(just to give a fascinating instance, consult
Kreisel's 1972 paper on recursive progressions through pi-1-1(I've got
no greek variables) and the very subtle issues involved here when we put
together ordinal logics together with the conjecture about HUMANLY
possible computations). I mean, if a crude outsider like myself
is trying to be cautious, surely the experts should know better,
even if all of us have a specific philosophical axe to grind.
This is not meant to be an argument. Its a FEELING, but nevertheless
perhaps an important one,that audacious claims should be pursued very slowly.
Perhaps I am slow here because of my naivete but still here is my feeling
that very strong conjectures,without LOGICAL basis, have been
put as self-evident. Do correct me if I am wrong.
[DIGRESSION.
My points above are absolutely independent of another topic of the
discussion where I tried to defend the claim that humans,
both in mathematical and (natural) language expressions of their thoughts,
seem to employ modes of thinking that bypass first order canonization.
I also tried to suggest that some of these modes of thinking
seem best formulable in second order systems that arent only incomplete
(for the standard interpretations), but,and this is
more important from my point of view,are not COMPACT.
Again,I have the feeling that we were much too quick here to make
claims on behalf of Montague's work.It is advisable,or so it seems to me,
to study Montague's SYSTEMS and see how careful HE was to distinguish
the compactness(and completeness) of SUB-fragments of intensional
logic(in particular, "predicative IL") from the non
compactness and failure of recursive enumerability of the set of logical
truths of the full second-order modal logic he was using. Do consult
"Pragmatics&Intensional logic" fn.13 and the text of which it is a footnote.
If you are really interested in relying on Montague for
your claims then, at least I feel, this very special man should be respected
for what he was.We have to read his serious work on high-order logic
and the repercussions of the breakdown between provability and logical
truth. In particular the paper in the "the theory of models" collection in
1965 and more so even pp.146-7 of "set theory and higher-order logic" where
things are stated rather clearly as to the PRICE we have to pay
to reestablish completeness in high order systems]
all the best, joseph
-------
I think I need to offer an excuse for having expressed my disappointment
about the direction the "retreat" is taking. I have never done such a
thing before, and I realize that there is no reason why a seminar should
go in the direction that any one participant prefers. However, for the
reasons I expressed, I believe that a connection between AI and philosophy
will not really be made if we continue this way. During the year
1979-80 at the Center for Advanced Study in the Behavioral Sciences,
the group on AI and philosophy took a similar path, and it was
my opinion at the time, and it remains my opinion, that it did not
succeed in maintaining a connection between what was discussed and AI.
This seemed to be more the fault of the AI people than of the
philosophers. I suppose most scientists find philosophical and
methodological discussion of their field congenial, but it usually
remains a mere distraction from work in it. However, I believe that
the benefits of interaction with philosophy for AI are potentially
large, and therefore I want to argue strongly for a more concrete
approach.
∂13-Nov-83 1613 @MIT-MULTICS.ARPA:"Tommy Ericson QZ"@ODEN.MAILNET addresses
Received: from MIT-MULTICS by SU-AI with TCP/SMTP; 13 Nov 83 16:13:44 PST
Received: from ODEN.MAILNET by MIT-MULTICS.ARPA with Mailnet id <2615069476215809@MIT-MULTICS.ARPA>; 13 Nov 1983 19:11:16 est
Date: 13-Nov-83 18:56-+0100
From: "Tommy Ericson QZ"@ODEN.MAILNET
Reply-to: "Tommy Ericson QZ"@ODEN.MAILNET
To: "John McCarthy" <JMC@SU-AI.ARPA>
Subject: addresses
Message-ID: <29862@QZCOM>
In-Reply-To: <29825@QZCOM>
"Postmaster@ODEN.Mailnet"@MIT-MULTICS.ARPA is the canonical name.
Internally that name is changed into "Postmaster at QZ" to simplify
name matching (we do record the names of other postmasters too).
As for the other two: (Review Presentation S T U)
Name: Sten-]ke T{rnlund UPMAIL
Address: Sturegatan 2A Uppsala University P.O. Box 2059 Uppsala
Telephone: 018 - 112559
(This name might be difficult for you to handle, can
be abbreviated to anything that is unambiguos, e.g.
"Sten- T UPMAIL" which does not contain any nationals.)
Name: Ken Kahn
Address: Box 2059, S-750 02 Uppsala
Telephone: 15 54 00 (1840)
Since we have a fuzzy name-matching you could have made your
guesses using heavy abbreviations which would have resulted in
a return letter from our MailHandler informing you about the
ambiguities (if any). Example: a letter sent to "KEN@ODEN..."
would have been rejected since there is a Ken Wilson Edunet too.
My name could be abbreviated to "T Er Q".
Furthermore, this system is a conference system called COM which
runs on the -10 and -20, I think DRF@SU-SCORE has installed a copy.
(Text 29862)------------------------------
∂13-Nov-83 2312 HST VISIT,LISP-CONF,ZEHE
NO I DON'T KNOW HIM.BUT IT'S EXTRAORDIARY THAT HE SPENDS SO MUCH YEARS
IN MEXICO.
IV HEM,IF I GET THE INVITED TALK I BELIEVE I CAN CGET MONEY.
I WILL SPEND SOME DAYS IN HONOLULU NEXT JANUARY (7. IS THE LAST).
I COULD MAKE STATION IN STANFORD IF YOU LIKE. I COULD OFFER THREE
TALKS:
1.THE PROGRAM TRANSFORMATION MODEL OF COMPILATION
2.PROGRAMMING STYLES, PROGRAMMING ;LANGUAGES, SEMANTICS
3.PRINCIPLES OF OBJECT-ORIENTED PROGRAMMING
(THE LAST WAS ACCEPTED FOR THE HICCS-CONFERENCE IN HAWAII.)
THIS COULD BE A CO GOOD CHANCE TO HAVE LOOK IN YOUR STOCK OF MATERIALS.
∂13-Nov-83 2344 POURNE@MIT-MC upcoming meeting
Received: from MIT-MC by SU-AI with TCP/SMTP; 13 Nov 83 23:44:34 PST
Date: 14 November 1983 02:50 EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
Subject: upcoming meeting
To: jmc @ SU-AI, llw @ SU-AI
Bring candidate conclusions and recommendations if possible. We
want to have a report out in time to get into Big Message if at
all possible. Have a chance, anyway...
JEP
∂14-Nov-83 1000 JMC*
when does winter quarter start
∂14-Nov-83 1000 JMC*
coop.xgp
∂14-Nov-83 1000 JMC*
labor to chudnovsky
∂14-Nov-83 1029 RPG
∂14-Nov-83 1025 PW YoW! ARe wE a neT, yEt?!?
To: RPG, ROD
The good news:
The 3600 on the 4th floor are on the screaming yellow 10Meg. They will
try to hook up the robotics 3600 later today. If they do, I can call SPA
to come and set it up.
The bad news:
Len Bosack said it would be December before they try to hook up the extended
console cable.
John: I find working in 433 intolerable. I don't see what is so difficult
about hooking up the extended cables that it takes longer than 2 weeks
to try it. Maybe we should ask Marty to do it? Or perhaps some heavy-handedness
from you towards Len would help?
-rpg-
∂14-Nov-83 1139 DFH student appointment
Lee Tarn wants to see you at 3 pm today. I told
her to come.
∂14-Nov-83 1450 DFH meeting
Please note that there is a meeting of the Search Committee for
a new Dept. Chairman at 2:30 pm on Thurs.
∂14-Nov-83 1503 MWALKER@SU-SCORE.ARPA Kurt Konolige Oral Exam
Received: from SU-SCORE by SU-AI with TCP/SMTP; 14 Nov 83 15:03:13 PST
Date: Mon 14 Nov 83 14:47:01-PST
From: Marilynn Walker <MWALKER@SU-SCORE.ARPA>
Subject: Kurt Konolige Oral Exam
To: JMC@SU-AI.ARPA
cc: mwalker@SU-SCORE.ARPA
The exam will be held in Bldg. 170 - Conference Room at 2:30 p.m.
-------
∂14-Nov-83 1647 ULLMAN@SU-SCORE.ARPA seminar
Received: from SU-SCORE by SU-AI with TCP/SMTP; 14 Nov 83 16:47:05 PST
Date: Mon 14 Nov 83 16:42:39-PST
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: seminar
To: CS440: ;
The speaker this Thursday is Al Davis of Fairchild.
Here is the abstract:
Title: Thoughts on Symbolic Supercomputers
Abstract: The talk will discuss the current status of the effort to
construct an ultra-concurrent symbolic processor at the Fairchild AI
Labs in Palo Alto. The work to date has concentrated on four areas:
1. A programming representation based on frames which
incorporates both logical and procedural behaviors.
2. Static task allocation methods which analyze the program
at compile time to produce a set of allocation modules which
will be assigned to physical resources at load time.
3. A machine architecture of what is hoped to be a single die
homogeneous processor amenable to a wafer scale implementation.
4. Some circuit experiments which are exploring memory structures
to support concurrent symbolic indexing via pattern matching.
The work is in its initial stages and the talk will provide the details of the
the current work, at least to the extent that they exist.
-------
∂14-Nov-83 1656 AAAI-OFFICE@SUMEX-AIM.ARPA Keynote speaker
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 14 Nov 83 16:56:45 PST
Date: Mon 14 Nov 83 16:34:45-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Keynote speaker
To: JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
At this year's banquet, we thought we have a keynote speaker. Can you
please suggest names of good, articulate speakers? Doug Lenat said he
would try to contact George Lucas (no word yet). Joshua Lederberg's
name has been suggested. We (Ron and I) decided that we did not want
a political figure -they are unreliable about showing up.
Look forwaard to hearing your suggestions!
Claudia
-------
I still think George Keyworth would be good to have with Frank Press as
an alternate. However, Keyworth declined because of a vacation which
might recur annually. I don't know whether Frank Press even replied.
My current suggestion is Bobby Inman. It would be worthwhile also to
arrange for a lunch for him with the executive committee. Then we can
interact with him concerning MCC's and its sponsors' interests in AI.
∂15-Nov-83 0212 YM Reminder: CSD-CF town meeting
To: students@SU-SCORE
CC: ME@SU-AI, mrc@SU-SCORE, bosack@SU-SCORE, REG@SU-AI, STR@SU-AI,
BS@SU-AI, LMG@SU-AI, JMC@SU-AI
Reply-To: bureaucrat@score
The CSD Computer Facilities town meeting will take place tomorrow (Wednesday 16
Nov) from 12 to 2pm in room 420-041 (Math/Psychology basement).
We'll try to discuss the most important issues between 12:20 and 1:15 so those
who have classes ending 12:15 or starting 1:15 can be there.
Oren & Yoni, bureaucrats
I will, alas, be in L.A. at the time of the Computer Facilities town meeting.
I would like to be informed of the main issues raised.
∂15-Nov-83 0219 YM
∂15-Nov-83 0216 JMC
I will, alas, be in L.A. at the time of the Computer Facilities town meeting.
I would like to be informed of the main issues raised.
---
I'm not sure that I'll be there but I'll make sure that someone will inform you.
∂15-Nov-83 1059 DEK search committee
Gene suggests maybe having a HPP person on the committee. What do you think?
So far we have lots of people already: You, me, Jeff, Gabriel, Cheriton,
student [probably Malachi], and Gene ex-officio...
I think the committee is big enough. More and we won't find meeting times.
∂15-Nov-83 1324 BOSACK@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 15 Nov 83 13:24:50 PST
Date: Tue 15 Nov 83 13:23:17-PST
From: Len Bosack <BOSACK@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 15 Nov 83 10:11:00-PST
Have a good trip.
Len
-------
∂15-Nov-83 1410 BMOORE@SRI-AI.ARPA Re: baby's comment on friday
Received: from SRI-AI by SU-AI with TCP/SMTP; 15 Nov 83 14:09:48 PST
Date: Mon 14 Nov 83 14:13:22-PST
From: Bob Moore <BMOORE@SRI-AI.ARPA>
Subject: Re: baby's comment on friday
To: Almog@SRI-AI.ARPA, stan@SRI-AI.ARPA, jrp@SRI-AI.ARPA, jmc@SU-AI.ARPA,
brattman@SRI-AI.ARPA, RPERRAULT@SRI-AI.ARPA
cc: BMOORE@SRI-AI.ARPA
In-Reply-To: Message from "Almog" of Sat 12 Nov 83 23:59:36-PST
I think that much of the misunderstanding on Friday was due to
a failure to appreciate (on both sides) what speech acts were being
performed. It is, as I understand it, the research program of AI to
investigate the working hypothesis that human intelligence can be
explained in computational terms. We do NOT take this as
self-evident; rather, that is the substantive hypothesis we are trying
to produce evidence for. If Joseph took me or anyone else to be
claiming that it was self-evident, I apologize for not being clear.
On the other hand, I took Joseph to be suggesting that
technical results about the limitations of first-order logic or formal
systems in general provide a knock-down argument AGAINST this research
program. If that is not what he was saying, then I apologize for
having misunderstood him.
My understanding of the logical results are that they support
neither position; that is, they are completely neutral between the two.
They do not support the AI position, because they show that their are
certain things formal systems cannot do. They do not support the
anti-AI position, because they do not show that people can do anything
that formal systems can't.
In particular, suppose we accept that Goedel's incompleteness
theorem gives us a way of "seeing" a truth of arithmetic not produced
by first-order Peano arithmetic (PA). Of course, we could produce a
stronger formal theory by adding that truth (the Goedel sentence for
PA) to PA. The anti-AI person might argue that he can form the Goedel
sentence of that theory, too, and thus always stay one jump ahead of
any formal system (this is Lucas's argument in his well-known paper).
But as far as I can tell, the only reason that a person can always
construct the Goedel sentence is that THERE IS A MECHANICAL PROCEDURE
FOR DOING SO. Therefore, so can a machine (or a formal system). Now,
somewhere as we ascend the ordinal hierarchy, we must loose mechanical
constructability--otherwise we would have a decision procedure for
arithmetic--but I know of no argument to show that people can go
beyond that point in THEIR ability to apprehend arithmetical truth.
Although it is not really relevant to my own research program,
since I have no committment to first-order logic, I can't resist
commenting on the interpretation of the theorems on the expressive
power of first-order logic. As far as I know, those theorems all
depend on placing restrictions on the non-logical vocabulary of the
logic. That is, it is not quite correct to say that there is no
first-order formula that has the same truth-conditions as a certain
branching quantifier formula; only that there is no such first-order
formula with THE SAME NON-LOGICAL VOCABULARY as the branching
quantifier formula. If we enrich the first-order theory to contain
first-order set theory, it is quite easy to obtain a formula with the
same truth-conditions as any branching quantifier formula. As Quine
has said, higher-order logic is just set theory in sheep's clothing.
Finally, I don't think I made any statement that relied on any
technical results of Montague. I merely refered to his work to point
out that when one hears the term "logic," one should not assume that
the system in question has a complete proof theory, giving Montague's
IL as an example of a logic that was defined as a language with a
model theory long before anyone thought of giving it a proof theory.
--Bob
-------
∂15-Nov-83 1428 ALMOG@SRI-AI.ARPA meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 15 Nov 83 14:28:39 PST
Date: 14 Nov 1983 1801-PST
From: Almog at SRI-AI
Subject: meeting
To: jmc at SU-AI
Could you tell me plaese where should I meet you tomorrow at 11?
I'll be very grateful. Joseph
-------
∂15-Nov-83 1428 ALMOG@SRI-AI.ARPA sorry
Received: from SRI-AI by SU-AI with TCP/SMTP; 15 Nov 83 14:28:49 PST
Date: 14 Nov 1983 2111-PST
From: Almog at SRI-AI
Subject: sorry
To: jmc at SU-AI
I'm sorry to bother but something urgent came up and I'll be grateful
if we can postpone tomorrow's meeting, I7ll have to accompany a friend
to the hospital. I'll be gone to see Kaplan in UCLA till saturday, but
would love to meet with you any time next week. Please set the date, any time will do for me.
I'm especially sorry to miss tomorrow's meeting since I've been thiking
about the "axioms" you have been writing down:
(I mean having seen them from the corner of my eye in your notebook). I
have some thoughts on the whole "methodology". Is the list going
to be exhaustive in any sense? why add this one and not that one? are
there lessons to be learned here from Kemeny's old way of saving Carnap's inductive logic by "meaning postulates"? etc. I hope this of interest to you.
Again, endless apologies for the cancellation tomorrow and please find time for us next week. Thanks again, joseph
-------
∂15-Nov-83 1521 DFH Dental appt.
To: JMC@SU-AI, ZM@SU-AI, yearwood@SU-SCORE
I have a dental appt. at 8 am tomorrow. Should be here by 9. -- Diana
∂15-Nov-83 1652 DFH Brooklyn college
Rohit Parikh called about the conference Carole Tretkoff
called you about previously. They are about to make up
their schedule and need to know if you are coming. His
phone number (home) is 914-833-0288--you can call him
until 8 pm our time tonight. I also gave him your home
phone. If you don't get ahold of each other tonight, and
there is a yes or no answer, could you leave me a message.
∂15-Nov-83 1656 DFH flight reservations
Do you need flights to and from LA again on Saturday?
If so, could you leave me a message with the times you
prefer?
The same flights would be possible, but it would suit me better to leave
after 9:30 and return as soon after 3:30 as possible.
∂15-Nov-83 2126 ALMOG@SRI-AI.ARPA msg
Received: from SRI-AI by SU-AI with TCP/SMTP; 15 Nov 83 21:26:09 PST
Date: 15 Nov 1983 2123-PST
From: Almog at SRI-AI
Subject: msg
To: jmc at SU-AI
I hope you got my message and again I apologuze, hoping you will
have time to meet.joseph
-------
∂15-Nov-83 2324 mailer@diablo add knowledge
Received: from SU-HNV by SU-AI with TCP/SMTP; 15 Nov 83 23:24:43 PST
From: mailer@diablo
Date: Tue, 15 Nov 83 23:24 PST
To: JMC@SU-AI
Subject: add knowledge
Response from Diablo Mail alias handler:
JMC@SU-AI successfully added to the knowledge mailing list.
∂16-Nov-83 0928 JK
(decl subst (type: |ground⊗ground⊗ground→ground|) (syntype: constant))
(defax subst |∀x y z z1 z2.(atom z ⊃ subst(x,y,z) = if z = y then x else z)
∧ subst(x,y,z1.z2) = subst(x,y,z1).subst(x,y,z2)|)
(label substdef)
;you forgot this assumption:
(axiom |∀x y z.sexp subst(x,y,z)|)
(label simpinfo)
;then the obvious attempt works
(ue (phi |λz.subst(x,y,subst(x,y,z)) = subst(subst(x,y,x),y,z)|)
sexpinduction (open subst))
∀X7.SUBST(X,Y,SUBST(X,Y,X7))=SUBST(SUBST(X,Y,X),Y,X7)
;the problem with definitions seems to be caused by a bug:
;by teasing it a little bit, I got PDL-overflows out of it.
;I will look into it.
∂16-Nov-83 2353 YOM Ignorance
I only mentioned it in passing. I have promised you a proof of the S and P
problem, and will deliver. About the wise men, I need to see your `non simple'
version, but I believe we can do it.
The simplest way to describe the non-simple version is to suppose
that the king asks his question to them all together. Thus the king
asks "Do you know the color of your spot?" three times. The hard problem is
to prove that the first two times they all answer "no", and the third
time they all answer "yes".
The easy problem is to prove that they answer yes the third time
assuming that they answer no the first two times.
∂17-Nov-83 0005 YOM ignorance
In that case I'm willing to contract both problems to be delivered
by the end of winter quarter.
Fine, but you may want to look at my solutions - or have you? I'm
not sure this got to you, because of a mailing error.
∂17-Nov-83 0657 STAN@SRI-AI.ARPA Tinlunch & CBCL
Received: from SRI-AI by SU-AI with TCP/SMTP; 17 Nov 83 06:56:29 PST
Date: 17 Nov 1983 0656-PST
From: Stan at SRI-AI
Subject: Tinlunch & CBCL
To: jmc at SAIL
cc: stan
John, I noticed that you are scheduled to select the Tinlunch paper of
December 8. Although it's not usually the case that selectors choose
their own papers (or papers in progress), it's not unheard of either,
and you would certainly get to hear the linguists' views on semantic
issues for CBCL. --Stan
-------
I didn't know about this at all. I'd have to think what paper to select
and would prefer to select something other than my own. At first thought,
perhaps I'd choose a couple chapters of Lincos by Hans Freudenthal.
∂17-Nov-83 0848 DFH
Terminals
Betty Scott mentioned to me a couple of weeks ago that she had once been
given a SAIL terminal which she was fairly sure was purchased on one of
your accounts, and which Arthur Keller subsequently traded with her for
another terminal. (This was about 3 years ago). After checking serial
numbers with Arthur and Martin Frost it appears that it was indeed
purchased on one of your NSF grants. At any rate, if/when you want it
back I presume we can get it somehow. Betty wanted you to be aware of
this.
∂17-Nov-83 0932 @MIT-MC:Hewitt%MIT-OZ@MIT-MC limitations of logic
Received: from MIT-MC by SU-AI with TCP/SMTP; 17 Nov 83 09:32:39 PST
Date: Thursday, 17 November 1983, 12:12-EST
From: Carl Hewitt <Hewitt%MIT-OZ@MIT-MC.ARPA>
Subject: limitations of logic
To: John McCarthy <JMC at SU-AI.ARPA>
Cc: phil-sci%oz at λSCRC|DMλ, psz at MIT-ML, Hewitt%MIT-OZ@MIT-MC.ARPA
In-reply-to: The message of 17 Nov 83 02:42-EST from John McCarthy <JMC at SU-AI>
Date: Thursday, 17 November 1983 02:42-EST
From: John McCarthy <JMC at SU-AI>
To: phil-sci%oz at MIT-MC
Re: limitations of logic
I will argue that lots of axiomatizations (note spelling) are consistent.
So far as I know, the statement that they are inconsistent is entirely
unsupported.
I believe that the thesis of the inconsistency of axiomatizations of expert knowledge in
all established branches of science and engineering is supported by the experience of
people working in fields like medical diagnosis. Perhaps we could get Peter Szolovits to
report on his experiences and those of his colleagues.
I assert, however, that axiomatizations of common sense
domains will require non-monotonic reasoning to be strong enough, and
this may be confused with inconsistency by the naive. Domains of
scientific physics will not require non-monotonic reasoning, because
they aspire to a completeness not realizable with common sense domains.
Hewitt, et. al., probably have a potentially useful intuition, but unless they
make the effort to make it as precise as possible, this potential will
not be realized.
Point of information: How well does circumscription work for inconsistent
axiomatizations?
Of course, I didn't hear Hewitt's lecture, but I did
read the "Scientific Community Metaphor" paper and didn't agree with
anything.
What points in the paper did you find particularly disappointing besides the use
of metaphor?
Indeed I didn't find the paper coherent, but then I don't
think metaphors should be offered as arguments; at most they are hints.
I believe that analogies and metaphors are fundamental to reasoning and argument.
To me logical inference alone (without analogies and metaphors) seems sterile and
incomplete.
My remark about non-monotonic reasoning being needed for formalizing
common sense is similar to DAM's remark about the need for making
closed world assumptions and taking them back. Circumscription generalizes
the usual ways of doing this. Incidentally, I now realize that I would
have found it more interesting to debate about the usefulness of logic
with Carl rather than with Roger Schank, who changed his mind about whether
he was willing to debate this subject. Perhaps at M.I.T. some time if
Carl is willing.
Sure! It would be fun.
Cheers,
Carl
∂17-Nov-83 1137 YOM
got block2 output, will be ready for class.
∂17-Nov-83 1448 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doc Program
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 17 Nov 83 14:48:36 PST
Date: Thu 17 Nov 83 14:50:30-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Post-Doc Program
To: JMC@SU-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I've done some research on post-doc fellowships programs. Below are
my findings.
Several variations on the structure of post-doctoral fellowship
exists.
Based on information from NSF's Fellowship Program Office and
Stanford's Post-Doc office in the College of Engineering , there are
three categories of post-doc programs, such as:
(1) Individual affiliates himself with a senior
scientist in a particular department; he is known
as a "post-doc research affiliate." That person
later is affiliated with a research project and his
benefits are paid by the insititution's staff benefits
pool.
(2) An outside agency sends the institution money to
cover n dollars and to cover student registration fee at
the institution. Hence, as a student, that person becomes
eligible for student benefits.
(3) An outside agency gives the individual n dollars;
that person is responsible in handling their own taxes,
student registration fee (did you know that the student
registration fee for post-docs at Stanford is $350/quarter?)
The NSF post-doc fellowships programs use the third option noted
above. Although there are many NSF post-doc fellowship programs,
there are several features shared by all the programs, such as:
* All arrangements for affiliation are the responsibility of the
Fellow.
* Appropriate Location of study are non-profit institutions and
institutions of higher education.
*Basic eligibility requirements are (1) US citizen; (2) should have
earned a Phd by the beginning of the fellowship tenure; and/or (3)
hold a doctorate for no more than five years.
* General evaluation process is similar; criteria for acceptance are
ability as evidenced by past research work, letters of recommendations,
suitability and availability of the sponsoring senior scientist and
other colleagues, conditions at the institution, likely impact on the
future scientific development of the applicant, and the likely
scientific quality of the research to emerge.
* Conditions of the application remain standard, such as rules
for additional suppliment of assistance.
Differences in the programs are:
* Tenure: 1 year (12 months), 2 academic years' full time, or 1
academic year and 2 academic years' half-time support. No tenure
should be less than 4 months.
* Stipends & Allowances: NSF gives Fellows stipends ranging from
$1500 (NATO post-docs) to $2300 (NSF Math post-doocs). Some
allowances include dependency allowances (say, $100/month) and
special allowances ($3600) for defraying costs associated with the
fellow's research.
According to our accountant, individuals that receive a stipend over
$300 per month are required to pay taxes on the remainder of the stipend.
That rule applies up to 2 or 3 years. Dependency allowances are taxable.
Allowances to defray research costs can be non-taxable if the AAAI establishes
a general account at the institution and prove that others will be using
the account besides the Fellow.
The cost of the establishing a fellowship program would be broken down
as follows:
* Direct Costs: ? number of stipends per year; amount
of each stipend and special allowances; approximate labor
cost to prepare and administer a program (if we decide
to keep it in-house)*; approximate costs for publicity
and the preparation and distribution of the application forms;
travel expenses for the selection committee
If we do decide to establish such a program, we will need to develop
eligibility requirements and selection criteria as well as policies
on supplimental income.
I thought, if the EXECOMM agrees with this idea, we could
establish a two year tenure position. The stipend would be approx
$2,500 per month (so to cover dependency allowances and registration
fees) or $60,000. Also, AAAI could establish an account at an
institution to cover research expenses (say, $3500/2 years). The AAAI
would have to budget per year $50,000 for the administration of the
program and the stipend.
I'd like to hear your thoughts.
Cheers,
Claudia
-------
1. The fellowships should be awarded to individuals, and they should
make the arrangements for place of tenure.
2. If we can afford it, there should be one two year fellowship awarded
each year.
3. What you say about costs seems reasonable to me. The stipend should
be one of the better ones.
4. We should abbreviate the administration to the point where we don't
spend more than $5,000 per year administering it.
5. You should make a report to the Executive Committee with your facts
and my comments.
∂17-Nov-83 1455 @SU-SCORE.ARPA,@MIT-MC:RICH%MIT-OZ@MIT-MC AAAI
Received: from SU-SCORE by SU-AI with TCP/SMTP; 17 Nov 83 14:54:14 PST
Received: from MIT-MC by SU-SCORE.ARPA with TCP; Thu 17 Nov 83 14:33:57-PST
Date: 17 Nov 1983 17:30 EST (Thu)
Message-ID: <RICH.11968430144.BABYL@MIT-OZ>
From: Charles Rich <RICH%MIT-OZ@MIT-MC.ARPA>
To: McCarthy@SU-SCORE.ARPA
CC: Nilsson@SRI-AI.ARPA, RICH%MIT-OZ@MIT-MC.ARPA
Subject: AAAI
John -
I am not quite sure of the etiquette of this
request, but I would like to make it known that I
would welcome being nominated for election as a
Councilor to the AAAI board next year. If I
understand correctly, nominations are made by the
executive committee and the outgoing Councilors.
Having been Tutorial Chair for the past two years, I
had the opportunity to become familiar and involved
with the needs and processes of the organization.
Given that I can no longer afford the large time
commitment which the Tutorial Chair entails, I would
nevertheless like to continue to serve.
Sincerely,
Chuck Rich.
∂17-Nov-83 1619 DEK thanks for your thoughts
I forgot to mention that the dean's office has to clear the ad
before we can publish it, but I expect them to do that tomorrow.
If you want to make sure that Joyce Friedman (or anybody else)
sees it, just send a note to PHY. I'll be writing directly to
Nils, as soon as the ad is cleared.
∂17-Nov-83 1638 DFH Inference Corp.
Chuck Williams confirms Sat. I gave him your flight
times. He will meet you at LAX at upper level of PSA.
∂17-Nov-83 2152 KUO message
∂17-Nov-83 2157 KUO
Prof. McCarthy:
1. I had read the article you gave to me.
2. Prof. Ma Xiwen wrote to me, he asks me to say hello to you
∂17-Nov-83 2218 KUO message
∂17-Nov-83 2228 KUO
Prof. McCarthy: I had read the article you gave to me.
∂17-Nov-83 2233 KUO
Thank you.
∂18-Nov-83 0920 BSCOTT@SU-SCORE.ARPA Comtex Agreement
Received: from SU-SCORE by SU-AI with TCP/SMTP; 18 Nov 83 09:20:30 PST
Date: Fri 18 Nov 83 09:03:02-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: Comtex Agreement
To: JMC@SU-AI.ARPA
cc: BScott@SU-SCORE.ARPA
John, this agreement will have to go to the Office of Technology Licensing
for approval. I have talked with Bill Osborne there, and his first comment
was that all reports belong to the authors, and unless a lot of money is
involved it's not worth getting permission from each one. Anyway, unless
you have some objection, I'll send a copy of the agreement to Bill.
Please let me know.
Betty
-------
Bruce Buchanan negotiated the deal. Please talk to him before doing anything.
∂18-Nov-83 1001 PATASHNIK@SU-SCORE.ARPA csd-cf town meeting
Received: from SU-SCORE by SU-AI with TCP/SMTP; 18 Nov 83 10:01:28 PST
Date: Fri 18 Nov 83 09:50:55-PST
From: Oren Patashnik <PATASHNIK@SU-SCORE.ARPA>
Subject: csd-cf town meeting
To: jmc@SU-AI.ARPA
cc: bureaucrat@SU-SCORE.ARPA
The main things that happened at the meeting (to the best of my
recollection) were:
1) Len talked about what's happening with the various
machines and with accounting for them and for the dover,
2) we talked about new terminals, terminal rooms, and their
distribution, and
3) we talked about computer security and privacy---many
students felt that the user community should at least know what
monitoring of their activities is routinely done, and should at
least be told (after the fact) when something nonroutine is done
involving them. There was also some sentiment that there should
be a user committee, akin to the congressional CIA oversight committee
(or whatever it's called), that would (after the fact, although a
few wanted it before the fact) decide on the appropriateness of
any nonroutine action. Some concerned users may contact some of the
facilites committee members about this.
--Oren
-------
∂18-Nov-83 1027 YM
To: DEK@SU-AI
CC: patashnik@SU-SCORE, cheriton@SU-HNV, JMC@SU-AI,
ullman@SU-SCORE, RPG@SU-AI
Student representative for the chairperson search committee
I did not know in advance when the first meeting was to be held and since I was
in a seminar when you were looking for me, I was not reachable. Oren
substituted for me and since he is willing to do it and I should start getting
out of non-academic activites, we decided that he will be the student member of
the committee.
-Yoni Malachi.
∂18-Nov-83 1028 JF@SU-SCORE.ARPA finding the room
Received: from SU-SCORE by SU-AI with TCP/SMTP; 18 Nov 83 10:28:48 PST
Date: Fri 18 Nov 83 10:28:47-PST
From: Joan Feigenbaum <JF@SU-SCORE.ARPA>
Subject: finding the room
To: jmc@SU-AI.ARPA
1) it was sent to you because you are on the BATS (Bay Area Theory
Seminar) mailing list. my guess is that you are actually on the mailing
list aflb.local, which is one of the addresses in bats@score.
2) most of the people who need directions to CERAS are those who are coming
from Berkeley, IBM, UCSC, and possibly Xerox, and they will be approaching
CERAS from a parking lot which is on the other side of the CERAS builing
from Jacks.
if there is anything you need to know about BATS, please ask me--i am
the stanford local coordinator and general factotum.
joan
-------
Betty, I apologize for the misprint.
∂18-Nov-83 1329 Mailer failed mail returned
The following message was undeliverable to recipient(s):
bsott@SU-SCORE
The remote host gave this response:
No such local mailbox as "bsott", recipient rejected
------- Begin undelivered message: -------
∂18-Nov-83 1129 JMC
To: bsott@SU-SCORE
Bruce Buchanan negotiated the deal. Please talk to him before doing anything.
------- End undelivered message -------
∂18-Nov-83 1632 DFH
Messages
1. Doug Ferguson, Univ. Library System, 7-9724. Wants to have lunch
some day next week.
2. David Chudnovsky called
3. Tom Zito, writer for the New Yorker will try to get ahold of you Monday,
wants to talk with you for an article he is doing.
4. A packet arrived by Federal Express from Inference. I gave it to
Carolyn to give to you.
∂18-Nov-83 1647 Brachman@SRI-KL.ARPA Keynote speakers for AAAI-84
Received: from SRI-KL by SU-AI with TCP/SMTP; 18 Nov 83 16:47:02 PST
Date: Fri 18 Nov 83 13:45:49-PST
From: Ron Brachman <Brachman at SRI-KL at SU-DSN>
Subject: Keynote speakers for AAAI-84
To: JMC at SU-AI
cc: AAAI-Office at SUMEX-AIM
Hi, John,
Claudia Mazzetti and I (as Program Chairman) have been considering the
keynote speaker issue for a while, and still aren't sure what the best
course of action is. I have a certain amount of hesitation at inviting
Bobby Inman to play too large a role at the conference - while in a
certain respect he is an obvious choice, with the conference in Austin,
as far as I can make out MCC is a private endeavor, and I am afraid of
serving private interests with our scientific organization. (In
particular, MCC is in competition with other AI institutions for people,
it is not a university operation, and it's hard to see why it should be
favored over other private supercomputer efforts.)
In addition, we have been making contact with a couple of other
candidates (although there's no reason in the world that the conference
couldn't use more than one invited speaker - would it be appropriate for
Inman to give an invited presentation on MCC?). In particular, we are
trying to get George Lucas to give an after-dinner address; failing
that, we would like to invite Josh Lederberg. Would you have any
objection to either of those two?
Please let us know if you have any particularly strong feelings on the
matter, and we will keep you posted on the status of our negotiations.
Thanks,
Ron Brachman
-------
I have no objection to Lederberg or Lucas, but also no special reason
to believe that they would be very informative. My motivation for
recommending Inman for something is to smoke out what this collective
of computer manufacturers has to say to an AI audience. I don't think
even being the keynote speaker would give MCC an unfair advantage.
∂18-Nov-83 2128 LLW@S1-A Missed Connections Regrets
Received: from S1-A by SU-AI with TCP/SMTP; 18 Nov 83 21:28:37 PST
Date: 18 Nov 83 2130 PST
From: Lowell Wood <LLW@S1-A>
Subject: Missed Connections Regrets
To: jmc@SU-AI
CC: LLW@S1-A
∂17-Nov-83 2204 JMC
See you tomorrow. I'll be in 10 to 5:30.
[John: Certainly sorry I missed you. I didn't see this note until now,
and had no idea that you would be leaving (relatively) early today. I got
tied up this PM with problems of an upcoming experiment, and didn't get
back to my office until 1800. I looked around for you, figuring that you
had gone out for dinner with S-1ers, and then decided that I'd definitely
see you when you returned. I was beginning to wonder to where you had
vanished, when I sat down to read my mail. . . . Hope to see you soon!
Lowell]
I had a concert to go to in San Francisco, which is why I deviated from
my usual custom. There's a message for you with a further note on
parallel LISP on Paula's desk. No doubt she'll get it to you. They
asked me to do the security forms again which I did. I hope it means
they intend to proceed with the clearance procedures this time.
Person Job Jobnam Idle Terminal
LLW Lowell Wood 10 E 4 !TTY-6 DSwitch Foonly6
PMF Mike Farmwald 3 E 413. TTY-12 DSwitch Foonly12
6 E 437 TTY-2 DSwitch Foonly2
RPG Dick Gabriel 12 E 427. TTY-3 DSwitch Foonly3
SYS system files 2 FINGER PTY21 SU-AI
↑C
.
∂19-Nov-83 2201 RPG Blocks
I looked at BLOCK2.LSP[F83,JMC] and decided that the programming
style per se was pretty modern as it was. If, however, you want to
include as `style' the overall solution, then it has to be completely
re-written, and the modern style is not as concise as what you have.
To be more specific, look at BLOCK2.LSP[1,RPG] in which I've defined
situation-plans, towers, etc abstractly (as macros in this case)
and then re-done the program using these macros. The effect is that
it is now easy to modify the underlying structure and it is easy to
understand the program. I also took the liberty of including the
on-table requirement, which also simpified the program by making it
more general (now there is no need to test for eq to TABLE and NULL).
The last page is the program being applied to some test cases in a
clearer setting than your test cases.
The new program is about 3 times as long due to the macros, but the
main program part is a little shorter.
The upshot is that you do have to pay a price for the ability to understand
what you're doing to be able to modify the code later.
-rpg-
∂19-Nov-83 2316 ALMOG@SRI-AI.ARPA meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 19 Nov 83 23:16:27 PST
Date: 19 Nov 1983 2313-PST
From: Almog at SRI-AI
Subject: meeting
To: jmc at SU-AI
I7m sorry not to have called on thursd. but I have been
down to ucla to see kaplan. Please tell me what time is good
for you. anytime will do for me(except tuesd.3-6 pm). Many
thanks, joseph
-------
∂20-Nov-83 0230 POURNE@MIT-MC Upcoming meeting
Received: from MIT-MC by SU-AI with TCP/SMTP; 20 Nov 83 02:29:52 PST
Date: 20 November 1983 05:22 EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
Subject: Upcoming meeting
To: DANNY @ MIT-MC, MINSKY @ MIT-MC, POURNELLE @ MIT-MC, LLW @ SU-AI,
JMC @ SU-AI
Idea;lly we want to connect tomorrow morning (or at least FY84)
with 30 years from now.
Please bring candidate Conclusions and Recommendations for
debate by full group.
Hans M will be t here, and we have good avenues into right
places.
It looks to be the most important yet.
∂20-Nov-83 0949 NILSSON@SRI-AI.ARPA (Response to message)
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Nov 83 09:49:41 PST
Date: 20 Nov 1983 0950-PST
From: Nilsson at SRI-AI
Subject: (Response to message)
To: JMC at SU-AI
cc: NILSSON
ok.
-------
∂20-Nov-83 1018 BSCOTT@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 20 Nov 83 10:18:42 PST
Date: Sun 20 Nov 83 10:18:48-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
cc: BSCOTT@SU-SCORE.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Fri 18 Nov 83 17:12:00-PST
Thanks, John. I will talk with Bruce about it.
Betty
-------
∂20-Nov-83 1501 CLT
i'm going to play with glb at the knoll, back around 5 or 6
∂21-Nov-83 0543 WONG@SU-SCORE.ARPA Rotation number
Received: from SU-SCORE by SU-AI with TCP/SMTP; 21 Nov 83 05:43:40 PST
Date: Mon 21 Nov 83 05:43:15-PST
From: Ping Wong <WONG@SU-SCORE.ARPA>
Subject: Rotation number
To: mccarthy@SU-AI.ARPA
I am deeply sorry that I have been so delinquent in the project, but
there always seem to be other things of greater urgency. And this quarter
I barely have any breathing time. Even though the project is not one that
I can sink my teeth into, still I would hate to abandon it. And there are
several things I want to try e.g. checking the gradients of the curves that I
have collected and see if there is any relation to be found or whether they
are related to any of the constants in Feigenbaum's paper. I hope to be
able to spend some time on it during the Christmas vacation.
-------
OK. Come and see me when you are ready to work on it again.
∂21-Nov-83 1033 AAAI-OFFICE@SUMEX-AIM.ARPA Reminder
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 21 Nov 83 10:33:05 PST
Date: Mon 21 Nov 83 09:03:45-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Reminder
To: JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
This is a reminder that we have a meeting today at 10:30 in rm 252 in the
CS dept regarding the on-line abstract/full text system.
Regards,
Claudia
-------
∂21-Nov-83 1216 RPG
block3.lsp[1,rpg exists with the updated functionality. Note that
FIND = DEST1.
∂21-Nov-83 1235 GENESERETH@SUMEX-AIM.ARPA rough draft
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 21 Nov 83 12:35:29 PST
Date: Mon 21 Nov 83 12:35:26-PST
From: Michael Genesereth <GENESERETH@SUMEX-AIM.ARPA>
Subject: rough draft
To: AAAI@SUMEX-AIM.ARPA, Amsler@SRI-AI.ARPA, Buchanan@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Genesereth@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA,
Walker@SRI-AI.ARPA
Dear Colleagues,
Since we didn't meet today, I'd like to circulate a rough draft of
the document I was going to distribute. It is obviously rough in
formas well as content. Please send me comments, and I'll try
to incorporate them before our next meeting. Thanks.
mrg
The Library of Artificial Intelligence
The field of Artificial Intelligence has grown extremely
rapidly of late, and the trend is likely to continue for years to
come. The expansion of research activity means a growth in the
literature of the field and a corresponding increase in the difficulty
of keeping abreast of important developments. In order to deal with
this problem, the AAAI is planning to create an online literature
service for AI researchers.
The key component of the plan is the development of an "online
library" of ALL AI literature -- the "Library of Artificial
Intelligence". The Library will include the full text of AI books,
journals, conference proceedings, theses, technical memos, technical
reports, etc. Hopefully, its collection will also include relevant
documents from closely allied fields like logic, linguistics, and
psychology.
[those we can get]
In order to assist its users in finding relevant documents,
the Library will provide a variety of search facilities. Initially,
there will be author, title, keyword, and citation indices for all
documents and a search program to retrieve documents satisfying an
arbitrary boolean combination of these features. Editing facilities
will be available to allow users to preview and abstract documents
before printing or transferring copies to their local machines.
Additional research and development will be conducted to improve these
capabilities.
Although housed on a central computer facility, the Library
will be remotely accesible via telephone and network connections.
The search and editing programs will be available to all users on the
host machine. Users will be able to get citations, abstracts, or the
full text of all documents both online and in hardcopy.
This paper describes the Library in detail and discusses some of
the key issues.
1. Hardware
The hardware for the Library includes appropriate storage to
house the collection, high quality printers, good telephone and
network connections, and an adequate host machine.
At this point in time, the most appropriate storage technology
appears to be disk or possibly videodisk. Semiconductor and bubble
memory are too expensive for the amount of information that must be stored.
Tapes require operators to mount and are slow to search. Disks have
adequate capacity and are reasonably economical. Consider the following
simple analysis.
[The number of bytes seems inflated.]
A typical page in an IJCAI proceedings contains about 10K
characters or approximately 10K bytes of information. Consequently,
the 1000 pages in a typical proceedings could be stored in 10 Mbytes.
A medium-sized textbook like Rich's Artificial Intelligence can be
stored in about 5 Mytes. The internal memos of a medium-sized
laboratory like the Stanford Heuristic Programming Project can be
stored in approximately 20 Mbytes. Adding up these figures for all
of the conferences held in 1983, all of the AI books published in 1983,
and the publications of the major laboratories in 1983 comes to a total
of approximately 200M bytes.
A DEC RP07 can store 600 Mbytes; in 1983 the list price for an
RP07 is $?. The forthcoming DEC RA81 can store 1.5 Gbytes and is
expected to cost around $50000. A videodisk can store ? Mbytes and
costs approximately $?.
For Library users requesting hardcopy, high quality printers
will be essential. A laser printer like the Dover and the Alphatype
are two strong possibilities. This needs more study.
[The Alphatype is not a possibility for printing single copies].
[The Xerox 2700 at $19,000 seems reasonable.]
Ease of access to the Library is critical to its success. The
current plan is to have the Library available on as many computer networks
as possible, at the very least including the Arpanet, Tymnet, and CSnet.
In addition, there will be several telephone ports for direct dial access.
A host machine for the Library must be able to support
its disks, printers, and terminal connections, and it must be able to
run all of the software. Since there is no current estimate on the
level of usage, it is difficult to know how much compute power this
machine must have.
At present a firm hardware plan has not been determined. One
possibility is to contract with a commercial like Dialog or Comtex.
Another alternative is to set up the Library on a community service
machine like NIC or ISI. Alternatively, it is possible to acquire a
new or old machine dedicated to Library.
2. Collection of Documents
Essential to the success of the Library is the assembly and
maintenance of a comprehensive and up-to-date collection. Appropriate
documents must be gathered and stored. In some cases, documents may
have to be reformatted to be readable online and compatible with the
Library's printers. The documents also need to be indexed for
retrieval by the search program.
[Richard M.]
Collecting future documents for the library should not be a
difficult problem. Most AI researchers prepare their papers online.
(For example, of the approximately 250 papers submitted to AAAI-83, all
but about a dozen were produced on computers.) The result is that
acquiring a document for the library is simply a matter of a network
file transfer or, in the worst case, a tape sent in the mail. (See
the comments below on copyright issues.)
The collection of past documents and documents from other
fields is more problematical. Many papers may have never been online;
the online versions of others may have been lost. In such cases the
only recourse is to have such papers retyped or to provide them only
in hardcopy.
[There is OCR.]
A key problem in providing an online Library is formatting.
- desires of author
- compatibility with printers and terminals
Possible Solutions
- standard format e.g. SCRIBE (but how do figures)
- Multiple formats e.g. scribe, TEX, etc.
- New format developed
At any rate get the documents on line one way or another.
Finally, there is the problem of indexing. It is not
difficult to automate the job of indexing papers by author, title, and
citation. Keyword indexing is more difficult.
In most cases, keywords are not available and must be obtained
from an analysis of the text. One solution is for the Library to
employ individuals to do the analysis and indexing. This is expensive
but assures a consistent result Another solution is to ask authors
to supply keywords with all documents entered in the Library. This is
inexpensive but is likely to yield inconsistent results and places an
extra burden on contributors to the Library. A third solution is
automatic indexing on all non-common words in the title and abstract.
This is also inexpensive but may not yield an index of acceptable
quality.
Another important part of indexing is the establishment of a
data base relating different keywords to each other. This data base
is important in helping users and the search program in finding
related keywords to use in searches. Setting up and maintaining this
data base will require the participation of AI experts.
3. Search Service
The initial library search program is intended to resemble any
of a variety of commercially available search programs. A user will
be able to specify an arbitrary boolean combination of features,
including author, title, keywords, and citations. The keyword
data base will be available to assist the user in selecting
appropriate terms.
Once a set of papers is found, a user will be able to examine
the list and any of its entries; and, if the list is inappropriate, it
will be possible to specify additional conjuncts or disjuncts and
continue the search. For each entry in a search list a user may
choose to look at a simple citation, an abstract, or the document's
full text.
In order to facilitate the examination of search lists and
documents, the Library will also provide an editing program like
Emacs. Of course, in order to be editable, the document must be
in an appropriate format.
[Simple title and author retrieval is useful for many purposes. It
is better to start with this.]
[Some of the plans seem more expensive than AAAI can afford.]
5. Examination of Documents
Once a user has completed his search, he may choose to examine
the documents on the host machine. In some cases he may only want to
browse through a document, extract a quote or citations, or retrieve a
particular section. Alternatively, he may wish to transfer
a copy of the document to his home machine or have a hardcopy sent to
him.
The possibility of online examination of documents raises the
issue of terminal compatibility.
Another issue is that of copyrights and royalties. Authors
and publishers may be unwilling to contribute their documents to the
library if they aren't reimbursed for all references. Consequently,
an appropriate charging mechanism must be set up. Also, anappropriate
royalty policy must be negotiated with contributing authors.
5. Financing
Startup Costs
Storage
Access software
Search software
Getting old documents on line
Indexing
Documentation
Maintenance Costs
Incremental Storage
Maintenance of data base
Indexing new documents
Sources
DARPA
AAAI
Fees (accounts, startup accounts at AAAI office)
possible yearly fees
6. Participants
At this stage several institutions are involved in setting
up the Library.
The primary institution is the American Association for
Artificial Intelligence. The AAAI plans to coordinate the
establishment of the Library including fundraising and, if necessary,
secretarial or programming support during its development and
thereafter.
The Network Information Center (NIC) why?
[NIC should for the present be a potential supplier, i.e. a competitor
with the commercial services.]
Several researchers from the Stanford Research Institute are
interested in investigating ways in which to improve its long range
organization and utility of the Library.
Several faculty members from Stanford University are involved
in overseeing the Library's development, and it may be appropriate to
establish the Library at Stanford. The school has a strong program in
AI, and there is an active library research community.
7. Request for Proposal
The AAAI is willing to entertain proposals to set up and
run the Library.
Proposals should describe the Library in detail, including
both hardware and software.
Proposals should include a discussion of all legal and business
issues.
Proposals should include full financial details for both startup
and maintenance of the Library.
Proposals should include a statement of long term commitment
to the Library. There should be evidence of the competence and
long term survivability of the company. A company history and a record
of other activities would be excellent.
-------
∂21-Nov-83 1443 @MIT-MC:RICKL%MIT-OZ@MIT-MC Re: limitations of logic
Received: from MIT-MC by SU-AI with TCP/SMTP; 21 Nov 83 14:43:13 PST
Date: Mon 21 Nov 83 17:27:20-EST
From: RICKL%MIT-OZ@MIT-MC.ARPA
Subject: Re: limitations of logic
To: JMC@SU-AI.ARPA
cc: phil-sci%oz@MIT-MC, phw%MIT-OZ@MIT-MC.ARPA, dughof%MIT-OZ@MIT-MC.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Thu 17 Nov 83 02:44:50-EST
Date: 16 Nov 83 2342 PST
From: John McCarthy <JMC@SU-AI>
Subject: limitations of logic
I will argue that lots of axiomatizations (note spelling) are consistent.
So far as I know, the statement that they are inconsistent is entirely
unsupported. I assert, however, that axiomatizations of common sense
domains will require non-monotonic reasoning to be strong enough, and
this may be confused with inconsistency by the naive. Domains of
scientific physics will not require non-monotonic reasoning, because
they aspire to a completeness not realizable with common sense domains.
Yes, lots of axiomatizations are consistent, but the existence of such
is not under debate. What is debated is "consistent axiomatization(s)
of expert knowledge of non-trivial domains". Below I will try to
support the claim.
In passing, note that in virtue of:
(a) Turing equivalence;
(b) the formal axiomatization of a Turing machine;
it follows that as much of science as is capturable by A.I. can also be
formally axiomatized. Since I believe that much of science is so
capturable, I am formally in complete agreement with you. Observe also
how this formal, axiomatic agreement obscures the real differences.
Hewitt, et. al., probably have a potentially useful intuition, but unless they
make the effort to make it as precise as possible, this potential will
not be realized.
I agree.
Of course, I didn't hear Hewitt's lecture,....
Carl announced at his talk that the gist of his remarks, in a somewhat
preliminary form, are in "Analyzing the Roles of Descriptions and
Actions in Open Systems", M.I.T. A.I. Memo 727, April 1983, and were also
presented at AAAI'83 and are in the proceedings.
-=*=- rick
================ longer message follows ================
PREAMBLE: The *attempt* to formalize is an essential and indispensable
component of science. To argue against the formal achievability of
consistent formal axiomatization is *not* to argue against the practical
utility of the attempt to achieve it. Logic is *not* useless.
Briefly, support for the claim:
(1) Empirical, as Carl suggested. In your reply you assert that the
claim is "entirely unsupported", but I notice that you stop short of
exhibiting a counter-example.
(2) The ubiquitious nature of real-world anomalies and exceptions.
Knowledge of the scientific facts of a domain must be counted as part of
the expert knowledge of a scientist. This includes knowledge of
anomalies and exceptions to theory. Any axiomatization which admits of
a known anomaly is inconsistent with what is known by experts to be
true, and so can hardly be called a consistent axiomatization of the
expert knowledge of that domain.
(3) The observed absence of an asymptotic convergence to a single
consistent axiom set.
Instead, science seems to progress by incremental refinement punctuated
by revolutionary revision, and shows no sign of doing otherwise. Thus,
even assuming that at any one point a consistent axiomatization of
scientific knowledge in a domain were to be produced, that axiomatization
would slowly become more and more inadequate until it was simply
discarded or superceded (with the same fate awaiting its successor).
In practical terms: this means that the surrounding theoretical
framework is likely to be replaced first, before consistency within that
framework can ever be achieved.
(4) The general failure of adjoining ceteris paribus clauses as a
strategy for achieving or maintaining scientific consistency.
Inconsistencies are patched by adjoining ceteris paribus clauses to the
axioms, but new anomalies always arise requiring new ceteris paribus
clauses. This I take to be the application to science of Carl's point
about Perpetual Inconsistency: if a particular inconsistency is patched
the resulting axiomatization will be inconsistent.
Non-monotonic reasoning is a structured mechanism for applying ceteris
paribus clauses, and it is not clear that it renders the resulting
system any more consistent than adjoining ceteris paribus clauses to a
scientific theory. For common-sense reasoning this is probably more
than adequate, however.
(5) The difficulty of *formally* separating "expert knowledge" from
"expert belief".
Any *inductive* law is only "believed with great strength". There is no
*formal* demarcation in the gradation of belief from "laws" to "accepted
wisdom". It is unlikely that any formal axiomatization which includes
the latter will be consistent, but unclear that it can be totally
excluded even in principle.
(footnote: this does *not* assert that "accepted wisdom" is acceptable
as the final goal of science. the fact of twilight does not mean that
day and night are indistinguishable. this does assert that there is not
a clearly defined separation of scientific knowledge and scientific
belief with respect to inductive laws.)
(6) The remarkable sparseness of attempts in science to do this at all.
(This may be more due to the fact that scientists really don't much care
about them in any practical sense.) Even the best-known attempt
(Newtonian mechanics) suffers from formal ambiguity. The second law may
be variously interpreted as a "law" (relating the primitive notions of
force, mass, and acceleration) or as a "definition" (of force, in terms
of the primitive notions of mass and acceleration).
(7) The frequency with which a new, initially less consistent and
successful theory displaces an older, more consistent one.
Consistency seems to be a poor measure of a theory. In particular,
consistency is not necessarily required in order for a theory to become
generally accepted. Galileo was never able to explain why the earth was
not subject to perpetual hurricane-force winds, if it really did move;
the prediction that oxygen and nitrogen should separate and settle out
in layers, if they really were a mixture instead of a compound, was an
embarassment to Dalton's theory for years. Examples like this abound.
(8) Inconsistencies in scientific knowledge itself.
For example, any attempt to formally axiomatize neurophysiology at this
point would be ridiculous, and also unhelpful. Which axiom set you
would get would depend not only on which scientist you asked, but also
on when you asked her. It is not at all clear that a plethora of
conflicting axiom sets is formally more consistent than a single
inconsistent one. Nor is it clear that individual scientists engaged in
active research always hold consistent beliefs about their domain.
For science to progress, indeed, it seems necessary that individual
scientists hold beliefs which do conflict with those of their
colleagues.
(9) It has not been found to be useful as a source of scientific
progress (recall the distinction between the axiomatization and the
attempt, above).
There is no important scientific discovery I am aware of which followed
as a direct result of the formal axiomatization of a scientific theory.
(Of course, "formal axiomatization" does *not* mean "application of
mathematical techniques to", but that much is obvious.)
One of several reasons EURISKO is fascinating is because it may provide
a counter-example. The 3-D VLSI discovery is impressive.
-=*=-
-------
∂21-Nov-83 1547 @SU-SCORE.ARPA:oliger@navajo Visit of Ben Gerber of Cray Res. and NSF Initiative
Received: from SU-SCORE by SU-AI with TCP/SMTP; 21 Nov 83 15:46:58 PST
Received: from Navajo by SU-SCORE.ARPA with TCP; Mon 21 Nov 83 15:41:23-PST
Date: Mon, 21 Nov 83 15:41:36 pst
To: super@Score
Cc: ferziger@Navajo, oliger@Navajo
Subject: Visit of Ben Gerber of Cray Res. and NSF Initiative
From: Joseph Oliger <oliger@navajo>
Ben Gerber of Cray Research wants to meet with parties who have an
interest in using/acquiring a large scale computer with the thought of
coordinating a responce to an NSF initiative to fund class VI machines
for the university community. I suggested that he might attend the
next Wed. lunch meeting of the CLaSSiC research group which will be
held this Wed. (11/23) at 12:15 in MJH 252. Any "Super" members
who are interested are welcome to participate.
Joe Oliger
∂21-Nov-83 1644 AAAI-OFFICE@SUMEX-AIM.ARPA Post-Doctorate program
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 21 Nov 83 16:43:50 PST
Date: Mon 21 Nov 83 16:42:14-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: Post-Doctorate program
To: AMAREL@RUTGERS.ARPA, BRACHMAN@SRI-KL.ARPA, BUCHANAN@SUMEX-AIM.ARPA,
BLEDSOE@UTEXAS-20.ARPA, KRD@MIT-MC.ARPA, BENGELMORE@SRI-KL.ARPA,
LERMAN@SRI-KL.ARPA, GENESERETH@SUMEX-AIM.ARPA, GROSZ@SRI-AI.ARPA,
HART@SRI-KL.ARPA, JMC@SU-AI.ARPA, MINSKY@MIT-MC.ARPA, NILSSON@SRI-AI.ARPA,
REDDY@CMU-CS-A.ARPA, STAN@SRI-AI.ARPA, STEFIK@PARC-MAXC.ARPA,
GJS@MIT-MC.ARPA, TENENBAUM@SRI-KL.ARPA, WALKER@SRI-AI.ARPA,
BONNIE.UPENN@UDEL-RELAY.ARPA, AAAI-OFFICE@SUMEX-AIM.ARPA,
FIKES%USC-ECLD@USC-ECL.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
Dear Colleagues,
In recent days, there's been some general discussion abou the AAAI establishinga Post-doctorate program. We would like to solicit your comments on this
proposal. Also, we've prepared a brief report on different types of post-doc
programs which we would also like you to read and comment.
Look forward to hearing your comments.
Regards,
Claudia
Several variations on the structure of post-doctorate fellowship programs
exists. Based on information from NSF's Fellowship Program Office and
Stanford's College of Engineering Post-Doc program office, there are
three general categories of post-doc programs, which are:
(1) Individual affiliates himself with a senior
scientist in a particular department; he is known
as a "post-doc research affiliate." That person
later is affiliated with a research project and his
benefits are paid by the insititution's staff benefits
pool.
(2) An outside agency sends the institution money to
cover n dollars and to cover student registration fee at
the institution. Hence, as a student, that person becomes
eligible for student benefits.
(3) An outside agency gives the individual n dollars;
that person is responsible in handling their own taxes,
student registration fee.
The NSF post-doc fellowships programs use the third option noted
above. Although there are many NSF post-doc fellowship programs,
there are several features shared by all the programs, such as:
* All arrangements for affiliation are the responsibility of the
Fellow.
* Appropriate location of study are non-profit institutions and
institutions of higher education.
*Basic eligibility requirements are (1) US citizen; (2) should have
earned a Phd by the beginning of the fellowship tenure; and/or (3)
hold a doctorate for no more than five years.
* General evaluation process is similar: criteria for acceptance are
ability as evidenced by past research work, letters of recommendations,
suitability and availability of the sponsoring senior scientist and
other colleagues, conditions at the institution, likely impact on the
future scientific development of the applicant, and the likely
scientific quality of the research to emerge.
* Conditions of the application remain standard, such as rules
for additional suppliment of assistance.
Differences in the programs are:
* Tenure: 1 year (12 months), 2 academic years' full time, or 1
academic year and 2 academic years' half-time support. No tenure
should be less than 4 months.
* Stipends & Allowances: NSF gives Fellows stipends ranging from
$1500 (NATO post-docs) to $2300 (NSF Math post-docs). Some
allowances include dependency allowances (say, $100/month) and
special allowances ($3600) for defraying costs associated with the
fellow's research.
According to our accountant, individuals that receive a stipend over
$300 per month are required to pay taxes on the remainder of the stipend.
That rule applies up to 2 or 3 years. Dependency allowances are taxable.
Allowances to defray research costs can be non-taxable if the AAAI establishes
a general account at the institution and prove that others will be using
the account besides the Fellow.
The cost of the establishing a fellowship program would be broken down
as follows:
* Direct Costs: ? number of stipends per year; amount
of each stipend and special allowances; approximate labor
cost to prepare and administer a program (if we decide
to keep it in-house); approximate costs for publicity
and the preparation and distribution of the application forms;
travel expenses for the selection committee
If we do decide to establish such a program, we will need to develop
eligibility requirements and selection criteria as well as policies
on supplimental income.
I thought, if the EXECOMM agrees with this idea, we could
establish a two year tenure position. The stipend would be approx
$2,500 per month (so to cover dependency allowances and registration
fees) or $60,000. Also, AAAI could establish an account at an
institution to cover research expenses (say, $3500/2 years). The AAAI
would have to budget per year $50,000 for the administration of the
program and the stipend.
I'd like to hear your thoughts.
Cheers,
Claudia
************************************************************
Here are John McCarthy's comments:
1.The fellowships should be awarded to individuals, and they should
make the arrangements for place of tenure.
2. If we can afford it, there should be one - two year fellowship awarded
each year.
3. What you say about costs seems reasonable to me. The stipend should one
of the better ones.
4. We should abbreviate the administration to the point where we don't spend
more than $5,000 per year administrating it.
5. You should make a report to the Executive Committee with your facts and my
comments.
-------
∂21-Nov-83 2034 @USC-ECL:FIKES%USC-ECLD%USC-ECL@SRI-NIC Re: Post-Doctorate program
Received: from USC-ECL by SU-AI with TCP/SMTP; 21 Nov 83 20:34:09 PST
Mail-from: DECNET site ECLD rcvd at 21-Nov-83 2022-PST
Date: 21 Nov 1983 2020-PST
From: Richard Fikes <FIKES%USC-ECLD%USC-ECL@SRI-NIC>
Subject: Re: Post-Doctorate program
To: AAAI-OFFICE@SUMEX-AIM, AMAREL@RUTGERS, BRACHMAN@SRI-KL, BUCHANAN@SUMEX-AIM,
BLEDSOE@UTEXAS-20, KRD@MIT-MC, BENGELMORE@SRI-KL, LERMAN@SRI-KL,
GENESERETH@SUMEX-AIM, GROSZ@SRI-AI, HART@SRI-KL, JMC@SU-AI, MINSKY@MIT-MC,
NILSSON@SRI-AI, REDDY@CMU-CS-A, STAN@SRI-AI, STEFIK@PARC-MAXC, GJS@MIT-MC,
TENENBAUM@SRI-KL, WALKER@SRI-AI, BONNIE.UPENN@UDEL-RELAY,
FIKES%USC-ECLD%USC-ECL@SRI-NIC
cc: FIKES%USC-ECLD%USC-ECL@SRI-NIC
In-Reply-To: Your message of 21-Nov-83 1648-PST
I want to take this opportunity to tell those of you who don't already
know that I have left Xerox PARC and joined IntelliGenetics Inc. to
head up their knowledge representation R&D group. IntelliGenetics is
now selling software tools and services to customers who want to build
knowledge-based systems using AI technology. Our first major software
product in that arena is KEE, a frame-based knowledge representation
system which provides a programming environment for building such
systems. I expect to play a major role in the continuing development
of KEE and of additional systems based on it.
regards,
richard
-------
∂21-Nov-83 2204 BUCHANAN@SUMEX-AIM.ARPA Re: rough draft
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 21 Nov 83 22:01:06 PST
Date: Mon 21 Nov 83 22:01:45-PST
From: Bruce Buchanan <BUCHANAN@SUMEX-AIM.ARPA>
Subject: Re: rough draft
To: GENESERETH@SUMEX-AIM.ARPA
cc: AAAI@SUMEX-AIM.ARPA, Amsler@SRI-AI.ARPA, Feinler@SRI-NIC.ARPA,
jmc@SU-AI.ARPA, Walker@SRI-AI.ARPA
In-Reply-To: Message from "Michael Genesereth <GENESERETH@SUMEX-AIM.ARPA>" of Mon 21 Nov 83 12:35:34-PST
Mike,
I'll get you more comments when I've had a chance to read
your draft carefully. One major question came to mind when
I scanned your document online: why assume one central machine?
If there were copies of a master catalog telling us where documents
are stored, then they could be distributed around the ARPAnet (and
other nets). This greatly reduces the storage costs at any one
site and could distribute the maintenace activity as well.
I have no strong preference toward this model, but would like
to see alternatives like this one considered.
thanks for your work in putting the draft together.
bgb
-------
∂21-Nov-83 2239 LEP Tower building in Prolog
I've finished a simple minded tower building program in Prolog. It takes
horrendous amounts of time to run, partly because I purposely ordered the
clauses badly in the file to make sure it didn't work fortuitously. I'd
appreciate any comments you have on it. Circumscription works naturally well,
but I think there are still difficulties with the ordering of situations.
Also, I need your signature on a change of study list, but I will be gone
from this Wednesday through Sunday. Are you available tomorrow, or shall we
set something up for next week?
Thanks,
L
I have some Lisp programs for block stacking that aren't sophisticated
but are satisfactorily fast. How about 3pm tomorrow (Tuesday)?
∂22-Nov-83 1020 vardi@Diablo Re: knowledge seminar
Received: from SU-HNV by SU-AI with PUP; 22-Nov-83 10:20 PST
Date: Tue, 22 Nov 83 08:41 PST
From: Moshe Vardi <vardi@Diablo>
Subject: Re: knowledge seminar
To: JMC@Sail, MYV@Sail
Sure. Thanks for volunteering.
Moshe
∂22-Nov-83 1434 AMSLER@SRI-AI.ARPA Re: rough draft
Received: from SRI-AI by SU-AI with TCP/SMTP; 22 Nov 83 14:30:53 PST
Date: Tue 22 Nov 83 14:31:07-PST
From: Robert Amsler <AMSLER@SRI-AI.ARPA>
Subject: Re: rough draft
To: GENESERETH@SUMEX-AIM.ARPA, AAAI@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, jmc@SU-AI.ARPA, Walker@SRI-AI.ARPA
cc: AMSLER@SRI-AI.ARPA
Don and I read through your draft and have several comments intended to
clarify certain issues or to raise questions that we need to consider as
a group.
The first has to do with hardware, in particular the use of Videodisc:
Videodisc technology is today largely an analogue mechanism.
That is, people record analogue images on Videodiscs (54,000
frames per disk) and can digitally address these frames--but
the information in the frame is not digital. Optical disk
technology is developing to the point where it will be commercially
practical, and provide digital recordings, but I am not aware
of any commercial vendors for this yet.
However, although Videodiscs cannot handle text, they might
be a companion to computer disk storage of the text, i.e.,
to handle figures and other graphic portions of the full-text.
We should investigate the use of Videodisc technology to store
the figures and other graphics of the document collection and
then sell the Videodiscs to be used in a hybrid system with a
terminal for digital text access (presumably over a network)
and a Videodisc player attached to the terminal which receives
disk addresses from the database retrievals and displays these
on the user's own Videodisc player. This means the user gets
the full document, with digital searching of the text, etc. as
well as capability to see figures (or even animation!!) in the
context of the retrievals. Such a system is capable of being
built around a Symbolics 3600 and we've seen a local version
of such running at Atari (Steve Weyer's group).
The latest issue of JASIS (Nov. '83) contains a couple of good
articles on the state of Videodisc technology and one, "Information
Providers and Videodisc/Optical Disk Technology" by Galloway
and Paris describes the procedure used for PATSEARCH. PATSEARCH
run by BRS (a database vendor) uses microprocessor controlled
videodisks located at the user's site but controlled from the
database accessed over a network connection.
With regard to the collection of past documents and those from other fields:
The appropriate technology for acquisition of old
documents is a Kurzweil Data Entry Machine (KDEM).
Literature on KDEM use indicates it is 1/3 the cost
of rekeying manually (once you acquire the KDEM itself)
Mead Data Central does an extensive amount of such
old data entry for full-text storage of journals.
I've been in contact with a KDEM representative from
San Francisco (Cal Nakanishi) and expect to be receiving their
literature shortly.
With regard to formats, there are two possible approaches:
First, we could simply accept a range of existing formats.
However, if the Library is to provide hardcopy, files would
have to be in a form appropriate for the output medium it
would be using.
It is unlikely that we can convert TeX, SCRIBE, RUNOFF,
TROFF, PUB, etc. into any one format. In addition, there
will be publisher phototypesetting tapes from books,
none of which are likely to correspond to any of these
conventions. (For a survey of how computer phototypesetting is
done in practice, see the recently published Annual Review of
Information Science and Technology's chapter on "Primary
Publication Systems and Scientific Text Processing" (ARIST,
Vol.18, 1983, ed. Martha E. Williams). In addition,
though, we must offer a lowest common denominator format for
CRT and lineprinter output. DIALOG, for instance, uses
elaborate means to encode things such as formulae, e.g.,
H/SUB 2/O instead of H2O, and it spells out Greek or other
non-ASCII characters. Spelling-out is the accepted practice
for characters which cannot be entered from computer
terminals.
We have just gotten information about a project to develop
industrywide standards for preparing and processing
electronic manuscripts. Sponsored by the Association of
American Publishers and coordinated through the Council
of Library Resources and the National Bureau of Standards,
it is building on the work of the the Graphics Communication
Association--which has developed a Standard Generalized Markup
Language, ANSI Committee X3J6, ISO TC97, and the UNISIST
International Centre for Bibliographic Descriptions. Many
publishers are involved, including some who produce AI
literature, notably the IEEE and the American Math. Society.
We have some descriptive material available.
In any case, it does not seem reasonable to develop our
own formats, although we may have to introduce some macro
conventions and perhaps limit the number of fonts allowed.
With regard to indexing:
The model to apply here is that used for the Educational
Resources Information Centers (ERIC). Before one can
effectively index material one must have a classification
system of the literature. Before one can build a classification
system, one must have a terminology list of the subjects in the
discipline, with definitions where needed. ERIC developed
a Thesaurus for this purpose.
We would therefore recommend that both a "Thesaurus of AI
Terminology" and a "Classification System for AI Literature" be
developed. As a beginning, we are processing the text of
the AI Handbook to derive such a set of terminology and a
classificatory system. Hopefully, this will be available
for use early next year.
One cautionary note: index keywords selected by authors
should be selected from a given set of pre-determined terms.
Having studied the unrestricted keywords used in the CACM
I can attest to the undesirability of author-invented keywords
as the sole indexing. Such terms are highly idiosyncratic
and never reach closure. Subsequent searchers can never
be sure the documents they find represent the contents of
the database if the indexing is inconsistent. DIALOG, for
instance, uses both "Descriptors" and "Identifiers" to
keyword index documents. The former are from a closed
subject list, the latter from an open set.
We are developing programs for automatic indexing as part
of our program in Natural-Language and Knowledge-Resource
Systems at SRI. We already have capabilities for determining
subject categories for text automatically from the text
itself given a dictionary of terminology in the discipline.
This is another reason to develop a Thesaurus of AI Terminology.
Some success in automatic indexing of full-text can be
achieved through localized frequency counts. That is,
if a frequency count is made of each subsection of a book,
the 3-5 highest frequency content bearing terms appear
to be useful index entries. I've only tried this on manuals
so far though. More testing is needed.
Programs to extract phrases and proper nouns from text
give additional indexing capabilities. Together all these
techniques could provide a powerful set of automated indexing
tools.
Now on to Search Services:
Additional search capabilities are both practical and
essential for a system representing artificial intelligence
literature. They are practical in the range of address-space
which can be accommodated on Symbolics 3600's or other
extended-addressing hardware. They are essential because
only via such a means can we achieve a database access
system which can advance the state of the art in information
access to a level appropriate for AI researchers and their
access needs.
The field of AI is less than 30 years old; however, most of
literature can be considered to represent roughly 20 years
worth of contributions (1963-1983). Twenty years at 200
Megabytes/year yields approximately 4 Gigabytes. At an
estimate of 1000-2000 AI documents per year this would
only be 20,000-40,000 bibliographic entries.
Work at SRI has already resulted in loading a 10,000 entry
bibliographic item database into ELISP on a DEC2060, and
a Symbolics 3600 can support an address space 1000 times
as large as that accessible via extended-addressing on the
DEC2060. Therefore, the complete bibliographic database for
the AI Literature could be supported on a single Symbolics
3600 in ZetaLisp.
The advantages for building such a system in LISP are
obvious to AI practitioners: capability for interconnection
of data entries, facilitating citation indexing and interactive
searching; simplification of the data base software to
capitalize on LISP's built-in hash-coding and symbolic data
management; as well as the development of additional
capabilities such as spelling correction, data authentication
at entry by checking input information against existing stored
information, inferencing to determine missing information in
user's queries, help systems, built-in multi-font and graphics
reproduction capabilities for expansion of document coverage,
windowing of displays for easy access to database contents,
networking of additional disks for access to original texts of
documents stored on other systems (even at other sites
accessed over computer networks).
It would seem that regardless of whether any commercial vendor
provides a suitable elementary service for the National Library
of AI Literature, that AAAI, ACL, and any other related (and
interested) societies ought to endorse the development
of a suitable LISP-based version of the system. We shouldn't
accept the state-of-THEIR-art when we already are using
capabilities in advance of that level of development.
For example, access formats should include bibliographic
output in something like SCRIBE input format such that
authors can use the system to assemble bibliographies for
papers they are writing.
Additional methods of access could include receipt of a
paragraph or more of text from an author with a request
that the system provide documents that closely approximate
that document's contents. Thus, access to the system
without dependence upon the user knowing the existing
classification system should be offered. Browsing
capabilities should at least include those of the CMU
BROWSE system (written in ZOG). This can at present be
accessed over the ARPANET at CMU and to offer a system
with less capabilities would invite it being discredited
as obsolete by the ARPANET AI community.
Additional access should be provided via full-sentence retrieval.
Copyrights and Royalties:
I believe we can get considerable cooperation from
publishers holding copyrights to current books and journals,
if, from the outset, we adhere to a policy of repayment for
access. In this light it is important to note several points.
Data Access is the same as Data Capture. It is not possible on
existing computer technology to assume that anything one has
seen on a computer screen has not also resulted in an electronic
copy passing into the hands of the reader. The only means of
control over this would be to control both the access lines
and the access terminals (e.g., as OCLC does). Since we do
not want to get into direct lines with customized terminals,
this means charging patrons for access in a manner identical
to that charged for data capture. This is a minimal problem for
bibliographic entries--but a considerable one for full-text
where existing paper copies of books are being sold
(e.g. North-Holland will have to be approached very carefully).
We do have an option to pursue special royalties based upon
number of bytes accessed rather than upon the cost of transfer
of the entire document. One shouldn't pay for a whole copy
of the AI Handbook if one only extracts one chapter.
Reimbursement is not just a matter for other publishers. The
development of this system and its widespread adoption can
result in a significant reduction to AAAI and other society
revenues due to a loss of subscriptions and sales of
conference proceedings. This reduction must be adequately
compensated for through royalties.
WE MUST FROM THE BEGINNING ASSUME WE ARE TRYING TO PROMOTE
ELECTRONIC ACCESS AS AN ALTERNATIVE TO PAPER ACCESS WHICH
MAY SOMEDAY COMPLETELY ELIMINATE PAPER ACCESS.
If we are successful, the AI Magazine, the AI Journal, the AJCL,
SIGART, and perhaps other journals could some day cease publishing
paper copies entirely--or offer free access to the electronic
copy in lieu of a paper copy subscription.
Some comments on the relevance of NIC:
(1) The NIC is already charged with establishing an electronic
library (of NIC documents) on the ARPANET and the development
of programs to permit access to them from remote sites over
the existing networks.
(2) The NIC has available, on its newly acquired DEC2060,
sufficient disk space, network connections, and computing
power to make it possible for it to donate resources
at least during the initial stages of this project. These initial
stages include acquisition of bibliographic entries, processing
of acquired documents for indexing, and access connections.
Currently the Network Information Center supports an average
of about 2 network users at all times already accessing the
NIC database (AI Resource Handbook, AI Directory) or other
on-line documents.
(3) NIC (as well as Don and I) is interested in the research
potential of this project to advance the state-of-the-art
in electronic information access.
(4) NIC is "politically" neutral and this could be an advantage
if government funding is perceived as being directed toward
Stanford vs. MIT, CMU, etc.
A final note:
"Stanford Research Institute" no longer exists, we're
"SRI International" where SRI stands at best for "formerly
Stanford Research Institute."
-------
∂22-Nov-83 1541 JJW Thesis topic
I would like to propose a thesis in the following area: proving facts about
parallelizing programs, and applying this to code generation in a compiler.
For example, given a loop, one could state the property that the computations
for each value of the index can be performed independently, and if this can be
proved true, then such code can be generated. Intensional properties, such as
the tradeoff involving the overhead in parallel computation, may also be
treated in this way.
By suitably generalizing various ways of making code parallel, the compiler
using this method can contain the "knowledge" to apply them whenever
appropriate.
Questions I have now are: (1) Is this suitable as a thesis topic? (2) Has
similar work already been done, or is anyone doing it right now?
1. It is suitable as a thesis topic. I would consider narrowing it so
as to not necessarily include writing a compiler and broadening it so
as to include more forms of parallelism. Pay attention to queue based
parallelism in general and RPG's qlambda in particular.
2. There is undoubtedly relevant work. I suggest you begin your literature
search by talking to Eric Gilbert, who has had the problem of locating
material in this area.
∂22-Nov-83 1703 DFH Alex Jacobson
is still trying to get ahold of you. He will be at
Inference until about 7 pm, 213-417-7997, or you
can call him at home later.
∂23-Nov-83 1045 CLT car
while your'e fixing, you might also
get new wipers.
∂23-Nov-83 1439 NILSSON@SRI-AI.ARPA Strips
Received: from SRI-AI by SU-AI with TCP/SMTP; 23 Nov 83 14:38:52 PST
Date: 23 Nov 1983 1439-PST
From: Nilsson at SRI-AI
Subject: Strips
To: jmc at SU-AI
cc: nilsson
The example that I give in my book for STRIPS (p 300-301) does
involve an explicit representation of the goal structure. On p.
302 -303 I describe another version of STRIPS in which "... the
exlicit use of a goal stack [is] supplanted by the built-in stack
mechanism of the language (such as LISP) in which recursive
STRIPS is implemented."
Also, I'd be interested in your comments on "commutative production
systems" pp. 35-37.
By the way, you may have an early printing of my book (rough-textured
maroon cover). The third printing (shiny maroon cover) has the
known errors corrected. I'm sending over a copy of the third printing
to you by ID mail. In particular, the first printing had an
error in the definition of commutative production systems (p. 35).
Here is how it should read:
a. Each member of the set of rules applicable to D is also
applicable to any database produced by applying an applicable rule to
D.
b. If the goal condition is satisfied by D, then it is also
satisfied by any database produced by applying any applicable rule to
D.
c. The database that results by applying to D any sequence composed
of rules that are applicable to D is invariant under permutations of
the sequence.
-Nils
-------
∂23-Nov-83 2110 ULLMAN@SU-SCORE.ARPA Winter CS440
Received: from SU-SCORE by SU-AI with TCP/SMTP; 23 Nov 83 21:10:00 PST
Date: Wed 23 Nov 83 21:07:24-PST
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Winter CS440
To: super@SU-SCORE.ARPA
Well as I expected, no one volunteered to do CS440 for the Winter.
I guess I'm left holding the
bag, but be warned that for the Spring, I won't even be here, so
if nobody takes it over, it goes down the tubes.
For starters, I'd like to get some volunteers to speak next quarter.
It seems that there are a number of people on campus who didn't
even get a chance to speak this quarter, and really have something to
say. Please send me mail--you know who you are.
-------
∂25-Nov-83 1518 @SU-SCORE.ARPA:cheriton@diablo POPL Paper on Concurrent Prolog
Received: from SU-SCORE by SU-AI with TCP/SMTP; 25 Nov 83 15:18:38 PST
Received: from Diablo by SU-SCORE.ARPA with TCP; Fri 25 Nov 83 15:16:36-PST
Date: Fri, 25 Nov 83 15:15 PST
From: David Cheriton <cheriton@diablo>
Subject: POPL Paper on Concurrent Prolog
To: super@score
There is a paper on Concurrent Prolog as a Multiprocessor's Kernel Language
by Ehud Shapiro. Any one have a copy or know anything about this?
∂26-Nov-83 0153 ME reply addr
∂25-Nov-83 0309 JMC
How do I reply to this person? Here are three lines from the header.
∂21-Nov-83 2212 @MIT-MC:perlis%umcp-cs@CSNET-CIC Re: limitations of logic
From: Don Perlis <perlis%umcp-cs@CSNet-Relay>
Return-Path: <perlis%umcp-cs@CSNet-Relay>
ME - With:
MAIL "perlis%umcp-cs"@CSNet-Relay
∂26-Nov-83 1305 perlis%umcp-cs@CSNet-Relay Re: logic.pro[f83,jmc] Proposal for logic in AI mailing list
Received: from CSNET-CIC by SU-AI with TCP/SMTP; 26 Nov 83 13:05:27 PST
Date: 26 Nov 83 15:38:41 EST (Sat)
From: Don Perlis <perlis%umcp-cs@CSNet-Relay>
Return-Path: <perlis%umcp-cs@CSNet-Relay>
Subject: Re: logic.pro[f83,jmc] Proposal for logic in AI mailing list
To: John McCarthy <JMC@SU-AI>
Via: UMCP-CS; 26 Nov 83 15:48-EST
Might be a good idea. But don't count me in on refereeing, for the present.
Maybe this problem won't arise for awhile?
∂26-Nov-83 1537 @MIT-MC:DAM%MIT-OZ@MIT-MC Edited Mailing List
Received: from MIT-MC by SU-AI with TCP/SMTP; 26 Nov 83 15:36:55 PST
Date: Sat, 26 Nov 1983 18:33 EST
Message-ID: <DAM.11970800974.BABYL@MIT-OZ>
From: DAM%MIT-OZ@MIT-MC.ARPA
To: JMC@SU-AI.ARPA
cc: phil-sci%MIT-OZ@MIT-MC.ARPA
Subject: Edited Mailing List
Date: Saturday, 26 November 1983 14:09-EST
From: John McCarthy <JMC at SU-AI>
It even might be worthwhile to have an edited discussion.
It would be much more tolerant than a journal, but not
every contribution would be accepted by the editor who
might use referees if he found it necessary.
I think an edited version of phil-sci is a grand idea. I
propose that there be three editors, one for logic, one for the
philosophy of science, and one for the philosophy of epistemology.
Rather than edit message content the primary function of the editing
should be to ensure that the messages are carefully written, concise,
relevant to the discussion, and non-redundant with other messages.
Perhaps redundant messages could be merged into co-authored messages
after a cycle of refereeing.
A potential author would send a message to one of the editors
(and thus place the messafge in one of the categories). The editor
would then forward the message to a referee. The referees would
then mail the messages back to the editor (so as to remain anonymous)
who would then either send them to the mailing list or return them to
the author with comments from the referee.
Messages should be short (a few pages at most) and turn around
time on refereeing should be about a day.
This would be much less formal than a journal. However it
would be much more useful than a large mailing list becuase both the
amount and quality of the mail could be somewhat controled. One
problem might be that the day delay caused by refereeing would keep
people form sending messages. However if a large enough readership
could be established this might not be a problem.
Would McCarthy volunteer to be the logic editor?
David Mc
∂26-Nov-83 2109 @MIT-MC:KDF%MIT-OZ@MIT-MC Logic in AI list
Received: from MIT-MC by SU-AI with TCP/SMTP; 26 Nov 83 21:09:20 PST
Date: Sun, 27 Nov 1983 00:07 EST
Message-ID: <KDF.11970861762.BABYL@MIT-OZ>
From: KDF%MIT-OZ@MIT-MC.ARPA
To: jmc@SU-AI.ARPA
Subject: Logic in AI list
I'd be quite interested in such a discussion. I think if the
group were small and serious (i.e., people who actually use logic in
their daily work) it probably wouldn't need to be edited. Since I'm
in thesis mode, I certainly can't volunteer any time to edit or
referee.
∂28-Nov-83 0151 HST VIXIT
HI,JOHN.I HAVE TO FIX MY TRAVEL ROUTE NOW.PLEASE TELL ME IF YOU ARE
INTERESTED.IF NOT,DON'T HESITATE TO SAY IT.I KNOW I'M NOT THE MOST
IMPORTANT GUEST OF STANFORD.HERBERT
∂28-Nov-83 0813 KJB@SRI-AI.ARPA talking
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Nov 83 08:13:04 PST
Date: Mon 28 Nov 83 08:05:39-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: talking
To: jmc@SU-AI.ARPA
John, when we spoke on Saturday I forgot that I have a meeting on
Mondays at 2 with Joan Bresnan, Ron Kaplan, Stan Peters and John Perry.
I also forgot that I have to give a talk tomorrow in Berkeley. So lets
make it later. Jon
-------
∂28-Nov-83 1000 JMC*
variable number of arguments of mapcar in ekl and seus.
∂28-Nov-83 1017 GENESERETH@SUMEX-AIM.ARPA
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 28 Nov 83 10:17:25 PST
Date: Mon 28 Nov 83 10:17:13-PST
From: Michael Genesereth <GENESERETH@SUMEX-AIM.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sun 27 Nov 83 15:45:00-PST
I have a lunch appointment on Wednesday already. How about today (Monday)
or Thursday if you can't make it today. Any particular topic?
mrg
-------
∂28-Nov-83 1133 HAKEN@SUMEX-AIM.ARPA Resolution Theorem Proving
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 28 Nov 83 11:33:25 PST
Date: Mon 28 Nov 83 11:33:17-PST
From: Armin Haken <HAKEN@SUMEX-AIM.ARPA>
Subject: Resolution Theorem Proving
To: jmc@SU-AI.ARPA
I have a proof that resolution on propositional formulas is
non-polynomial on worst case.
If you want a copy I can put it in your mailbox.
I'll be in Stanford this week.
Armin Haken,
Haken@sumex
-------
Yes, I'd like a copy of your paper.
∂28-Nov-83 1212 ULLMAN@SU-SCORE.ARPA Next Meeting
Received: from SU-SCORE by SU-AI with TCP/SMTP; 28 Nov 83 12:12:46 PST
Date: Mon 28 Nov 83 12:08:31-PST
From: Jeffrey D. Ullman <ULLMAN@SU-SCORE.ARPA>
Subject: Next Meeting
To: CS440: ;
Here is the abstract of Jay Misra's talk this Thursday 12/1/83:
Title: Distributed Simulation
Abstract: The problem of system simulation is typically solved in a sequential
manner because events are sequentially scheduled using an event list. We propose
a distributed solution in which many processes may simultaneously simulate
different parts of the system. Processes operate autonomously; they send messages
to each other to synchronize simulations of dependent events.
We show different schemes for avoiding deadlock in the simulator.
An implementation on a multiple processor system is sketched.
-------
∂28-Nov-83 1546 CLT supper tonight
Some part of PERSEUS including me taking Lucca Cardelli
to supper - to indoctrinate.
∂28-Nov-83 1705 LGC
Sure; see you at about 17:15...
∂28-Nov-83 2300 JMC*
Louis wants to see AAAI and IJCAI.
∂29-Nov-83 0939 @SU-SCORE.ARPA:CHAPPELL@SRI-AI.ARPA Tinlunch
Received: from SU-SCORE by SU-AI with TCP/SMTP; 29 Nov 83 09:39:30 PST
Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Tue 29 Nov 83 09:39:36-PST
Date: 29 Nov 1983 0933-PST
From: Chappell at SRI-AI
Subject: Tinlunch
To: jmc at SU-SCORE
cc: chappell
Please let me know if you will lead Tinlunch discussion on
December 8, 1983. If you have questions regarding Tinlunch
please contact me at 859-2312 or contact Stan Rosenschein at 859-4167.
Thank you.
--Elsie
-------
∂29-Nov-83 1309 JK
;the schema involving mapcar with a number of arguments could be stated
;as follows, in a slightly APL'ish style:
;does not work yet
(get-proofs lispax prf prf jk)
(proof mapcar)
(decl (pars1 pars2) (type: |ground*|))
;define a few functions on list types
;generating tuples of the form (nil,nil,...,nil)
(decl nilgen (type: |ground→ground*|))
(defax nilgen |nilgen(nil)=nil∧∀u.nilgen(nil.u)=(nil,nilgen(u))|)
(label nilgendef)
;consing tuples elementwise
(decl tcons (type: |ground*⊗ground*→ground*|) (syntype: constant) (infixname: |⊗|)
(prefixname: tcons) (bindingpower: 850))
(defax tcons |()⊗()=()∧
∀pars1 pars2 x y.(x,pars1)⊗(y,pars2)=((x.y),pars1⊗pars2)|)
(label tconsdef)
;definition of extended mapcar α
(decl α (type: |(ground*→ground)⊗ground*→ground|))
(decl lfun (type: |ground*→ground|))
(defax α |∀u.α(lfun,nilgen(u))=nil∧
∀pars1 pars2.α(fun,pars1⊗pars2)=fun(pars1).α(fun,pars2)|)
(label alphadef)
;now one can instantiate this to arbitrary number of variables
(decl fun2 (type: |ground⊗ground→ground|))
(ue ((u.|'(nil nil)|) (lfun.|λpars.fun2(π1(pars),π2(pars))|)
(pars1.|(x,u)|) (pars2.|(x1,u1)|))
alphadef
(use (tconsdef alphadef) mode: always))
;this should return:
α(fun2,nil,nil)=nil∧
∀x x1 u u1.α(fun2,(x,u),(x1,u1))=fun2(x,x1).fun2(u,u1)
∂29-Nov-83 1524 DFH appointment
I gave Bob Givan (student in your CS206 class) an
appointment Mon. Dec. 5 at 2 pm. If this needs to
be changed, let me know.
∂29-Nov-83 1614 CHAPPELL@SRI-AI.ARPA Tinlunch 12/15
Received: from SRI-AI by SU-AI with TCP/SMTP; 29 Nov 83 16:14:22 PST
Date: 29 Nov 1983 1615-PST
From: Chappell at SRI-AI
Subject: Tinlunch 12/15
To: jmc at SU-AI
cc: chappell
John,
I had originally scheduled you for TINLunch discussion on 12/8, since
I have not heard from you, I asked Bob Moore to be discussion leader
for that date. Will you select a paper for 12/15? Please let me
know what the paper will be for that date ASAP.
Thanks for your cooperation.
--Elsie
-------
∂29-Nov-83 1733 SJG British public opinion
To: JMC
CC: JMC
Sally confirms my reporting of British public opinion and of the opinion of the
academics there. Is this how MAIL works?
Also, can you tell me the name of yesterday's restaurant? Many thanks.
I got the message twice, because you mailed it to me and also cc'ed me.
Many people cc themselves, but since you have an outgo.msg file, that
isn't necessary. The restaurant is the Normandie on University Ave.
Louis Lerman left some cash for you, and it's in Diana's office.
∂30-Nov-83 0006 100 (from: golu⎇) address
What's Ginsberg's address? GENE$
∂30-Nov-83 0423 @MIT-MC:JCMA%MIT-OZ@MIT-MC Edited Sub-List on philosophy-of-logic
Received: from MIT-MC by SU-AI with TCP/SMTP; 30 Nov 83 04:23:12 PST
Received: from MIT-APIARY-5 by MIT-OZ via Chaosnet; 30 Nov 83 07:22-EST
Date: Wednesday, 30 November 1983, 07:23-EST
From: JCMA%MIT-OZ@MIT-MC.ARPA
Subject: Edited Sub-List on philosophy-of-logic
To: DAM@MIT-MC
Cc: JMC@SU-AI.ARPA, cstacy@MIT-MC, gavan@MIT-MC, batali@MIT-MC,
phil-sci-request%MIT-OZ@MIT-MC.ARPA
In-reply-to: <DAM.11970800974.BABYL@MIT-OZ>
Date: Sat, 26 Nov 1983 18:33 EST
From: DAM@MIT-OZ
Subject: Edited Mailing List
Date: Saturday, 26 November 1983 14:09-EST
From: John McCarthy <JMC at SU-AI>
It even might be worthwhile to have an edited discussion.
It would be much more tolerant than a journal, but not
every contribution would be accepted by the editor who
might use referees if he found it necessary.
I think an edited version of phil-sci is a grand idea. I
propose that there be three editors, one for logic, one for the
philosophy of science, and one for the philosophy of epistemology.
I think the proposal is fine. It looks like you definitely want one more list
for the philosophy of logic. I don't know if there is enough demand for the
others topics at present. We can always create them as needed. One list also
makes a better experiment, and involves less maintenance.
Rather than edit message content the primary function of the editing
should be to ensure that the messages are carefully written, concise,
relevant to the discussion, and non-redundant with other messages.
Perhaps redundant messages could be merged into co-authored messages
after a cycle of refereeing.
I think it would also be useful to make the editor fully responsible for the
quality of the publication. If we are going to bother with an editor, the
editor should earn his/her keep by filtering out redundancy and randomness
while retaining unconventional ideas. I assume that what we are interested in
is a high quality source of well-thought-out views on the philosophy of logic.
A high concentration of good ideas/debates makes the publication worth reading.
It seems that there is little point in making too much effort unless a large
payoff in terms of quality/concentration can be expected.
A potential author would send a message to one of the editors
(and thus place the messafge in one of the categories). The editor
would then forward the message to a referee. The referees would
then mail the messages back to the editor (so as to remain anonymous)
who would then either send them to the mailing list or return them to
the author with comments from the referee.
Messages should be short (a few pages at most) and turn around
time on refereeing should be about a day.
This all sounds fine. It seems like what one might term a "psuedo-refereed
virtual journal."
This would be much less formal than a journal. However it
would be much more useful than a large mailing list becuase both the
amount and quality of the mail could be somewhat controled. One
problem might be that the day delay caused by refereeing would keep
people form sending messages. However if a large enough readership
could be established this might not be a problem.
The following arrange looks like what you want:
1) People wishing to contribute to the philosophy-of-logic list
send mail to that address.
2) That mail is reviewed an compiled into digests according to
subject-matter/debate by the editor according to some refereeing
scheme.
3) The resulting digests are periodically mailed to the phil-sci public.
What we would have here is a sub-digest to an imediate redistribution mailing
list. That is an interesting concept (although not new) which provides a way
of controlling the volume on Phil-sci, while not really undermining the rate
of interchange. The main problem is finding competent and conscientious
people to do the editing. I know I am too busy for that.
It is also important to note that there is a prologue digest, and that we
don't want to compete directly with that. I assume that what you have in mind
does not overlap with the major thrust of the prologue list.
Would McCarthy volunteer to be the logic editor?
Good question. If not, perhaps he would like to designate DAM as his stand-in.
John Mallery
∂30-Nov-83 0939 @MIT-MC:DAM%MIT-OZ@MIT-MC Edited Sub-List on philosophy-of-logic
Received: from MIT-MC by SU-AI with TCP/SMTP; 30 Nov 83 09:39:03 PST
Date: Wed, 30 Nov 1983 12:38 EST
Message-ID: <DAM.11971784807.BABYL@MIT-OZ>
From: DAM%MIT-OZ@MIT-MC.ARPA
To: JCMA%MIT-OZ@MIT-MC.ARPA
Cc: batali@MIT-MC, cstacy@MIT-MC, gavan@MIT-MC, JMC@SU-AI.ARPA,
phil-sci-request%MIT-OZ@MIT-MC.ARPA
Subject: Edited Sub-List on philosophy-of-logic
In-reply-to: Msg of 30 Nov 1983 07:23-EST from JCMA%MIT-OZ at MIT-MC.ARPA
I would be willing to edit a philosophy-of-logic and
logic-in-AI mailing list (unless JMC wants to do it). I think
the distribution list should be seperate from phil-sci so that
people can be on one and not the other (I am afraid of the
logic messages getting lost in the flood of unedited phil-sci
messages). I also think it would be appropriate to solicit new
members, people interested specifically in logic and AI (there are
lots of people interested in logic who have not participated
in phil-sci). Furthermore I think it is important to have a good
archive and to service requests for reprints of discussions.
I will write a charter for the new edited mailing list and send it
out for comments.
David Mc
logic in AI electronic journal
I entirely agree with David's proposal especially emphasizing
logic-in-AI. I think there are quite a few additional people
who would want to join and who would make good contributions.
I'm looking forward to seeing the charter, and I vote for David
as editor. I think the editor should take full responsibility
and should use referees only in so far as he needs expertise
he doesn't have.
∂30-Nov-83 1338 BOSACK@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 13:38:38 PST
Date: Wed 30 Nov 83 13:37:03-PST
From: Len Bosack <BOSACK@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 12:11:00-PST
We were thinking of adding two more Sail terminals to the second floor:
one in with the Dover (so people can do the last minute fixes to their
work) and one either in 225 or in the area under the stairs up to the
lounge.
Len
-------
The Dover room would be ok provided there is no shortage of terminals,
but it's an impossible place for doing anything else than last minute
fixes. I have now a few people whose use of SAIL I am sponsoring but
who have trouble finding a public SAIL terminal. The only one's I
know about are the one near the 4th floor terrace and the one next
to the xgp. Therefore, I would recommend putting at least one in 225
which is presumably a reasonable place to work.
∂30-Nov-83 1423 JJW CS 206
Janet Lee, who was in 206 last year, is just finishing her work to make up her
incomplete. She would like us to certify that she passed the course as soon as
possible, "due to budget allocation constraints at SRI" whatever that means.
Can we do this, and then assign a grade after seeing her project?
∂30-Nov-83 1423 DFH Woody Bledsoe
Just called. Please call him at 512 471 1242. He
will be there for a couple more hours.
∂30-Nov-83 1425 BOSACK@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 14:25:36 PST
Date: Wed 30 Nov 83 14:20:26-PST
From: Len Bosack <BOSACK@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:47:00-PST
I think I have 3 Sail terminals available; would you like another one
in an office somewhere?
-------
∂30-Nov-83 1514 WORLEY@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 15:14:39 PST
Date: Wed 30 Nov 83 15:14:48-PST
From: Pat Worley <WORLEY@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:29:00-PST
I've seen it in one of the Pink Panther movies, (#3 or #4).
It's a line by Inspector Dreyfus he learned in the asylum.
Whether this is derived from somewhere else again (being
a subtle, to me, joke by the script writers) I have no idea.
-------
Thanks, it's older than that. It comes from some 19th century or
early 20th century positive thinker, but I have no idea how to find
it, since my Oxford Book of Quotations doesn't have it. I need it
to illustrate a heuristic, I'm proposing.
∂30-Nov-83 1548 DEW
∂30-Nov-83 JMC 30-Nov-83 DEW
>30-Nov-83 1329 JMC
To: su-bboards@SU-AI
Can anyone give a citation for the following:
"Every day in every way I get better and better" and perhaps a
corrected version if that isn't correct?
dew - That is a lyric from a song on the last album John Lennon produced
before he was shot. I'll provide the complete and correct lyrics to the
song tomorrow or later tonight.
∂30-Nov-83 1619 LIBRARY@SU-SCORE.ARPA Every day....
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 16:19:37 PST
Date: Wed 30 Nov 83 16:10:34-PST
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: Every day....
To: jmc@SU-AI.ARPA
Prof. McCarthy:
The exact quote is "Every day in every way I am getting better and better".
It is attributed to a self-proclaimed French psychoterapist named Emil
Coue(withe accent on the e) who lived 1857-1926. Information found in
Home book of Qutations by Stevenson, page 872.
Harry Llull
-------
Harry,
Thanks for the reference.
∂30-Nov-83 1627 SJG office
have office, terminal, phone etc. many thanks ...
∂30-Nov-83 1630 ANDY@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 16:29:50 PST
Date: Wed 30 Nov 83 16:18:56-PST
From: Andy Freeman <ANDY@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:29:00-PST
I heard it in a Pink Panther movie. I don't remember the name but
Inspector Dreyfus escaped/was released from an asylum, kidnapped a
scientist to make him a destruction ray, destroyed the UN building
with it, demanded that the nations of the world kill Clouseau (sic)
(Peter Sellers), etc. There were several attempts to kill Clouseau in
a beer garden in Munich during Oktoberfest. Omar Sharif appears as
an Egyptian spy who kills an American dressed up (accidently) as
Clouseau and then (Omar) seduces (in the dark) a Soviet spy who
thinks she's with Clouseau, falls in love, and doesn't kill him.
After Dreyfus' castle is destroyed (Closeau goes undercover as a
dentist but his disguise melts, various other strangeness also occurs)
the Soviet spy and Closeau go back to his apartment in Paris, he goes
through his macho routine (he catches his tie on his head while
ripping his clothes off) and the bed folds up through the wall when
Kato attacks and they are all thrown into a fountain.
I believe that the correct quote is "... I am getting better ...",
possibly with "Every day and in ..." at the front.
-andy
-------
∂30-Nov-83 1632 LIBRARY@SU-SCORE.ARPA The Artificial Intelligence Report
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 16:32:09 PST
Date: Wed 30 Nov 83 16:32:24-PST
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: The Artificial Intelligence Report
To: JMC@SU-AI.ARPA
cc: Feigenbaum@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA
I have just received a sample copy of the first issue of The Artificial
Intelligence Report, v. 1 no. 1, January 1984. It will be published 10
times a year by William Kaufmann, Inc. It reports AI applications in
the U.S. Issue number 2 will focus on current and projected AI applications
in the US air force,navy, army, marine corps, and Nasa. The following is in
the first issue: AI at DEC, Delta/Cats-1, AI abroad, IntelliGenetics, and
New Companies. Editor is Lou Robinson, Advisory board includes Avron Barr,
Mark Fox, Tom Garvey, Rick Hayes-Roth, Kaufmann, and Thomas Kehler.
Would you expect this to be in the library? The subscription cost is $200
however I have a note on the blurb I received that we can get for $150.
Generally I do not have very much of this type of material addressed more
for the applied areas or for those in industry/marketing etc. Need your
advice as to whether we should spend the money for this type of serial
thanks
Harry
-------
∂30-Nov-83 1701 BUCHANAN@SUMEX-AIM.ARPA Re: The Artificial Intelligence Report
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 30 Nov 83 17:00:54 PST
Date: Wed 30 Nov 83 17:00:22-PST
From: Bruce Buchanan <BUCHANAN@SUMEX-AIM.ARPA>
Subject: Re: The Artificial Intelligence Report
To: LIBRARY@SU-SCORE.ARPA
cc: JMC@SU-AI.ARPA, Feigenbaum@SUMEX-AIM.ARPA
In-Reply-To: Message from "C.S./Math Library <LIBRARY@SU-SCORE.ARPA>" of Wed 30 Nov 83 16:32:27-PST
Harry,
I would rather have you spend your money on publications of more
lasting interest in AI. This material is dated and not of as much
research interest as many other things you could buy. I would
read it sometimes out of curiosity, but I don't see that it would
help my students.
bgb
-------
∂30-Nov-83 2205 GROSOF@SUMEX-AIM.ARPA quote
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 30 Nov 83 22:05:13 PST
Date: Wed 30 Nov 83 22:05:05-PST
From: Benjamin Grosof <GROSOF@SUMEX-AIM.ARPA>
Subject: quote
To: jmc@SU-AI.ARPA
The version I've heard is a bit more active in tense: "I'm getting" rather
than "I get". Have fun!
Benjamin
-------
∂30-Nov-83 2255 AAAI@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Nov 83 22:55:20 PST
Date: Wed 30 Nov 83 22:55:28-PST
From: AAAI-OFFICE <AAAI@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:29:00-PST
Mr. (AAAI) President:
The quote (in a secondary source) is ``Day by day in every way I am
getting better and better.''
The author is Emile Coue', a French hypnotist ; the citation is ``Self
Mastery Through Conscious Autosuggestion'' (American Library Service).
The phrase is Coue' favorite affirmation for individuals who wish to
use self-hypnosis to cure their ills.
Your article in Psych Today (``Little thoughts...) was great, by the
way. It counter-balances a Tom Moran paper: ``Analogy Reconsidered...''
David King
AAAI: AI Magazine TeX-er
-------
∂01-Dec-83 0046 LLW@S1-A Weekend Meeting RSVP
Received: from S1-A by SU-AI with TCP/SMTP; 1 Dec 83 00:46:29 PST
Date: 01 Dec 83 0045 PST
From: Lowell Wood <LLW@S1-A>
Subject: Weekend Meeting RSVP
To: pourne@MIT-MC
CC: LLW@S1-A, RAH@S1-A, jmc@SU-AI
Dear Jerry:
I've got to be in DC through Saturday PM, so will not be able to show up
at this weekend's meeting until Saturday evening late (if it's still
going on), or Sunday AM (if I get there too late Saturday). Rod is up
to his posterior in alligators, trying to put some major technical
substance behind the President's strategic defense proposals vis-a-vis a
critical upcoming experiment, and won't be able to show at all. I've
not heard from John re his plans.
Sorry for the late RSVP--life has been even more hectic than usual recently!
Lowell
∂01-Dec-83 0429 JLH.BRADFORD@SU-SIERRA.ARPA
Received: from SU-SIERRA by SU-AI with TCP/SMTP; 1 Dec 83 04:29:18 PST
Date: Thu 1 Dec 83 04:29:33-PST
From: Ethan Bradford <JLH.BRADFORD@SU-SIERRA.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:29:00-PST
I don't know if it is a quote from an older source, but the place I remember
it from is the third sequel to The Pink Panther (I confuse the titles of the
sequels, but it may have been The Pink Panther Strikes Back). Inspector
Clouseau (sp?) (the hero) had in a previous episode driven his boss insane
and in the start of this episode he was about to be released from the
mental hospital. That quote is what his doctor taught him to say to develop
a positive attitude about his recovery.
-- Ethan
-------
∂01-Dec-83 0747 LIBRARY@SU-SCORE.ARPA another version of quote
Received: from SU-SCORE by SU-AI with TCP/SMTP; 1 Dec 83 07:46:55 PST
Date: Thu 1 Dec 83 07:47:08-PST
From: C.S./Math Library <LIBRARY@SU-SCORE.ARPA>
Subject: another version of quote
To: jmc@SU-AI.ARPA
Prof. McCarthy:
Encyclopaedia of Britannica has a slightly different version of the quote:
"Every day, and in every way, I am becoming better and better."
Harry
-------
∂01-Dec-83 0840 TAJNAI@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 1 Dec 83 08:40:12 PST
Date: Thu 1 Dec 83 08:40:21-PST
From: Carolyn Tajnai <TAJNAI@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Thu 1 Dec 83 01:10:00-PST
Thanks, John.
Carolyn
-------
∂01-Dec-83 0911 CC.CLIVE@UTEXAS-20.ARPA quotation
Received: from UTEXAS-20 by SU-AI with TCP/SMTP; 1 Dec 83 09:11:10 PST
Date: Thu 1 Dec 83 11:11:21-CST
From: Clive Dawson <CC.Clive@UTEXAS-20.ARPA>
Subject: quotation
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 30 Nov 83 13:29:00-CST
This is probably not what you're looking for, but your quote triggered
my recall of a song sung by John Denver on a random tape of mine:
Everyday it's a'getting closer
Going faster than a roller coaster
Love like yours will surely come my way
Hey, hey.
Everyday it's a'getting faster
Everyone said go ahead and ask her
Love like yours will surely come my way
Hey, hey.
Everyday seems a little longer
and every way, love's a little stronger
come what may, do you ever long for
true love from me?
etc.
Unfortunately I'm not sure of the title or even if John Denver is the
original writer. If this is what you were looking for, let me know and
I'll dig deeper.
Clive Dawson
-------
∂01-Dec-83 0912 RJW symetrizing
the same problem you mentioned occurs in the tower of hanoi,
and even in showing properties of programs
eg you cant prove gcd(x y)|x unless you decide to
prove gcd(x y)|x ∧ gcd(x, y)| y.
one possible approach is a good approach to generalization
anyway. if you develop a goal that is a replica of an earlier
goal, then you can use induction (getting a recursive plan).
if you develop a goal that is almost but not quite a replica
of the previous goal, this inspires a generalization of the
previous goal, in the hope that you can then redo the proof
and the induction will go through.
richard
It is probably even better to detect the symmetries in advance,
and this can probably be done systematically. Certainly it is
so in map coloring, gcd and the tower of Hanoi. It is harder if
the problem isn't stated in a symmetric way, but
has subproblems that can be stated in a symmetric way.
So far as I know, no-one has really studied symmetry in AI or
combinatorial computations.
∂01-Dec-83 1008 WATROUS@RUTGERS.ARPA [COMEAU@RU-BLUE.ARPA: Re: [John McCarthy <JMC@SU-AI>:]]
Received: from RUTGERS by SU-AI with TCP/SMTP; 1 Dec 83 10:07:57 PST
Date: 1 Dec 83 13:07:25 EST
From: Don <WATROUS@RUTGERS.ARPA>
Subject: [COMEAU@RU-BLUE.ARPA: Re: [John McCarthy <JMC@SU-AI>:]]
To: JMC@SU-AI.ARPA
Mail-From: COMEAU@BLUE created at 1-Dec-83 12:59:28
Date: 1 Dec 83 13:00:47 EST
From: COMEAU@RU-BLUE.ARPA
Subject: Re: [John McCarthy <JMC@SU-AI>:]
To: WATROUS@RUTGERS.ARPA
In-Reply-To: Message from "Don <WATROUS@RUTGERS.ARPA>" of 1 Dec 83 12:55:51 EST
Cite: Emile Coue, used as motto for Silva Mind Control. Coue
wrote around 1916 I believe - My father heard about him when in
school and my father is 84.
Hope that's helpful.
Judy
-------
-------
∂01-Dec-83 1119 JJW
To: JMC, YOM
Janet Coursey has sent me a message giving several useful suggestions for the
EKL manual, corrections, and index entries. She also found a bug in EKL last
week, that Jussi has now fixed.
∂01-Dec-83 1421 @MIT-MC:DAM%MIT-OZ@MIT-MC LOGIC-IN-AI edited mailing list.
Received: from MIT-MC by SU-AI with TCP/SMTP; 1 Dec 83 14:18:46 PST
Date: Thu, 1 Dec 1983 17:13 EST
Message-ID: <DAM.11972097117.BABYL@MIT-OZ>
From: DAM%MIT-OZ@MIT-MC.ARPA
To: DAM%MIT-OZ@MIT-MC.ARPA, JMC@SU-AI.ARPA, BATALI%MIT-OZ@MIT-MC.ARPA,
AGRE%MIT-OZ@MIT-MC.ARPA, KDF%MIT-OZ@MIT-MC.ARPA, WELD%MIT-OZ@MIT-MC.ARPA,
JCMA%MIT-OZ@MIT-MC.ARPA
Subject: LOGIC-IN-AI edited mailing list.
The proposed charter for the new mailing list is presented
below. If you have any comments or suggestions please let me know.
I have constructed the new mailing list as a superset of phil-sci
(I copied the local phil-sci list so that individuals can be
deleted from logic-in-ai if they wish, however non-local phil-sci
mailing lists are currently included under logic-in-ai). I would
also be interested in suggestions for new readers with specific
interest in logic and AI.
CHARTER:
The LOGIC-IN-AI edited mailing list will provide a format for
high quality but informal discussions relating to logic and artificial
intelligence. The LOGIC-IN-AI mailing list will be edited; messages
should be sent to the editor (currently DAM@MIT-MC.ARPA) who will be
responsible responsible for the quality of the distributed messages
(anonymous refereeing may be used if the number of messages submitted
becomes large). Messages will be judged on the extent to which they
are technically well informed and are of interest to the readership;
messages need not contain new ideas or report on original research.
Rejected messages may be resubmitted after modification. Messages on
the following topics will be considered to be of general interest:
Specific logics and axiomatizations. For example logics of
time and action, logics of belief and knowledge, and reflective (or
meta) logics.
Mathematical models of reasoning domains. For example
mathematical models of computation, or mathematical models of time,
choice, and action. Mathematical models should be precisely defined
but need not be related to any formal syntax or "logic" for reasoning
about them.
Competence-Performance issues. Is completeness an important
pragmatic issue in inference systems? What should be done when the
inference problems are provably NP-Complete, or worse yet
non-recursive, or still worse what if the valid inferences are not
even recursively enumerable? Should an inference system know that
some questions are "unaswerable" (neither the statement nor its
negation follows)?
Non-Monotonic belief revision. Should inference systems jump
to conclusions while being able to retract these conclusions in the
light of further evidence? Should such systems be called "inference"
systems? Is non-monotonic belief revision the same as making
assumptions which can be later retracted (hypothetical reasoning)?
How is non-monotonic belief revision related to the scientific
induction of theory (belief) from data?
The general nature of logic and inference. What is "logic"?
What is "inference"? Is inference by definition monotonic? How is
inference related to other aspects of cognition such as perception,
scientific induction, and value judgment.
Foundations of mathematics. Why does set theory seem to be a
"special" collection of first order axioms? Does set theory express
universal innate principles of human thought, or does it describe a
particular actually existing universe of mathematical objects? The
formal foundation of mathematics is usually taken to be set theory
rather than unconstrained first order logic, does this mean that set
theory has more useful inference principles than those of
unconstrained first order logic?
Logic and belief. How is hypothetical inference related to
actual belief about the world? Human mathematics seems to involve
purely hypothetical inference. Thus mathematics is world-independent.
How can logic, as a subfield of mathematics, be related to the real
world? How is mathematics in general related to the real world?
Messages concerning PROLOG, or other issues surrounding
programming technology, will not be considered to be of interest to
the readership. Messages which are purely mathematical and which are
judged to be irrelevant to artificial intelligence will also be
considered not to be of interest to the readership.
FORMAT:
The format will be similar to that of a normal mailing
list with each author's message distributed individually (as opposed
to collecting messages into periodical issues or digests). In addition
to the normal header fields, each message should contain a "keywords"
field with keyword phrases separated by commas. For example:
To: DAM@MIT-MC
From: Author
Re: The Importance of Completeness.
Keywords: Importance of Completeness, Completeness
Keywords: Performance, Semantic Specification of Inference
I would like to submit the following to LOGIC-IN-AI:
The completeness of an inference system is generally irrelevant
to any pragmatic AI issues. ....
ARCHIVING AND DISCUSSION REPRINTS:
Messages set to LOGIC-IN-AI will be archived. An automated
system will be developed for extracting reprints of particular
discussions. This system will take a Boolean predicate on keywords
(and perhaps more complex predicates on messages) and extract the time
sequence of all messages satisfying that predicate. Initially
requests for discussion reprints should be sent directly to the editor
(DAM@MIT-MC.ARPA). Authors should attempt to make keywords specific
to discussion so that discussions can be extracted effectively by keyword
reference. Eventually the reprint service will be entirely automated.
COPYWRITING:
Authors should consider the messages sent to LOGIC-IN-AI to be
in the public domain. All requests for reprints will be serviced and
the editor will make no attempt to prevent messages sent to LOGIC-IN-AI
from being distributed on other mailing lists or in other formats.
Now that I have taken another look, I have only one more suggestion which
is to weaken the prohibition of Prolog and mathematical logic. I would
suggest
"Matters more appropriate for the Prolog Digest will not usually
be accepted for Logic-in-AI nor will matters of purely mathematical
logical or philosophical interest".
It seems to me that it would be appropriate for someone to enter a
discussion with a statement, "but that's just a Prolog program".
Further remark on logic in AI. Keep life simple for yourself and
for contributors. The biggest problem is that there won't be enough
submissions.
∂01-Dec-83 1654 DFH Fri. schedule
Will you be here tomorrow afternoon? Daryl Pratt,
CS206, wants to see you for a few minutes.
∂01-Dec-83 1753 WALDINGER@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 1 Dec 83 17:53:26 PST
Date: Thu 1 Dec 83 17:50:30-PST
From: WALDINGER@SRI-AI.ARPA
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Thu 1 Dec 83 09:53:00-PST
the aprroach to generalization in which you discover that
subgoals are near replicas of higher-level goals applies to
many sorts of generalizations, not just symmetrizing. in general
i think it is difficult to try to guess what the appropriate
generalization is without first trying to do the specific
problem.
-------
∂01-Dec-83 2327 vardi@Diablo Knowledge Seminar
Received: from SU-HNV by SU-AI with PUP; 01-Dec-83 23:27 PST
Date: Thu, 1 Dec 83 23:26 PST
From: Moshe Vardi <vardi@Diablo>
Subject: Knowledge Seminar
To: knowledge@Diablo
The first meeting of the Knowledge Seminar will be held at IBM
Research Lab on Friday, Dec. 9, 10:00am. Here are directions:
Take 101 south and exit at Ford Road
(the exit is on the left, and comes after Capitol Expressway and
Hellyer). Make a right off the exit ramp and after 100 yards or so turn
right onto a cloverleaf (there is a sign there marked IBM). Soon after
you exit this cloverleaf there will be yet another cloverleaf entrance
with a sign marked IBM. Ignore this sign! Instead, get into the left
lane and continue to the second traffic light, which is Beswick Rd. Make
a left on Beswick. Beswick ends at Cottle Rd., in front of Gate 3 to IBM.
Enter by this gate. The Research building (#28) is the large one on your
right. There is parking both in front and behind this building. Enter
by the front door, and register with the receptionist. She will point
out the auditorium where the seminar will be held.
The second meeting is planned for Friday, Dec. 16, 2:00pm.
∂02-Dec-83 1135 LUNDSTROM@SU-SCORE.ARPA [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 2 Dec 83 11:35:00 PST
Date: Fri 2 Dec 83 11:29:14-PST
From: Stephen Lundstrom <LUNDSTROM@SU-SCORE.ARPA>
Subject: [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
To: super@SU-SCORE.ARPA
Please note the following meeting which will be at Skilling Aud. on
Wed. next. Steve Lundstrom
---------------
Return-Path: <WEEKS@Ames-VMSB>
Received: from Ames-VMSB.ARPA by SU-SCORE.ARPA with TCP; Thu 1 Dec 83 19:01:00-PST
Date: 1 Dec 1983 1745-GMT
From: WEEKS@Ames-VMSB
Subject: ACM SIGBIG Meeting
To: lundstrom at su-score
Reply-To: WEEKS at Ames-VMSB
ASSOCIATION FOR COMPUTING MACHINERY
SAN FRANCISCO GOLDEN GATE CHAPTER
SPECIAL INTEREST COMMITTEE "SIGBIG"
For Large High Speed Computers
Wednesday, December 7,1983
7:00 Business Meeting (Volunteers are needed for Seminar,
Program and other committees.)
7:30 Speaker
Dr. George Michael
Lawrence Livermore Lab
SIGBIG Co-founder
Location: Stanford University, Skelling Auditorium
(Stanford campus maps are posted on AMES bulletin boards in N-233.)
Please let Mary Fowler know if you plan to attend.
For more information, contact Mary Fowler, TDC ACF, (415)965-6515
Arpanet fowler @ ames-vmsa
Ames DECnet CER::fowler
Ride Sharing, contact Frank Olken, LBL, (415)486-5891
Arpanet olken at lbl-csam
------
-------
∂03-Dec-83 0900 JMC*
Call Raj
∂03-Dec-83 1326 SJG general
John -
(1) Where can I get a copy of Winograd's thesis (or the generic equivalent)?
(2) Can I please have the name again of the `better' chess thesis?
(3) Winston's book on LISP seems to have a lot more examples than the document
you gave me, but a lot of them are riddled with prog's and go's. Am I best
off ignoring these?
(4) I am leaving for Oxford on Wednesday, and these silly messages will stop ...
Matt
(1) I have it somewhere, but the library is your best bet. There is less to
it than meets the eye.
(2) The author is Barbara Huberman. (Her present name is Liskov, and she is
a professor at M.I.T., but she works on less interesting problems.) Again
try the library.
(3) Progs and gotos have their place, but Winston uses them too much. Part
of the problem is that Maclisp doesn't compile pure lisp efficiently.
(4) We should get together again before Wednesday. Send me a note saying
when you will be back from Oxford.
∂03-Dec-83 1420 SJG travel
leaving for Oxford PM 12/7; returning PM 1/18; hopefully, between Huberman &
Winograd's thoughts (not to mention LISP generally and a few of my own perhaps)
I will have easily enough to keep me busy, but anything additional would be
appreciated.
∂03-Dec-83 1706 CHAPPELL@SRI-AI.ARPA TINLunch
Received: from SRI-AI by SU-AI with TCP/SMTP; 3 Dec 83 17:05:55 PST
Date: 3 Dec 1983 1705-PST
From: Chappell at SRI-AI
Subject: TINLunch
To: jmc at SAIL
cc: chappell
Stan suggest I try sending you mail at this netaddress. I've been
trying to contact you to see if you are going to be discussion leader and
'Paper Selector' for 12/15 TINLunch. Please let me know what the paper will be
ASAP, so the title can be included in the next tinlunch annoucement and
CSLI newsletter. If you have any questions please call me at 859-2312.
Thanks,
--Elsie Chappell
-------
∂03-Dec-83 1750 TREVOR@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Dec 83 17:50:47 PST
Date: Sat 3 Dec 83 17:50:49-PST
From: Trevor Hastie <TREVOR@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sat 3 Dec 83 16:59:00-PST
Thanks for the info. I think I would like to read a possibly thick book
that covors his whole life up till the end of the war. Im off to the
printers Inc right now to have a look at what is there.
-------
∂03-Dec-83 2232 FEIGENBAUM@SUMEX-AIM.ARPA Re: The Artificial Intelligence Report
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 3 Dec 83 22:32:38 PST
Date: Sat 3 Dec 83 22:32:12-PST
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM.ARPA>
Subject: Re: The Artificial Intelligence Report
To: LIBRARY@SU-SCORE.ARPA, JMC@SU-AI.ARPA
cc: Buchanan@SUMEX-AIM.ARPA
In-Reply-To: Message from "C.S./Math Library <LIBRARY@SU-SCORE.ARPA>" of Wed 30 Nov 83 16:32:28-PST
Harry,
Lou Robinson asked me to be on the Advisory Board, which I will probably do,
so will make it a quid-pro-quo that I get an extra copy for the Library.
In any event, there is too little in this minor publication to warrant the
library spending $150.
Ed
-------
∂04-Dec-83 1104 @MIT-MC:BATALI%MIT-OZ@MIT-MC Comments on your Charter
Received: from MIT-MC by SU-AI with TCP/SMTP; 4 Dec 83 11:03:58 PST
Date: Sun, 4 Dec 1983 14:00 EST
Message-ID: <BATALI.11972848436.BABYL@MIT-OZ>
From: BATALI%MIT-OZ@MIT-MC.ARPA
To: DAM%MIT-OZ@MIT-MC.ARPA
Cc: AGRE%MIT-OZ@MIT-MC.ARPA, JCMA%MIT-OZ@MIT-MC.ARPA, JMC@SU-AI.ARPA,
KDF%MIT-OZ@MIT-MC.ARPA, WELD%MIT-OZ@MIT-MC.ARPA
Subject: Comments on your Charter
In-reply-to: Msg of 1 Dec 1983 17:13-EST from DAM
1. I would prefer a weekly digest format, rather than individual
messages. This might help edit the mailing list -- you could impose
some limit on the size of the digest and use that limit to help decide
what gets included. There is no reason to hold to a perfectly fixed
limit of course. Also a weekly receipt would make the digest a bit
more welcomed -- it would be something to keep around and read at
leisure. Also it might tend to make people more careful in
submissions.
2. For people who want to discuss some issue brought up by the digest
in "real time" you could establish some sort of way to link people in
small mailing lists. You could then collect and edit the discussion
and include it in the next week's digest. Here's what I have in mind:
Suppose that there is a message in the digest titled "Why Logic
Sucks". You could announce the existence of a special "logic-sucks"
mailing list for the responses to that message. The author would be
on it, so would anyone who expressed an interest. Some people might
want to be on all such mailing lists, or only ones with certain
keywords. This would be some work, I suppose, but it would allow the
sort of fast-turnaround discussion we like, without overloading
everyone elses mail.
3. Rather than have submissions sent to you, make a mailing list
called LOGIC-IN-AI-SUBMISSIONS. Put yourself on it. This insulates
you from the unwashed masses, is easier to remember, and allows
minimal fuss if you ever resign, or adopt have some method of forwarding
submissions to alternate editors.
∂04-Dec-83 1453 KONOLIGE@SRI-AI.ARPA Re: ignorance
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Dec 83 14:53:15 PST
Date: Sun 4 Dec 83 14:55:02-PST
From: Kurt Konolige <Konolige@SRI-AI.ARPA>
Subject: Re: ignorance
To: JMC@SU-AI.ARPA
cc: YOM@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 16 Nov 83 23:43:00-PST
John, I showed how to formalize the ignorance part of the "hard"
wise man problem in my AAAI82 paper. I have a complete automatic proof of
the "hard" wise man problem, including subproofs of the ignorance of the
first two wise men, using a decision procedure (based on the Davis-Putnam
method) for one of the logics in my thesis. I have not yet tried to do the
S+P problem, because that involves using a more complicated nonpropositional
base for the modal logic. --kk
-------
∂04-Dec-83 2309 HST visit and compilation
To: CLT
CC: JMC
hi carolyn.thanks for your interest.yes i would like to visit stanford
but john did not give his final agreement.i'm on a conference in
honolulu and come back on 7.of januar.i could stay about a week at
stanford.the question(for me)would be how to get a cheap room and
a possibility to travel from s.f. to stanford.may be the room would
be enough.(I was only one time in stanford for 3 days.)i offered john
talks,at least: 1.compilation viewed as program transformation
2.programming styles in ai
but he semms not to be interested.
therefore,if you are interested and could help me to live somewhere
around stanford with my small rest of money(honolulu is not cheap)
then you can fix a date for my talk (between 7. and,say 12.jan.)
herbert
Carolyn, who is my wife, and I will arrange something involving
both my project and the company belonging to Richard Weyhrauch and
her. The details are uncertain, because she and I may go to Paris
early in January, but will be back before January 12. We don't
know yet whether and how long we'll be in Paris. Till when will
you be reachable by such messages, i.e. when are you leaving
Erlangen?
∂04-Dec-83 2343 HST visit
i leave erlangen on 2.januar.but spend 2 weeks in the wallis(alpen
mountains)therefore i'm reachable until 24.december.herbert
We'll make arrangements definite well before December 24 then.
∂05-Dec-83 0754 @MIT-MC:DAM%MIT-OZ@MIT-MC Comments on your Charter
Received: from MIT-MC by SU-AI with TCP/SMTP; 5 Dec 83 07:53:54 PST
Date: Mon, 5 Dec 1983 10:51 EST
Message-ID: <DAM.11973076081.BABYL@MIT-OZ>
From: DAM%MIT-OZ@MIT-MC.ARPA
To: BATALI%MIT-OZ@MIT-MC.ARPA
Cc: DAM%MIT-OZ@MIT-MC.ARPA, AGRE%MIT-OZ@MIT-MC.ARPA, JCMA%MIT-OZ@MIT-MC.ARPA,
JMC@SU-AI.ARPA, KDF%MIT-OZ@MIT-MC.ARPA, WELD%MIT-OZ@MIT-MC.ARPA,
ZVONA%MIT-OZ@MIT-MC.ARPA
Subject: Comments on your Charter
In-reply-to: Msg of 4 Dec 1983 14:00-EST from BATALI
Everyone has suggested that messages be sent to something
like LOGIC-IN-AI-EDITOR; this is a good idea and I will do it.
Several people have also suggested that we use a digest
format, collecting messages for a week or so and then sending them
out together. The main argument in favour of this idea is that
people would be less bothered by a single digest message than by
a flood of individual messages. However I think that in an edited
format the number of distributed messages could be kept down to
one or two a day (remember someone has to read every message
before it is distributed). Thus I don't think that there will
be a flood of messages.
I think there are a couple good reasons in favour of
a single message format. First I think it increases the
likelyhood that a given message will be read. It is easier
to look at a couple messages day than to look at ten messages
which arrive in single digest once a week. Furthermore the
more carefully and fully messages are read the more likely
it is that readers will respond to them; I am afraid of the list
dying for lack of messages. I also think
that the one day turn around time on message distribution is
important. Responses to messages can be recieved while the
original message is still fresh in the mind of its author.
The notion of an "issue" should be replaced by the
notion of a "discussion". A discussion reprint can be gotten
by asking for a reprint of all those messages satisfying a
certain keyword predicate.
David Mc
∂05-Dec-83 0847 RPG Capital Equipment
I got a message from Ohlander suggesting that he could possibly get the
Common Lisp project a 3600. Then we could use the money that would have bought
a third 3600 for us to upgrade the current ones with more memory.
What do you think?
-rpg-
I defer to your judgment on this, but it seems reasonable.
∂05-Dec-83 0900 JMC*
Reddy
∂05-Dec-83 0900 JMC*
Call Lee Douglas 781-9700, Robertson,Colman,Stephens
∂05-Dec-83 1026 DFH Tues. a.m. off
I'm planning to take tomorrow morning off and come
in at 1 pm.
∂05-Dec-83 1103 CLT Sato and proposal
To: "@SATO.DIS[1,CLT]"
Richard talked to Sato (some time ago)
and told him we are interested in the project
and will produce a proposal soon - so the March
deadline can be met.
Sato had in mind that they would come to Stanford
twice for around a month each time.
He was a little disappointed that we were thinking in terms
of a week or two.
You should compose a paragraph or two
- something about your current research interests,
and what you would do as part of the project -
Send this to JK@SU-AI - by December 15.
Jussi will make a draft proposal and we will go from there.
(I have Sato's letter and proposal if anyone wants to look at it.)
∂05-Dec-83 1953 YOM
tomorrow we should hand out the take-home exam. Do you have material?
∂05-Dec-83 2141 POURNE@MIT-MC maximum effort
Received: from MIT-MC by SU-AI with TCP/SMTP; 5 Dec 83 21:41:36 PST
Date: 6 December 1983 00:38 EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
Subject: maximum effort
To: REM @ MIT-MC, TAA @ MIT-MC, RMS @ MIT-MC, POURNELLE @ MIT-MC,
LLW @ SU-AI, JMC @ SU-AI, PDL @ MIT-XX
A maximum effort mission is under way within L-5 ,
Space-pac, NSI, and other space groups.
The decision window in the White House is short, three
to five weeks at MOST for State of Union. There is good reason
to believe the President favors an expanded space program, but
is being persuaded by advisors that this would not be popular.
Letters requesting President:
Dear Mr. President:
Lead us back to space again.
Other messages can include educational, commercial, etc.
advantages, fact that space is investment and thus not true
deficit. Projects should include LUNAR SETTLEMENT BEFORE END OF
DECADE, as well as space station.
We have an opportunity to affect the space program; this
is unique opportunity taht may not arise again in this decade,
due to special election, persuaion,economic, other circumstances
not likely to be duplicated.
This is the moment that space organizations were formed
to exploit.
Please pass this message on to nets, mail lists, anything you
thnk appropriate.
∂06-Dec-83 1410 ME news wires
∂05-Dec-83 1745 JMC
If it's our news service being copied, we are in violation of our agreements.
PKR - I think news service is also available at SRI, so you have your choice
of machines to try to get accounts on.
ME - Our news is not being copied. SRI has their own wire or two. They
are running some of our NS software, however.
∂06-Dec-83 1600 JMC*
Another ekl manual for home.
∂06-Dec-83 1707 HALPERN.SJRLVM1.IBM-SJ@Rand-Relay knowledge seminar
Received: from SU-HNV by SU-AI with PUP; 06-Dec-83 17:06 PST
Received: from rand-relay.ARPA by Diablo with TCP; Tue, 6 Dec 83 17:05:42 pst
Date: 6 Dec 1983 14:17:11-PST (Tuesday)
From: Joe Halpern <HALPERN.IBM-SJ@Rand-Relay>
Return-Path: <HALPERN.SJRLVM1.IBM-SJ@Rand-Relay>
Subject: knowledge seminar
To: KNOWLEDGE@SU-HNV
Via: IBM-SJ; 6 Dec 83 16:17-PST
The first meeting of the knowledge seminar will take place on
Friday, Dec. 9 at 10 A.M. in the main auditorium of Building 28
(the research building) at IBM. Part of this meeting will be
devoted to planning the shape of this series of seminars. There will
also be a talk by Joe Halpern. Here is the abstract:
APPLYING MODAL LOGIC TO REASON ABOUT KNOWLEDGE AND LIKELIHOOD
One way of capturing reasoning about knowledge in a formal
way is by means of modal logic. We illustrate this approach
by outlining a modal logic for reasoning about likelihood.
In the process, we will discuss some themes which will crop
up throughout this series of seminars: what exactly are we
trying to capture and how can we tell that we have captured it.
Note: the talk is self-contained. No background in modal logic
will be assumed.
There was also a small error in the directions to IBM circulated
last week. There is no sign to IBM at the first cloverleaf. The
sign just says "Blossom Hill Rd. and Cottle Rd." There is a sign
to IBM at the second cloverleaf. This should be ignored as stated in
the instructions. Here are the (corrected)
instructions again for those who missed them the first time around:
Take 101 south and exit at Ford
Rd. (the exit is on the left, and comes after Capitol Expressway and
Hellyer). Make a right off the exit ramp and after 100 yards or so turn
right onto a cloverleaf. Soon after
you exit this cloverleaf there will be yet another cloverleaf entrance
with a sign marked IBM. Ignore this sign! Instead, get into the left
lane and continue to the second traffic light, which is Beswick Rd. Make
a left on Beswick. Beswick ends at Cottle Rd., in front of Gate 3 to IBM.
Enter by this gate. The Research building is the large one on your
right. There is parking both in front and behind this building. Enter
by the front door, and register with the receptionist. She will point
out the auditorium where the seminar will be held.
∂06-Dec-83 1842 SJG contact
I will usually be with Sally in Oxford:
c/o Sally Greenhalgh
6, Wesley House Cottages
New Inn Hall Street
Oxford OX1 2DW
England 011-44-865-251255
or with Sally's family:
c/o Greenhalgh
Rosemary House
16 Church Street
Hampton
Middlesex TW12 2EG
England 011-44-1-979-5809
In the telephone numbers, 011 means international
44 means England
865 means Oxford, 1 means near London
and the rest is the local stuff.
The time difference is 8 hours.
Do have a happy holiday; I look forward to seeing you in January. And
thanks very much for welcoming me so warmly to Stanford. Makes a big difference.
Matt
∂07-Dec-83 0008 BACH@SUMEX-AIM.ARPA re: To the person with the red pencil and nasty temperament: (from SAIL's BBOARD)
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 7 Dec 83 00:08:07 PST
Date: Wed 7 Dec 83 00:08:03-PST
From: Rene Bach <BACH@SUMEX-AIM.ARPA>
Subject: re: To the person with the red pencil and nasty temperament: (from SAIL's BBOARD)
To: JMC@SU-AI.ARPA
cc: bach@SUMEX-AIM.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 5 Dec 83 17:41:00-PST
It's not so much the printing time which is relevant, but to spool such a file
takes a long time, builds up the queue, prevents others from spooling, and since
the DOVER is not able to print as it is spooling (hey, computer are so stupid,
(:-) ) there is some snowballing effect.
Rene
PS. I am not the person in question. I haved used pen and pencils because I
think it is reasonable to try to inforce a policy which is reasonable. If there
is no enforcement, then the policy is, unfortunately, useless. Adding a note
is a harmless and effective mean to "reinforce" (in the psychological meaning)
such policy. (see the BBOARD). Anynonymous is the only objectionable (as far as
I am concerned) point.
-------
Dover spooling
According to RPG a 157 page file took 7 minutes. If this is really all the
time it took, then the complaints have to be about something else. Either
that small files don't take proportionally less or simply that one doesn't
know when one's file is really going to be printed. When the xgp at the
AI Lab was overloaded (it's much slower), there wasn't the same level of
complaint, so I suspect that the lack of information is a major part of
the reason for so much complaint.
∂07-Dec-83 0900 JMC*
Lee Douglas 781-9700
Call Lee Douglas 781-9700, Robertson,Colman,Stephens
∂07-Dec-83 1017 DFH Lee Douglas
phoned again, has apparently tried several times.
This is in reference to Inference. He is planning
to have lunch with Doug Lenat today, and will keep
trying to reach you. His number is 781-9700.
∂07-Dec-83 1123 DFH flight reservations
for next week are on your calendar. Let me know
if you want anything changed.
∂07-Dec-83 1458 SJG right track
I have read your note and am no less than ever convinced that I am on
it, but have no new arguments. We can discuss it further when I return.
(I suspect that the proof for me of this pudding may be in the eating.)
Merry Christmas!
∂07-Dec-83 1557 BSCOTT@SU-SCORE.ARPA Len Bosack
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Dec 83 15:57:15 PST
Date: Wed 7 Dec 83 15:57:28-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: Len Bosack
To: JMC@SU-AI.ARPA
cc: BScott@SU-SCORE.ARPA
John, we FINALLY have approval to make a formal offer to Len Bosack. Gene's
5% salary recommendation was holding it up.
Anyway, is it o.k. with you if I write a letter to Len in Gene's behalf, offer-
ing the position to him, confirming the salary and the higher classification?
The classification changes from Computer Systems Specialist to University
Manager III, at the same C09 curve range, effective 11/14.
Please let me know as soon as possible.
Thanks,
Betty
-------
Ok about Len.
∂07-Dec-83 1651 BSCOTT@SU-SCORE.ARPA
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Dec 83 16:51:36 PST
Date: Wed 7 Dec 83 16:50:59-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
To: JMC@SU-AI.ARPA
cc: BSCOTT@SU-SCORE.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Wed 7 Dec 83 16:21:00-PST
Thanks for the speedy reply, John. I'll take care of it.
Betty
-------
∂08-Dec-83 0851 CLT
stoyan
∂08-Dec-83 0914 @MIT-MC:DAM%MIT-OZ@MIT-MC Comments
Received: from MIT-MC by SU-AI with TCP/SMTP; 8 Dec 83 09:14:49 PST
Date: Thu, 8 Dec 1983 11:54 EST
Message-ID: <DAM.11973873966.BABYL@MIT-OZ>
From: DAM%MIT-OZ@MIT-MC.ARPA
To: JMC@SU-AI.ARPA, JCMA%MIT-OZ@MIT-MC.ARPA
Subject: Comments
I was just wondering if you are still planning to send
comments on the charter for LOGIC-IN-AI.
David Mc
∂08-Dec-83 1241 @SU-SCORE.ARPA:VLSI.SLNDSTRM@SU-SIERRA.ARPA CS440 TODAY - Bob Keller -Rediflow
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Dec 83 12:41:07 PST
Received: from SU-SIERRA.ARPA by SU-SCORE.ARPA with TCP; Thu 8 Dec 83 12:38:47-PST
Date: Thu 8 Dec 83 12:38:45-PST
From: Steve Lundstrom <VLSI.SLNDSTRM@SU-SIERRA.ARPA>
Subject: CS440 TODAY - Bob Keller -Rediflow
To: super@SU-SCORE.ARPA
cc: vlsi.slndstrm@SU-SIERRA.ARPA
Following is an abstract for the CS440 meeting at 4:15pm TODAY
(room 200-034, history corner) Y'ALL COME!
(Bob will be arriving on campus around 2pm, anyone who wants a few
extra minutes with him, send me mail or call 7-0140 Steve Lundstrom)
Rediflow Multiprocessing
Robert M. Keller
Computing Research Group
Lawrence Livermore National Laboratory
and
Department of Computer Science
University of Utah
The goal of Rediflow (= "reduction" + "dataflow") is the development of a
programmable, scalable, multiprocessing system based upon communication and
data-structuring concepts from the realm of functional programming. More
specifically, Rediflow aims at providing concurrent execution for
irregularly-structured problems of medium and larger granularity (e.g.
knowledge bases, many AI applications, adaptive mesh numerical problems,
etc.), which are not treatable by "regular" approaches (e.g. SIMD machines,
processor arrays, etc.)
In addition to its reduction and dataflow basis, the "flow" metaphor also
appears in the technique for physical distribution of units of work to
processors. We use a "distributed" model in which every unit of memory
is dedicated to some processor, but which uses a common logical address space
to manage information across such memories. The load distribution and
balancing aspects of this model are achieved by a technique analogous to
"fluid flow" over a "surface" of processor-memories, with units of work as
molecules of the fluid.
The talk will elaborate on the above ideas and indicate preliminary
validation of Rediflow concepts via simulation.
-------
∂08-Dec-83 1340 REGES@SU-SCORE.ARPA TA for 258?
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Dec 83 13:39:58 PST
Date: Thu 8 Dec 83 13:38:25-PST
From: Stuart Reges <REGES@SU-SCORE.ARPA>
Subject: TA for 258?
To: mccarthy@SU-SCORE.ARPA
Office: Margaret Jacks 210, 497-9798
Do you think you will need a TA for 258? What kind of enrollment do you
anticipate?
-------
∂08-Dec-83 1412 TW
To: JMC, DEK, ZM, TW
To the committee:
I have put into your boxes the resumes of the candidates for consulting
professorships. I would like to reach a recommendation on the matter with
the minimum of meetings. I propose to do the following:
1) On the basis of the previous discussions and the resumes, I would like
a note back from each of you as to which of the candidates you would
be willing to approve without further discussion. If you would approve
them at a different rank from that proposed (associate for all but Kay,
who is full), let me know that.
2) On those for which there is unanimous agreement, no further discussion is
needed. For each of the others, let me know if there is any further information
(e.g., current research work, recommendations, etc.) that would help decide,
and I will try to get it as soon as possible.
3) I will schedule a meeting for sometime during the break to discuss the
hard cases, so that we will have a proposal ready for the faculty meeting on
January 10. I need to know when you will be available for such a meeting.
--------
To start, I need a response from you within a week (by DEC 15) as described
above. If I don't hear anything from you, I will assume you are happy
to approve all candidates as proposed, and proceed accordingly.
--t
----------------------------
Response form:
I approve the following appointments:
(Candidates: Grosz, Kay, Moore, Perrault, Rosenschein, Smith)
I would like the following information before going further:
I will NOT be available for meetings on the following days between Dec 15 and
January 6:
----------------------------
∂08-Dec-83 1458 SHORTLIFFE@SUMEX-AIM.ARPA new AAAI-M liaison
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 8 Dec 83 14:58:05 PST
Date: Thu 8 Dec 83 14:57:51-PST
From: Ted Shortliffe <Shortliffe@SUMEX-AIM.ARPA>
Subject: new AAAI-M liaison
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: psz@MIT-ML.ARPA, jmc@SU-AI.ARPA
Office: Room TC-117, Stanford Med Center; Phone: (415) 497-6979
Claudia,
We've counted all the ballots for the AAAI-M liaison election, and
Peter Szolovits is the new liaison. He is at MIT; I'm sure you have his
address in the files. I'll brief him on how we've handled mailing labels,
postage charges, xeroxing charges, etc. I hope it can be worked out
for him as easily at MIT as it was for us here at Stanford.
Pete will hold the office through the end of 1985.
Regards,
Ted
-------
∂08-Dec-83 1855 @MIT-MC:MACQUEEN@USC-ISIF bibliography and McCarty paper
Received: from MIT-MC by SU-AI with TCP/SMTP; 8 Dec 83 18:55:47 PST
Received: from MIT-MC by MIT-OZ via Chaosnet; 8 Dec 83 21:54-EST
Date: 8 Dec 1983 1854-PST
Subject: bibliography and McCarty paper
From: Dave MacQueen <MACQUEEN@USC-ISIF>
To: JMC@MIT-MC
Hi, John
By the way, thanks for you hospitality during my visit to MIT last week.
I had an interesting time, and I have to see if I can't make it back again
soon. I'll have to pick a better time to talk with Albert, and there
seems to be a lot to discuss with the Scheme people. λλ
I forgot to get a copy of Albert's famous types bibliography. Could you
send me a copy please? Also, I meant to ask you about that McCarty paper
you mentioned some time ago. If you
-------
Curiously enough, JMC@MIT is me at Stanford. Who was your host at M.I.T.,
since I sometimes get mail to JMC@MC that obviously wasn't addressed to me?
∂08-Dec-83 1900 @MIT-MC:MACQUEEN@USC-ISIF oops...
Received: from MIT-MC by SU-AI with TCP/SMTP; 8 Dec 83 19:00:35 PST
Received: from MIT-MC by MIT-OZ via Chaosnet; 8 Dec 83 21:59-EST
Date: 8 Dec 1983 1858-PST
Subject: oops...
From: Dave MacQueen <MACQUEEN@USC-ISIF>
To: JMC@MIT-MC
oops... (once again I forgot that I wasn't in emacs and sent the mail with
an <escape>). I was saying that I would like to get a copy if it isn't too
much trouble. Or you could give me a reference and I could contact McCarty
(do you have a net address for him?).
Merry Christmas, etc...
Dave
-------
∂09-Dec-83 0614 RMS@MIT-MC
Received: from MIT-MC by SU-AI with TCP/SMTP; 9 Dec 83 06:13:49 PST
Date: 9 December 1983 09:11 EST
From: Richard M. Stallman <RMS @ MIT-MC>
To: POURNE @ MIT-MC, jmc @ SU-AI
This is the latest word I got through the phone tree.
JMC, maybe you can post it on various machines at Stanford.
DISTRIB: *BBOARD
EXPIRES: 12/14/83 09:09:06
RMS@MIT-MC 12/09/83 09:09:06 Re: Update on action for space program
I now hear that President Reagan discussed three options for
the space program with the cabinet last week:
* nothing
* space station
* space station and lunar station
The third option would advance fastest, so that is what
space advocates are supporting. If you wish to add your support,
the thing to do is phone these five people (their staffs, actually, I suspect):
Malcolm Baldridge Secretary of Commerce
(202) 377-2112
William French Smith Attorney General
(202) 633-2001
William Clarke Secretary of the Interior
(202) 343-7351
William Brock US Trade Representative
(202) 395-3204
President Reagan
(202) 456-7639
I'm told it is important to call by Monday the 12th.
∨
∂09-Dec-83 1034 uucp@Shasta Lisp for IBM
Received: from SU-SHASTA by SU-AI with TCP/SMTP; 9 Dec 83 10:34:26 PST
Received: from decwrl by Shasta with UUCP; Fri, 9 Dec 83 09:17 PST
Date: 9 Dec 1983 0903-PST (Friday)
Sender: uucp@Shasta
From: decwrl!baskett (Forest Baskett) <decwrl!baskett@Shasta>
Subject: Lisp for IBM
Message-Id: <8312091703.AA22621@DECWRL>
Received: by DECWRL (3.327/4.09) 9 Dec 83 09:03:04 PST (Fri)
To: jmc@sail
I hear you might be planning a new project? Are any details public?
Forest
∂09-Dec-83 1048 avg@diablo
Received: from SU-HNV by SU-AI with TCP/SMTP; 9 Dec 83 10:37:38 PST
Date: Fri, 9 Dec 83 10:14 PST
From: Allen VanGelder <avg@diablo>
To: JMC@SU-AI
I think that's an est chant ... so Werner Erhart gets "credit".
∂09-Dec-83 1403 JJW Janet Lee's incomplete
Janet Lee has given me her project from last year's 206 class, to make
up her incomplete. If it's OK with you, I'll look it over this weekend,
suggest a grade, and give the necessary form to Diana on Monday for your
signature. I'm going to be away from Monday afternoon until Friday, but
would like to get this taken care of as soon as possible.
Agreed about Janet Lee.
∂09-Dec-83 1625 minker%umcp-cs@CSNet-Relay Gregory Minc
Received: from CSNET-CIC by SU-AI with TCP/SMTP; 9 Dec 83 16:25:54 PST
Date: 9 Dec 83 16:40:00 EST (Fri)
From: JACK MINKER <minker%umcp-cs@CSNet-Relay>
Return-Path: <minker%umcp-cs@CSNet-Relay>
Subject: Gregory Minc
To: henschen@ANL-MCS
Cc: Wos@ANL-MCS
Via: UMCP-CS; 9 Dec 83 18:23-EST
Dear Larry,
Wolfgang Bibel was just in Leningrad to attend a workshop on AI. He met
Gregory Minc who has done work in theorem proving. Minc informed Bibel that
he is a refusenik. That is, that he has requested an exit visa and was
denied one by the Soviet Government. Minc has lost his job and has no means of
support. Would it be possible to send him a complemenary copy of the
Theorem Proving Newsletter? If you cannot do this, I will try to obtain
support to have him funded either from the ACM or from the Committee of
Concerned Scientists. I am not sure that I can obtain the support and that is
why I am asking you first.
As a second point with respect to Minc, assuming that you are going to have a
journal of theorem proving, would it be appropriate to have him as one of the
editors? I am not familiar enough with his work in theorem proving. I am sure
that Wolfgang Bibel knows his work and could comment on this matter. Assuming
that he is appropriate, it would be immensely helpful to him to be an editor
of an international journal.
Please let me hear from you on this matter. For your information, Minc's
address is:
Gregory Minc
USSR 192242
Leningrad
Poste Restante, Box 16
Best regards,
Jack Minker
∂10-Dec-83 1015 JK
∂09-Dec-83 1514 JMC motivation on "Algebra of Types"
I haven't got far into the paper and don't know whether I will,
but I think the paper would be read much more if you would supply
a motivational introduction. It should mention the main examples of
functions with variable numbers of arguments, e.g. LIST, APPEND,
MAPLIST and MAPCAR and whatever others you know whose formalization
require require extended notions of type. I know that it is customary
in mathematical writing to leave this out, but it will make a big
difference here, not least of all to our patient sponsors.
------------
O.K. - I will put in some examples of EKL proofs on MAPCAR, once
I got the type stuff fixed.
I'm not sure examples of EKL proofs are required, although they
will be helpful. What I think is more important is a brief discussion
of a few functions with variable numbers of arguments, etc. and
why they motivate your more complex types.
∂10-Dec-83 1427 CLT alice
has plans for tonight
we are going to play trios with glb tomorrow late afternoon
then we wanted to go to supper with rww and yasuko
is that ok with you?
∂11-Dec-83 1201 JJW Janet Lee
I looked over Janet Lee's EKL assignment and final project. She didn't
get very far with EKL; the project (an algebraic simplifier) was fair but
not good, since it would make errors on some fairly simple cases that
obviously weren't tested. I gave the project a B-.
I would recommend an overall grade of B-, comparable to others in her
situation. Actually that might be somewhat generous since she scored
quite low in the exams, but you gave nothing lower than B- to anyone who
completed all of the work.
Diana has the assignments, and I've asked her to get a grade form for you
to sign.
∂11-Dec-83 2025 RESTIVO@SU-SCORE.ARPA Re: What is the moral of this story?
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Dec 83 20:25:03 PST
Date: Sun 11 Dec 83 20:25:14-PST
From: Chuck Restivo <RESTIVO@SU-SCORE.ARPA>
Subject: Re: What is the moral of this story?
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sun 11 Dec 83 17:42:00-PST
That my entire emotional stability is relational to a garlic pizza.
-------
∂11-Dec-83 2345 EJH@SU-SCORE.ARPA Re: What is the moral of this story?
Received: from SU-SCORE by SU-AI with TCP/SMTP; 11 Dec 83 23:45:41 PST
Date: Sun 11 Dec 83 23:45:56-PST
From: Eric J. Horvitz <EJH@SU-SCORE.ARPA>
Subject: Re: What is the moral of this story?
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sun 11 Dec 83 17:42:00-PST
Insanity Prevails Even In The Worst of Situations
-------
∂12-Dec-83 1008 TW
To: ZM@SU-AI, JMC@SU-AI, golub@SU-SCORE
∂10-Dec-83 1951 DEK consulting profs
As I mentioned yesterday, it seems OK to go along with the
proposals, now that all the re'sume's are available.
However, I think you made a mistake when you said "associate for all
by Kay"; there was at least one proposed for assistant rank.
Namely, I'm pretty sure Smith was to be Consulting Assistant Professor
[especially because he is less than two years past the PhD].
With that amendment, I approve all the appointments unless somebody
else on our committee has reservations; in the latter event, I
want to hear about them.
Between 15 Dec and 6 Jan, however, I guess I don't want to hear about
them, because I won't be around!
To me the ranks are not important, and therefore, I approve the proposal
as amended. Kay consulting full, Smith consulting asst., Rosenschein,
Moore, Grosz and Perrault consulting assoc.
∂12-Dec-83 1409 AAAI@SRI-AI.ARPA next Presidential message
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Dec 83 14:08:35 PST
Date: Mon 12 Dec 83 14:08:54-PST
From: name AAAI-OFFICE <AAAI@SRI-AI.ARPA>
Subject: next Presidential message
To: JMC@SU-AI.ARPA
cc: aaai@SRI-AI.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
John,
We're getting the next issue of the AI Magazine ready. Do you
want to include another message to the membership in this issue?
Claudia
-------
Yes, I intend to do one each quarter. What is the deadline?
∂12-Dec-83 1507 AAAI@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 12 Dec 83 15:07:25 PST
Date: Mon 12 Dec 83 15:06:31-PST
From: name AAAI-OFFICE <AAAI@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
cc: AAAI@SRI-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 12 Dec 83 14:59:00-PST
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
The deadline is January 3, 1984.
-------
∂12-Dec-83 1847 MACQUEEN@USC-ISIF
Received: from USC-ISIF by SU-AI with TCP/SMTP; 12 Dec 83 18:47:44 PST
Date: 12 Dec 1983 1846-PST
From: Dave MacQueen <MACQUEEN@USC-ISIF>
To: John McCarthy <JMC@SU-AI>
In-Reply-To: Your message of 09 Dec 83 0055 PST
Sorry about the misaddressed msλessage. I was intending to send it to
John Mitchell, a graduate student styλudying with Albert Meyer. His address
is JCM@MC.
Dave MacQueen
-------
∂13-Dec-83 1610 BSCOTT@SU-SCORE.ARPA ARPA Funding
Received: from SU-SCORE by SU-AI with TCP/SMTP; 13 Dec 83 16:10:34 PST
Date: Tue 13 Dec 83 16:10:29-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: ARPA Funding
To: JMC@SU-AI.ARPA, ZM@SU-AI.ARPA, DCL@SU-AI.ARPA, TOB@SU-AI.ARPA,
Wiederhold@SUMEX-AIM.ARPA
cc: BScott@SU-SCORE.ARPA
I sent a message to Ohlander and Machado on Friday concerning the progress of
the negotiations on the new umbrella contract. Ohlander responded by net
message and Machado called me. Here is the latest information.
The contract is not going to have a December 1 start date--negotiations have
not begun. And when I talked to Machado he had not yet delivered the task
orders to the contracts people; he said he would do that in a day or so, but
this further delays the start date. I was also told that we could not have
preaward costs, and that the contract could not be backdated. Machado said
that we could work toward a January start date, but it seems highly unlikely
that negotiations between Stanford and Navelex, which have not even begun,
can be completed by the end of the month.
Ohlander told me that it may be possible to get an approximate $50K modification
to the current contract to enable the continuation of the research. I told
him that $50K would keep everyone going for about 1.5 months, subject to the
understanding that Binford and Luckham agree to further postpone the purchase
of equipment. Machado also mentioned this $50K modification, and said that he
would get detailed information to me in the next day or so. I have not heard
further from him. If he doesn't call by tomorrow, I'll try to reach him.
One further note concerning the current contract: you will all receive an
indirect cost rebate on expenditures beginning 9/l/82, at which time the
rate went from 58% to 69%. I will have the precise amount of each rebate
sometime next week and will let you know the amounts. These $$ will be added
to your unrestricted accounts by Stanford.
I will keep you all informed as negotiations on the new contract get under
way. And will certainly let you know the details about the $50K modification
as soon as I have more information.
Betty
-------
∂13-Dec-83 1818 Newman.ES@PARC-MAXC.ARPA Re: New Generation computing: Japanese and U.S. views
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 13 Dec 83 18:18:19 PST
Date: 13 Dec 83 18:18:23 PST (Tuesday)
Subject: Re: New Generation computing: Japanese and U.S. views
In-reply-to: TREITEL's, Rogers's, and McCarthy's messages of Tue, 13 Dec
83
To: Richard Treitel <TREITEL@SUMEX-AIM.ARPA>
To: David Rogers <DRogers@SUMEX-AIM.ARPA>
To: John McCarthy <JMC@SU-AI.ARPA>
cc: Newman.es@PARC-MAXC.ARPA, SU-BBOARDS@SU-AI.ARPA
From: Ron Newman <Newman.es@PARC-MAXC.ARPA>
My juxtaposition of quotations is intended to demonstrate the difference
in priorities between the Japanese and U.S. "next generation" computer
research programs. Moto-Oka is a prime mover behind the Japanese
program, and DARPA's Robert Kahn is a prime mover behind the American
one. Thus I consider the quotations comparable.
To put it bluntly: the Japanese say they are developing this technology
to help solve human and social problems. The Americans say they are
developing this technology to find more efficient ways of killing
people.
The difference in intent is quite striking, and will undoubtedly produce
a "next-generation" repetition of an all too familiar syndrome. While
the U.S. pours yet more money and scientific talent into the military
sinkhole, the Japanese invest their monetary and human capital in
projects that will produce profitable industrial products.
Here are a couple more comparable quotes, both from IEEE Spectrum, Vol.
20, No. 11, November 1983:
"DARPA intends to apply the computers developed in this program to a
number of broad military applications...
"An example might be a pilot's assistant that can respond to spoken
commands by a pilot and carry them out without error, drawing upon
specific aircraft, sensor, and tactical knowledge stored in memory and
upon prodigious computer power. Such capability could free a pilot to
concentrate on tactics while the computer automatically activated
surveillance sensors, interpreted radar, optical, and electronic
intelligence, and prepared appropriate weapons systems to counter
hostile aircraft or missiles....
"Such systems may also help in military assessments on a battlefield,
simulating and predicting the consequences of various courses of
military action and interpreting signals acquired on the battlefield.
This information could be compiled and presented as sophisticated
graphics that would allow a commander and his staff to concentrate on
the larger strategic issues, rather than having to manage the enormous
data flow that will[!] characterize future battles."
--Robert S. Cooper and Robert E. Kahn, DARPA, page 53.
"Fifth generation computers systems are exptected to fulfill four
major roles: (1) enhancement of productivity in low-productivity areas,
such as nonstandardized operations in smaller industries; (2)
conservation of national resources and energy through optimal energy
conversion; (3) establishment of medical, educational, and other kinds
of support systems for solving complex social problems, such as the
transition to a society made up largely of the elderly; and (4)
fostering of international cooperation through the machine translation
of languages."
--Tohru Moto-Oka, University of Tokyo, page 46
Which end result would *you* rather see?
/Ron
∂13-Dec-83 2032 TREITEL@SUMEX-AIM.ARPA Comparable quotes
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 13 Dec 83 20:31:51 PST
Date: Tue 13 Dec 83 20:31:43-PST
From: Richard Treitel <TREITEL@SUMEX-AIM.ARPA>
Subject: Comparable quotes
To: su-bboard@SUMEX-AIM.ARPA, newman.es@PARC-MAXC.ARPA
cc: treitel@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA
Ron,
If you think that I, or anyone else on this system, regards the use of advanced
computing technology for military vs. civilian purposes as mutually exclusive,
you are out to lunch. The question "Which outcome would you rather see?" is
thus bogus. We will definitely see both, or neither. I prefer both.
Those who criticise expenditure on military-oriented research should at least
consider what it would cost to maintain an equally effective force without the
benefits of such research. If you think NATO's military forces are too
effective, that is a separate issue.
- Richard
-------
∂13-Dec-83 2204 KJB@SRI-AI.ARPA Semantics of computer languages seminar
Received: from SRI-AI by SU-AI with TCP/SMTP; 13 Dec 83 22:04:28 PST
Date: Tue 13 Dec 83 21:59:45-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Semantics of computer languages seminar
To: Goguen@SRI-AI.ARPA, Meseguer@SRI-AI.ARPA
cc: Briansmith@PARC-MAXC.ARPA, tw@SU-AI.ARPA, jmc@SU-AI.ARPA
Dear Pepe and Joe,
This quarter's seminar on the semantics of computer languages
came to an end today. We had a series of speakers that included,
off the top of my head, Brain, Carolynn Talcott, Glynn Winskel,
Yiannis Moschovakis, Joe Halpern, Henson Graves, and me. The basic
issues seemed to revolve around just what ones to do with a semantics,
and what the alternatives are, and how they are related. I think it
was fairly useful, but also a bit superficial, as we never did dig into
anything. There is a general feeling that if we are to contine next
quarter, we should dig in. Do either of you have a suggestion of
how we might proceed in a useful way to build up our expertise in this
area? John McCarthy has one talk that we would like to have, but
otherwise the slate is clean.
Jon
-------
∂14-Dec-83 1038 vardi@diablo Knowledge Seminar
Received: from SU-HNV by SU-AI with TCP/SMTP; 14 Dec 83 10:37:55 PST
Date: Wed, 14 Dec 83 09:33 PST
From: Moshe Vardi <vardi@diablo>
Subject: Knowledge Seminar
To: knowledge@Diablo
We do not meet this Friday. The next meeting will be in January.
∂14-Dec-83 1050 JK proposals for japan
To: "@SATO.DIS[1,CLT]"
I am still waiting for responses from most of you -
deadline dec 15.
∂14-Dec-83 1055 JK
∂14-Dec-83 0111 JMC
To: JK, CLT
See japan[f83,jmc].
-------
Thanks. I have copied it into sato.tex[1,jk]
∂14-Dec-83 1206 MOGUL@SU-SCORE.ARPA Re: Comparable quotes
Received: from SU-SCORE by SU-AI with TCP/SMTP; 14 Dec 83 12:05:58 PST
Date: Wed 14 Dec 83 11:58:31-PST
From: Jeffrey Mogul <MOGUL@SU-SCORE.ARPA>
Subject: Re: Comparable quotes
To: TREITEL@SUMEX-AIM.ARPA, newman.es@PARC-MAXC.ARPA
cc: jmc@SU-AI.ARPA
In-Reply-To: Message from "Richard Treitel <TREITEL@SUMEX-AIM.ARPA>" of Tue 13 Dec 83 20:32:20-PST
I don't want to get into a debate on the morality of military research,
but I would like to point out a seemingly counter-intuitive result
of applying extremely high-tech to military systems:
The Israelis have one of the most effective military organizations in
the world, and while they are not "backward" their weapons have a
reputation for simplicity, as opposed to state-of-the-art-hood.
On the other hand, the US, which arguably has applied the most advanced
technology to weaponry (well, maybe France is ahead on some things and
Britain on others) has come up with the following:
A Navy which, having spent all its money on new, flashy ships
and planes and almost none on spare parts, finds that it has
too few to maintain readiness
The Aegis cruiser (?) which apparently is a high-tech failure
The F-18, which doesn't meet goals for distance and payload,
yet costs many times more than estimated
The list goes on. I believe that if the US wants to maintain a strong
conventional defense, it must stop buying new toys and start maintaining
the proven equipment it has.
And, to toss out another thought: the more advanced a strategic
weapon system is, the more destabilizing (some) people perceive it to
be.
-Jeff
-------
∂14-Dec-83 1357 TREITEL@SUMEX-AIM.ARPA Re: Comparable quotes
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 14 Dec 83 13:57:39 PST
Date: Wed 14 Dec 83 13:56:53-PST
From: Richard Treitel <TREITEL@SUMEX-AIM.ARPA>
Subject: Re: Comparable quotes
To: MOGUL@SU-SCORE.ARPA
cc: newman.es@PARC-MAXC.ARPA, jmc@SU-AI.ARPA, treitel@SUMEX-AIM.ARPA
In-Reply-To: Message from "Jeffrey Mogul <MOGUL@SU-SCORE.ARPA>" of Wed 14 Dec 83 12:05:39-PST
Jeff,
The Israelis apply pretty high technology to their armed forces, but having had
far more experience with actual combat than the U.S., and being plagued with
far less pork-barrel syndrome (would you be willing, as a Congressbeing, to
balance the jobs of your voters against the survival of your entire district?),
they do a better job of it than the U.S. Remember the enormous importance of
ECM in what they did to the Syrian Air Force last year (or whenever). The
problems you mention, while very real, are rooted in the procurement and budget
processes, and are only exacerbated by the cost of high tech. The basic cause
of the trouble is that the U.S. armed forces have been made to believe that
they will never be required to do any real fighting (or that Congress will
never allow this to happen), so all they have left to rejoice in is their
gold-plated toys, which don't even need to work right as long as they keep
Lockheed employees on the job.
Since you mention Britain, in a recent campaign which I need hardly name, the
computer-guided Sea Wolf missile and the AIM-9L did their jobs far better
than the older, cheaper Sea Cat and the Rapier. I'm willing to admit that
none of the weapons we used were up-to-the-minute state-of-the-art, but close.
Regarding strategic weapons, I guess you're mainly talking accuracy. However,
modern technology can also improve the chances of a warhead reaching its
target, which makes it less destabilising (better deterrence).
- Richard
-------
∂14-Dec-83 1401 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 14 Dec 83 14:00:53 PST
Date: Wed 14 Dec 83 14:00:31-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: next meeting
To: JMC@SU-AI.ARPA, Buchanan@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I'd like to schedule the next meeting regarding the on-line abstract/library
service sometime during the first week in January (Jan 3-6). Jake's colleague
at DIALOG will not be back into her office until that week. So, what's
the best time for you to meet during that week or the following week?
Regards,
Claudia
-------
∂14-Dec-83 1458 WALKER@SRI-AI.ARPA Re: next meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 14 Dec 83 14:57:55 PST
Date: Wed 14 Dec 83 14:57:26-PST
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: next meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: JMC@SU-AI.ARPA, Buchanan@SUMEX-AIM.ARPA, Genesereth@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Amsler@SRI-AI.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of Wed 14 Dec 83 14:01:44-PST
it will have to be toward the end of the week at the earliest
-------
∂14-Dec-83 2308 JJW Let-by-need
I'm looking at the let-by-need assignments for Yoram. Just starting, it
occurred to me that the handout doesn't mention a case that it should handle.
(Although it does hint at how to do it.) The case is something like
(let-by-need ((x exp))
(cons (if p x a) (if q x b)))
The outermost expression is not a conditional, which is what the handout
spends most of its time talking about. The first person's solution that I
looked at would notice this and then treat the arguments separately, which
would cause X to be computed twice if both P and Q are true.
The correct way to do it, it seems, is to push the inner conditionals to
the outside before anything else. Actually only one has to be done, and
then the rest is easy.
Some people have been careful enough to handle this case. I wonder how
much I should count against the others who don't do it right.
Certainly the people who do it right should get more credit. How much
I leave to you.
∂16-Dec-83 0413 YOM Finals
i
Well, my plane is leaving in a few hours. I've graded half of the questions
(numbers 1, 3 and 6). For the other questions I suggest the following point
distribution:
#2 - 15pt (5 5 5)
#4 - 25pt (12 2 11)
#5 - 10pt (single section.)
2 students asked for an incomplete - Bryan Sayre (who said he'd be happy with
any B flavored or even C grade), and Samson Tu. They want to complete HW
assignments. Bryan Sayre had missinterpreted the instructions on the final
and spent time proving facts about partition (#3) and claims to have therefore
lacked time to conclude other sections. I suggested adding 5 pts to his total
for that work. Ian Mason gave an especially complete proof of the cflat
question, showing 2cflat ≤ cflatten +1. I suggest adding 5 pts to his total.
Otherwise it seems that most students have a handle on the material. Although
they might still mess up the match-sublis question (#4). A number of students
asked me during the final whether proving that match ≠ NO ⊃ sublis[match] is
the identity (a precise formulation of this) was enough, and I approved.
The 4th part of #1 ended up to be longer than I expected, and perhaps worth more
than the 5pts alloted. I gave 20pts for #1 (5 5 5 5), 15pts for #3, 15pts for
#6 (5 10). #6 b became fairly easy with the change.
Joe (jjw) will have the list of grades for HW assignments. If you need me
desperately Sunday or Monday, my home phone # is (ISRAEL) -57- 690946.
The time difference is +10 hours (later there). I return on Thursday, Jan 12th.
Happy holidays
Yoram.
I couldn't find a numerical grade for Robert Given's prob. 6, but all
parts were marked with check, so I assumed he got 15.
∂16-Dec-83 0742 MCCORDUCK@COLUMBIA-20.ARPA Re: Doctor's dilemma
Received: from COLUMBIA-20 by SU-AI with TCP/SMTP; 16 Dec 83 07:42:31 PST
Date: Fri 16 Dec 83 10:42:54-EST
From: Pamela McCorduck <MCCORDUCK@COLUMBIA-20.ARPA>
Subject: Re: Doctor's dilemma
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Thu 15 Dec 83 08:53:00-EST
John,
What I have in manuscript form right now is a brief summary, a sample
scenario or two, and then some details about the answer your piece
provides. All fully cited, of course. Since this is for a book that
won't even be delivered to the publisher until September, and then will
take another year (assuming the usual publishing sloth) it won't see
the light of day in my book until fall 1985.x Hope that's okay. Good
luck publishing the piece in its hilarious entirety.
Pamela
-------
There is no problem with what you plan to publish. Any suggestions on
where to try for whole piece?
∂16-Dec-83 0910 WALKER@SRI-AI.ARPA Tutorials at IJCAI-85
Received: from SRI-AI by SU-AI with TCP/SMTP; 16 Dec 83 09:09:51 PST
Date: Fri 16 Dec 83 09:09:18-PST
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Tutorials at IJCAI-85
To: Lenat@SU-SCORE.ARPA
cc: aaai-office@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA, bledsoe@UTEXAS-20.ARPA
Doug,
I know that you have hardly begun tutorializing for AAAI-84, but
we desperately need information about plans for the IJCAI-85 UCLA meeting.
Claudia tells me that she sent you materials regarding rooms and locations
there. Can you provide her with the appropriate guidance immediately?
Or is it the case that you do not expect to be continuing in the tutorial
role in 1985? If so, we need to know that right away so we can make
other arrangements.
Don
-------
∂16-Dec-83 1438 LENAT@SU-SCORE.ARPA Re: Tutorials at IJCAI-85
Received: from SU-SCORE by SU-AI with TCP/SMTP; 16 Dec 83 14:38:24 PST
Date: Fri 16 Dec 83 14:37:35-PST
From: Doug Lenat <LENAT@SU-SCORE.ARPA>
Subject: Re: Tutorials at IJCAI-85
To: WALKER@SRI-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA, bledsoe@UTEXAS-20.ARPA
In-Reply-To: Message from "Don Walker <WALKER@SRI-AI.ARPA>" of Fri 16 Dec 83 09:09:41-PST
I will be back in about 6 days, and can get you a
list of desired rooms then. Sorry for the delay.
Doug
-------
∂16-Dec-83 1648 DFH Messages
1. 12/15 John Funk, First Interstate Capital, 213-622-1922. Referred by
Alex Jacobson
2. 12/15 Claudia Mazzetti called.
∂16-Dec-83 2231 POURNE@MIT-MC
Received: from MIT-MC by SU-AI with TCP/SMTP; 16 Dec 83 22:31:26 PST
Date: 17 December 1983 01:32 EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
To: RMS @ MIT-MC
cc: jmc @ SU-AI
In-reply-to: Msg of 9 Dec 1983 09:11 EST from Richard M. Stallman <RMS>
It appears we have got to about 75% CHANCE OF GETTING THE LUNAR
SETTLEMENT. ANY PUSH TO CONVINCE WHITE HOUSE STAFF THAT THIS IS
VOICE OF THE PEOPLE WILL HELP A LOT.
Keep trying. Can you repeat this message to all of network
land?
Date: 9 December 1983 09:11 EST
From: Richard M. Stallman <RMS>
To: POURNE, jmc at SU-AI
This is the latest word I got through the phone tree.
JMC, maybe you can post it on various machines at Stanford.
DISTRIB: *BBOARD
EXPIRES: 12/14/83 09:09:06
RMS@MIT-MC 12/09/83 09:09:06 Re: Update on action for space program
I now hear that President Reagan discussed three options for
the space program with the cabinet last week:
* nothing
* space station
* space station and lunar station
The third option would advance fastest, so that is what
space advocates are supporting. If you wish to add your support,
the thing to do is phone these five people (their staffs, actually, I suspect):
Malcolm Baldridge Secretary of Commerce
(202) 377-2112
William French Smith Attorney General
(202) 633-2001
William Clarke Secretary of the Interior
(202) 343-7351
William Brock US Trade Representative
(202) 395-3204
President Reagan
(202) 456-7639
I'm told it is important to call by Monday the 12th.
↑←
∂17-Dec-83 1951 PYLYSHYN@CMU-CS-C.ARPA The Frame Problem and other enigmas
Received: from CMU-CS-C by SU-AI with TCP/SMTP; 17 Dec 83 19:50:53 PST
Received: ID <PYLYSHYN@CMU-CS-C.ARPA>; Sat 17 Dec 83 22:46:44-EST
Date: Sat 17 Dec 83 22:46:43-EST
From: PYLYSHYN@CMU-CS-C.ARPA
Subject: The Frame Problem and other enigmas
To: jmc@SU-AI.ARPA, BMOORE@SRI-AI.ARPA
Last spring I spoke with a number of people at the Cog Sci meeting in Rochester
and out of that came an idea to put together a collection of papers on the so-
called Frame problem and related basic problems of AI. Dennett has written a
piece on it called Cognitive Wheels (which is nice but goes off on a bit of an
irrelevance) and .Clark Glymour (Phil@Pitts) and Dreyfus wrote replies. In
addition a fellow named Janlert from Sweden wrote a nice paper on the
general Frame problem (whic is not published) and Fodor has some stuff in his
modularity book on it. Pat Hayes and John Haugeland are both enthusiastic
about writing something on it (Pat has been chomping at the bit for years to
"straighten out" the various people who have talked about the Frame Problem)
and John Haugeland has agreed to give a talk on it soon. Would you be willing
to write a piece about either the Frame Porblem as it was desribed by McCarthy
and Hayes or about the general class of holism problems of which it is a
member? think it would make a nice collection. Since almost all the
papers are ready now (or, in Haugeland's case will be ready in a few months) it would be
useful to have the papers by spring. Interested?
-------
I could have a short paper on using circumscription to solve the
frame problem. It would be based on a section of a larger paper on
applying circumscription to formalizing common sense facts.
∂19-Dec-83 0900 JMC*
Griffiths
∂19-Dec-83 0924 PYLYSHYN@CMU-CS-C.ARPA Re: frame problem
Received: from CMU-CS-C by SU-AI with TCP/SMTP; 19 Dec 83 09:24:31 PST
Delivery-Notice: While sending this message to SU-AI.ARPA, the
CMU-CS-C.ARPA mailer was obliged to send this message in 50-byte
individually Pushed segments because normal TCP stream transmission
timed out. This probably indicates a problem with the receiving TCP
or SMTP server. See your site's software support if you have any questions.
Received: ID <PYLYSHYN@CMU-CS-C.ARPA>; Mon 19 Dec 83 12:26:04-EST
Date: Mon 19 Dec 83 12:26:00-EST
From: PYLYSHYN@CMU-CS-C.ARPA
Subject: Re: frame problem
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 19 Dec 83 00:45:00-EST
That would be fine: If circumscription can solve the frame problem it would
be of enormous interest -- especially if people who say that the frame problem
is isomorphic to the problem of induction are right. Please send along
whatever you have and I may have some comment. Have a good holiday.
-------
∂19-Dec-83 1051 BSCOTT@SU-SCORE.ARPA Incentive Funds for 81-82
Received: from SU-SCORE by SU-AI with TCP/SMTP; 19 Dec 83 10:50:54 PST
Date: Mon 19 Dec 83 10:50:57-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: Incentive Funds for 81-82
To: JMC@SU-AI.ARPA
cc: DFH@SU-AI.ARPA, BScott@SU-SCORE.ARPA
Arnice Streit has just processed salary incentive savings for 81-82. The total
going into your unrestricted account for that year is $3,857.
We are not sure how soon the 82-83 savings can be processed. We have to locate
the funds.
Betty
-------
∂19-Dec-83 1529 MCCORDUCK@COLUMBIA-20.ARPA
Received: from COLUMBIA-20 by SU-AI with TCP/SMTP; 19 Dec 83 15:29:23 PST
Date: Mon 19 Dec 83 18:29:53-EST
From: Pamela McCorduck <MCCORDUCK@COLUMBIA-20.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Mon 19 Dec 83 14:11:00-EST
John, it's absolutely a delicious piece, and deserves a wide readership.
Most especially, it ought to be seen where readers of "literature" read,
because the most interesting notion, in my view, is the point you make
that good science (and by extension, good technology) doesn't make good
literature, at least by today's canons of literature. Let me think about
this, and get back to you.
Pamela
-------
∂19-Dec-83 2200 JMC*
coathanger
∂20-Dec-83 0439 HST visit
please don't forget to fix the date or confoirm with the 7th of jan.
somebody here told that during the turing lecture of allen newell he
got the information that garbage collection was used before lisp.
but really he couldn't make his claim definite.Do you have any idea
about that?What is newells net address?
fing newell@cmu-cs-a
PN Who Job What Pages State TTY Where
AN02 Allen Newell 39 RDMAIL 36+186 ↑C :23 4 Newell house 421-3668
Office: 4202 Science Hall (412) 578-2602 Secretary: Betsy Herk
Home: (412) 421-3668
Computer Science and Psychology faculty member
I think someone told me shortly after LISP was started that garbage collection
had been used before. My impression is that this was in a specific program
rather than in a programming system. I don't remember whether I have read
Newell's Turing lecture.
Please co-ordinate your visit with Carolyn.
∂20-Dec-83 1359 ZM RE: Consulting Profs
To: TW@SU-AI, DEK@SU-AI, JMC@SU-AI, GHG@SU-AI, ZM@SU-AI
Each one of the six candidates is qualified for the proposed title
(with the amendment). However, I am against double appointments; I
really don't understand why they need the title from two departments.
Also, I don't believe we should establish six new Consulting
Professorships all at once.
According to Terry's message:
1. The appointments of Moore, Perrault and Smith have been signed
off by Philosophy.
2. The appointments of Kay and Grosz have been signed off by Linguistics.
This leaves us with Stan Rosenschein. I approve Stan's appointment as
a Consulting Associate Professor.
∂20-Dec-83 1532 KJB@SRI-AI.ARPA Lunch?
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Dec 83 15:32:39 PST
Date: Tue 20 Dec 83 15:33:33-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Lunch?
To: Jmc@SU-AI.ARPA
How about tomorrow? I suppose the faculty club is closed. Any ideas.
I'll meet you anywhere near at 12 +/- .5hrs.
-------
The Faculty Club is open till the 23rd. How about there at 12? I'll
make the reservation if you agree.
In fact I have made the reservation for 12 and will change it if you
prefer something else.
∂20-Dec-83 1538 KJB@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 20 Dec 83 15:38:27 PST
Date: Tue 20 Dec 83 15:39:43-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 20 Dec 83 15:34:00-PST
Faculty club, tomorow, 12. Great.
-------
∂20-Dec-83 2226 CLT visit
To: HST, JMC, RWW
I propose the following arrangement -
We will arrange for you to give seminars on the 10th and 11th of January
(one at stanford, one to the Perseus group).
You may stay at our house the nights of the 10th and 11th.
The current plan is for John and I to be in Paris until the 9th.
However, if you would like to come earlier so as to talk to other
people or whatever, I could try to arrange a motel that is not
too, expensive.
(I think I might find something for about $20 per night).
Let me know what you would like to do.
Carolyn
∂21-Dec-83 0913 DFH Room 252
Has classes in it during the afternoon on the days you requested. The
best I was able to come up with is room 301 MJH at 4 pm. (There are
classes in 252, 352, and 301 at 3 pm). Please let me know if this is
acceptable. Also, will you want an overhead projector?
301 is acceptable, and I'll want an overhead.
∂21-Dec-83 1039 STAN@SRI-AI.ARPA Ehud Shapiro
Received: from SRI-AI by SU-AI with TCP/SMTP; 21 Dec 83 10:38:58 PST
Date: 21 Dec 1983 1037-PST
From: Stan at SRI-AI
Subject: Ehud Shapiro
To: jmc at SAIL, nix at PARC
cc: stan
I just got a call from Udi, and he plans to visit the Palo Alto area
the week after POPL. If possible, he would like to give a talk at
Xerox and a talk at Stanford, and he asked me to coordinate his visit.
The talks should be sometime between Mon., Jan. 23 and Wed., Jan. 25,
and the subject he would like to speak on (at Stanford, at least) is:
"The Bagel: A Systolic Concurrent Prolog Machine"
(I can get an abstract from Fernando Pereira, who has some papers Udi
sent ahead). He is anxious to know the dates, so if we can decide
soon, I will let him know what was decided. Thanks.
--Stan Rosenschein
-------
∂21-Dec-83 1159 Nix.pa@PARC-MAXC.ARPA Re: Ehud Shapiro
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 21 Dec 83 11:58:49 PST
Date: 21 Dec 83 11:35:32 PST
From: Nix.pa@PARC-MAXC.ARPA
Subject: Re: Ehud Shapiro
In-reply-to: "Stan@SRI-AI.ARPA's message of 21 Dec 83 10:37 PST"
To: Stan@SRI-AI.ARPA
Cc: jmc@SAIL.ARPA, Nix.pa@PARC-MAXC.ARPA
...He is anxious to know the dates, so if we can decide soon, I will let
him know what was decided.
Monday morning or Tuesday afternoon (Jan 23 or 24) would be best here at
PARC; Wednesdays tend not to be as good.
...(I can get an abstract from Fernando Pereira, who has some papers Udi
sent ahead).
Could you send me a copy of his abstract? Do you know if Udi would be
up for giving another gossip talk on ICOT as well?
Thanks,
Bob
I think it would be very desirable to have Shapiro give the CSD Colloquium
on January 24. I want to repropose him for a faculty position.
I have proposed to Rod Brooks that he give the CSD Colloquium on
Jan 24, but I see Rod hasn't logged in for several days, so I'm not
confident of a quick response.
∂21-Dec-83 1248 TW Consulting professorships
To: TW@SU-AI, DEK@SU-AI, JMC@SU-AI, ZM@SU-AI
CC: golub@SU-SCORE
Everyone has responded with respect to the consulting professorships,
with the following results:
WINOGRAD: approve all, with Kay as full, others as associate
KNUTH:
As I mentioned yesterday, it seems OK to go along with the
proposals, now that all the resumes are available.
However, I think you made a mistake when you said "associate for all
by Kay"; there was at least one proposed for assistant rank.
Namely, I'm pretty sure Smith was to be Consulting Assistant Professor
[especially because he is less than two years past the PhD].
With that amendment, I approve all the appointments unless somebody
else on our committee has reservations; in the latter event, I
want to hear about them.
MCCARTHY:
To me the ranks are not important, and therefore, I approve the proposal
as amended. Kay consulting full, Smith consulting asst., Rosenschein,
Moore, Grosz and Perrault consulting assoc.
MANNA:
Each one of the six candidates is qualified for the proposed title
(with the amendment). However, I am against double appointments; I
really don't understand why they need the title from two departments.
Also, I don't believe we should establish six new Consulting
Professorships all at once.
According to Terry's message:
1. The appointments of Moore, Perrault and Smith have been signed
off by Philosophy.
2. The appointments of Kay and Grosz have been signed off by Linguistics.
This leaves us with Stan Rosenschein. I approve Stan's appointment as
a Consulting Associate Professor.
--------------------
I see the summary as follows:
1) Having seen the vitaes, everyone is satisfied with the qualifications
of all of the candidates.
2) Everyone is happy with ranks of Full for Kay, associates for everyone else
but Smith
3) Manna is against giving joint appointments, and would therefore appoint
only Rosenschein, who does not have one.
We can either simply present this as the committee's findings to the
faculty meeting, or have a meeting to try to come to consensus.
May own opinion is that we should go ahead with the joint appointments, since
in fact the candidates have played and will continue to play a real role in
the department (teaching courses, advising students, etc,). Also, it would
be a little strange that out of the entire set of consulting appointments
in all of the departments, exactly one would be singled out as assistant
professor. Brian Smith is indeed fresh out out graduate school, but he
is in fact a major intellectual force in CSLI and at Xerox, and I think
there is broad agreeement that he is qualified far beyond what his resume
shows. His appointment in philosophy is already as an associate.
Let me know before Monday the 9th whether you feel we need to discuss things
further, and when you could meet on that day or on the 10th before the
faculty meeting.
Happy holidays. --t
Let's go ahead with Rosenschein, since there is no objection to him, and
having the privileges may make some difference. I see some force in
Zohar's view that multiple consulting professorship appointments
violate Occam's razor. Smith is certainly primarily a computer scientist,
at least I hope so, but for that reason I think we should not give
him a higher consulting rank than we would give him if he had an
ordinary appointment. While I would have no hesitation about a regular
assistant professorship for him had we the slot, I wouldn't suppose that
he would rate higher than our present assistant professors in similar
fields.
∂21-Dec-83 1440 YEARWOOD@SU-SCORE.ARPA Grades for 206
Received: from SU-SCORE by SU-AI with TCP/SMTP; 21 Dec 83 14:40:33 PST
Date: Wed 21 Dec 83 14:40:53-PST
From: Marlene Yearwood <YEARWOOD@SU-SCORE.ARPA>
Subject: Grades for 206
To: jmc@SU-AI.ARPA
Stanford-Phone: (415) 497-2266
The Registrar's office just called and said that if you could get
the grades for your 206 class in by 3 p.m. it would still be in
time to include grades for student transcripts for autumn quarter.
Otherwise, students who are transferring or graduating are held up.
Sorry to have to rush you, but thought you would want to know.
Thanks.
marlie
-------
Thanks, but I won't be able to make it with the CS206 grades, although
I can have my other grades in 10 minutes.
∂21-Dec-83 1423 KJB@SRI-AI.ARPA room
Received: from SRI-AI by SU-AI with TCP/SMTP; 21 Dec 83 14:22:57 PST
Date: Wed 21 Dec 83 14:18:17-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: room
To: Jmc@SU-AI.ARPA
cc: bmacken@SRI-AI.ARPA, IGONI@SRI-AI.ARPA
The room is already booked every afternoon except Friday. We could
a) have it Friday
b) have it in the smaller room 6, which will hold about 10 people
c) have it in a morning
d) have it elsewhere, say at MJ.
Which would you prefer?
-------
I prefer to have it Friday afternoon, as early as possible after 2pm.
Many thanks.
∂21-Dec-83 1515 Nix.pa@PARC-MAXC.ARPA Re: Ehud Shapiro
Received: from PARC-MAXC by SU-AI with TCP/SMTP; 21 Dec 83 15:15:27 PST
Date: 21 Dec 83 15:14:55 PST
From: Nix.pa@PARC-MAXC.ARPA
Subject: Re: Ehud Shapiro
In-reply-to: "Your message of 21 Dec 83 15:01 PST"
To: John McCarthy <JMC@SU-AI.ARPA>
Cc: Nix.pa@PARC-MAXC.ARPA, Stan@SRI-AI.ARPA
...I have proposed to Rod Brooks that he give the CSD Colloquium on Jan
24...
For the moment, then, I'll schedule Udi to give a talk here at PARC on
Monday morning, January 23. I won't announce this until the 10th or so,
so it'll be easy to make any last minute adjustments.
-- Bob
∂21-Dec-83 2220 ENGELMORE@SUMEX-AIM.ARPA Re: Winter 1983 or Winter 1984
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 21 Dec 83 22:20:29 PST
Date: Wed 21 Dec 83 22:24:47-PST
From: Bob Engelmore <ENGELMORE@SUMEX-AIM.ARPA>
Subject: Re: Winter 1983 or Winter 1984
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Tue 20 Dec 83 22:49:00-PST
Home Phone: (415) 322-0627
Office Phone: (415) 327-6600
John,
We haven't been consistent in our labeling of "Winter" issues. Last
year we decided to associate one calendar year with one volume number of the
magazine, and to start the volume with the Spring issue. Hence, Volume 4,
Number 4 is linked to Winter 1983. (Unfortunately, Volume 4, Number 1 was
labelled Winter/Spring 1983 because it was a catch-up issue.)
If you have a strong feeling about how to label the Winter issues,
I'm easily persuaded to change the current scheme.
Bob
-------
∂22-Dec-83 1007 STAN@SRI-AI.ARPA Udi's abstract
Received: from SRI-AI by SU-AI with TCP/SMTP; 22 Dec 83 10:07:16 PST
Date: 22 Dec 1983 1007-PST
From: Stan at SRI-AI
Subject: Udi's abstract
To: jmc at SAIL, nix at PARC
cc: stan
John, I gathered from Bob's reply to my message that you have proposed
Udi give the CSD colloquim on Jan. 24. Could you forward his abstract
to the relevant person(s)? Thanks.
Udi's "Bagel" abstract is as follows:
The Bagel: A Systolic Concurrent Prolog Machine
Ehud Shapiro
The Weizmann Institute of Science
It is argued that explicit mapping of processes to processors is
essential to effectively program a general-purpose parallel computer,
and, as a consequence, that the kernel language of such a computer
should include a process-to-processor mapping notation.
The Bagel is a parallel architecture that combines concepts of
dataflow, graph-reduction and systolic arrays. The Bagel's kernel
language is Concurrent Prolog, augmented with Turtle programs as a
mapping notation.
Concurrent Prolog, combined with Turtle programs, can easily implement
systolic systems on the Bagel. Several systolic process structures are
explored via programming examples, including linear pipes (sieve of
Erasthotenes, merge sort, natural-language interface to a database),
rectangular arrays (rectangular matrix multiplication, band-matrix
multiplication, dynamic programming, array relaxation), static and
dynamic H-trees (divide-and-conquer, distributed database), and
chaotic structures (a herd of Turtles).
All programs shown have been debugged using the Turtle graphics Bagel
simulator, which is implemented in Prolog.
-------
Stan, the relevant person is Rod Brooks (ROD@SU-AI) to whom I have
proposed it by net mail and to whom I have forwarded the abstract.
However, he hasn't logged in since Friday, so I suppose he's travelling,
and so I can't be sure he hasn't already committed the relevant date.
∂22-Dec-83 1046 STAN@SRI-AI.ARPA Ehud Shapiro
Received: from SRI-AI by SU-AI with TCP/SMTP; 22 Dec 83 10:45:55 PST
Date: 22 Dec 1983 1034-PST
From: Stan at SRI-AI
Subject: Ehud Shapiro
To: rod at SU-AI
cc: stan, jmc at SU-AI
John McCarthy has suggested that Ehud Shapiro give the CSD colloquium on
January 24. If this date has already been committed, please let me know
so that an alternate time can be found for his talk. (I am coordinating
his visit to the area.) --Stan Rosenschein
-------
∂22-Dec-83 1117 TW Joint appointments
To: JMC@SU-AI, DEK@SU-AI, TW@SU-AI, ZM@SU-AI
CC: golub@SU-SCORE
In light of the questions about joint appointments, I asked John Perry why
they were pushing for the them, and got the following reply:
****
While it may seem like having consulting faculty appointed in two departments
is redundant, there are two reasons that make this desirable. These two
reasons relate to the two main points of going through the appointment process.
The first is to get some degree of departmental review over the main
participants in the CSLI program. The second, related purpose is to give the
consulting faculty and the students they are likely to be dealing with an
indication that the consulting faculty will be deemed reasonable choices for
the dissertation committees of those students, should their research projects
be in the appropriate areas.
Since CSLI is a part of IMSSS, an independent research institute, its
appointments do not routinely go through Humanities and Sciences, or any other
school, for faculty review. The primary motivation for the requirement that
the CSLI principals be consulting faculty, rather than simply research
associates, is to assure that there will be some degree of departmental
scrutiny, by the departments that are likely to be asked to have these persons
serve on student's committees. Given the interdisciplinary emphasis at CSLI,
most of the principals are expected to play some educational role in the lives
of at least two departments, and that is the reason we have asked two
departments to administer scrutiny in the case of most of the principals.
We expect to provide support for a number of students in computer science over
the next few years, and already are providing support for some. These students
will be expected to become involved with the projects at CSLI, and we expect
that some of them will write dissertations focussed on such projects to a
greater or lesser degree. Thus we would like an indication for the principals
such students are likely to be working with that their department regards these
principals as suitable mentors for students involved in such projects.
Consulting faculty will not direct dissertations without further approval not
only of the student's departments, but also the dean of graduate studies. But
their service as members of committees should be routine. If any of the
principals we have proposed to Computer Science are not so qualified, we should
know about it now. So I hope that any negative decisions would be on this
basis.
The dual requests served one other purpose, largely bookkeeping, which was
intended to be a convenience to computer science. Consulting faculty, altough
not paid by Stanford, are recorded as serving at some percentage of their time.
50% reflects the degree of involvement the consulting faculty have committed
themselves to, and is also the cut-off percentage for some privileges of a
relatively minor sort bestowed by the university. Although the other
departments involved do not have many requests for consulting faculty, we
assume that computer science does. By recording the involvement as 25% with
two departments, the official number of consulting faculty charged against
computer science is reduced.
I would be happy to discuss these matters further with anyone that has
questions.
John Perry
-------
∂22-Dec-83 1125 DMC meeting about partial pass of AI qual.
When would be a good time to catch you either this afternoon or tomorrow
morning?
-- Dave Chelberg (DMC@SAIL)
Try 3pm.
∂22-Dec-83 1147 KJB@SRI-AI.ARPA McCarthy's talks
Received: from SRI-AI by SU-AI with TCP/SMTP; 22 Dec 83 11:47:15 PST
Date: Thu 22 Dec 83 11:42:05-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: McCarthy's talks
To: Igoni@SRI-AI.ARPA
cc: bmacken@SRI-AI.ARPA, jmc@SU-AI.ARPA
Please schedule the seminar room for John Mc Carthy as early as possible
after 2 pm for an hour on Fridays, starting Jan 20. 4 weeks.
Thnks, Jon
-------
∂22-Dec-83 1147 KJB@SRI-AI.ARPA oops
Received: from SRI-AI by SU-AI with TCP/SMTP; 22 Dec 83 11:47:26 PST
Date: Thu 22 Dec 83 11:43:15-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: oops
To: igoni@SRI-AI.ARPA, jmc@SU-AI.ARPA
Francis, start that on Jan 13, not 20th. Sorry.
-------
∂22-Dec-83 1152 CLT tonight
I tentatively told Sarah we would take her out to supper
tonight. Is that ok? She will call around 6 to find out for sure.
Taking Sarah to supper will be fine.
Here are the flights my secretary, Diana Hall, found. I have decided
on the Pan Am. I would like Hotel Bristol reservations through
Saturday night, because we might stay one more day and leave on
Sunday.
∂22-Dec-83 1324 DFH
Flight possibilities
On Pan Am: Jan 1 SFO/JFK PA72 8:30am/4:40pm
JFK/Paris (Orly) PA 114 6:45pm/7:35am next day
Jan 7 Paris(Orly)/JFK PA 115 1pm/3:05pm
JFK/SFO PA 67 4:30pm/7:22pm
On Air France: LAX/Paris 10 pm/5:30pm next day
Paris/LAX 5 pm/7:35pm
Operates every day except Monday and Tuesday
Fares are the same on each (Air France common rates SFO and LAX).
Regular all-year fare (coach): $939 each way
Executive class: $1,259 each way
I have made reservations on the Pan Am flight listed above. They will be
automatically cancelled on Dec. 28 if they are not notified that a ticket
has been done.
Let me know if there is some other option you would like me to check.
∂22-Dec-83 1754 DEK change my vote
To: TW@SU-AI
CC: ZM@SU-AI, JMC@SU-AI, GHG@SU-AI
After seeing Zohar's response, which contained information that I hadn't
known, I now am totally in agreement with his views. Namely, I think
it is best to make just one appointment.
∂23-Dec-83 0442 HALPERN.SJRLVM1.IBM-SJ@Rand-Relay knowledge seminar
Received: from SU-HNV by SU-AI with TCP/SMTP; 23 Dec 83 04:42:48 PST
Received: from rand-relay.ARPA by Diablo with TCP; Fri, 23 Dec 83 04:41:38 pst
Date: 22 Dec 1983 16:59:46-PST (Thursday)
From: Joe Halpern <HALPERN.IBM-SJ@Rand-Relay>
Return-Path: <HALPERN.SJRLVM1.IBM-SJ@Rand-Relay>
Subject: knowledge seminar
To: KNOWLEDGE@SU-HNV
Via: IBM-SJ; 23 Dec 83 2:09-PST
Because of holidays and scheduling conflicts, the next meeting of
the knowledge seminar will not be until Friday, January 27, 1984 (!),
at 2 PM, when John McCarthy will speak. Tentatively, we will then
meet next on Friday, February 4, and then every second Friday
afternoon after that. It looks like we'll have a series of at
least six or seven talks; Moshe Vardi, Bob Moore, Kurt Konolige,
and Hector Levesque have all agreed to speak on their research.
Given that IBM is such a long drive for some people, I would rather
have a 2-3 hour meeting every second Friday afternoon rather than
a 1 hour meeting every Friday. The hope is that these seminars
will also provide a stimulus for further research, so many of the
speakers will be presenting open problems and partial results, as well
as published research.
If anyone else would like to volunteer to give a talk or has ideas
on how the seminar should be run, please send a message to
me at HALPERN.IBM-SJ@RAND-RELAY.
Happy holidays!
Regards,
Joe Halpern
∂24-Dec-83 1021 CLT turkey
Could you put the turkey and squash in the oven at 5?
They are ready and the oven is set at the desired temp.
Put the squash in the back. They will fit if both are crosswise.
Thanks -
∂24-Dec-83 1700 JMC*
turkey
∂24-Dec-83 1700 CLT*
turkey?
∂26-Dec-83 1000 JMC*
call Reddy
∂26-Dec-83 1000 JMC*
amex
∂27-Dec-83 0900 JMC*
7-0601 to set up appointment with Teller.
∂27-Dec-83 1000 JMC*
Lifschitz 915 581-4261
∂27-Dec-83 1022 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 27 Dec 83 10:22:05 PST
Date: Tue 27 Dec 83 10:26:28-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: next meeting
To: JMC@SU-AI.ARPA, Feinler@SRI-NIC.ARPA, Genesereth@SUMEX-AIM.ARPA,
Buchanan@SUMEX-AIM.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
Well, it looks like now the next meeting regarding the on-line abstract/
library service can not be held until the week of the Jan. 16th. What's
that week like for everyone?
Claudia
-------
next meeting
Wednesday is out for me. Otherwise any morning, preferably late.
∂27-Dec-83 2323 LLW@S1-A Regrets and Greetings
Received: from S1-A by SU-AI with TCP/SMTP; 27 Dec 83 23:23:19 PST
Date: 27 Dec 83 2224 PST
From: Lowell Wood <LLW@S1-A>
Subject: Regrets and Greetings
To: jmc@SU-AI
CC: LLW@S1-A
∂24-Dec-83 0045 JMC@SU-AI
Received: from SU-AI by S1-A with TCP/SMTP; 24 Dec 83 00:45:05 PST
Date: 24 Dec 83 0049 PST
From: John McCarthy <JMC@SU-AI>
To: llw@S1-A
Pick up the phone if you can talk.
[John: Sorry--I had irresponsibly left my terminal logged in, while I
wandered off; I just now found this message. A belated Merry Christmas,
and a (barely) timely Happy New Year to you! Lowell]
∂28-Dec-83 0020 @SU-SCORE.ARPA:PARK@SRI-AI.ARPA ISIR14
Received: from SU-SCORE by SU-AI with TCP/SMTP; 28 Dec 83 00:20:23 PST
Received: from SRI-AI.ARPA by SU-SCORE.ARPA with TCP; Wed 28 Dec 83 00:20:47-PST
Date: 28 Dec 1983 0015-PST
From: Park at SRI-AI
Subject: ISIR14
To: JMC%SU-AI at SU-SCORE
cc: PARK
John, can you tell me the address to which to send abstracts for the
14th International Symposium on Industrial Robotics (Sweden, Oct '84).
They are due Jan 1, I believe. No one around here seems to have the
call for papers.
Thanks,
Bill Park (PARK@SRI-AI)
-------
No idea. Try Tom Binford (TOB@SU-AI). No need to go through SCORE.
∂28-Dec-83 0338 CL.BOYER@UTEXAS-20.ARPA Space
Received: from UTEXAS-20 by SU-AI with TCP/SMTP; 28 Dec 83 03:38:31 PST
Date: Wed 28 Dec 83 05:39:07-CST
From: Bob Boyer <CL.BOYER@UTEXAS-20.ARPA>
Subject: Space
To: JMC@SU-AI.ARPA
I've got one of those nifty academic jobs: write a report
on how much space our department needs. Could I bother you
on two matters?
1. How much space is it appropriate to consider offering
someone whom we want to take a chair here and do AI? I would
appreciate some breakdown such as the following:
Chair's office: 300 sq ft
2 Faculty associates: 400 sq ft
1 conference room: 300 sq ft
1 secy: 150 sq ft
12 grad students: 600 sq ft
1 lab 500 sq ft.
1 machine room 500 sq ft.
2. SAIL had always struck me as an ideal setup for CS.
What was the space and occupancy at SAIL? If I remember the
AI wing, it had something like 20,000 square feet for 4
faculty, 4 administration, 6 research fellows, and 30
graduate students.
-------
On the whole, you're ideas are more concrete than mine ever were and
seem very reasonable. The one difference that I see is that mature
research projects tend to have a few non-faculty research associates,
although many of these people might have faculty positions at U.T.
Because there was always infinite unoccupied space at D.C. Power
Lab to expand into, I never had to do calculations. Les Earnest,
(home: 941-3984) did some, I believe.
∂28-Dec-83 0903 MEYER@MIT-MC dirty Lisp
Received: from MIT-MC by SU-AI with TCP/SMTP; 28 Dec 83 09:03:21 PST
Date: 28 December 1983 12:05 EST
From: Albert R. Meyer <MEYER @ MIT-MC>
Subject: dirty Lisp
To: JMC @ SU-AI
cc: "(FILE [AR4:MEYER;MCARTH MES])" @ MIT-MC, JCM @ MIT-MC,
halpern.ibm-sj @ RAND-RELAY
Date: 27 Dec 83 1357 PST
From: John McCarthy <JMC@SU-AI>
I have been think about the problems of formalizing impure Lisp (with setq)
and dirty Lisp (with rplac). The temptation is simply to call this kind of
usage bad, especially when something is written in functional style and
has side effects. If the functional expression is written only for its
side effects, then this seems indeed unnecessarily obscure and can be
replaced by a construction with setq (more generally setf). However,
if both the side effects and the value are important, there is often no
obvious other way to write the program concisely. In that case, we theorists
should endeavor to provide suitable formalisms, and I have made a little
progress, not written up. Carolyn Talcott has some reformulations that
will be in her forthcoming thesis.
I'm sorry I missed your talk. If there is a written version, I would like
to see it.
--------------------------------------------
I hope Talcott will send me a copy of her thesis.
An overview of our approach to separating side-effects from purely functional
expressions is contained in my IFIP paper which I am sending you. I'm
considering submitting a paper to the LISP/functioanl programming symposium
this summer arguing that ALGOL (more precisely Reynold's and our idealization
thereof) is largely a purely functional language. If it gets written, I'll
send you a copy.
In general, I respect the intuition of sophisticated programmers, and agree we
theoreticians ought to try to capture and clarify it, rather than reject it.
Still, there are places where contrasting judgements seem appropriate, and the
pun of treating expressions as notations for sequential procedures with
side-effects is such a place in my view. I'd be interested in seeing your
examples where there is ``no obvious other way to write the program
concisely.'' I have also asked Gerry Sussman to offer some basic examples
supporting this mixed style, but he hasn't convinced me yet that there is
something especially natural about this view of algorithms. It could be
interesting to develop a new pun-free notation for describing algorithms in
which the mixture of side-effects and value passing is ``important''. (Perhaps
all that is needed is an expression notation in which the intended order of
evaluation is made explicit.)
It was nice to have a chance to chat with you again, even if only briefly.
Best wishes fr the New Year.
Regards, A..
∂28-Dec-83 1032 CLT stoyan
when do you want HST to speak at stanford?
(Jan 10 or 11? what time??)
Make it Jan 10 (Tuesday) at 1pm.
∂28-Dec-83 1607 decvax!mulga!yorick.moncskermit@Berkeley Received: from
UCB-VAX by SU-AI with TCP/SMTP; 28 Dec 83 16:06:46 PST Received: by
UCB-VAX.ARPA (4.22/4.18)
id AA12003; Wed, 28 Dec 83 16:05:06 pst Received: by decvax.UUCP
(4.12/4.13)
id AA19216; Sat, 24 Dec 83 07:06:52 est From:
decvax!mulga!yorick.moncskermit@Berkeley (Prof. Yorick Wilks) Message-Id:
<8312241206.AA19216@decvax.UUCP> Received: by mulga.Melbourne (3.326)
id AA08210; 20 Dec 83 20:11:04 EST (Tue) Date: Tuesday, 20 Dec
1983 19:40-EST To: jmc%su-ai.arpa.mulga@Berkeley
.ls 2
Computer Science Dept.,
Monash University,
Clayton, Vic. 3168.
AUSTRALIA.
21 December
Dear John,
I am writing to you about the new Edinburgh University Press
series in Artificial Intelligence and Cognitive Science, with the aim of
trying to get a manuscript out of you. The Press is a high quality one,
with a history of close association with the field. It will offer the
normal royalties, rapid publication for accepted volumes and a guarantee
of large scale distribution in the US with an appropriate American
publisher.
I well remember your pieces on the Stanford system on social
issues, and the good effect they had, though I dont remember seeing many
of them published. How would you react to the idea of collecting them
together in a volume, with either yourself doing it, or someone else if
you dont want the bother?
My address will be the one above until (air mail day) 10th
February, and after that the Cognitive Studies Centre, University of
Essex, Colchester, Essex, UK. A manuscript should go directly to the
Press: Ms. V. Bone, EUP, 22 george Square, Edinburgh.
I look forward to hearing from you,
With best regards,
Yorick Wilks.
Professor of Computer Science.
(ARPA net address till 23 Feb.,with quotes important:
"decvax!mulga!yorick.basservax"@ucb-vax )
Yorick,
I'm not sure what pieces on social issues you are referring to, although
nothing on the system is lost. I have a large number of "Technology
essays", but only a few are on computer technology and of these only
one or two (as I recall) relate to cognitive science. Whatever they
are, they're still on disk, and I'd like to have them published.
Is your move to Australia permanent?
Best Regards,
John
∂28-Dec-83 1802 WALKER@SRI-AI.ARPA Re: next meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 28 Dec 83 18:02:02 PST
Date: Wed 28 Dec 83 17:57:08-PST
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: next meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: JMC@SU-AI.ARPA, Feinler@SRI-NIC.ARPA, Genesereth@SUMEX-AIM.ARPA,
Buchanan@SUMEX-AIM.ARPA, Amsler@SRI-AI.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of Tue 27 Dec 83 10:23:10-PST
I think i will be out of town that week, but I am not sure yet. Be back to you
later.
Don
-------
∂29-Dec-83 1000 JMC*
Stanford benefits
∂29-Dec-83 1000 JMC*
>Reddy (home: 412 621-2617) (work: 578-2597)
* REDDY@CMU-CS-A, secretary (1983 july) is Janet Asbury
* ASBURY@CMU-CS-A
Thanks for the tickets which I now have. (In future it would be better
to prepay via airline directly (if possible) rather than through a
travel agency. That way I can have my agency write the ticket and
deliver it rather than wait for an hour at Pan Am while they write
it). I assume Centre is making reservations at the Bristol, and I
hope someone will be there to talk to about state of project.
∂29-Dec-83 1427 CLT tonight
i will come home to practice in time to cook the lamb
if sarah wants to stay / return for supper that is fine
i expect it will be ready around 7:30
∂30-Dec-83 0057 LGC Terminal Return
To: JMC
CC: DFH
If you have no objection, I'll bring my terminal in on Monday, Jan. 9.
-- Lew
That's fine. Give it to Diana.
∂30-Dec-83 1145 ME long mail addresses
∂28-Dec-83 1620 JMC MAILing from file
What is the most convenient way to MAIL to single addresses
stored in a file? Addresses like
"decvax!mulga!yorick.basservax"@ucb-vax
are getting so long that the probability of error in copying
them is such that one prefers to do it electronically.
ME - You can put that address in a separate file, and use @file as
the destination, although that is wasteful of disk space. For
mailing from E, you can keep one file with all addresses in it
(say, each preceded by the real name, for ease in finding them).
Then you can just switch to this index file (using a new window),
copy the line with the mail address in it, and put it down just
ahead of your message (or subject) text.
Maybe someday we'll invent a syntax that allows users to have
their own forwarding list, so that you can use an abbreviated
form of the address and have the real address looked up in
your file by MAIL.
me
mailing from file
Thanks Marty. I inferred from your note that if I omitted the address
from the MAIL command it would use the first line of the message. That's
actually what I needed.
∂30-Dec-83 1325 BSCOTT@SU-SCORE.ARPA Indirect Cost Rebate
Received: from SU-SCORE by SU-AI with TCP/SMTP; 30 Dec 83 13:24:53 PST
Date: Fri 30 Dec 83 13:24:12-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: Indirect Cost Rebate
To: JMC@SU-AI.ARPA, TOB@SU-AI.ARPA, ZM@SU-AI.ARPA, DCL@SU-AI.ARPA,
Wiederhold@SUMEX-AIM.ARPA
cc: DFH@SU-AI.ARPA, RBA@SU-AI.ARPA, MAS@SU-AI.ARPA, Atkinson@SU-SCORE.ARPA,
BScott@SU-SCORE.ARPA
You have received indirect cost rebates on your current ARPA contract for the
period 9/1/82 through 11/30/83 as listed below. These amounts have been
credited to your respective unrestricted research accounts.
Binford $ 26,537
Luckham 33,404
Manna 5,062
McCarthy 22,435
Wiederhold 22,515
If you have questions, please send message.
Betty
-------
∂31-Dec-83 0430 ME
To: ARK, JMC
NS is working again.
∂31-Dec-83 2110 CLT ?
do you know tony hearns net address? hearn@rand-ai yields no such host
contrary to the 1982 arpanet book
∂31-Dec-83 2149 CLT calendar item
Sun. 22 Jan 19:30 Pippin Offenbach - Bridge of Sighs
Sun. 5 Feb 19:30 Pippin Smetana - Two Widows
Sun. 19 Feb 19:30 Pippin Nicolai - The Merry Wives of Windsor
Sun. 26 Feb 19:30 Pippin Weber - Der Freischutz
Sun. 11 Mar 19:30 Pippin Schubert - The Conspirators
∂01-Jan-84 0054 WIEDERHOLD@SRI-AI.ARPA
Received: from SRI-AI by SU-AI with TCP/SMTP; 1 Jan 84 00:54:18 PST
Date: Sun 1 Jan 84 00:54:44-PST
From: Gio <Wiederhold@SRI-AI.ARPA>
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI>" of Sun 1 Jan 84 00:50:00-PST
Thanks, and the same to you! ( I was off the terminal for a half hour ... )
Gio
-------
∂01-Jan-84 1743 ME host name rand-cs
∂31-Dec-83 2118 JMC
finger hearn%rand-cs
Unknown host name
ME - Seems to work now. I don't know if that host name suddenly
disappeared and reappeared in the host table, but it's there now.
On the other hand, it appears they don't support remote finger,
since they reset our finger connection attempts.
∂02-Jan-84 1827 TARNLUND@SRI-AI.ARPA a visit to Palo Alto
Received: from SRI-AI by SU-AI with TCP/SMTP; 2 Jan 84 18:27:46 PST
Date: Mon 2 Jan 84 17:21:48-PST
From: TARNLUND@SRI-AI.ARPA
Subject: a visit to Palo Alto
To: jmc@SU-AI.ARPA
I'll be around Palo Alto for about a week from
about January 19. If you can spare some time I would
like to see you. Best regards -Sten-ake
-------
∂03-Jan-84 1130 FIKES%USC-ECLD%USC-ECL%MARYLAND.ARPA@USC-ECL.ARPA Executive Committee Meeting Minutes
Received: from USC-ECL by SU-AI with TCP/SMTP; 3 Jan 84 11:30:15 PST
Mail-from: DECNET site ECLD rcvd at 2-Jan-84 2154-PST
Date: 2 Jan 1984 2149-PST
From: Richard Fikes <FIKES%USC-ECLD%USC-ECL@MARYLAND>
Subject: Executive Committee Meeting Minutes
To: aaai@SRI-AI, jmc@SU-AI, walker@SRI-AI
cc: FIKES%USC-ECLD%USC-ECL@MARYLAND
The following are the exerpted minutes of the '83 executive committee meeting
that I plan to publish in the AI magazine. Since I was not present at
the meeting, please check them for misrepresentations or other errors.
The magazine needs the minutes this week, so please send me your
comments asap.
thanks,
richard
American Association for
Artificial Intelligence
Minutes
Fourth Annual Meeting, 24 August 1983
Washington Hilton Hotel, Washington, D.C.
Staff Reports, chaired by Claudia Mazzetti, Executive Director
FINANCES. AAAI is financially solvent. The conference revenues
this year were outstanding; the gross revenues were approximately
$391,000 for the tutorials and Technology Transfer Symposium, and
$187,000 for the remainder of the conference. Assets of over $384,000
are projected for the end of the year.
LOCAL ARRANGEMENTS. The local arrangements committee will remain
active throughout the year.
AAAI OFFICE. A temporary employee will be hired several months prior
to the conference to assist in preparation for the conference.
Committee Reports, chaired by Nils Nilsson, President (1982-1983)
ELECTION. Marvin Minsky announced that Woody Bledsoe is the
President-Elect for 1984-85. Mark Stefik, Stanley Rosenschein, Eugene
Charniak, and Randall Davis will serve on the Council for 1983-86.
'83 CONFERENCE PROGRAM. Michael Genesereth reported that approximately
250 papers were submitted and 92 accepted. Innovations in this year's
program included the debate between John McCarthy and Roger Schank,
the addition of technical workshops prior to the conference, and the
Technology Transfer Symposium. The winner of the Publisher's Prize
was a paper by Etherington and Reiter from the University of British
Columbia. The prize will be sponsored by MIT Press in 1984 and
Addison-Wesley in 1985.
'84 CONFERENCE PROGRAM. The program chairman will be Ronald Brachman
from Fairchild. The tutorial chair will be Douglas Lenat from
Stanford University. The policy preventing authors from submitting
more than one paper will no longer apply to papers having multiple
authors. The program and tutorial chairs will determine who will be
responsible for the Technology Transfer Symposium.
AI MAGAZINE. Robert Engelmore, substituting for Lee Erman, commended
Claudia Mazzetti for her outstanding support in the production and
editing of the AI Magazine. He reported that the magazine currently
has a sustaining number of articles, and that a Presidential Message
will be included in the magazine each quarter.
MEMBERSHIP. Bruce Buchanan reported we have approximately 3,000
members. A AAAI membership directory will be produced; members will
be listed in the directory only if they give their explicit permission.
IJCAI. Saul Amarel reported on the agreement between IJCAI and AAAI
in which AAAI agrees to provide services in support of IJCAI
conferences. The agreement will be reevaluated to assure that AAAI is
adequately reimbursed for its services. In 1986, IJCAI and AAAI will
hold a joint conference at UCLA. Italy will likely be the site
for the 1987 IJCAI conference.
FUTURE CONFERENCE SITES. Marty Tenenbaum announced that The University
of Texas at Austin will be next year's conference site. AAAI will
provide funds to support the work of the Austin local arrangements
committee, chaired by Elaine Rich. The proposed site for 1986 is
Philadelphia. Tenenbaum is charged to study this matter and future
sites beyond 1986.
ON-LINE ABSTRACT SERVICE. John McCarthy will prepare a proposal for
AAAI to provide an on-line abstract service for its membership.
TRANSFER OF THE PRESIDENCY. Nilsson ended his term as president by
turning the platform over to John McCarthy.
John McCarthy, President (1983-1984), presiding
INDEPENDENT WORKSHOPS. $20,000 was allocated as seed money to support
independent workshops.
FUNDING POLICY WORKSHOP. John McCarthy will prepare a proposal for AAAI
participation in a workshop to consider what areas of AI should be
funded in the U.S. in response to the Japanese threat.
The meeting was adjourned.
Richard E. Fikes, Secretary-Treasurer
-------
∂03-Jan-84 1159 LUNDSTROM@SU-SCORE.ARPA [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Jan 84 11:59:13 PST
Date: Tue 3 Jan 84 11:56:03-PST
From: Stephen Lundstrom <LUNDSTROM@SU-SCORE.ARPA>
Subject: [WEEKS@Ames-VMSB: ACM SIGBIG Meeting]
To: super@SU-SCORE.ARPA
Following meeting may be of interest to you. Happy New Year. Steve Lundstrom
---------------
Return-Path: <WEEKS@Ames-VMSB>
Received: from Ames-VMSB.ARPA by SU-SCORE.ARPA with TCP; Tue 3 Jan 84 11:09:40-PST
Date: 3 Jan 1984 1047-GMT
From: WEEKS@Ames-VMSB
Subject: ACM SIGBIG Meeting
To: lundstrom at su-score
Reply-To: WEEKS at Ames-VMSB
ASSOCIATION FOR COMPUTING MACHINERY
SAN FRANCISCO GOLDEN GATE CHAPTER
SPECIAL INTEREST COMMITTEE "SIGBIG"
For Large High Speed Computers
Wednesday, January 4, 1984
7:00 Business Meeting
7:30 Speaker
George Leedon
Noetics
"Super Mini to Super Architecture"
(He has worked on Cray Architecture.)
Location: Lawrence Berkley Laboratory, Building 50B.
Get room number from guard.
For more information, contact Mary Fowler, TDC ACF, (415)965-6515
Arpanet fowler @ ames-vmsa
Ames DECnet CER::fowler
Ride Sharing, contact Frank Olken, LBL, (415)486-5891
Arpanet olken at lbl-csam
------
-------
∂03-Jan-84 1357 GOTELLI@SU-SCORE.ARPA Gavan Duffy
Received: from SU-SCORE by SU-AI with TCP/SMTP; 3 Jan 84 13:56:54 PST
Date: Tue 3 Jan 84 13:56:10-PST
From: Lynn Gotelli <GOTELLI@SU-SCORE.ARPA>
Subject: Gavan Duffy
To: JMC@SU-AI.ARPA
John, Currently CF is paying for a Sail computer account for
Gavan Duffy. I don't recall why CF is paying for a guest of
yours but unless you are willing to pay for his Sail computer
usage perhaps his account should go away. Would appreciate
hearing from you regarding this user.
Thanks,
Lynn
-------
∂04-Jan-84 0929 AAAI-OFFICE@SUMEX-AIM.ARPA next meeting
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 4 Jan 84 09:28:55 PST
Date: Wed 4 Jan 84 09:33:25-PST
From: AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>
Subject: next meeting
To: JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Walker@SRI-AI.ARPA, Amsler@SRI-AI.ARPA
cc: aaai-office@SUMEX-AIM.ARPA
Telephone: (415) 328-3123
Postal-Address: 445 Burgess Drive, Menlo Park, CA 94025
I'd like to schedule the next meeting on the on-line abstract/library
proposal for Monday, Jan. 23 at 10:30 am in room 252, MJH. At that time,
Jake's colleague from DIALOG will discuss the DIALOG's capabilities and
services. If you cannot make it, pls send me a msg. Thanks.
Claudia
-------
∂04-Jan-84 1027 ROD Shapiro.
To: DFH@SU-AI
CC: JMC@SU-AI, stan@SRI-AI
Jan 24th is ok for Shapiro to give the CS colloquium. I'll let you
know the exact time and place soon.
(Sorry for the delay in reply -- I've been out of ARPA-land for
a couple of weeks.)
∂04-Jan-84 1110 WALKER@SRI-AI.ARPA Re: next meeting
Received: from SRI-AI by SU-AI with TCP/SMTP; 4 Jan 84 11:10:08 PST
Date: Wed 4 Jan 84 11:08:30-PST
From: Don Walker <WALKER@SRI-AI.ARPA>
Subject: Re: next meeting
To: AAAI-OFFICE@SUMEX-AIM.ARPA
cc: JMC@SU-AI.ARPA, Genesereth@SUMEX-AIM.ARPA, Buchanan@SUMEX-AIM.ARPA,
Feinler@SRI-NIC.ARPA, Amsler@SRI-AI.ARPA
In-Reply-To: Message from "AAAI <AAAI-OFFICE@SUMEX-AIM.ARPA>" of Wed 4 Jan 84 09:31:36-PST
Bob Amsler and I will both be out of town on that date and probably not back
until late in the week.
Don
-------
∂04-Jan-84 1152 TW Consulting professorship committee meeting
To: DEK@SU-AI, JMC@SU-AI, TW@SU-AI, ZM@SU-AI
CC: golub@SU-SCORE
My own opinion at this point is that we are getting pretty petty. The
whole reason for the appontments is to establish some connection and good
will between CSLI and CSD. The reason for people to be appointed in CSD
is that they in fact teach courses and advise students in our department,
regardless of what other appointments they hold. The joint appointment
strategy may not have been the ideal one for Perry and Barwise to pursue,
but they have, and not going along with it creates a good deal of hassle
and paperwork for them (i.e., going back to the other departments to get
official approval of changing the already-assigned percentages there),
plus ill-will among the people we wouldn't appoint (even if we assure them
it was nothing personal).
When there was a question about individual qualifications it seemed to me
totally appropriate that we as a department take an active role in
deciding whether they were in fact people we wanted listed as our
colleagues. There has been a complete consensus that that on examination,
there is no problem and that we would be happy to appoint any of them on
that basis. When the only issue is the somewhat abstract one of whether
it is a good idea for consulting people to have joint appointments, I
think we put ourself in the position of being obstructionist for no
particular reason but our unwillingness to go along with a way of dealing
with Stanford red tape that we didn't ourselves invent.
Since my vote is clearly not in consensus with the rest, we do need to
meet and come up with a report for the faculty meeting. I would like to
have a meeting of the committee sometime on Monday to put together a
recommendation for the faculty meeting on Tuesday. Let me know what by
Friday noon what times between 9 and 5 Monday you are NOT available.
Thanks --t
∂04-Jan-84 1622 ASHOK@SU-SCORE.ARPA CS 206 grades
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Jan 84 16:22:10 PST
Date: Wed 4 Jan 84 16:22:53-PST
From: Ashok Subramanian <ASHOK@SU-SCORE.ARPA>
Subject: CS 206 grades
To: jmc@SU-AI.ARPA
The people at the Registrar's office tell me that my grades have not
been forwarded to them. The chart on Yoram's door reports a grade, however.
When I picked up my transcript, I was told to contact you to find out
what had happened. Please do let me know what has happened.
--ashok subramanian
-------
The CS206 grades were taken to the registrar's office on Jan. 3. Probably
they just haven't finished whatever their recording process is. -- Diana
∂04-Jan-84 1704 LUNDSTROM@SU-SCORE.ARPA [olken@LBL-CSAM (Frank Olken [csam]): Re: SIGBIG meeting]
Received: from SU-SCORE by SU-AI with TCP/SMTP; 4 Jan 84 17:04:50 PST
Date: Wed 4 Jan 84 17:03:02-PST
From: Stephen Lundstrom <LUNDSTROM@SU-SCORE.ARPA>
Subject: [olken@LBL-CSAM (Frank Olken [csam]): Re: SIGBIG meeting]
To: super@SU-SCORE.ARPA
Here are directions to tonights meeting (sorry it's late coming, but
I just got it). Steve
---------------
Return-Path: <olken@LBL-CSAM>
Received: from lbl-csam.ARPA by SU-SCORE.ARPA with TCP; Wed 4 Jan 84 16:38:42-PST
Date: Wed, 4 Jan 84 16:40:35 pst
From: olken@LBL-CSAM (Frank Olken [csam])
Return-Path: <olken@LBL-CSAM>
Message-Id: <8401050040.AA01187@lbl-csam.ARPA>
Received: by lbl-csam.ARPA ; Wed, 4 Jan 84 16:40:35 pst
To: LUNDSTROM@SU-SCORE.ARPA
Subject: Re: SIGBIG meeting
Instructions to reach the January SIGBIG meeting at LBL.
LBL is located on a hill east of UC Berkeley, overlooking the campus.
To reach Lawrence Berkeley Lab take Highway 17 (Nimitz Freeway) northbound
to the University Ave. turnoff in Berkeley. Go east on University Ave.
until you reach the university at Oxford St. Turn left (north) onto
Oxford. Go about 1 block to a traffic light and then turn right
(east) onto Hearst Ave. Go up the hill on Hearst Ave past the traffice light at
Hearst and Euclid to the traffic light at Hearst & Gayley (about 7 blocks).
Continue on Hearst. A block beyond the traffic light Hearst veers right,
commencing a switchback. Continue up Hearst to the guardhouse.
The guards have been notified of the meeting and will have parking passes and
maps available.
You must stop at the guardhouse to pick up a parking pass, otherwise
you will get a Berkeley parking ticket. There should be a parking pass
waiting for you at the guardhouse.
Continue up the hill to the next bend (to the right). On your left
is the visitors parking lot. You should be able to find a parking
space there.
The SIGBIG meeting is in the new director's conference room, Bldg 50A, Rm 5132.
Bldg 50A is the northern wing of the large brown U-shaped administration
building across the road
from the visitors parking lot. Cross the street, go up the stairs,
down the corridor to the elevator. Take the elevator to the fifth
floor and turn left as you exit.
The director's conference room is near the end of corridor on the left.
If the visitors parking lot is full, park anyplace that is marked
as a parking place, but not reserved. Unmarked parking places are
illegal. There is parking both on the uphill and downhill side of
building 50. If all else fails you may have to park in Blackberry
Canyon, a pit reached by continuing beyond the visitors parking lot
and turning left at every opportunity. There are stairs back to the
visitors parking lot.
Frank Olken
CSAM Dept.
50B/3238
415-486-5891
-------
∂05-Jan-84 2053 KJB@SRI-AI.ARPA Thomason paper
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Jan 84 20:53:43 PST
Date: Thu 5 Jan 84 14:37:56-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Thomason paper
To: csli-d2@SRI-AI.ARPA, csli-b1@SRI-AI.ARPA, csli-b5@SRI-AI.ARPA
Rich Thomason sent me about 10 copies of a paper called "Accomodation,
conversational planning and implicature", which is based on his talk
in the seminar here last month, I think. He asked that I give it
to interested parties. I will leave them in the reading room downstaris.
Jon
-------
∂05-Jan-84 2053 KJB@SRI-AI.ARPA Thomason paper
Received: from SRI-AI by SU-AI with TCP/SMTP; 5 Jan 84 20:53:43 PST
Date: Thu 5 Jan 84 14:37:56-PST
From: Jon Barwise <KJB@SRI-AI.ARPA>
Subject: Thomason paper
To: csli-d2@SRI-AI.ARPA, csli-b1@SRI-AI.ARPA, csli-b5@SRI-AI.ARPA
Rich Thomason sent me about 10 copies of a paper called "Accomodation,
conversational planning and implicature", which is based on his talk
in the seminar here last month, I think. He asked that I give it
to interested parties. I will leave them in the reading room downstaris.
Jon
-------
∂05-Jan-84 2200 DEK proxy
To: TW, JMC, ZM
I'll be out of town next week. You have all brought up valid points.
I have decided to vote with JMC on whatever position he finally
decides to adopt.
∂05-Jan-84 2314 @SU-SCORE.ARPA:cheriton@diablo Distributed Prolog article
Received: from SU-SCORE by SU-AI with TCP/SMTP; 5 Jan 84 23:14:23 PST
Received: from Diablo by SU-SCORE.ARPA with TCP; Thu 5 Jan 84 23:05:28-PST
Date: Thu, 5 Jan 84 23:04 PST
From: David Cheriton <cheriton@diablo>
Subject: Distributed Prolog article
To: super@score
Has anyone read "Executing Distributed Prolog Programs on Broadcast
Networks" by David Warren, to appear in Int. Conference on Logical
Programming. I am interested in an opinion on same as well as a copy
of the paper - if it is worth looking at.
∂06-Jan-84 1543 DFH Stoyan accommodations
I got him a room at the faculty club for $65/night,
which will be charged to your faculty club account.
When you get this bill I can do an expense report to
charge to your unrestricted.
∂06-Jan-84 1646 DFH Messages
1. Prof. Rohit Parikh (Brooklyn College) will be coming to the West Coast
1/10, at Stanford 1/12. He will try to get ahold of you then. He
wants to know if you want your presentation included as part of the
Proceedings for the conference (he thinks there will be enough
papers to do one). Also he wanted to let you know that there will
be a slight delay in reimbursing your expenses because not everyone
had filed their expense reports yet. If you want to call him before
he leaves (Monday only), no. is 914-833-0288.
2. Rosa Chang of FMC. 1/4. (408) 289-2850
3. John Weiner. 1/5. NY Acad. of Sciences. Wants to see if you would
write an article for their magazine. 212-838-0230 X257.
4. Gina Maranto, Discover Magazine. 1/6. Doing a story on the "mind-brain"
problem and would like to talk with you. She will call again.
∂07-Jan-84 1442 BSCOTT@SU-SCORE.ARPA ARPA Umbrella Contract
Received: from SU-SCORE by SU-AI with TCP/SMTP; 7 Jan 84 14:42:22 PST
Date: Sat 7 Jan 84 14:41:52-PST
From: Betty Scott <BSCOTT@SU-SCORE.ARPA>
Subject: ARPA Umbrella Contract
To: JMC@SU-AI.ARPA
cc: BScott@SU-SCORE.ARPA
John, the negotiations on the new contract are really bogged down in the
bureaucracy--they haven't even begun to negotiate yet.
John Machado and Ron Ohlander said we would be given a $50K modification
to the old contract to keep the research going. The Navelex contracts people
said verbally that we could spend the $50K pending receipt of the modification,
but that if the big contract was not negotiated and awarded, Stanford would
be responsible for the $50K. What they are saying is that we can spend the
money at our own risk.
Anyway, I talked with Gene Golub and Tom Rindfleisch about this at our staff
meeting this week. Gene says he doesn't want to take the $50K risk, and they
both said that probably Bob Kahn is not aware of this situation. Gene suggests
that you call or message Kahn to see if he can get some action for us.
I will appreciate anything you can do--our Sponsored Projects Office is having
difficulty getting the Navelex contracts people to respond to anything.
And Gio is really without any funds at all at the present time.
Thanks for any help you can give.
Betty
-------
∂07-Jan-84 2239 RWW STOYAN
To: CLT, JMC
HE DOES NOT SEEM TO HAVE ARRIVED. ALSO THE MOTEL DOESN'T SEEM TO
EXPECT HIM!!!!!! I AM CONFUSED ANY IDEAS WOULD BE HELPFUL.
THANKS
RICHARD
Check the Faculty Club. I decided to be unstingy and put him up
there.
∂07-Jan-84 2251 RWW THANKS
HE WAS THERE. DONT WORRY ILL TAKE CARE OF HIM.
RICHARD
∂08-Jan-84 2035 KEDES@SUMEX-AIM.ARPA McCarthy on Bacteria
Received: from SUMEX-AIM by SU-AI with TCP/SMTP; 8 Jan 84 20:35:30 PST
Date: Sun 8 Jan 84 20:40:16-PST
From: Larry Kedes <KEDES@SUMEX-AIM.ARPA>
Subject: McCarthy on Bacteria
To: su-bboard@SUMEX-AIM.ARPA
cc: jmc@SU-AI.ARPA
While intervening sequences may well be comments in the genetic code,
God only put them in the code of eukaryotes. Bacterial genes are
characteristically devoid of intervening sequences.
-------
∂08-Jan-84 2203 JVC@SU-SCORE.ARPA Re: McCarthy on Bacteria
Received: from SU-SCORE by SU-AI with TCP/SMTP; 8 Jan 84 22:03:41 PST
Date: Sun 8 Jan 84 22:02:31-PST
From: Janet Coursey <JVC@SU-SCORE.ARPA>
Subject: Re: McCarthy on Bacteria
To: su-bboard@SU-SCORE.ARPA
cc: kedes@SUMEX-AIM.ARPA, jmc@SU-AI.ARPA
In-Reply-To: Message from "Larry Kedes <KEDES@SUMEX-AIM.ARPA>" of Sun 8 Jan 84 20:35:16-PST
I think that was (one of) the point(s): "When God *finished* the bacteria..."
(my italics). But I do not know what an Edsger is. Anyone know?
-------