perm filename MSG.MSG[1,JMC] blob
sn#885247 filedate 1990-06-14 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00398 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00048 00002 ∂01-Apr-90 1236 pjd@riacs.edu Re: AI and strong AI
C00050 00003 ∂01-Apr-90 1507 underdog@portia.stanford.edu re: gopher
C00051 00004 ∂01-Apr-90 1831 Mailer failed mail returned
C00053 00005 ∂01-Apr-90 2143 hoffman@csli.Stanford.EDU a family emergency
C00054 00006 ∂01-Apr-90 2155 vera@fanaraaken.STANFORD.EDU Re: Japanese Space Effort
C00055 00007 ∂02-Apr-90 0024 ME alberta.bitnet
C00056 00008 ∂02-Apr-90 1155 pjd@riacs.edu Letters
C00057 00009 ∂02-Apr-90 1253 shoham@Hudson.Stanford.EDU meeting
C00058 00010 ∂02-Apr-90 1301 @Score.Stanford.EDU,@ai.ai.mit.edu,@mc.lcs.mit.edu,@central.cis.upenn.edu:dale@linc.cis.upenn.edu ICLP90 conference information and registration
C00090 00011 ∂02-Apr-90 1352 AI.LENAT@MCC.COM
C00091 00012 ∂02-Apr-90 1755 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
C00094 00013 ∂02-Apr-90 1857 @Score.Stanford.EDU,@ai.ai.mit.edu,@mc.lcs.mit.edu,@relay.cdnnet.ca:joyce@cs.ubc.ca Cambridge HOL System Course
C00109 00014 ∂02-Apr-90 2000 JMC
C00110 00015 ∂02-Apr-90 2000 JMC
C00111 00016 ∂02-Apr-90 2151 @IGNORANT.Stanford.EDU:RDZ@SCORE.Stanford.EDU Question
C00113 00017 ∂03-Apr-90 1327 searle@cogsci.berkeley.edu Re: Andreas Dorschel adresss
C00114 00018 ∂03-Apr-90 1515 ingrid@russell.Stanford.EDU your presentation at the IAP meeting next week
C00116 00019 ∂03-Apr-90 1827 CN.MCS@Forsythe.Stanford.EDU
C00120 00020 ∂04-Apr-90 0737 chandler@Sunburn.Stanford.EDU Re: committee on computer readable phd theses
C00122 00021 ∂04-Apr-90 1041 mps@Gang-of-Four.Stanford.EDU committee meeting
C00123 00022 ∂04-Apr-90 1046 ingrid@russell.Stanford.EDU Re: title and handout
C00125 00023 ∂04-Apr-90 1058 ingrid@russell.Stanford.EDU Re: title and handout
C00127 00024 ∂04-Apr-90 1110 ingrid@russell.Stanford.EDU Re: title and handout
C00129 00025 ∂04-Apr-90 1118 ingrid@russell.Stanford.EDU Re: title and handout
C00131 00026 ∂04-Apr-90 1140 VAL message from Moscow
C00132 00027 ∂04-Apr-90 1206 Mailer re: death penalty
C00133 00028 ∂04-Apr-90 1217 VAL
C00134 00029 ∂04-Apr-90 1442 gray@Neon.Stanford.EDU Re: Cobol question
C00137 00030 ∂04-Apr-90 1431 jutta@coyote.stanford.edu AI Division Lunch
C00139 00031 ∂04-Apr-90 1545 weening@Gang-of-Four.Stanford.EDU Cobol question
C00142 00032 ∂04-Apr-90 2015 sreerang@portia.stanford.edu term paper (cs323)
C00144 00033 ∂04-Apr-90 2028 sreerang@portia.stanford.edu re: term paper (cs323)
C00145 00034 ∂05-Apr-90 0029 ME end of a long system
C00148 00035 ∂05-Apr-90 0932 MPS Files
C00149 00036 ∂05-Apr-90 1001 MPS
C00150 00037 ∂05-Apr-90 1037 hiller@Hudson.Stanford.EDU AI Day-On-Campus Brochure
C00152 00038 ∂05-Apr-90 1342 MPS Meeting
C00153 00039 ∂05-Apr-90 1404 MPS
C00154 00040 ∂05-Apr-90 1514 rathmann@eclipse.stanford.edu Authoritative partitions
C00158 00041 ∂06-Apr-90 0903 JMC
C00159 00042 ∂06-Apr-90 0927 mps@Gang-of-Four.Stanford.EDU
C00160 00043 ∂06-Apr-90 1229 slagle@cs.umn.edu Distinguished Visitor
C00164 00044 ∂06-Apr-90 1359 @IBM.COM:LEORA@YKTVMH AAAI Symposium Proposal
C00170 00045 ∂06-Apr-90 1533 shoham@Hudson.Stanford.EDU meeting
C00172 00046 ∂06-Apr-90 1605 CN.MCS@Forsythe.Stanford.EDU
C00174 00047 ∂06-Apr-90 1619 shoham@Hudson.Stanford.EDU re: meeting
C00176 00048 ∂06-Apr-90 1719 ingrid@russell.Stanford.EDU your IAP presentations
C00178 00049 ∂07-Apr-90 1221 shoham@Hudson.Stanford.EDU time
C00180 00050 ∂08-Apr-90 1422 VAL re: time
C00181 00051 ∂08-Apr-90 1445 shoham@Hudson.Stanford.EDU ok
C00182 00052 ∂08-Apr-90 2307 rdz@Gang-of-Four.Stanford.EDU 15
C00184 00053 ∂09-Apr-90 0104 @IGNORANT.Stanford.EDU:RDZ@SCORE.Stanford.EDU puzzle
C00186 00054 ∂09-Apr-90 0124 @Score.Stanford.EDU,@PRECARIOUS.Stanford.EDU:RDZ@SCORE.Stanford.EDU Current problem
C00188 00055 ∂09-Apr-90 0755 mps@Gang-of-Four.Stanford.EDU
C00190 00056 ∂09-Apr-90 0800 JMC
C00191 00057 ∂09-Apr-90 0919 GA.JRG@Forsythe.Stanford.EDU Harley, care to help out Prof. McCarthy on this?
C00193 00058 ∂09-Apr-90 1024 MPS Chairmanship of CS
C00194 00059 ∂09-Apr-90 1104 shoham@Hudson.Stanford.EDU ai doc
C00197 00060 ∂09-Apr-90 1225 Mailer re: Death penalty
C00198 00061 ∂09-Apr-90 1345 jfetzer@ub.d.umn.edu Books for Review
C00200 00062 ∂09-Apr-90 1432 ingrid@russell.Stanford.EDU handouts for tomorrow's meeting
C00201 00063 ∂09-Apr-90 1557 guha@sumex-aim.stanford.edu mid term report
C00203 00064 ∂09-Apr-90 1600 winograd@loire.stanford.edu Re: ai doc
C00204 00065 ∂09-Apr-90 1601 boyer@CLI.COM References
C00206 00066 ∂09-Apr-90 1920 boyer@CLI.COM References
C00209 00067 ∂10-Apr-90 0907 littell@Sunburn.Stanford.EDU Fellowship supplement
C00210 00068 ∂10-Apr-90 1147 WALDINGER@Warbucks.AI.SRI.COM
C00212 00069 ∂10-Apr-90 1221 AI.GUHA@MCC.COM
C00213 00070 ∂10-Apr-90 1257 jfetzer@ub.d.umn.edu re: Books for Review
C00215 00071 ∂10-Apr-90 1417 rdz@Gang-of-Four.Stanford.EDU [maja@ai.mit.edu: seminar this Thursday 4p.m.]
C00220 00072 ∂10-Apr-90 1538 MPS
C00221 00073 ∂11-Apr-90 1229 U.UNDERDOG@Macbeth.Stanford.EDU being a gopher
C00222 00074 ∂11-Apr-90 1312 slagle@cs.umn.edu Draft--McCarthy's Schedule
C00225 00075 ∂11-Apr-90 1358 jones@Sunburn.Stanford.EDU undergraduate colloquium
C00227 00076 ∂11-Apr-90 1546 VAL Texas
C00230 00077 ∂11-Apr-90 1715 shoham@Hudson.Stanford.EDU follow up
C00231 00078 ∂11-Apr-90 1721 VAL re: follow up
C00232 00079 ∂11-Apr-90 1900 JMC
C00233 00080 ∂12-Apr-90 0835 ito@ito.ecei.tohoku.ac.jp 'Foreword'
C00236 00081 ∂12-Apr-90 1416 MRC@CAC.Washington.EDU re: Death Penalty Alternative
C00239 00082 ∂12-Apr-90 1503 Mailer Re: housing policy for unmarried cohabitants
C00242 00083 ∂12-Apr-90 1739 jones@Sunburn.Stanford.EDU undergrad colloquium
C00244 00084 ∂12-Apr-90 2220 Mailer re: housing policy for unmarried cohabitants
C00247 00085 ∂12-Apr-90 2257 vera@fanaraaken.STANFORD.EDU Re: housing policy for unmarried cohabitants
C00251 00086 ∂13-Apr-90 0800 JMC
C00255 00087 ∂13-Apr-90 1140 bthomas@Neon.Stanford.EDU Re: housing policy for unmarried cohabitants
C00258 00088 ∂13-Apr-90 1231 bthomas@Neon.Stanford.EDU Re: housing policy for unmarried cohabitants
C00261 00089 ∂13-Apr-90 1445 VAL re: phone call on my line
C00262 00090 ∂13-Apr-90 1708 Mailer re: housing policy for unmarried cohabitants
C00264 00091 ∂13-Apr-90 1738 mogul@wrl.dec.com Re: Death Penalty Alternative
C00267 00092 ∂13-Apr-90 2307 beeson@ucscd.UCSC.EDU a mathematical problem for your amusement
C00271 00093 ∂14-Apr-90 0049 beeson@ucscd.UCSC.EDU yes indeed
C00272 00094 ∂14-Apr-90 0141 LES re: Spider contest
C00275 00095 ∂15-Apr-90 0427 ito@ito.ecei.tohoku.ac.jp R
C00279 00096 ∂16-Apr-90 0958 weening@Gang-of-Four.Stanford.EDU Re: Am I truthful?
C00287 00097 ∂16-Apr-90 1302 rpg@lucid.com Truth
C00297 00098 ∂16-Apr-90 1445 gini@cs.umn.edu your visit to Minneapolis
C00298 00099 ∂16-Apr-90 1625 mps@Gang-of-Four.Stanford.EDU Trip to Minn
C00299 00100 ∂16-Apr-90 1749 rick@hanauma.stanford.edu automobile hydrogen fuel
C00301 00101 ∂17-Apr-90 0527 cross@vax.darpa.mil Re: appointments
C00303 00102 ∂17-Apr-90 0601 rapaport@cs.Buffalo.EDU UB Center for Cognitive Science Workshop
C00319 00103 ∂17-Apr-90 0620 rapaport@cs.Buffalo.EDU Alonzo Church Symposium
C00330 00104 ∂17-Apr-90 1122 scherlis@vax.darpa.mil Request for help
C00334 00105 ∂17-Apr-90 1152 S.SALUT@Macbeth.Stanford.EDU [liza@ernie.Berkeley.EDU (Liza C. Gabato): CS Division, U.C.B. Seminar for the week of 4/16/90 (long).]
C00341 00106 ∂17-Apr-90 1427 ito@ito.ecei.tohoku.ac.jp R
C00343 00107 ∂17-Apr-90 1552 VAL Commonsense and Nonmonotonic Reasoning Seminar
C00345 00108 ∂18-Apr-90 0607 CLT qlisp
C00346 00109 ∂18-Apr-90 1124 korf@CS.UCLA.EDU Your visit
C00348 00110 ∂18-Apr-90 1404 shoham@Hudson.Stanford.EDU next meeting
C00350 00111 ∂18-Apr-90 1422 VAL re: next meeting
C00351 00112 ∂19-Apr-90 0812 shekhar@cs.umn.edu request for a paper
C00353 00113 ∂19-Apr-90 0848 jes@cs.brown.edu Request for help
C00384 00114 ∂19-Apr-90 2142 harnad@clarity.Princeton.EDU Dynamic Behavioral Modeling: BBS Call for Commentators
C00389 00115 ∂20-Apr-90 0904 shoham@Hudson.Stanford.EDU
C00390 00116 ∂20-Apr-90 0905 shoham@Hudson.Stanford.EDU
C00391 00117 ∂20-Apr-90 1012 jutta@coyote.stanford.edu Spring Quarter AI Division Lunch
C00393 00118 ∂20-Apr-90 1042 VAL Bondarenko on Yale shooting
C00395 00119 ∂20-Apr-90 1112 jutta@coyote.stanford.edu URGENT AI Division business
C00397 00120 ∂20-Apr-90 1527 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
C00399 00121 ∂20-Apr-90 1648 korf@CS.UCLA.EDU Re: May
C00401 00122 ∂22-Apr-90 1524 harnad@clarity.Princeton.EDU SPP Symposium/Workshop
C00405 00123 ∂22-Apr-90 1819 harnad@clarity.Princeton.EDU Neuropsychology of Schizophrenia: BBS Call for Commentators
C00411 00124 ∂22-Apr-90 2011 skitodie@med.stanford.edu PDP
C00414 00125 ∂23-Apr-90 1419 VAL bug in Bondarenko's solution
C00417 00126 ∂23-Apr-90 1646 casley@Neon.Stanford.EDU Origins of Lisp
C00419 00127 ∂23-Apr-90 1646 guha@sumex-aim.stanford.edu meeting
C00421 00128 ∂24-Apr-90 1027 @Score.Stanford.EDU:james@cs.rochester.edu a question
C00423 00129 ∂24-Apr-90 1118 korf@CS.UCLA.EDU Oops
C00425 00130 ∂24-Apr-90 1212 VAL Commonsense and Nonmonotonic Reasoning Seminar
C00427 00131 ∂24-Apr-90 1920 harnad@clarity.Princeton.EDU Movshon Abstract for SPP
C00435 00132 ∂24-Apr-90 1952 harnad@clarity.Princeton.EDU Granger Abstract for SPP
C00439 00133 ∂25-Apr-90 1425 lin@Neon.Stanford.EDU Reading committee
C00441 00134 ∂25-Apr-90 1442 VAL re: Reading committee
C00442 00135 ∂25-Apr-90 1618 cdp!kpeters@labrea.stanford.edu RISKS contribution
C00446 00136 ∂25-Apr-90 1900 rathmann@eclipse.stanford.edu new draft reference request
C00449 00137 ∂25-Apr-90 2138 lin@Neon.Stanford.EDU re: Reading committee
C00450 00138 ∂25-Apr-90 2143 rathmann@eclipse.stanford.edu Re: new draft reference request
C00452 00139 ∂25-Apr-90 2148 rathmann@eclipse.stanford.edu Re: new draft reference request
C00454 00140 ∂25-Apr-90 2202 rathmann@eclipse.stanford.edu Re: new draft reference request
C00457 00141 ∂26-Apr-90 0623 @MCC.COM:ai.guha@MCC.COM re: meeting
C00459 00142 ∂26-Apr-90 0714 james@cs.rochester.edu re: a question
C00462 00143 ∂26-Apr-90 1130 JMC
C00463 00144 ∂26-Apr-90 1145 korf@CS.UCLA.EDU Re: Oops
C00465 00145 ∂26-Apr-90 1255 nilsson@Tenaya.Stanford.EDU Re: renewal of courtesy and consulting faculty
C00467 00146 ∂26-Apr-90 1335 cdp!kpeters@labrea.stanford.edu re: RISKS contribution
C00469 00147 ∂26-Apr-90 1846 nilsson@Tenaya.Stanford.EDU Principia Schedule
C00471 00148 ∂26-Apr-90 2218 eaf@sumex-aim.stanford.edu Re: quote
C00473 00149 ∂26-Apr-90 2219 eaf@sumex-aim.stanford.edu [John McCarthy <JMC@sail.stanford.edu> : quot]
C00475 00150 ∂27-Apr-90 0113 hoffman@csli.Stanford.EDU IEEE Expert Interview (reprint from last summer)
C00481 00151 ∂27-Apr-90 0915 harnad@clarity.Princeton.EDU SPP Abstract: Powers
C00490 00152 ∂27-Apr-90 1019 hoffman@csli.Stanford.EDU re: IEEE Expert Interview (reprint from last summer)
C00492 00153 ∂27-Apr-90 1634 jlm@lucid.com PhD program
C00495 00154 ∂28-Apr-90 1613 hoffman@csli.Stanford.EDU edited Interview (with a few questions embeddedin [...]
C00506 00155 ∂28-Apr-90 1615 hoffman@csli.Stanford.EDU unedited interview
C00518 00156 ∂28-Apr-90 1637 hoffman@csli.Stanford.EDU two things
C00520 00157 ∂28-Apr-90 1959 mommyduc@med.stanford.edu Reid Hoffman and the Dinkelspiel Award
C00523 00158 ∂29-Apr-90 1730 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
C00525 00159 ∂29-Apr-90 2157 hoffman@csli.Stanford.EDU previous messages
C00527 00160 ∂29-Apr-90 2158 hoffman@csli.Stanford.EDU fairly soon in the previous letter means tomorrow evening.
C00529 00161 ∂30-Apr-90 0800 JMC
C00530 00162 ∂30-Apr-90 0800 JMC
C00531 00163 ∂30-Apr-90 0824 carlstea@sierra.Stanford.EDU re: Parking
C00532 00164 ∂30-Apr-90 1040 susan@nessus.stanford.edu Parking
C00536 00165 ∂30-Apr-90 1330 JMC
C00537 00166 ∂30-Apr-90 1403 harnad@clarity.Princeton.EDU SPP: Powers abstract (shortened)
C00541 00167 ∂30-Apr-90 1525 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Signatures
C00543 00168 ∂30-Apr-90 2054 hoffman@csli.Stanford.EDU IEEE Interview
C00545 00169 ∂30-Apr-90 2130 weening@Gang-of-Four.Stanford.EDU
C00551 00170 ∂30-Apr-90 2209 hoffman@csli.Stanford.EDU re: IEEE Interview
C00552 00171 ∂01-May-90 0000 JMC
C00553 00172 ∂01-May-90 0800 JMC
C00554 00173 ∂01-May-90 0850 wheaton@Sunburn.Stanford.EDU
C00556 00174 ∂01-May-90 0900 shoham@Hudson.Stanford.EDU reimnder
C00557 00175 ∂01-May-90 1100 JMC
C00558 00176 ∂01-May-90 1110 RPG Lowell
C00559 00177 ∂01-May-90 1337 jutta@Sunburn.Stanford.EDU AI Division Meeting May 2
C00561 00178 ∂01-May-90 1614 lum@portia.stanford.edu SURF
C00565 00179 ∂01-May-90 1617 guha@Neon.Stanford.EDU Thursdays appointment.
C00567 00180 ∂01-May-90 1631 jutta@Sunburn.Stanford.EDU Time for AI Division Meeting
C00569 00181 ∂01-May-90 2158 @MCC.COM:ai.guha@MCC.COM re: Thursdays appointment.
C00571 00182 ∂01-May-90 2321 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
C00574 00183 ∂02-May-90 0856 CLT bing today
C00575 00184 ∂02-May-90 0927 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU Cognology
C00577 00185 ∂02-May-90 0957 MPS
C00578 00186 ∂02-May-90 1100 JMC
C00579 00187 ∂02-May-90 1202 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU re: Cognology
C00581 00188 ∂02-May-90 1433 rlm@Hudson.Stanford.EDU AIDOC ABSTRACTS
C00583 00189 ∂02-May-90 1506 gjohn@Neon.Stanford.EDU summer research
C00586 00190 ∂02-May-90 1522 MPS Toy
C00587 00191 ∂02-May-90 1602 harnad@clarity.Princeton.EDU Re: SPP and representations
C00596 00192 ∂02-May-90 1642 carlstea@sierra.Stanford.EDU re: "The Meeting"
C00598 00193 ∂02-May-90 1746 weening@Gang-of-Four.Stanford.EDU
C00600 00194 ∂02-May-90 2041 jones@Sunburn.Stanford.EDU undergraduate lunch
C00602 00195 ∂02-May-90 2058 jones@Sunburn.Stanford.EDU re: undergraduate lunch
C00603 00196 ∂02-May-90 2335 shoham@Hudson.Stanford.EDU teaching ai
C00608 00197 ∂03-May-90 0748 HF.SXM@Forsythe.Stanford.EDU re: There is more to it
C00609 00198 ∂03-May-90 1258 jim@kaos.Stanford.EDU Re: Bikes and Bikes
C00612 00199 ∂03-May-90 1300 LES re: Parking
C00616 00200 ∂03-May-90 1456 rabin%humus.huji.ac.il@Forsythe.Stanford.EDU fall term visit
C00621 00201 ∂03-May-90 1638 Mailer Motorcycle safety
C00629 00202 ∂03-May-90 1722 LES re: Bikes and Bikes
C00631 00203 ∂03-May-90 1848 Mailer re: Motorcycle safety
C00635 00204 ∂04-May-90 0700 JMC
C00636 00205 ∂04-May-90 0800 MPS
C00637 00206 ∂04-May-90 0800 JMC
C00638 00207 ∂04-May-90 0800 JMC
C00639 00208 ∂04-May-90 1003 ginsberg@Sunburn.Stanford.EDU Re: teaching ai
C00644 00209 ∂04-May-90 1502 eaf@sumex-aim.stanford.edu Re: teaching ai
C00647 00210 ∂04-May-90 1546 VAL Commonsense and Nonmonotonic Reasoning Seminar: No meeting
C00648 00211 ∂04-May-90 1633 shoham@Hudson.Stanford.EDU courses, round 2
C00655 00212 ∂04-May-90 1717 ginsberg@Sunburn.Stanford.EDU Re: courses, round 2
C00659 00213 ∂04-May-90 1757 winograd@loire.stanford.edu Re: courses, round 2
C00663 00214 ∂05-May-90 0602 qphysics-owner@neat.cs.toronto.edu in search of a moderator
C00673 00215 ∂06-May-90 0707 harnad@clarity.Princeton.EDU SPP Abstract
C00675 00216 ∂06-May-90 0723 harnad@clarity.Princeton.EDU SPP Abstracts needed
C00678 00217 ∂06-May-90 2156 jones@Sunburn.Stanford.EDU Re: courses, round 2
C00680 00218 ∂07-May-90 0020 ok@coyote.stanford.edu AI Qual Schedule - About to be made official
C00683 00219 ∂07-May-90 1004 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Dissertation
C00685 00220 ∂07-May-90 1305 larson@unix.sri.com Re: Bikes and Bikes
C00688 00221 ∂07-May-90 1616 VAL Commonsense and Nonmonotonic Reasoning Seminar
C00690 00222 ∂07-May-90 2105 GLB
C00692 00223 ∂07-May-90 2110 harnad@clarity.Princeton.EDU Hayes's Abstract for SPP
C00696 00224 ∂08-May-90 1730 hayes@roo.parc.xerox.com LISP syntax
C00700 00225 ∂09-May-90 0905 CLT TTM ss#
C00701 00226 ∂09-May-90 1354 @IBM.COM:LEORA@YKTVMH response to workshop proposal
C00710 00227 ∂09-May-90 1518 harnad@clarity.Princeton.EDU Optimality: BBS Call for Commentators
C00715 00228 ∂09-May-90 1805 weening@Gang-of-Four.Stanford.EDU names for parallel iteration functions
C00720 00229 ∂09-May-90 1912 arg@lucid.com names for parallel iteration functions
C00723 00230 ∂09-May-90 1951 weening@Gang-of-Four.Stanford.EDU Re: names for parallel iteration functions
C00727 00231 ∂09-May-90 2034 weening@Gang-of-Four.Stanford.EDU parallel iteration functions
C00735 00232 ∂10-May-90 1121 pehoushe@Gang-of-Four.Stanford.EDU parallel iteration functions
C00739 00233 ∂10-May-90 1138 rpg@lucid.com parallel iteration functions
C00743 00234 ∂10-May-90 1201 pehoushe@Gang-of-Four.Stanford.EDU parallel iteration functions
C00746 00235 ∂10-May-90 1447 bergman@Sunburn.Stanford.EDU US-Japan travel grant
C00748 00236 ∂10-May-90 1613 ok@coyote.stanford.edu AI Qual Schedule
C00751 00237 ∂11-May-90 0957 korf@CS.UCLA.EDU visit
C00752 00238 ∂11-May-90 1201 jfetzer@ub.d.umn.edu Current Developments
C00756 00239 ∂11-May-90 1305 jlm@lucid.com names for parallel iteration functions
C00759 00240 ∂11-May-90 1431 MPS
C00760 00241 ∂11-May-90 1507 pehoushe@Gang-of-Four.Stanford.EDU Iteration funcations, mainly Qdotimes
C00766 00242 ∂11-May-90 1521 iam@Gang-of-Four.Stanford.EDU typep
C00769 00243 ∂11-May-90 1521 franz@cs.washington.edu A new home for the qualitative physics list
C00771 00244 ∂11-May-90 1526 rpg@lucid.com DYNAMIC
C00773 00245 ∂11-May-90 1724 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
C00775 00246 ∂11-May-90 2018 arg@lucid.com typep
C00779 00247 ∂12-May-90 0730 weening@Gang-of-Four.Stanford.EDU Measuring parallel processor performance
C00782 00248 ∂12-May-90 1636 ok@coyote.stanford.edu Young's Depth Area
C00783 00249 ∂13-May-90 1054 harnad@clarity.Princeton.EDU Re: abstract
C00788 00250 ∂13-May-90 1055 shoham@Hudson.Stanford.EDU michael young quals
C00793 00251 ∂14-May-90 0911 MPS
C00794 00252 ∂14-May-90 0930 JMC
C00795 00253 ∂14-May-90 1000 JMC
C00796 00254 ∂14-May-90 1025 VAL
C00797 00255 ∂14-May-90 1039 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Signatures
C00799 00256 ∂14-May-90 1100 JMC
C00800 00257 ∂14-May-90 1100 JMC
C00801 00258 ∂14-May-90 1208 MPS Mileage miles
C00802 00259 ∂14-May-90 1211 MPS Mileage
C00803 00260 ∂14-May-90 1347 shoham@Hudson.Stanford.EDU qual depth area
C00805 00261 ∂14-May-90 1353 MPS
C00806 00262 ∂14-May-90 1445 RWF history of LISP
C00808 00263 ∂14-May-90 1747 young@Neon.Stanford.EDU re: qual depth area
C00810 00264 ∂14-May-90 1918 jwl%compsci.bristol.ac.uk@NSFnet-Relay.AC.UK Symposium on Computational Logic
C00819 00265 ∂15-May-90 0021 ME re: interesting finger "bug"
C00822 00266 ∂15-May-90 0319 nsuzuki%jpntscvm.bitnet@Forsythe.Stanford.EDU my resume
C00824 00267 ∂15-May-90 0621 rapaport@cs.Buffalo.EDU CALL FOR PAPERS
C00836 00268 ∂15-May-90 0822 mps@Gang-of-Four.Stanford.EDU
C00837 00269 ∂15-May-90 0928 mps@Gang-of-Four.Stanford.EDU
C00839 00270 ∂15-May-90 0929 mps@Gang-of-Four.Stanford.EDU
C00840 00271 ∂15-May-90 0945 damon@Sunburn.Stanford.EDU re: interesting finger "bug"
C00843 00272 ∂15-May-90 0946 rathmann@eclipse.stanford.edu did you get a chance to look at the section?
C00845 00273 ∂15-May-90 1149 Mailer re: delay in aid to Nicaragua and Panama
C00847 00274 ∂15-May-90 1413 PHY Knuth
C00848 00275 ∂16-May-90 1423 @IBM.COM:LEORA@YKTVMH Patel-Schneider
C00852 00276 ∂18-May-90 0227 ME your mail
C00853 00277 ∂17-May-90 2234 linvill@sierra.Stanford.EDU Hennessy Nomination
C00863 00278 ∂15-May-90 1425 korf@CS.UCLA.EDU Re: Let's postpone getting together
C00865 00279 ∂16-May-90 0935 MPS
C00866 00280 ∂16-May-90 1108 weening@Gang-of-Four.Stanford.EDU Top Level Common Lisp
C00868 00281 ∂16-May-90 2312 @MCC.COM:ai.guha@MCC.COM between the devil and the deep sea
C00873 00282 ∂17-May-90 1402 harnad@clarity.Princeton.EDU Re: spp abstract
C00886 00283 ∂17-May-90 1649 pehoushe@Gang-of-Four.Stanford.EDU A good way to implement qlambda
C00892 00284 ∂18-May-90 0550 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU
C00903 00285 ∂18-May-90 0729 boyer@CLI.COM Bledsoe Symposium
C00905 00286 ∂18-May-90 1004 VAL Commonsense and Nonmonotonic Reasoning Seminar
C00906 00287 ∂18-May-90 1559 visikka@cmx.npac.syr.edu
C00909 00288 ∂18-May-90 2128 hamilton@cs.sfu.ca Abstracts for AAAI?
C00911 00289 ∂19-May-90 2034 weld@cs.washington.edu Abstracts for AAAI?
C00914 00290 ∂21-May-90 0740 hogge@m2.csc.ti.com Qphysics Applications
C00916 00291 ∂21-May-90 0844 jc@cs.utexas.edu ABS: QPC: A Compiler from Physical Models into Qualitative Differential Equations
C00920 00292 ∂21-May-90 1741 arg@lucid.com Qlisp at toplevel
C00929 00293 ∂22-May-90 0608 @grenada:wellman@wrdc.af.mil Abstracts for AAAI?
C00932 00294 ∂22-May-90 0851 mkatz@garlic.Stanford.EDU Qlisp at toplevel
C00936 00295 ∂22-May-90 1432 arg@lucid.com Qlisp at toplevel
C00939 00296 ∂22-May-90 1835 @UWAVM.ACS.WASHINGTON.EDU:nishida@kuis.kyoto-u.ac.jp ABSTRACT: Reasoning with Model Lattices
C00944 00297 ∂24-May-90 1033 mps@Gang-of-Four.Stanford.EDU
C00945 00298 ∂24-May-90 1625 mps@Gang-of-Four.Stanford.EDU Farewell Dinner for Joe Weening
C00948 00299 ∂24-May-90 1643 mps@Gang-of-Four.Stanford.EDU Farewell Dinner for Joe Weening (correction)
C00951 00300 ∂25-May-90 0648 harnad@clarity.Princeton.EDU SPP Abstract
C00952 00301 ∂25-May-90 2009 beeson@ucscd.UCSC.EDU Pure Bargaining
C00955 00302 ∂28-May-90 0845 harnad@clarity.Princeton.EDU Urgent: SPP Abstract
C00957 00303 ∂28-May-90 1129 rathmann@eclipse.stanford.edu Welcome back
C00959 00304 ∂28-May-90 1709 harnad@clarity.Princeton.EDU Confirmation
C00960 00305 ∂28-May-90 1723 harnad@clarity.Princeton.EDU J. McCarthy's Abstract for SPP Searle Symposium
C00965 00306 ∂28-May-90 2120 rathmann@eclipse.stanford.edu new "almost final" draft
C00967 00307 ∂28-May-90 2317 young@Neon.Stanford.EDU re: qual depth area
C00969 00308 ∂29-May-90 0000 JMC
C00970 00309 ∂29-May-90 1100 rlm@Hudson.Stanford.EDU AIDOC ABSTRACTS NEEDED
C00972 00310 ∂29-May-90 1121 mps@Gang-of-Four.Stanford.EDU reply to message
C00974 00311 ∂29-May-90 1124 harnad@clarity.Princeton.EDU Biederman Abstract: SPP Symbol Grounding Workshop
C00980 00312 ∂29-May-90 1436 kuipers@cs.utexas.edu research job opportunity
C00985 00313 ∂29-May-90 1533 VAL Going to Tahoe
C00986 00314 ∂30-May-90 0736 mps@Gang-of-Four.Stanford.EDU Joe's Dinner
C00988 00315 ∂30-May-90 0915 CLT calenday
C00989 00316 ∂30-May-90 0953 jutta@Sunburn.Stanford.EDU reminder
C00991 00317 ∂30-May-90 1007 CLT revision/calendar item
C00992 00318 ∂30-May-90 1147 linvill@sierra.Stanford.EDU re: Hennessy Nomination
C00994 00319 ∂30-May-90 1210 gangolli@Theory.Stanford.EDU Rides to Grand China tonight
C00997 00320 ∂30-May-90 1215 "JC_._POSTMAST_@_THPINYC_(JC)%THPINYC"@mcimail.com John C. Dvorak 6/26/90 Editorial
C00999 00321 ∂30-May-90 1433 VAL Message from Arkady
C01001 00322 ∂30-May-90 1519 harnad@clarity.Princeton.EDU SPP Abstract: Pattee
C01005 00323 ∂30-May-90 1656 ok@coyote.stanford.edu Re: Depth Examiner
C01007 00324 ∂30-May-90 2336 @MCC.COM:ai.guha@MCC.COM re: between the devil and the deep sea
C01013 00325 ∂31-May-90 0900 scherlis@vax.darpa.mil ANNUAL REPORTING
C01020 00326 ∂31-May-90 0959 MPS
C01021 00327 ∂31-May-90 1001 weening@Gang-of-Four.Stanford.EDU SAIL -> Gang-of-Four
C01022 00328 ∂31-May-90 1013 mps@Gang-of-Four.Stanford.EDU
C01023 00329 ∂31-May-90 1131 scherlis@vax.darpa.mil ISTO SOFTWARE TECHNOLOGY COMMUNITY MEETING
C01040 00330 ∂31-May-90 1347 jfeldman%icsib2.Berkeley.EDU@jade.berkeley.edu Chinese Room
C01042 00331 ∂31-May-90 1358 scherlis@vax.darpa.mil Re: telephone appointment
C01044 00332 ∂31-May-90 1501 harnad@clarity.Princeton.EDU SPP Workshop: Plunkett Abstract
C01047 00333 ∂31-May-90 1609 mps@Gang-of-Four.Stanford.EDU Renewed book
C01048 00334 ∂31-May-90 1633 mps@Gang-of-Four.Stanford.EDU Rathman Thesis
C01050 00335 ∂31-May-90 1826 hanson@csli.Stanford.EDU invitation
C01052 00336 ∂31-May-90 1943 rathmann@eclipse.stanford.edu Any comments on my thesis?
C01054 00337 ∂01-Jun-90 0103 ME SAIL printing
C01055 00338 ∂01-Jun-90 1041 MPS
C01056 00339 ∂01-Jun-90 1500 harnad@clarity.Princeton.EDU Synopsis of SG Problem & 11 SPP Abstracts
C01098 00340 ∂01-Jun-90 1545 pratt@coraki.stanford.edu notes for CS 350
C01099 00341 ∂03-Jun-90 1247 rathmann@eclipse.stanford.edu
C01101 00342 ∂03-Jun-90 1636 rathmann@eclipse.stanford.edu circumscribing equality (your formulation)
C01107 00343 ∂03-Jun-90 1738 rathmann@eclipse.stanford.edu Re: circumscribing equality (your formulation)
C01110 00344 ∂04-Jun-90 0901 JMC
C01111 00345 ∂04-Jun-90 1023 rww@ibuki.com Hungary
C01112 00346 ∂04-Jun-90 1101 mostow@cs.rutgers.edu letter from Stefanuk
C01115 00347 ∂04-Jun-90 1107 rww@ibuki.com Soviet Commerce
C01117 00348 ∂04-Jun-90 1121 wood%hpmeg@LANL.GOV Preprint Request
C01119 00349 ∂04-Jun-90 1214 cphoenix@csli.Stanford.EDU Paper on agreement
C01121 00350 ∂04-Jun-90 1219 cphoenix@csli.Stanford.EDU re: Paper on agreement
C01122 00351 ∂04-Jun-90 1224 cphoenix@csli.Stanford.EDU re: Paper on agreement
C01123 00352 ∂04-Jun-90 1508 ME X terminals
C01124 00353 ∂04-Jun-90 1534 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
C01127 00354 ∂04-Jun-90 1653 VAL
C01129 00355 ∂04-Jun-90 1807 VAL Travel plans
C01130 00356 ∂04-Jun-90 2225 ME X terminal logout
C01132 00357 ∂05-Jun-90 0744 bundy@aipna.edinburgh.ac.uk Carolyn Talcott
C01133 00358 ∂05-Jun-90 0753 mps@Gang-of-Four.Stanford.EDU Rathman Thesis
C01134 00359 ∂05-Jun-90 1006 beeson@ucscd.UCSC.EDU software law (an idle question)
C01135 00360 ∂05-Jun-90 1658 ok@coyote.stanford.edu AI Qual Schedule (Reminder)
C01137 00361 ∂05-Jun-90 2107 ash@sumex-aim.stanford.edu Re: AI Qual Schedule (Reminder)
C01139 00362 ∂06-Jun-90 0950 ok@coyote.stanford.edu Re: AI Qual Schedule (Reminder)
C01141 00363 ∂06-Jun-90 1002 tajnai@Hudson.Stanford.EDU AI Day on Campus -- RSVP for lunch
C01143 00364 ∂06-Jun-90 1031 ME Tip crashes
C01146 00365 ∂06-Jun-90 1033 ME Tip crashes cont'd
C01147 00366 ∂06-Jun-90 1401 MPS Free Software
C01148 00367 ∂06-Jun-90 1553 MPS
C01149 00368 ∂07-Jun-90 0700 JMC
C01150 00369 ∂07-Jun-90 0730 mumford@sancho.harvard.edu ATP committee
C01152 00370 ∂07-Jun-90 1040 MPS
C01153 00371 ∂07-Jun-90 1348 MPS
C01154 00372 ∂07-Jun-90 1500 ME SAIL font
C01157 00373 ∂07-Jun-90 1901 H.HARPER@Macbeth.Stanford.EDU re: speaking of seccesionist states from federations
C01159 00374 ∂08-Jun-90 0854 njacobs@vax.darpa.mil ISTO Software Technology Community Meeting
C01163 00375 ∂08-Jun-90 1332 @coraki.stanford.edu:pratt@cs.stanford.edu Re: notes for CS 350
C01165 00376 ∂08-Jun-90 1415 betsy@russell.Stanford.EDU Information for Keith
C01168 00377 ∂11-Jun-90 0930 rlm@Hudson.Stanford.EDU
C01178 00378 ∂11-Jun-90 1035 mps@Gang-of-Four.Stanford.EDU Phone messages
C01179 00379 ∂11-Jun-90 1204 harnad@clarity.Princeton.EDU To all Spp Workshop/Symposium participants
C01182 00380 ∂11-Jun-90 1322 LISTSERV@VM1.NoDak.EDU Message
C01183 00381 ∂11-Jun-90 1601 xanadu!peterson@uunet.UU.NET book quote(s)
C01186 00382 ∂12-Jun-90 1151 E1.I85@Forsythe.Stanford.EDU
C01188 00383 ∂12-Jun-90 1233 tajnai@Hudson.Stanford.EDU
C01190 00384 ∂12-Jun-90 1418 mps@Gang-of-Four.Stanford.EDU Fed Ex
C01192 00385 ∂12-Jun-90 1622 tajnai@Hudson.Stanford.EDU
C01194 00386 ∂13-Jun-90 0821 tajnai@Hudson.Stanford.EDU Re: handout
C01195 00387 ∂13-Jun-90 0950 shoham@Hudson.Stanford.EDU aidocjmc
C01197 00388 ∂13-Jun-90 1139 tajnai@Hudson.Stanford.EDU Re: aidocjmc
C01199 00389 ∂13-Jun-90 1138 MPS Ablex
C01200 00390 ∂13-Jun-90 1141 tajnai@Hudson.Stanford.EDU re: aidocjmc
C01202 00391 ∂13-Jun-90 1324 tajnai@Hudson.Stanford.EDU re: aidocjmc
C01203 00392 ∂13-Jun-90 1420 MPS
C01204 00393 ∂13-Jun-90 1427 @IBM.COM:LEORA@YKTVMH patel-schneider's address
C01205 00394 ∂13-Jun-90 1643 harnad@clarity.Princeton.EDU Incest Avoidance: BBS Call for Commentators
C01211 00395 ∂14-Jun-90 0636 jdr@vax.darpa.mil test
C01212 00396 ∂14-Jun-90 1316 me@Gang-of-Four.Stanford.EDU files moving from SAIL
C01214 00397 ∂14-Jun-90 1351 MPS Sarah called
C01215 00398 ∂14-Jun-90 1617 me@Gang-of-Four.Stanford.EDU JMC files moved from SAIL to Go4
C01217 ENDMK
C⊗;
∂01-Apr-90 1236 pjd@riacs.edu Re: AI and strong AI
Received: from icarus.riacs.edu ([128.102.16.8]) by SAIL.Stanford.EDU with TCP; 1 Apr 90 12:36:06 PDT
Received: from hydra.riacs.edu by icarus.riacs.edu (5.59/2.1G)
id AA23352; Sun, 1 Apr 90 12:35:23 PDT
Received: by hydra.riacs.edu (4.12/2.0N)
id AA15026; Sun, 1 Apr 90 12:37:46 pdt
Message-Id: <9004011937.AA15026@hydra.riacs.edu>
Date: Sun, 1 Apr 90 12:37:46 pdt
From: Peter J. Denning <pjd@riacs.edu>
To: JMC@SAIL.Stanford.EDU
Subject: Re: AI and strong AI
John,
Thanks for your comments. I can't please everyone, obviously,
and therefore don't even try. I would very much like to see
your Penrose review and your letter to Scientific American.
If they are available electronically, send them that way, otherwise
send to me c/o RIACS, 230-5, NASA Ames Research Center, Moffett
Field, CA 94035.
Best wishes
Peter
∂01-Apr-90 1507 underdog@portia.stanford.edu re: gopher
Received: from portia.stanford.edu by SAIL.Stanford.EDU with TCP; 1 Apr 90 15:06:56 PDT
Received: by portia.stanford.edu (5.59/25-eef) id AA01706; Sun, 1 Apr 90 15:07:39 PDT
Date: Sun, 1 Apr 90 15:07:39 PDT
From: Dwight Joe <underdog@portia.stanford.edu>
Message-Id: <9004012207.AA01706@portia.stanford.edu>
To: JMC@SAIL.Stanford.EDU, underdog@PORTIA.STANFORD.EDU
Subject: re: gopher
Okay.
∂01-Apr-90 1831 Mailer failed mail returned
The following message was undeliverable to recipient(s):
marsland%alberta.bitnet@FORSYTHE.STANFORD.EDU
Here is how the remote host replied to this mail address:
marsland%alberta.bitnet@FORSYTHE.STANFORD.EDU
550 Unknown Bitnet system: alberta.bitnet
------- Begin undelivered message: -------
01-Apr-90 1831 JMC Please surface mail
To: marsland%alberta.bitnet@FORSYTHE.STANFORD.EDU
me a copy of the final version of my paper.
Also I would like a bibiographical reference to
the volume in which it will appear. Also please
acknowledge this message, since I'm guessing
on the address.
------- End undelivered message -------
∂01-Apr-90 2143 hoffman@csli.Stanford.EDU a family emergency
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 Apr 90 21:43:12 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA16794; Sun, 1 Apr 90 21:45:16 PDT
Date: Sun 1 Apr 90 21:45:16-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: a family emergency
To: jmc@sail.stanford.edu
Message-Id: <639035116.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
A family emergency called me away this week: when would be good to find you
this week?
reid
-------
∂01-Apr-90 2155 vera@fanaraaken.STANFORD.EDU Re: Japanese Space Effort
Received: from fanaraaken.STANFORD.EDU by SAIL.Stanford.EDU with TCP; 1 Apr 90 21:55:26 PDT
Received: from localhost by fanaraaken.STANFORD.EDU (4.0/4.7); Sun, 1 Apr 90 21:55:01 PDT
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: Japanese Space Effort
In-Reply-To: Your message of 01 Apr 90 18:07:00 -0700.
<TB#kD@SAIL.Stanford.EDU>
Date: Sun, 01 Apr 90 21:54:58 -0700
From: James S. Vera <vera@fanaraaken.STANFORD.EDU>
thanks for the info
James Vera
∂02-Apr-90 0024 ME alberta.bitnet
∂01-Apr-90 1833 JMC Any suggestion?
marsland%alberta.bitnet@FORSYTHE.STANFORD.EDU
550 Unknown Bitnet system: alberta.bitnet
ME - Here are a couple of names (the first column) of Bitnet nodes
from our Bitnet list (dated '87). Substitute the first column
name for "alberta" in your address. (Do you have a message from
this person that you are replying to?)
UALTAMTS U Alberta Comp Svcs MTS MTS
UALTAVM U Alberta Comp Svcs VM VM/SP
∂02-Apr-90 1155 pjd@riacs.edu Letters
Received: from icarus.riacs.edu by SAIL.Stanford.EDU with TCP; 2 Apr 90 11:55:26 PDT
Received: from hydra.riacs.edu by icarus.riacs.edu (5.59/2.1G)
id AA15793; Mon, 2 Apr 90 11:54:44 PDT
Received: by hydra.riacs.edu (4.12/2.0N)
id AA23822; Mon, 2 Apr 90 11:57:05 pdt
Message-Id: <9004021857.AA23822@hydra.riacs.edu>
Date: Mon, 2 Apr 90 11:57:05 pdt
From: Peter J. Denning <pjd@riacs.edu>
To: JMC@SAIL.Stanford.EDU
Subject: Letters
John, I received your letters. Thanks.
Peter
∂02-Apr-90 1253 shoham@Hudson.Stanford.EDU meeting
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 Apr 90 12:53:21 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA14943; Mon, 2 Apr 90 12:52:58 -0700
Date: Mon, 2 Apr 1990 12:52:57 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: meeting
Message-Id: <CMM.0.88.639085977.shoham@Hudson.Stanford.EDU>
John,
Are you free tomorrow, Tuesday, between facultylunch and faculty meeting,
i.e. at ~ 1:15?
Yoav
∂02-Apr-90 1301 @Score.Stanford.EDU,@ai.ai.mit.edu,@mc.lcs.mit.edu,@central.cis.upenn.edu:dale@linc.cis.upenn.edu ICLP90 conference information and registration
Received: from Score.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 Apr 90 13:00:53 PDT
Received: from lcs.mit.edu (MINTAKA.LCS.MIT.EDU.#Internet) by SCORE.STANFORD.EDU with TCP; Mon 2 Apr 90 13:00:28-PDT
Received: from ai.ai.mit.edu by mintaka.lcs.mit.edu id ac19684;
2 Apr 90 15:43 EDT
Received: from MC.LCS.MIT.EDU (CHAOS 3131) by AI.AI.MIT.EDU; 2 Apr 90 15:22:24 EDT
Received: from lcs.mit.edu (CHAOS 15044) by MC.LCS.MIT.EDU; 2 Apr 90 13:10:42 EDT
Received: from CENTRAL.CIS.UPENN.EDU by mintaka.lcs.mit.edu id aa12624;
2 Apr 90 13:09 EDT
Received: from LINC.CIS.UPENN.EDU by central.cis.upenn.edu
id AA20536; Mon, 2 Apr 90 13:08:54 -0400
Return-Path: <dale@linc.cis.upenn.edu>
Received: by linc.cis.upenn.edu
id AA04330; Mon, 2 Apr 90 13:08:36 EDT
Date: Mon, 2 Apr 90 13:08:36 EDT
From: Dale Miller <dale@linc.cis.upenn.edu>
Posted-Date: Mon, 2 Apr 90 13:08:36 EDT
Message-Id: <9004021708.AA04330@linc.cis.upenn.edu>
To: clp@cs.cmu.edu, theorem-provers@mc.lcs.mit.edu
Subject: ICLP90 conference information and registration
The Association for Logic Programming is sponsoring the Seventh
International Conference on Logic Programming, to be held in
the Hyatt-Regency Hotel in Jerusalem, on June 18-20, 1990.
Preconference workshops will be held in Neptune Hotel in Eilat on June
14-15.
This message contains the conference programme, information on
preconference workshops and registration forms.
IMPORTANT: Book your flights to Israel as soon as possible.
Further information and a conference brochure can be obtained from
ORTRA Ltd., PO Box 50432, Tel Aviv 61500, Israel
phone: +972-3-664825, Fax +972-3-660952, Telex 361142
===============================================================================
ICLP90 CONFERENCE PROGRAMME
AT A GLANCE
JUNE 18 JUNE 19 JUNE 20
MONDAY TUESDAY WEDNESDAY
08.30 Welcome Address Tutorial3:|Session3B: Tutorial5:|Session6B:
to Invited Lecture: Logic | Language Parallel | Program
10.00 Ray Reiter Databases| Issues Systems | Termination
10.30 Tutorial1:|Session1B: Session3A:|Tutorial4: Session6A:|Tutorial6:
to Concurr. | Misc. Logic | Higher IndepAnd-| Types
12.00 Constr. | Databases| Order Logic Parallel |
13.30 Invited Lecture: Invited Lecture: Invited Lecture:
to Alain Colmerauer Moshe Vardi David Harel
14.30
15.00 Session1A:|Tutorial2: Session4A:|Session4B: Session7A:|Session7B:
to Concurr. | Semantics Implemen-| HO Logic, And- | Program
16.30 Langs. | tation| Abduction Parallel | Synthesis
17.00 Session2A:|Session2B: Session5A:|Session5B: Session8A:|Session8B:
to Architec-| Semantics CLP,etc | Negation Parallel | Program
18.30 tures| ---(to 18.00)--- Systems | Analysis
General Assembly
Evening Reception at Conference Dinner Farewell reception
David's Tower
____________________________________________________________________________
MONDAY, JUNE 18
08.30-10.00
WELCOME ADDRESS
INVITED LECTURE - What Should a Database Know?
Ray Reiter [University of Toronto, Canada]
10.30-12.00
TUTORIAL1 - Concurrent Constraint Programming Languages
Vijay Saraswat [Xerox, USA]
SESSION 1B - GRAMMARS, PROGRAM MAPPINGS, PROOF CONSTRUCTION
Extending Definite Clause Grammars with Scoping Constructs
Remo Pareschi, Dale Miller [West Germany, USA]
A Notion of Map between Logic Programs
A J Power, Leon Sterling [USA]
A Logic Programming Approach to Specifying Logics and Constructing Proofs
H Sawamura, T Minami, K Yokota, K Ohashi [Japan]
13.30-14.30
INVITED LECTURE - Prolog-III as it actually is
Alain Colmerauer [University of Marseille, France]
15.00-16.30
SESSION 1A - CONCURRENT LOGIC LANGUAGES
A New Implementation Technique for Flat GHC
Kazunori Ueda, Masao Morita [Japan]
A Detection Algorithm of Perpetual Suspension in KL1
Yu Inamura, Satoshi Onishi [Japan]
Kernel Andorra Prolog and its Computation Model
Seif Haridi, Sverker Janson [Sweden]
TUTORIAL2 - Semantics
Haim Gaifman [Hebrew University, Israel]
17.00-18.30
SESSION 2A - ARCHITECTURES
Sequential Architecture Models for Prolog: A Performance Comparison
M Korsloot, J M Mulder [Netherlands]
An Extended RISC Methodology and its Application to FCP
Arie Harsat, Ran Ginosar [Israel]
Evaluation of MRB Garbage Collection on Parallel Logic Programming
Architectures
Kenji Nishida, Yasunori Kimura, Akira Matsumoto, Atsuhiro Goto [Japan]
SESSION 2B - SEMANTICS
Generalized Stable Models, Truth Maintenance and Conflict Resolution
Laura Giordano, Alberto Martelli [Italy]
A New Fixpoint Semantics for General Logic Programs Compared with the
Well-Founded and the Stable Model Semantics
Fran\c{c}ois Fages [France]
Three-Valued Stable Semantics For Normal and Disjunctive Logic Programs
Teodor Przymusinski [USA]
____________________________________________________________________________
TUESDAY, JUNE 19
08.30-10.00
TUTORIAL3 - Logic Databases
Catriel Beeri [Hebrew University, Israel]
SESSION 3B - LANGUAGE ISSUES
Inheritance in Logic Programming
Lu\'{\i}s Monteiro, Ant\'{o}nio Porto [Portugal]
>From Objects to Logic
Jean-Marc Andreoli, Remo Pareschi [West Germany]
Representing Objects in a Logic Programming Language with Scoping Constructs
Joshua S Hodas, Dale Miller [USA]
10.30-12.00
SESSION 3A - LOGIC DATABASES
Incremental Re-evaluation of LDL Queries
Oded Shmueli, Shalom Tsur [USA]
Logical Diagnosis of LDL Programs
Oded Shmueli, Shalom Tsur [USA]
Top-Down Integrity Constraint Checking for Deductive Databases
Ulrike Griefahn, Stefan L\"{u}ttringhaus [West Germany]
TUTORIAL4 - Higher Order Logic Programming
Dale Miller [University of Pennsylvania, USA]
13.30-14.30
INVITED LECTURE - Global Optimization Problems for Database Logic Programs
Moshe Vardi [IBM, USA]
15.00-16.30
SESSION 4A - IMPLEMENTATION
An Algorithm for Optimal Back-Striding in Prolog
V M Malhotra [Australia]
A Matching Tree Oriented Abstract Machine for Prolog
Neng-Fa Zhou, Toshihisa Takagi, Kazuo Ushijima [Japan]
LIPS on a MIPS: Results from a Prolog Compiler for a RISC
Andrew Taylor [Australia]
SESSION 4B - HIGHER ORDER LOGIC, ABDUCTION
HIFUNLOG: Logic Programming with Higher-order Relational Functions
Yeh-Heng Sheng [USA]
On Warren's method for Functional Programming in Logic
M H M Cheng, M H van Emden, B E Richards [Canada]
Intensional Updates: Abduction via Deduction
Francois Bry [West Germany]
17.00-18.00
SESSION 5A - CONSTRAINTS, ATTRIBUTE GRAMMARS
Incremental Constraint Satisfaction in Logic Programming
Pascal van Hentenryck [West Germany]
A Logic-Based Modification of Attribute Grammars for Practical Compiler Writing
Jukka Paakki [Finland]
SESSION 5B - NEGATION
Logic Programs with Classical Negation
Michael Gelfond, Vladimir Lifschitz [USA]
Logic Programs with Exceptions
Robert A Kowalski, Fariba Sadri [UK]
____________________________________________________________________________
WEDNESDAY, JUNE 20
08.30-10.00
TUTORIAL5 - Parallel Logic Programming Systems
Jacques Chassin [CAP-SESA-Innovation, France]
SESSION 6B - PROGRAM TERMINATION
Acyclic Programs (extended abstract)
Krzysztof Apt, Marc Bezem [Netherlands]
Termination Proofs for Logic Programs based on Predicate Inequalities
Lutz Pl\"{u}mer [West Germany]
A Practical Technique for Detecting Non-terminating Queries for a
Restricted Class of Horn Clauses, using Directed, Weighted Graphs
Danny De Schreye, Kristof Verschaetse, Maurice Bruynooghe [Belgium]
10.30-12.00
SESSION 6A - INDEPENDENT AND-PARALLELISM
The CDG, UDG and MEL Methods for Automatic Compile-time Parallelization of
Logic Programs for Independent And-parallelism
K Muthukumar, M Hermenegildo [USA]
Non-Strict Independent And-Parallelism
Manuel Hermenegildo, Francesca Rossi [USA]
\&-Prolog and its Performance: Exploiting Independent And-Parallelism
M V Hermenegildo, K J Greene [USA]
TUTORIAL6 - Types in Logic Programming
Frank Pfenning [Carnegie-Mellon University, USA]
13.30-14.30
INVITED LECTURE - How Hard is it to Reason About Propositional Programs?
David Harel [Weizmann Institute, Israel]
15.00-16.30
SESSION 7A - AND-PARALLELISM
Architected Failure Handling for AND-Parallel Logic Programs
David M Meyer, John S Conery [USA]
Pipeline Optimizations in AND-Parallelism by Abstract Interpretation
Roberto Giacobazzi, Laura Ricci [Italy]
AND Parallelism without Shared Variables
Antonio Brogi [Italy]
SESSION 7B - PROGRAM SYNTHESIS AND OPTIMISATION
Top-Down Synthesis of Recursive Logic Procedures from First-order Logic
Specifications
K.K. Lau, S.D. Prestwich [UK, West Germany]
Extracting Logic Programs from Proofs that use Extended Prolog Execution
and Induction
Laurent Fribourg [France]
Generation and Compilation of Efficient Computation Rules
Kristof Verschaetse, Danny De Schreye, Maurice Bruynooghe [Belgium]
17.00-18.30
SESSION 8A - PARALLEL SYSTEMS AND ALGORITHMS
Logic and Functional Programming on Distributed Memory Architectures
P G Bosco, C Cecchi, C Moiso, M Porta, G Sofi [Italy]
Or-Parallel Prolog and Search Problems in Artificial Intelligence
Applications
T J Reynolds, P Kefalas [UK]
Efficient Parallel Term Matching and Anti-Unification
Arthur L Delcher, Simon Kasif [USA]
SESSION 8B - PROGRAM ANALYSIS AND TRANSFORMATION
Multiple Specialization of Logic Programs with Run-time Tests
Dean Jacobs, Anno Langen, Will Winsborough [USA]
The Derivation of an Algorithm for Program Specialisation
John Gallagher, Maurice Bruynooghe [Belgium]
Analysis of Shared Data Structures for Compile-Time Garbage Collection in
Logic Programs
Anne Mulkers, William Winsborough, Maurice Bruynooghe [Belgium, USA]
____________________________________________________________________________
Programme Chairmen Conference Chairman
David H.D. Warren and Peter Szeredi Ehud Shapiro
Department of Computer Science Department of Applied Mathematics
University of Bristol and Computer Science
Bristol BS8 1TR The Weizmann Institute of Science
United Kingdom Rehovot 76100, Israel
warren@compsci.bristol.ac.uk udi@wisdom.weizmann.ac.il
Programme Committee
Maurice Bruynooghe, Belgium Takashi Chikayama, Japan
Seif Haridi, Sweden Manuel Hermenegildo, USA
Chris Hogger, UK Feliks Kluzniak, Poland
Jean-Louis Lassez, USA John Lloyd, UK
Ewing Lusk, USA Maurizio Martelli, Italy
Chris Mellish, UK Shamim Naqvi, USA
Vijay Saraswat, USA Kazunori Ueda, Japan
Jeff Ullman, USA David S. Warren, USA
Local arrangements and exhibits by ORTRA Ltd., Israel.
phone: +972-3-664825, Fax +972-3-660952, Telex 361142
=======================================================================
PRECONFERENCE WORKSHOPS
Several informal one-day workshops will be held on June 14-15
in Neptune Hotel, Eilat.
Registration to the workshop events are according to the enclsoed
forms. Acceptance to each individual workshop requires, in addition
to registering, an approval by the workshop organizer(s).
AT A GLANCE: Dates, Titles, and Organizers
Wednesday Evening, June 13: Arrival and reception
Thursday, June 14
Workshop 1: Semantics of Concurrent Logic Languages
Moreno Falaschi and Michael Maher
Workshop 2: Abstract Interpretation and Program Analysis
Nevin Heintze and Kim Marriott
Workshop 3: Logic Programming Environments
Jacob Levy and Anthony Kusalik
Friday, June 15
Workshop 4: Parallel Logic Programming
Seif Haridi
Workshop 5: Partial Deduction and Partial Evaluation
Jan Komorowski
Workshop 6: Structuring Disciplines for Logic Programming
Dale Miller
Friday Evening: Dinner Cruise at the Gulf of Eilat
Satuday, June 16: Tour in the vicinity of Eilat
Sunday, June 17: Tour to Jerusalem via the Negev, Dead Sea, and
Massada.
Please contact the workshop organizers for detailed Call for
Participation, as well as for permission to participate in each
individual workshop.
Workshops coordinator: Kenneth Kahn, Xerox PARC, USA, kahn.pa@xerox.com
Contacts:
Workshop 1:
Moreno Falaschi
Department of Applied Mathematics and Computer Science
The Weizmann Institute of Science
Rehovot 76100
Israel
moreno@wisdom.bitnet or moreno@wisdom.weizmann.ac.il
(+972)-8-342186
Fax: +972-8-466966
Telex: 361900 (please specify: c/o dept. appl. math.)
Workshop 2:
Kim Marriott,
IBM T-J Watson Research Center,
PO Box 704,
Yorktown Heights, NY 10598,
USA
kimbal@ibm.com
(914)784-7133
Workshop 3:
Jacob Levy Anthony Kusalik
Dept. of Electrical Engineering Dept. of Computational Science
Technion Univ. of Saskatchewan
Haifa Saskatoon, Saskatchewan
32-000 Israel S7N 0W0 Canada
jlevy@ee.technion.ac.il kusalik@sklpl.usask.ca
(+972) 4 294657 (306)966-4904
Workshop 4:
Seif Haridi
Swedish Institute of Computer Science
Box 1263, S-164 28 Kista, Sweden
seif@sics.se
Workshop 5:
Jan Komorowski
Department of Computer Science
Abo Akademi University
SF-20520 Abo Finland
JKomorowski@finabo.abo.fi or JKomorowski@finabo.bitnet
(+358)-21-654679 (Time zone: -2 GMT)
Fax: (+358)-21-654732
Workshp 6:
Dale Miller
Computer and Information Science Dept
University of Pennsylvania
Philadelphia, PA 19104-6389 USA
dale@cis.upenn.edu
(215)-898-1593
Fax: (215)-898-0587
========================================================================
SEVENTH INTERNATIONAL CONFERENCE ON LOGIC PROGRAMMING
REGISTRATION FORM
(To be returned to ORTRA, P.O. Box 50432, Tel-Aviv 61500, Israel)
Name: _________________________ First Name: _________________________
Affiliation: ___________________________________________________________
Address: _______________________________________________________________
Country: _______________ Tel: _______________ Fax: _______________
( ) I am a member of ALP - Membership No. ______________________________
( ) Attached is:
( ) Check # _______________ made out to Ortra Ldt.
( ) Copy of Bank Transfer to account #193-47233 (Ortra Ldt.)
Bank Hapoalim, Branch #554 (Namir Square)
in the amount of $ _____________________ to cover the following:
Eilat Workshop: 13-17 June 1990
-------------------------------
( ) Participation Fee, Member of ALP $ 95/$ 85* __________
( ) Participation Fee, Non-Member $110/$100* __________
( ) Participation Fee, Student** $ 35/$ 30* __________
( ) Optional Dinner and Cruise per $40 ___ ticket(s) __________
( ) Optional Eilat Tour per $28 ___ ticket(s) __________
Optional Tour to Jerusalem per $50 ___ ticket(s) __________
Conference: 17-21 June 1990
---------------------------
( ) Participation Fee, Member of ALP $280/$255* __________
( ) Participation Fee, Non-Member $300/$275* __________
( ) Participation Fee, Student* $150/$125* __________
( ) Participation Fee, Accompanying person $150/$125* __________
( ) Optional Conference Dinner per $35 ___ ticket(s) __________
Total: $ __________
* For payments reaching ORTRA before May 10, 1990
** Attached is letter from University
Sinagure: _______________________________ Date: ____________________
========================================================================
SEVENTH INTERNATIONAL CONFERENCE ON LOGIC PROGRAMMING
TOURIST SERVICES REQUEST
(To be returned to ORTRA, P.O. Box 50432, Tel-Aviv 61500, Israel)
Name: _________________________ First Name: _________________________
Affiliation: ___________________________________________________________
Address: _______________________________________________________________
Country: _______________ Tel: _______________ Fax: _______________
Eilat Workshop (13-17 June 1990)
--------------------------------
( ) Flight to Eilat from Tel-Aviv (per $64) ___ ticket(s)
Date: ________________ / Preferred departure time: _____________
( ) Flight to Jerusalem from Eilat (per $64) ___ ticket(s)
Date: ________________ / Preferred departure time: _____________
Accommodation at the Neptune Hotel:
From _______________ to _______________
( ) Room for two ($48 per person/per night): _____ persons _____ nights
( ) I will share a room with ____________________ who will pay by
her/himself. Attached is her/his Tourist Service Request Form.
( ) Room for one ($81 per night) _____ nights
Conference (17-20 June 1990)
----------------------------
( ) Accommodation from _______________ to _______________ at:
( ) Hyatt ( ) Caesar
( ) Room for two ($ 58*) ($33*) * per person/per night
( ) Room for one ($102*) ($53*)
( ) Transportation from the airport to Jerusalem
I/we shall arrive by flight # ________________ from ________________
on ________________
Jerusalem Conference Package and Galillee Tour
----------------------------------------------
( ) With room for two ($535 per person)**
( ) With room for one ($815)
** I shall share a room with ____________________ who will pay by
her/himself. Attached is her/his Tourist Services Request Form.
( ) I/we require the following additional services:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
Please send information and quote prices.
( ) Attached is a $100 deposit per person
( ) Check # _______________ made out to Ortra Ldt.
( ) Copy of Bank Transfer to account #193-47233 (Ortra Ldt.)
Bank Hapoalim, Branch #554 (Namir Square)
Remarks: _______________________________________________________________
Sinagure: _______________________________ Date: ____________________
========================================================================
∂02-Apr-90 1352 AI.LENAT@MCC.COM
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 2 Apr 90 13:52:38 PDT
Date: Mon 2 Apr 90 15:52:48-CDT
From: Doug Lenat <AI.LENAT@MCC.COM>
To: JMC@SAIL.Stanford.EDU
cc: AI.LENAT@MCC.COM
In-Reply-To: <1DCr9q@SAIL.Stanford.EDU>
Message-ID: <12578683616.59.AI.LENAT@MCC.COM>
I'm innocent!
-------
∂02-Apr-90 1755 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 Apr 90 17:55:17 PDT
Received: from Neon.Stanford.EDU by Sunburn.Stanford.EDU with SMTP (5.61+IDA/25-eef) id AA10170; Mon, 2 Apr 90 17:56:04 -0700
Date: Mon, 2 Apr 90 17:56:04 -0700
Message-Id: <9004030056.AA10170@Sunburn.Stanford.EDU>
From: The Bill Program <pony-bills@Neon.Stanford.EDU>
Subject: JMC Prancing Pony Bill
Reply-To: pony-bills@Neon.Stanford.EDU
To: John McCarthy <JMC@SAIL.STANFORD.EDU>
Prancing Pony Bill of John McCarthy (JMC) for March 1990 (4/2/1990)
Previous Balance 2.13
Monthly Interest at 1.00% 0.02
Current Charges NONE
---------
TOTAL AMOUNT DUE 2.15
PAYMENT DELIVERY LOCATION: CSD Receptionist.
Make checks payable to: STANFORD UNIVERSITY.
Please deliver payments to the Computer Science Dept receptionist, Bldg. 460.
To ensure proper crediting, please include your ACCOUNT NAME on your
check. If you pay by cash, use the small yellow envelopes provided
and write both your ACCOUNT NAME and the AMOUNT on outside.
Note: The recording of a payment may take up to three weeks after the payment
is made, but never beyond the next billing date. Please allow for this delay.
Bills are payable upon presentation. Interest of 1.00% per month will be
charged on balances remaining unpaid 25 days after bill date above.
An account with a credit balance earns interest of 0.33% per month,
based on the average daily balance.
You haven't paid your Pony bill since 10/1989.
Accounts with balances remaining unpaid for more than 55 days are
considered delinquent and are subject to reduction of credit limit.
Please pay your bill and keep your account current.
∂02-Apr-90 1857 @Score.Stanford.EDU,@ai.ai.mit.edu,@mc.lcs.mit.edu,@relay.cdnnet.ca:joyce@cs.ubc.ca Cambridge HOL System Course
Received: from Score.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 Apr 90 18:57:05 PDT
Received: from lcs.mit.edu (MINTAKA.LCS.MIT.EDU.#Internet) by SCORE.STANFORD.EDU with TCP; Mon 2 Apr 90 18:56:38-PDT
Received: from ai.ai.mit.edu by mintaka.lcs.mit.edu id aa07948;
2 Apr 90 21:49 EDT
Received: from MC.LCS.MIT.EDU (CHAOS 3131) by AI.AI.MIT.EDU; 2 Apr 90 21:16:41 EDT
Received: from lcs.mit.edu (CHAOS 15044) by MC.LCS.MIT.EDU; 2 Apr 90 21:16:17 EDT
Received: from relay.cdnnet.ca by mintaka.lcs.mit.edu id aa06459;
2 Apr 90 21:14 EDT
Received: by relay.CDNnet.CA (4.1/1.14)
id AA17715; Mon, 2 Apr 90 17:14:47 PDT
Date: 2 Apr 90 17:16 -0700
From: Jeffrey Joyce <joyce@cs.ubc.ca>
To: theorem-provers@mc.lcs.mit.edu
MMDF-Warning: Parse error in original version of preceding line at lcs.mit.edu
Message-Id: <310*joyce@cs.ubc.ca>
Subject: Cambridge HOL System Course
The Cambridge HOL System:
========================
An Introduction to Interactive Machine-Assisted
Theorem-Proving in Higher-Order Logic
Monday-Friday, June 4-8, 1990
FIVE-DAY INTENSIVE COURSE
Seminars and Lab Workshops to be held on
The University of British Columbia Campus
(Vancouver Canada)
Sponsored by: The UBC Centre for Continuing Education and
The UBC Centre for Integrated Computer Systems Research
Course Description:
-------------------
This five-day course is an introduction to higher-order logic and use of
the Cambridge HOL System. The HOL system is an interactive environment
for machine-assisted theorem-proving in higher-order logic. This system
is currently in use at sites in North America, Europe, China and Japan.
It is used for hardware verification, software verification and basic
research into machine-assisted formal proof. Users include both companies
doing contract proofs and researchers.
This course will be of particular interest to contractors developing
hardware/software systems to meet increasingly tough standards such as
the UK MoD Interim Defence Standard 00-55 for the development of safety-
critical software and the US DoD Trusted Computer Systems A1 security
classification (both requiring the use of formal verification techniques).
Morning lectures will describe the HOL formulation of higher-order logic
along with some examples of using higher-order logic to specify digital
hardware.
Afternoon laboratory sessions will provide instruction on using the HOL
system to create formal specifications and generate formal proofs.
Course participants will have workstation access to the HOL system for
course exercises.
For more details on using the HOL system to verify digital hardware, see
"The Notion of Proof in Hardware Verification" (by Avra Cohn, Journal of
Automated Reasoning, Vol. 5, May 1989, pp. 127-139) which gives an account
of using the HOL system to partially verify the commercially-available
VIPER microprocessor.
Course Instructors:
-------------------
Rachel Cardell-Oliver, Ph.D. Candidate, Cambridge University/Australian
Defence Science and Technology Organization, (Protocol Specification
and Verification).
John Herbert, Research Associate, Cambridge University/SRI Cambridge,
(Specification and Verification of digital hardware).
Jeff Joyce, Assistant Professor, University of British Columbia,
(Specification and Verification of microprocessor-based systems).
Who will benefit:
-----------------
This course will benefit researchers in both industrial and academic
settings with an interest in the development of reliable hardware/software
systems. A background in formal logic is NOT required, however, some
familiarity with the notation of predicate calculus would be helpful.
Experience with the use of an interactive functional programming language
(such as Lisp) would be very helpful.
Other Information:
------------------
The Cambridge HOL system is available as a public domain system. Copies
of the system will be available at the course (at cost for magnetic tape).
Pre-ordered copies of the three-volume HOL System Manual will also be
available (at cost for copying).
For further course information contact:
Jeff Joyce
Computer Science Department Telephone: (604) 228-4327
The University of British Columbia Fax: (604) 228-5485
6356 Agricultural Road Email: joyce@cs.ubc.ca
Vancouver, B.C. V6T 1W5
Canada
For information regarding registration contact:
Vicki Ayerbe
Computer Science Programs Telephone: (604) 222-5251
Centre for Continuing Education Fax: (604) 222-5283
5997 Iona Drive
Vancouver, B.C. V6T 2A4
Canada
Registration Information:
-------------------------
Course Location: University of British Columbia Campus,
Vancouver, B.C., Canada
Pre-registration deadline: May 7, 1990
Enrollment limited.
Before May 7 After May 7
------------ -----------
Registration Fees: Regular $295.00 $345.00
Student $150.00 $175.00
Also available by order: Software (magnetic tape): $25.00*
Three Volume Manual: $110.00*
* - public domain (copying costs only)
Ways to Register !
Mail your cheque (payable to UBC) along with yhour registration form
(attached below).
In Canada only: Use your VISA or MASTERCARD and register by telephone,
(604) 222-5251, or by FAX at (604) 222-5283, 10am - 4pm, Monday-Friday.
Cancellation Policy: Notice of participant cancellation must be in
writing and must be received by the Centre for Continuing Education
two weeks prior to the course starting date to qualify for a refund
(a full refund less $50.00 cancellation fee). No refund will be
made for cancellation received after that time. The course may
be cancelled by May 14 due to insufficient registrations.
-------------------- CUT HERE --------------------------------------
Accommodation Information and Registration Form
-----------------------------------------------
Walter Gage Residence
Single bedrooms are arranged apartment style (each apartment contains
six private bedrooms with a single bed) with one large shared washroom,
a lounge area with refrigerator and a balcony.
Accommodation payment is requested upon check-in by cash, traveller's
cheque, VISA or MASTERCARD.
NAME: ____________________________________________________
ADDRESS: ____________________________________________________
____________________________________________________
TELEPHONE: ____________________________________________________
ARRIVAL DATE: __________________ DEPARTURE DATE: _________________
Check-in time: 1400 hours (2 pm) Check-out time: 1100 hours (11 am)
Accommodation:
Single Bedroom (sharing washroom and lounge) $28.00/night
A deposit is not required.
A provincial hotel tax, currently 8%, is added to the above rate.
Send this reservation request form to:
UBC Conference Centre
5961 Student Union Mall
Vancouver, B.C., V6T 2C9 Canada
Telephone: (604) 228-2963
Fax: (604) 228-5297
G00602A
-------------------- CUT HERE --------------------------------------
Course Registration Form
------------------------
The Cambridge HOL System CS 5012-290
____________________________________________________________________
LAST NAME FIRST NAME
____________________________________________________________________
ADDRESS CITY/TOWN
Registration Fee: $______________
Software (optional): $______________
Manual (optional): $______________
TOTAL: $______________
Method of Payment
Cheque/Money Order (Payable to UBC)
Cash (For Registration in person)
VISA Card __________________________
MASTERCARD _________________________
Valid Date: ______________ Expiry Date: ______________
(if MASTERCARD)
I authorize UBC to charge the above amount to my credit card
Signature _____________________________________________________
Name of person paying for program (s) (if different than registrant)
Mail completed registration form to:
REGISTRATIONS,
CENTRE FOR CONTINUING EDUCATION,
THE UNIVERISTY OF BRITISH COLUMBIA
5997 IONA DRIVE,
VANCOUVER, B.C., V6T 2A4
CANADA
∂02-Apr-90 2000 JMC
Korf for Dan and Emacs
∂02-Apr-90 2000 JMC
Rubik
∂02-Apr-90 2151 @IGNORANT.Stanford.EDU:RDZ@SCORE.Stanford.EDU Question
Received: from IGNORANT.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 Apr 90 21:51:35 PDT
Received: from PRECARIOUS.Stanford.EDU by IGNORANT.Stanford.EDU via CHAOS with CHAOS-MAIL id 2905; Mon 2-Apr-90 22:03:10 PDT
Date: Mon, 2 Apr 90 22:02 PDT
From: Ramin Zabih <RDZ@SCORE.Stanford.EDU>
Subject: Question
To: jmc@SAIL.Stanford.EDU
Message-ID: <19900403050246.1.RDZ@PRECARIOUS.Stanford.EDU>
My notes from our last conversation claim that the new better is: if
the next tile is nearer where it eventually wants to be, OR if the blank
is nearer to the next tile. Is this correct?
∂03-Apr-90 1327 searle@cogsci.berkeley.edu Re: Andreas Dorschel adresss
Received: from cogsci.berkeley.edu by SAIL.Stanford.EDU with TCP; 3 Apr 90 13:27:39 PDT
Received: by cogsci.berkeley.edu (5.61/1.29)
id AA06673; Tue, 3 Apr 90 13:28:31 -0700
Date: Tue, 3 Apr 90 13:28:31 -0700
From: searle@cogsci.berkeley.edu (John R. Searle)
Message-Id: <9004032028.AA06673@cogsci.berkeley.edu>
To: JMC@SAIL.Stanford.EDU, searle@cogsci.berkeley.edu
Subject: Re: Andreas Dorschel adresss
Sorry , I dont know his address.
best
john
∂03-Apr-90 1515 ingrid@russell.Stanford.EDU your presentation at the IAP meeting next week
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 3 Apr 90 15:15:13 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA13447; Tue, 3 Apr 90 15:17:44 PDT
Message-Id: <9004032217.AA13447@russell.Stanford.EDU>
To: jmc@sail.stanford.edu, kay.pa@xerox.com, etch@russell.Stanford.EDU,
winograd@russell.Stanford.EDU
Subject: your presentation at the IAP meeting next week
Date: Tue, 03 Apr 90 15:17:42 PDT
From: ingrid@russell.Stanford.EDU
Please remember to let me have your title and handouts as soon as
possible. The deadline was yesterday.
Ingrid
∂03-Apr-90 1827 CN.MCS@Forsythe.Stanford.EDU
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 3 Apr 90 18:27:49 PDT
Date: Tue, 3 Apr 90 18:20:11 PDT
To: jmc@sail
From: "REBECCA LASHER" <CN.MCS@Forsythe.Stanford.EDU>
John Sack and John McCarthy,
I have the meeting arranged for tomorrow for 5 p.m. in the Math/CS
Library. Bill Arms from CMU will arrive at 1 p.m. so he should get
here on time.
I have given each of you a draft statement of the meeting in
Washington D.C. last week. Briefly, this is a project that five
institutions plan to submit to DARPA to obtain funding. There are
basically two parts. The first is to make bibliographic data
for technical reports available in some standard format and to share
this data. The second part of the project is to store images of the
technical reports in a database which could be accessed by the
participating institutions. The plan is for each institution to
store only images from the reports it produces. Participating
institutions would be able to obtain images from each other using
NSFnet.
There are options with the project and I am hoping that the four of
us can meet and get some sense of the scope and feasibility here at
Stanford.
Rebecca Lasher
Math/CS Library
3-0864
∂04-Apr-90 0737 chandler@Sunburn.Stanford.EDU Re: committee on computer readable phd theses
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 07:37:00 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA27649; Wed, 4 Apr 90 07:38:51 -0700
Date: Wed, 4 Apr 1990 7:38:50 PDT
From: "Joyce R. Chandler" <chandler@Sunburn.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: committee on computer readable phd theses
In-Reply-To: Your message of 03 Apr 90 1654 PDT
Message-Id: <CMM.0.88.639239930.chandler@Sunburn.Stanford.EDU>
The only committee member you left out was Genesereth.
∂04-Apr-90 1041 mps@Gang-of-Four.Stanford.EDU committee meeting
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 10:41:25 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA24832; Wed, 4 Apr 90 10:42:52 -0700
Date: Wed, 4 Apr 90 10:42:52 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9004041742.AA24832@Gang-of-Four.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 04 Apr 90 1038 PDT <1DD9FD@SAIL.Stanford.EDU>
Subject: committee meeting
what day would you like
∂04-Apr-90 1046 ingrid@russell.Stanford.EDU Re: title and handout
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 10:46:29 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA18933; Wed, 4 Apr 90 10:49:05 PDT
Message-Id: <9004041749.AA18933@russell.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: title and handout
In-Reply-To: Your message of 04 Apr 90 10:40:00 PDT.
<1DD9Hg@SAIL.Stanford.EDU>
Date: Wed, 04 Apr 90 10:49:04 PDT
From: ingrid@russell.Stanford.EDU
Thanks for that. Is this going to be the only handout you will have?
Do you want me to make a transparancy of it or will you bring your
own?
Ingrid
∂04-Apr-90 1058 ingrid@russell.Stanford.EDU Re: title and handout
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 10:58:53 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA19083; Wed, 4 Apr 90 11:01:30 PDT
Message-Id: <9004041801.AA19083@russell.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: title and handout
In-Reply-To: Your message of 04 Apr 90 10:53:00 PDT.
<1XD9U3@SAIL.Stanford.EDU>
Date: Wed, 04 Apr 90 11:01:30 PDT
From: ingrid@russell.Stanford.EDU
Will you let me have a copy of your transparancies? I'm preparing
ring binders for the IAP representatives and would like to include
them in that.
Ingrid
∂04-Apr-90 1110 ingrid@russell.Stanford.EDU Re: title and handout
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 11:10:06 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA19159; Wed, 4 Apr 90 11:12:43 PDT
Message-Id: <9004041812.AA19159@russell.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: title and handout
In-Reply-To: Your message of 04 Apr 90 11:07:00 PDT.
<1rD9ax@SAIL.Stanford.EDU>
Date: Wed, 04 Apr 90 11:12:42 PDT
From: ingrid@russell.Stanford.EDU
It would be great if I could have them Friday, but I understand if
that's not possible. Just let me know as soon as they are ready so I
can pick them up.
Ingrid
∂04-Apr-90 1118 ingrid@russell.Stanford.EDU Re: title and handout
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 11:18:51 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA19368; Wed, 4 Apr 90 11:21:27 PDT
Message-Id: <9004041821.AA19368@russell.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: title and handout
In-Reply-To: Your message of 04 Apr 90 11:15:00 PDT.
<1XD9iF@SAIL.Stanford.EDU>
Date: Wed, 04 Apr 90 11:21:25 PDT
From: ingrid@russell.Stanford.EDU
My office is Ventura 24. Or you can leave them in my mailbox in
Ventura (Deiwiks). If you can't find anyone with a key to Ventura,
leave them on your desk marked for me and I'll come looking there
Monday morning.
Ingrid
∂04-Apr-90 1140 VAL message from Moscow
Vadim Sadovsky, who arrived on Sunday, said that your invitation had been
sent out.
∂04-Apr-90 1206 Mailer re: death penalty
To: JMC@SAIL.Stanford.EDU, su-etc@SAIL.Stanford.EDU
From: Robert W Floyd <RWF@SAIL.Stanford.EDU>
[In reply to message from JMC rcvd 03-Apr-90 17:50-PT.]
*Another possibility is that he could offer some of them to any countries
*whose leaders protest, provided they guarantee they not return to
*California.
Tut, tut, John, that's unconstitutional. States may not make treaties
with foreign powers. Keep that construction strict.
∂04-Apr-90 1217 VAL
Vadim Sadovski: 3-0549 (Ventura Hall), 857-0333 x106 (hotel).
∂04-Apr-90 1442 gray@Neon.Stanford.EDU Re: Cobol question
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 14:42:45 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA05700; Wed, 4 Apr 90 14:43:30 -0700
Date: Wed, 4 Apr 90 14:43:30 -0700
From: Cary G. Gray <gray@Neon.Stanford.EDU>
Message-Id: <9004042143.AA05700@Neon.Stanford.EDU>
To: JMC@SAIL.STANFORD.EDU
Subject: Re: Cobol question
Newsgroups: su.etc
References: <dDp5c@SAIL.Stanford.EDU>
In su.etc you write:
>For an illustration I need the most Englishy Cobol form of the
>pseudo-Algol statements
>bill := 123.45;
>print("Please pay $",bill," by May 10, 1990.");
>Is there anyone around who knows enough Cobol to do that?
>I don't need the decorations required to make a complete
>Cobol program out of that. I'd like it as soon as possible,
>but don't bother after Monday. I'll fake it somehow.
>I promise not to tell anyone you know Cobol.
It's not all that simple, since the COBOL-ish way to do formatting
is to use the data division. So you get something like this (omitting
lots of other decorations you'd need to make it a complete program):
DATA DIVISION.
WORKING-STORAGE SECTION.
01 OUTPUT-LINE.
02 FILLER VALUE 'Please pay ' PICTURE X(11).
02 BILL PICTURE $$$9.99.
02 FILLER VALUE ' by May 10, 1990.' PICTURE X(17).
PROCEDURE DIVISION.
START.
... other processing
MOVE 123.45 TO BILL.
DISPLAY OUTPUT-LINE.
...
I'm not sure that I've used DISPLAY correctly (it may need to be
DISPLAY SOME-OTHER-RECORD FROM OUTPUT-LINE, plus declarations for
SOME-OTHER-RECORD and its associated file/device).
It's been a long time since I last wrote any COBOL; I confess to peeking
at my programming languages text to get the syntax right.
Cary
∂04-Apr-90 1431 jutta@coyote.stanford.edu AI Division Lunch
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 4 Apr 90 14:31:47 PDT
Received: by coyote.stanford.edu; Wed, 4 Apr 90 13:30:51 PST
Date: 4 Apr 1990 1330-PST (Wednesday)
From: Jutta McCormick <jutta@coyote.stanford.edu>
To: latombe@coyote.stanford.edu, binford@coyote.stanford.edu,
eaf@SUMEX-AIM.Stanford.EDU, mrg@sunburn.stanford.edu,
ok@coyote.stanford.edu, jmc@sail.stanford.edu, nilsson@cs.stanford.edu,
shoham@cs.stanford.edu
Cc: jutta@coyote.stanford.edu
Subject: AI Division Lunch
The next AI Division lunch is scheduled for Wednesday, April 11, noon,
at the Faculty Club. Topic: AI in the manufacturing program at
Stanford. Please let me know ASAP whether or not you will be able to
attend.
--Jutta
∂04-Apr-90 1545 weening@Gang-of-Four.Stanford.EDU Cobol question
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Apr 90 15:45:04 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA26416; Wed, 4 Apr 90 15:46:24 -0700
Date: Wed, 4 Apr 90 15:46:24 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9004042246.AA26416@Gang-of-Four.Stanford.EDU>
To: JMC@SAIL.STANFORD.EDU (John McCarthy)
In-Reply-To: JMC@SAIL.STANFORD.EDU's message of 4 Apr 90 18:28:00 GMT
Subject: Cobol question
I used to know COBOL, but that was long ago, and I didn't keep any
manuals. You can probably get more verbosity if there is some
computation, i.e.,
bill := 123.45;
is something like
SET BILL TO 123.45.
while
bill := purchases * (1 + tax_rate) - credits;
might be
ADD 1 TO TAX_RATE AND MULTIPLY BY PURCHASES, SUBTRACTING
CREDITS, GIVING BILL.
or something like that, though that sentence is probably not correct.
Formatted output like
print("Please pay $",bill," by May 10, 1990.");
is often accomplished by a combination of record descriptions (in the
data division of the program) and code (in the execution division), so
I'm not sure it can be expressed just in code. I don't remember the
syntax of this at all, though I recall that for some purposes it was
actually clearer than the equivalent code in Fortran or Algol. (PL/I
tried to combine the best features of both styles, but didn't do it
very well in my opinion.)
∂04-Apr-90 2015 sreerang@portia.stanford.edu term paper (cs323)
Received: from portia.stanford.edu by SAIL.Stanford.EDU with TCP; 4 Apr 90 20:15:29 PDT
Received: by portia.stanford.edu (5.59/25-eef) id AA12286; Wed, 4 Apr 90 20:15:45 PDT
Date: Wed, 4 Apr 90 20:15:45 PDT
From: Sreeranga Rajan <sreerang@portia.stanford.edu>
Message-Id: <9004050315.AA12286@portia.stanford.edu>
To: jmc@sail.stanford.edu
Subject: term paper (cs323)
Cc: sreerang@portia.stanford.edu
I was wondering if you had the time to go through my cs323 term paper
on Language for Design. I would be interested in having your comments
and recommendations to pursue it further after you have read it.
Regards,
-- Sree
∂04-Apr-90 2028 sreerang@portia.stanford.edu re: term paper (cs323)
Received: from portia.stanford.edu by SAIL.Stanford.EDU with TCP; 4 Apr 90 20:27:56 PDT
Received: by portia.stanford.edu (5.59/25-eef) id AA13091; Wed, 4 Apr 90 20:28:24 PDT
Date: Wed, 4 Apr 90 20:28:24 PDT
From: Sreeranga Rajan <sreerang@portia.stanford.edu>
Message-Id: <9004050328.AA13091@portia.stanford.edu>
To: JMC@SAIL.Stanford.EDU
Subject: re: term paper (cs323)
I would like to come over and discuss about that. Will tomorrow be
convenient for you sometime in the afternoon?
--Sree
∂05-Apr-90 0029 ME end of a long system
To: JJW@SAIL.Stanford.EDU, STU@SAIL.Stanford.EDU, TD@SAIL.Stanford.EDU,
JMC@SAIL.Stanford.EDU, CLT@SAIL.Stanford.EDU, ME@SAIL.Stanford.EDU
SAIL died tonight when the KL powered itself off with an Air Flow Cpu
fault at 22:13. In all likelihood, this was the result of a flakey air
flow sensor or possibly dust in the sensor. (The fans seems to be working
and the fault hasn't repeated.)
Anyway, SAIL was up over 1136 hours (47+ days) in the fourth longest
running SAIL system. During that time, it even survived delicate surgery,
as it were, when I replaced the entire magtape driver with a new version
to run the 7-track tape drive from MIT -- replaced with the system running.
Actually, I replaced the driver about four times as I fixed a few minor
bugs. This had never been done before on SAIL (replacing a large chunk of
code in the running system by overwriting it with new code). And the new
driver has been running the tape drive OK.
In case you're interested, here are the four longest SAIL systems and
the causes of their demises (no software crashes in the bunch).
Lifetime Loaded ;days up
1702:07 11/07/86 1754 ;70.9 CRAM parity error
1621:58 10/13/88 1906 ;67.6 Fire alarm tech tripped alarm, cut power
1323:11 02/20/85 1256 ;55.1 MGB cont 1 core stack died
1136:23 02/16/90 1132 ;47.3 Air flow cpu fault, prob. bad sensor
The longest running WAITS system was at LLL, on S1-A -- all winter.
S1-A record uptimes
Lifetime Loaded ;days up
2293:54 12/21/84 0936 ;95.5 crashed 03/27/85 (IMP problem really)
∂05-Apr-90 0932 MPS Files
How do you want your chron files filed?
Working from Jan 88 in front of drawer to Jan 89
or
working from Jan 89 in front of drawer back to Jan 88
∂05-Apr-90 1001 MPS
Pratt and Khatib, both agree to the 9th. I havent been
able to get Mike yet. Do you want to go with just the 2 of them?
∂05-Apr-90 1037 hiller@Hudson.Stanford.EDU AI Day-On-Campus Brochure
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 5 Apr 90 10:36:51 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA00552; Thu, 5 Apr 90 10:35:57 -0700
Date: Thu, 5 Apr 1990 10:35:56 PDT
From: "Bonnie R. Hiller" <hiller@Hudson.Stanford.EDU>
To: AI-DOC:;@Hudson.Stanford.EDU
Cc: wheaton@cs.Stanford.EDU, rlm@Hudson.Stanford.EDU,
prior@Hudson.Stanford.EDU
Subject: AI Day-On-Campus Brochure
Message-Id: <CMM.0.88.639336956.hiller@Hudson.Stanford.EDU>
The brochures for AI Day-On-Campus arrived yesterday. I'm sending 10
copies to each of you today by ID mail. The announcement to our
member companies and the rest of the people on our mailing list is
going out today as well.
If you need more brochures, please send email to John Prior at
prior@hudson or call him at 723-9689.
-Bonnie Hiller
Computer Forum
∂05-Apr-90 1342 MPS Meeting
Hi,
Patty Nogales, grad student, Philosophy, minor in cognitive
science, 325-7964 would like to come in and see you. 1:40pm
She will be at the above number in about an hour.
∂05-Apr-90 1404 MPS
Lunch on Monday the 9th at 12:00 with your committee.
Pat
∂05-Apr-90 1514 rathmann@eclipse.stanford.edu Authoritative partitions
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 5 Apr 90 15:14:26 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA25934; Thu, 5 Apr 90 15:15:59 PDT
Date: Thu, 5 Apr 90 15:15:59 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9004052215.AA25934@eclipse.stanford.edu>
To: JMC@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: Authoritative partitions
At my oral exam, you expressed the need to handle the case where even
if a predicate is not full by my syntactic definition, it may be
"semantically full" meaning that all the sentences which define the
predicate are contained in the partition. Your example dealt with
McCathy's salary being defined in the CS partition, but being used by
the faculty club accounting partiton in order to compute paycheck
deductions.
Verifying such semantic fullness is difficult, but as you suspected,
it can be legislated with a few well placed abs.
The basic idea uses two predicate names, SALARY and SALARY_EXPORT.
Make sure that all mentions of SALARY are contained in the CS
partition, so that SALARY is full by the simple definition. The
faculty club uses SALARY_EXPORT for its calculations. The predicates
are glued together using an abnormality predicate:
\forall x SALARY(x)=SALARY_EXPORT(x) \lor AB(x)
As long as we make sure that AB is minimized with lower priority than
any predicate minimized in the CS partition, nothing said in another
partition about SALARY_EXPORT will affect the preferred models of
SALARY.
Effectively, we have deferred the cost of verifying semantic fullness.
If it becomes necessary to ensure that everyone has a consistent view,
we have to verify not AB(x), which has the usual cost of a
circumscriptive deduction.
I am writing up a theorem to prove that all this works. It is pretty
straightforward. It will also give some useful results about
localizing inconsistency.
-Peter
∂06-Apr-90 0903 JMC
slides for elephant talk
∂06-Apr-90 0927 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Apr 90 09:26:34 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03043; Fri, 6 Apr 90 07:36:00 -0700
Date: Fri, 6 Apr 90 07:36:00 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9004061436.AA03043@Gang-of-Four.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 05 Apr 90 1657 PDT <LDzjD@SAIL.Stanford.EDU>
does this message refer to petrio.1, which is my next message?
∂06-Apr-90 1229 slagle@cs.umn.edu Distinguished Visitor
Received: from cs.umn.edu by SAIL.Stanford.EDU with TCP; 6 Apr 90 12:29:41 PDT
Received: by cs.umn.edu (5.59/1.14)
id AA00586; Fri, 6 Apr 90 14:30:26 CDT
Date: Fri, 6 Apr 90 14:30:26 CDT
From: "James Slagle" <slagle@cs.umn.edu>
Message-Id: <9004061930.AA00586@cs.umn.edu>
To: jmc@sail.stanford.edu
Subject: Distinguished Visitor
Please respond (in any way) to the invitation.
I've updated the invitation. In particular, please tell us your travel
plans as soon as possible. I am very much looking forward to seeing you
again.
---Jim Slagle
---------------------------------------------------------------------------
.mn
.LT
.SZ 12
Professor John McCarthy
Computer Science Department
Stanford University
Stanford, CA 94305-2140
Dear Professor McCarthy:
.pp
This is to confirm our invitation to you to be a Distinguished Visitor and
to give lectures at 2:30-3:30pm on April 18 (Prospects for Artificial
Intelligence: Some Expert Systems Need Common Sense) and 20, (Elephant
2000: A Programming Language with Speech Acts) 1990 and
11:15-12:05 on April 19, (Non-Monotonic Reasoning) 1990 in room 3-230
in the EE/CS Building at the University. Your talks
will be videotaped. After each
lecture, there is a small reception and discussion, which lasts a half
hour. Within one week, please write to verify your acceptance of this
invitation. At the same time, please send us titles and abstracts of your
talks. We shall pay a Distinguished Visitor honorarium of $600/day
for the three days of your visit.
.pp
We shall reimburse all your travel expenses and provide hotel
accommodations and meals. It will be easier for us if you make your own
arrangements, and we shall reimburse you after your visit. We shall
reserve a room for you at the Radisson University Hotel on Washington Ave.
in Minneapolis. If we can be of further assistance in making any other
arrangements, please let us know. When you arrive at the airport, you
should take a taxi to the Radisson University Hotel. Please let us know
in advance of your travel plans including flight numbers and times of
arrival/departure, so that we can make the hotel reservation.
.pp
It is a great pleasure for us that you so kindly accepted our invitation,
and we are looking forward to your visit. Your host will be Dr. James
Slagle ((612) 625-0329).
.in 25
Sincerely,
.sp 4
James Slagle
.br
Professor
∂06-Apr-90 1359 @IBM.COM:LEORA@YKTVMH AAAI Symposium Proposal
Received: from IBM.COM by SAIL.Stanford.EDU with TCP; 6 Apr 90 13:58:55 PDT
Received: from YKTVMH by IBM.COM (IBM VM SMTP R1.2.1MX) with BSMTP id 7301; Fri, 06 Apr 90 14:00:30 PDT
Date: Fri, 6 Apr 90 16:59:01 EDT
From: LEORA@IBM.COM
To: jmc@sail.stanford.edu, val@cs.stanford.edu
Subject: AAAI Symposium Proposal
I've sent off the proposal (enclosed) to Peter Patel-Schneider. This
is essentially the proposal that Vladimir mailed me, with a change of
title and very minor stylistic changes. Hope this is okay.
As I mentioned to John, I'll be away for the next 3 weeks (Israel).
However, I'll be having someone forward my email to me, so I will
be reachable if need be (Also, I'll be checking my answering machine,
if that's possible from Israel - I don't know if it is, because I'm
afraid they are still on pulse)
Leora
=============================================================
Peter,
Enclosed please find our proposal for the 1991 AAAI Spring Symposium Series.
Please let us know if you have any comments. Thank you very much,
John McCarthy
Vladimir Lifschitz
Leora Morgenstern
===========================================================================
Proposal for the 1991 AAAI Spring Symposium Series
Logical Formalizations of Commonsense Reasoning:
Evaluating Existing Formalisms and Moving on to New Domains
We solicit papers describing original ideas and new results on
expressing commonsense knowledge in formal declarative languages
and on formalizing commonsense reasoning. We are particularly
interested in:
1. Formalizations aimed at better solutions of known difficulties
or raising new ones.
Possible topics include:
- causality,
- relations between knowledge and action,
- knowledge and communication,
- illuminating small domains like the blocks world and extensions,
- common sense physics,
- common sense psychology and
- applications of nonmonotonic reasoning to the above.
2. Extending formalization to new domains.
3. Evaluation and mathematical analysis of the soundness and generality of
formalizations.
4. Mathematical results and counterexamples clarifying relations
between different formalisms.
Relevance to artificial intelligence will be the primary consideration.
New logics advanced for their own sakes will be regarded with suspicion.
Theorems only marginally related to the problem of formalizing common
sense, as well as papers investigating primarily computational problems
(algorithms, complexity, implementation), would not be appropriate for
this symposium either. People considering submissions may wish to
contact one of the organizers by telephone, e-mail or mail to discuss
appropriateness of the topic.
Organizers:
John McCarthy (Chairman)
Computer Science Department
Stanford University
Stanford, CA 94305
Phone: (415)723-4430
E-mail: JMC@CS.STANFORD.EDU
Vladimir Lifschitz
Computer Science Department
Stanford University
Stanford, CA 94305
Phone: (415)723-3334
E-mail: VAL@CS.STANFORD.EDU
Leora Morgenstern
IBM Watson Research Center
P.O.Box 704
Yorktown Heights, NY 10598
Phone: (914)784-7151
E-mail: LEORA@IBM.COM
∂06-Apr-90 1533 shoham@Hudson.Stanford.EDU meeting
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Apr 90 15:32:40 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA06997; Fri, 6 Apr 90 14:56:25 -0700
Date: Fri, 6 Apr 1990 14:56:24 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: meeting
Message-Id: <CMM.0.88.639438984.shoham@Hudson.Stanford.EDU>
You were right about Wednesday noon. I have therefore changed my group's
meeting time next week. If you're still willing to come and participate in
the discussion of AOP and Elephant, then I propose one of the following:
Tuesday at 3:15
Wednesday at 9:30
Friday at 12:15
This is roughly also my order of preference, which is very weak.
Yoav
PS I will try to have an image of what the airline reservation system might
look like by then
∂06-Apr-90 1605 CN.MCS@Forsythe.Stanford.EDU
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Apr 90 16:05:08 PDT
Date: Fri, 6 Apr 90 16:05:51 PDT
To: jmc@sail
From: "REBECCA LASHER" <CN.MCS@Forsythe.Stanford.EDU>
John,
I am very glad that you took the time to attend the meeting on this
DARPA project. I had been worried myself about not using phone
lines for retrieval so I was very pleased when you spoke up. If we
go forward with a proposal to DARPA, I hope you will stay interested
in this project and help us look at this from the user's
perspective. Your experience is invaluable.
Rebecca Lasher
∂06-Apr-90 1619 shoham@Hudson.Stanford.EDU re: meeting
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Apr 90 16:18:53 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA07406; Fri, 6 Apr 90 16:18:21 -0700
Date: Fri, 6 Apr 1990 16:18:20 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: val@sail.Stanford.EDU
Subject: re: meeting
In-Reply-To: Your message of 06 Apr 90 1539 PDT
Message-Id: <CMM.0.88.639443900.shoham@Hudson.Stanford.EDU>
Then Wednesday at 9:30 it is. How about the robotics (Cedar Hall) conference
room? Rather modest, I might warn you.
Yoav
∂06-Apr-90 1719 ingrid@russell.Stanford.EDU your IAP presentations
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Apr 90 17:19:44 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA07080; Fri, 6 Apr 90 17:22:24 PDT
Message-Id: <9004070022.AA07080@russell.Stanford.EDU>
To: jmc@sail.stanford.edu, kay.pa@xerox.com, sells@russell.Stanford.EDU,
etch@russell.Stanford.EDU, winograd@russell.Stanford.EDU,
poser@russell.Stanford.EDU, peters@russell.Stanford.EDU,
schuetze@russell.Stanford.EDU
Subject: your IAP presentations
Date: Fri, 06 Apr 90 17:22:23 PDT
From: ingrid@russell.Stanford.EDU
Please remember to let me have a hard copy of the transparancies for
your IAP presentations on Monday. I need to copy them and then place
them in the ring binders I'm preparing for our visitors.
Ingrid
∂07-Apr-90 1221 shoham@Hudson.Stanford.EDU time
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 7 Apr 90 12:21:28 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA10460; Sat, 7 Apr 90 12:20:56 -0700
Date: Sat, 7 Apr 1990 12:20:55 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: time
Message-Id: <CMM.0.88.639516055.shoham@Hudson.Stanford.EDU>
It turns out that I have to leave the lab at 10:15 on Wednesday; I'd
forgottten about an appointment. I suggest bringing the meeting up to 9, or,
preferably, doing it at 12:15, Friday. I understand it conflicts with your
regular lunch. If you'd like a late lunch, though, I'll buy.
Yoav
∂08-Apr-90 1422 VAL re: time
To: shoham@HUDSON.STANFORD.EDU, JMC@SAIL.Stanford.EDU
[In reply to message sent Sat, 7 Apr 1990 12:20:55 PDT.]
9am is fine.
--Vladimir
∂08-Apr-90 1445 shoham@Hudson.Stanford.EDU ok
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 8 Apr 90 14:45:30 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA13061; Sun, 8 Apr 90 14:44:56 -0700
Date: Sun, 8 Apr 1990 14:44:56 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: ok
Message-Id: <CMM.0.88.639611096.shoham@Hudson.Stanford.EDU>
WE're all set of Sednesday at 9, then.
Yoav
∂08-Apr-90 2307 rdz@Gang-of-Four.Stanford.EDU 15
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 8 Apr 90 23:07:12 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12947; Sun, 8 Apr 90 23:08:33 -0700
Date: Sun, 8 Apr 90 23:08:33 -0700
From: rdz@Gang-of-Four.Stanford.EDU (Ramin Zabih)
Message-Id: <9004090608.AA12947@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
In-Reply-To: John McCarthy's message of Sat, 7 Apr 90 17:43:29 -0700 <9004080043.AA10590@Gang-of-Four.Stanford.EDU>
Subject: 15
He's out of the intensive care unit, but I haven't dealt with
*anything* else since Wednesday evening (I've logged in twice to read
my mail). He's still in the hospital and, frankly, it's likely to
occupy a lot of my time for the next week.
It might be that the right thing is for me to teach Dan about how to
make changes to the program. I'll try to be in tomorrow afternoon
sometime and we can chat about this.
Ramin
∂09-Apr-90 0104 @IGNORANT.Stanford.EDU:RDZ@SCORE.Stanford.EDU puzzle
Received: from IGNORANT.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 01:04:16 PDT
Received: from PRECARIOUS.Stanford.EDU by IGNORANT.Stanford.EDU via CHAOS with CHAOS-MAIL id 3201; Mon 9-Apr-90 01:16:28 PDT
Date: Mon, 9 Apr 90 01:15 PDT
From: Ramin Zabih <RDZ@SCORE.Stanford.EDU>
Subject: puzzle
To: jmc@SAIL.Stanford.EDU
Message-ID: <19900409081531.1.RDZ@PRECARIOUS.Stanford.EDU>
The files are in go4:/u/rdz/Puzzle. The file
load-puzzle, when loaded, creates the package and
loads all the files. The key file is
measures.lisp, which contains all the worse (and
the unique better) measures.
Ramin
∂09-Apr-90 0124 @Score.Stanford.EDU,@PRECARIOUS.Stanford.EDU:RDZ@SCORE.Stanford.EDU Current problem
Received: from Score.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 01:24:45 PDT
Received: from PRECARIOUS.Stanford.EDU by SCORE.STANFORD.EDU with TCP; Mon 9 Apr 90 01:24:22-PDT
Date: Mon, 9 Apr 90 01:36 PDT
From: Ramin Zabih <RDZ@SCORE.Stanford.EDU>
Subject: Current problem
To: jmc@SAIL.Stanford.EDU
Message-ID: <19900409083603.2.RDZ@PRECARIOUS.Stanford.EDU>
The blank adjacency heuristic has a horrible time
with the board whose top two rows are:
1 2 3 11
9 10 X 4
(where X is the blank). I can't convince myself
at the moment that this is a bug in my
implementation. Even 14 moves deep it finds no
better solution, and the only pruning is done by
BLANK-NOT-ADJACENT.
Ramin
∂09-Apr-90 0755 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 07:55:32 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA14385; Mon, 9 Apr 90 07:56:51 -0700
Date: Mon, 9 Apr 90 07:56:51 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9004091456.AA14385@Gang-of-Four.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 08 Apr 90 2238 PDT <vFCia@SAIL.Stanford.EDU>
Good morning
I was still working on some changes late Friday. I will put them in
today. Also, do you want me to not leave any lines between Seattle,
Minn, Boston, Wash, etc. in your calendar listing. Do you like the
way I entered it for your trip to Spain?
∂09-Apr-90 0800 JMC
call nixon principal
∂09-Apr-90 0919 GA.JRG@Forsythe.Stanford.EDU Harley, care to help out Prof. McCarthy on this?
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 09:19:50 PDT
Date: Mon, 9 Apr 90 09:20:38 PDT
To: jmc@sail
From: "June Genis" <GA.JRG@Forsythe.Stanford.EDU>
Subject: Harley, care to help out Prof. McCarthy on this?
John, it doesn't look to me like Harley copied you on this. He was
out sick last week. Hope it's still in time to be of some help.
/June
To: JMC@SAIL
FORWARDED MESSAGE 04/09/90 08:40 FROM GE.HEG "Harley Goes": Harley, care to
help out Prof. McCarthy on this?
REPLY TO 04/04/90 13:08 FROM GA.JRG "June Genis": Harley, care to help out
Prof. McCarthy on this?
June,
This response may be too late, but here is a short example.
Harley
01 BILL PIC $$$$.99.
MOVE 123.45 TO BILL.
DISPLAY 'Please pay ' BILL ' by May 10, 1990.'.
To: GA.JRG
∂09-Apr-90 1024 MPS Chairmanship of CS
Dean Gibbons had a memo handcarried concerning interviewing senior faculty
this week. He and C. Kruger will be doing the interviewing. He would like
you to meet with Kruger this week for a 1 hour meeting at your earliest
convenience.
Pat
∂09-Apr-90 1104 shoham@Hudson.Stanford.EDU ai doc
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 11:04:46 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA15744; Mon, 9 Apr 90 11:04:07 -0700
Date: Mon, 9 Apr 1990 11:04:06 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, nilsson@cs.Stanford.EDU, winograd@cs.Stanford.EDU,
genesereth@cs.Stanford.EDU, bahyes-roth@SUMEX-AIM.Stanford.EDU
Cc: tajnai@cs.Stanford.EDU
Subject: ai doc
Message-Id: <CMM.0.88.639684246.shoham@Hudson.Stanford.EDU>
Looking at the impressive brochure that Carolyn put together a thought
occurred to me that I wish had occurred earlier. Several of our talks will
overlap in content, as does our research. The audience will hear several of
us use the word agents, several talk of systems and environments, several
about social metaphores for interaction. It would have been nice to provide a
brief overview of these issues, and place the various work on some sort of
map. I don't know if one can touch the schedule now, but if Carolyn decides
one can squeeze in another 15-20 minutes, how do you all feel about it? As
I've put some time into understanding your work I'm willing to do it, but it
doesn't have to be me.
Yoav
∂09-Apr-90 1225 Mailer re: Death penalty
To: JMC@SAIL.Stanford.EDU, tran@PORTIA.STANFORD.EDU,
su-etc@SAIL.Stanford.EDU
From: Robert W Floyd <RWF@SAIL.Stanford.EDU>
[In reply to message from JMC rcvd 09-Apr-90 11:21-PT.]
When it is possible to freeze and revive a person, that will not be
a punishment. Clever people will put a little money in the bank at interest,
then have themselves frozen to be thawed several hundred years later,
at which time they will be wealthy. Please don't throw me in that briar patch!
∂09-Apr-90 1345 jfetzer@ub.d.umn.edu Books for Review
Received: from ub.d.umn.edu by SAIL.Stanford.EDU with TCP; 9 Apr 90 13:45:47 PDT
Received: by ub.d.umn.edu (5.59/UMD-891211)
id AA16377; Mon, 9 Apr 90 15:47:52 CDT
From: jfetzer@ub.d.umn.edu (james fetzer)
Message-Id: <9004092047.AA16377@ub.d.umn.edu>
Subject: Books for Review
To: jmc@sail.stanford.edu
Date: Mon, 9 Apr 90 15:47:51 CDT
Cc: rapaport@cs.buffalo.edu, jfetzer@ub.d.umn.edu
X-Mailer: Elm [version 2.1 PL1]
John,
Unless you sent the information directly to Bill Rapaport,
I would appreciate having bibliographical data on Manna
and Waldinger's two books for David Nelson to review.
Thanks.
Jim
∂09-Apr-90 1432 ingrid@russell.Stanford.EDU handouts for tomorrow's meeting
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 Apr 90 14:32:51 PDT
Received: from localhost by russell.Stanford.EDU (4.1/inc-1.0)
id AA18890; Mon, 9 Apr 90 14:35:27 PDT
Message-Id: <9004092135.AA18890@russell.Stanford.EDU>
To: poser@russell.Stanford.EDU, winograd@russell.Stanford.EDU,
kay.pa@xerox.com, jmc@sail.stanford.edu
Subject: handouts for tomorrow's meeting
Date: Mon, 09 Apr 90 14:35:26 PDT
From: ingrid@russell.Stanford.EDU
I MUST HAVE A SET OF YOUR TRANSPARANCIES BY 4:00 P.M. TODAY!
Ingrid
∂09-Apr-90 1557 guha@sumex-aim.stanford.edu mid term report
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 9 Apr 90 15:57:36 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA00778; Mon, 9 Apr 90 16:00:39 PDT
Date: Mon, 9 Apr 1990 16:00:38 PDT
From: "Ramanathan V. Guha" <guha@sumex-aim.stanford.edu>
To: jmc@sail.stanford.edu
Subject: mid term report
Message-Id: <CMM.0.88.639702038.guha@sumex-aim.stanford.edu>
Hi,
I wanted to make sure that you had gotten the newer version of the cyc
mid term report draft. I had dropped it off with you secretary about a week
and a half ago.
Thanks
Guha
∂09-Apr-90 1600 winograd@loire.stanford.edu Re: ai doc
Received: from loire.stanford.edu by SAIL.Stanford.EDU with TCP; 9 Apr 90 16:00:00 PDT
Received: by loire.stanford.edu (5.59/25-eef) id AA01166; Mon, 9 Apr 90 16:02:34 PDT
Date: Mon, 9 Apr 90 16:02:34 PDT
Message-Id: <9004092302.AA01166@loire.stanford.edu>
From: Terry Winograd <Winograd@csli.stanford.edu>
To: shoham@Hudson.Stanford.EDU
Subject: Re: ai doc
Cc: bahyes-roth@SUMEX-AIM.Stanford.EDU, genesereth@cs.Stanford.EDU,
jmc@sail.Stanford.EDU, nilsson@cs.Stanford.EDU
Yoav, I'd be glad to have you do it. Thanks for offering. --t
∂09-Apr-90 1601 boyer@CLI.COM References
Received: from CLI.COM by SAIL.Stanford.EDU with TCP; 9 Apr 90 16:01:31 PDT
Received: by CLI.COM (4.0/1); Mon, 9 Apr 90 17:59:59 CDT
Date: Mon, 9 Apr 90 17:59:59 CDT
From: Robert S. Boyer <boyer@CLI.COM>
Message-Id: <9004092259.AA18460@CLI.COM>
To: jmc@sail.stanford.edu
Subject: References
Reply-To: boyer@CLI.COM
Here is Moore's and my most recent book, which describes our prover,
its logic, and its applications: R. S. Boyer and J. S. Moore, "A
Computational Logic Handbook", Academic Press, Newy York, 1988.
The special issue of the Journal of Automated Reasoning is the
"Special Issue on System Verification", volume 5, number 4, 1989. It
includes articles by Moore, Hunt, Bevier, and Young with the titles:
"An Approach to Systems Verification", "Microprocessor Design
Verification", "A Mechanically Verified Language Implementation", "A
Mechanically Verified Code Generator", and "Kit and the Short Stack".
These articles are quite unified in describing the greatest
achievement of verification to date.
Fetzer knows about the JAR special issue. I have seem him quote from
it in some email messages.
∂09-Apr-90 1920 boyer@CLI.COM References
Received: from CLI.COM by SAIL.Stanford.EDU with TCP; 9 Apr 90 19:20:27 PDT
Received: by CLI.COM (4.0/1); Mon, 9 Apr 90 21:18:54 CDT
Date: Mon, 9 Apr 90 21:18:54 CDT
From: Robert S. Boyer <boyer@CLI.COM>
Message-Id: <9004100218.AA19109@CLI.COM>
To: JMC@SAIL.Stanford.EDU
Subject: References
Reply-To: boyer@CLI.COM
Look forward to a visit soon.
∂10-Apr-90 0907 littell@Sunburn.Stanford.EDU Fellowship supplement
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 Apr 90 09:07:19 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA29836; Tue, 10 Apr 90 09:08:09 -0700
Date: Tue, 10 Apr 90 09:08:09 -0700
From: Angelina M. Littell <littell@Sunburn.Stanford.EDU>
Message-Id: <9004101608.AA29836@Sunburn.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 10 Apr 90 0906 PDT <dG80v@SAIL.Stanford.EDU>
Subject: Fellowship supplement
What account should I charge it to?
Thanks.
--Angie
∂10-Apr-90 1147 WALDINGER@Warbucks.AI.SRI.COM
Received: from Warbucks.AI.SRI.COM by SAIL.Stanford.EDU with TCP; 10 Apr 90 11:47:51 PDT
Date: Tue 10 Apr 90 11:48:08-PST
From: WALDINGER@Warbucks.AI.SRI.COM (Richard Waldinger)
To: JMC@SAIL.STANFORD.EDU
Message-ID: <639776888.0.WALDINGER@WARBUCKS.AI.SRI.COM>
In-Reply-To: <hG9$u@SAIL.Stanford.EDU>
Mail-System-Version: <VAX-MM(229)+TOPSLIB(126)@WARBUCKS.AI.SRI.COM>
sorry, the message i got said you needed the book's bibliography.
the reference is
Zohar Manna and Richard Waldinger, The Logical Basis for
Computer Programming. Volume 1: Deductive Reasoning.
Addison-Wesley, Reading, Massachussetts, 1985.
Volume 2: Deductive Systems. Addison-Wesley, Reading,
Massachussetts, 1990.
-------
∂10-Apr-90 1221 AI.GUHA@MCC.COM
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 10 Apr 90 12:21:26 PDT
Date: Tue 10 Apr 90 14:22:03-CDT
From: R. V. Guha <AI.GUHA@MCC.COM>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: <hGpEG@SAIL.Stanford.EDU>
Message-ID: <12580764248.37.AI.GUHA@MCC.COM>
We made most of the changes you suggested in your earlier comments. I dropped
it by just so that you would have a copy of the latest version. If you think
we should give someone a copy please tell us and we will send him/her one
(or if they are around there, please feel free to give them a copy).
Thanks
Guha
-------
∂10-Apr-90 1257 jfetzer@ub.d.umn.edu re: Books for Review
Received: from ub.d.umn.edu by SAIL.Stanford.EDU with TCP; 10 Apr 90 12:57:09 PDT
Received: by ub.d.umn.edu (5.59/UMD-891211)
id AA14978; Tue, 10 Apr 90 14:59:15 CDT
From: jfetzer@ub.d.umn.edu (james fetzer)
Message-Id: <9004101959.AA14978@ub.d.umn.edu>
Subject: re: Books for Review
To: JMC@SAIL.Stanford.EDU (John McCarthy)
Date: Tue, 10 Apr 90 14:59:14 CDT
Cc: jfetzer@ub.d.umn.edu
In-Reply-To: <1HFxNg@SAIL.Stanford.EDU>; from "John McCarthy" at Apr 09, 90 1412
X-Mailer: Elm [version 2.1 PL1]
John,
Good suggestions! I like the idea of doing the handbook and the
special issue of the Journal of Automated Reasoning. I also like
the idea of having two reviews from different points of view. So
I will look forward to having the relevant bibliographical infor-
mation.
Incidentally, could you send me copies of some of your early papers
on mathematics and computing from 1962 and 1963? I would also like
to have copies of any other papers that are available, although I
imagine all of them together could fill a truck. Please send what
you can.
Jim
∂10-Apr-90 1417 rdz@Gang-of-Four.Stanford.EDU [maja@ai.mit.edu: seminar this Thursday 4p.m.]
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 Apr 90 14:17:11 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA19652; Tue, 10 Apr 90 14:18:28 -0700
Date: Tue, 10 Apr 90 14:18:28 -0700
From: rdz@Gang-of-Four.Stanford.EDU (Ramin Zabih)
Message-Id: <9004102118.AA19652@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: [maja@ai.mit.edu: seminar this Thursday 4p.m.]
Return-Path: <@ai.ai.mit.edu,@REAGAN.ai.mit.edu:maja@ai.mit.edu>
>From: "Maja J. Mataric" <maja@ai.mit.edu>
Date: Mon, 9 Apr 90 22:13:24 EDT
To: ai-seminar@ai.mit.edu
Subject: seminar this Thursday 4p.m.
--- AI REVOLVING SEMINAR ---
The Role of Academics in Public Policy
an Open Discussion
Ron Chaney
Historically, many members of the academic community have become
influential dissidents for political reform. Perhaps the most notable
example is the Russian Physicist Andrei Sakharov who won the Nobel
Prize for Peace as a result of his protest of human rights policies in
the Soviet Union. More recently academics have been influential in
the pro-democracy movement in China as well as the restructuring of
Central and Eastern European countries.
In the United States, however, the political participation of members
of the academic community has declined in the past decade. Some argue
that this is desirable since politics are inherently subjective and
participation in the political system reduces the ability of academics
to be objective. Furthermore, academics are often out of touch with
the concerns of the common man. Others disagree claiming that the
participation of academics in the political system helps to maintain
some degree of objectivity and that it is desirable for academics to
infuse radical and idealistic values into the process.
For members of the MIT community this raises a number of interesting
issues. To what degree should we be involved in public policy issues?
Would such involvement compromise our objectivity? More specifically,
what responsibility, if any, do we have for the use of technology
after we have developed it? Should we attempt to influence the use of
the technology? If so, what are reasonable and effective strategies?
I will make some introductory remarks. Then, I will moderate a
discussion of these issues while making every attempt to keep it
focused and objective. Flaming on current political topics will be
kept to a minimum. Come with a carefully considered opinion, an open
mind, or both.
Thursday, April 12, 4p.m. NE43 8th Floor Playroom.
Good Debatable Food
∂10-Apr-90 1538 MPS
Laura from Van Nostrand called
619 561-3408
∂11-Apr-90 1229 U.UNDERDOG@Macbeth.Stanford.EDU being a gopher
Received: from Macbeth.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 Apr 90 12:29:04 PDT
Date: Wed 11 Apr 90 12:30:42-PDT
From: D. J. <U.UNDERDOG@Macbeth.Stanford.EDU>
Subject: being a gopher
To: jmc@sail.stanford.edu
Message-ID: <12581027966.25.U.UNDERDOG@Macbeth.Stanford.EDU>
So, what did you decide, John?
You mentioned that would send me a message, but I haven't gotten
it yet.
--Dwight
-------
∂11-Apr-90 1312 slagle@cs.umn.edu Draft--McCarthy's Schedule
Received: from cs.umn.edu by SAIL.Stanford.EDU with TCP; 11 Apr 90 13:11:35 PDT
Received: by cs.umn.edu (5.59/1.14)
id AA17073; Wed, 11 Apr 90 15:12:11 CDT
Date: Wed, 11 Apr 90 15:12:11 CDT
From: "James Slagle" <slagle@cs.umn.edu>
Message-Id: <9004112012.AA17073@cs.umn.edu>
To: jmc@sail.stanford.edu, kessler@cs.umn.edu, rosen@cs.umn.edu
Subject: Draft--McCarthy's Schedule
Note: Everything is in the EE/CSci Building unless otherwise noted.
Wednesday, April 18, 1990
1:00pm Arrival EE/CSci 4-192
1:05-1:45 James Slagle, 5-189
1:45-2:15
2:30-3:30 Lecture 1, 231 Smith Hall
3:30-4:00 Reception, 5th Floor Faculty Lounge
4:00-4:30
4:30-5:00 Slagle, 5-189
6:00 Dinner. Ben Rosen, James Slagle, and ?, Holiday
Metrodome
Thursday, April 19, 1990
9:00-9:30am
9:30-10:00
10:00-11:00 Discussion with graduate students. Large
conference room (4-192a).
11:15-12:05 Lecture 2, 2-230
12:15-1:00 Reception, 5th Floor Faculty Lounge
1:00-2:00 Lunch, Coffman Union
2:00-3:00 Discussion with graduate students. Large
conference room (4-192a).
3:00-3:30 Slagle, 5-189
3:30-4:00
4:00-4:30
4:30-5:00
Friday, April 20, 1990
9:00-9:30am
9:30-10:00
10:00-11:00 Graduate/Faculty Coffee/Cookie Hour, 2-212
11:00-12:00 Discussion with graduate students. Small
conference room (4-192e).
12:00-1:00 Lunch. Slagle and ?, Coffman Union.
1:00-1:30
1:30-2:00
2:00-2:30
2:30-3:30 Lecture 3, 2-230
3:30-4:00 Reception, 5th Floor Faculty Lounge
4:00-4:30 Slagle, 5-189
4:30 Departure for Airport
∂11-Apr-90 1358 jones@Sunburn.Stanford.EDU undergraduate colloquium
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 Apr 90 13:57:57 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA00721; Wed, 11 Apr 90 13:58:49 -0700
Date: Wed, 11 Apr 1990 13:58:48 PDT
From: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: undergraduate colloquium
Message-Id: <CMM.0.88.639867528.jones@Sunburn.Stanford.EDU>
John,
Just wanted to confirm that you'll be coming to the undergrad colloquium this
Thusday, 4/12, from 3:30-5:00. The class meets in 60-62A, which is in the
symbolic systems building to the right of the church as you approach it from
mjh.
Roy
∂11-Apr-90 1546 VAL Texas
∂11-Apr-90 1540 pbw@cs.utexas.edu Texas
Received: from cs.utexas.edu by SAIL.Stanford.EDU with TCP; 11 Apr 90 15:40:43 PDT
Posted-Date: Wed, 11 Apr 90 17:41:27 CDT
Message-Id: <9004112241.AA01441@cs.utexas.edu>
Received: by cs.utexas.edu (5.61/1.56)
id AA01441; Wed, 11 Apr 90 17:41:28 -0500
From: pbw@cs.utexas.edu (Paul Woodruff)
Date: Wed, 11 Apr 90 17:41:27 CDT
X-Mailer: Mail User's Shell (6.5.6 6/30/89)
To: val@sail.stanford.edu
Subject: Texas
Cc: aldale@cs.utexas.edu, pbw@cs.utexas.edu
Dear Vladimir,
Both departments have now met, and I can tell you that enthusiasm
among the logic people in both department runs very high, and both departments
have voted for offering you the position. We are now in negotiation
with the two deans about the final details of the offer, and I expect
it will be pretty much what we talked about when you were here. Nothing
can be official, however, until we have the administration's official
OK. This normally takes two or three weeks. I'm sorry it is so slow.
Did I tell you that I talked to Michael Katz of Slavic Languages?
His interest in inviting Elena to join his department is real and quite
enthusiastic.
I very much hope that we will be able to make Texans of you once
again. Austin is, from my perspective, a dream of a city for easy and
pleasant living. If there is anything we can tell you about it,
please give me a call. I know that Lucia would be happy to talk to Elena
about the way things are here.
Once again, I am delighted to report this result to you. Your
coming here would be a great thing for both departments.
Yours,
Paul
∂11-Apr-90 1715 shoham@Hudson.Stanford.EDU follow up
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 Apr 90 17:15:10 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA28457; Wed, 11 Apr 90 17:14:31 -0700
Date: Wed, 11 Apr 1990 17:14:30 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: follow up
Message-Id: <CMM.0.88.639879270.shoham@Hudson.Stanford.EDU>
Since next week seems bad, how about the week after, the 25th, at noon?
Yoav
∂11-Apr-90 1721 VAL re: follow up
To: shoham@HUDSON.STANFORD.EDU, JMC@SAIL.Stanford.EDU
[In reply to message from shoham@Hudson.Stanford.EDU sent Wed, 11 Apr 1990 17:14:30 PDT.]
The 25th is fine with me. - Vladimir
∂11-Apr-90 1900 JMC
Decide about Dwight Joe.
∂12-Apr-90 0835 ito@ito.ecei.tohoku.ac.jp 'Foreword'
Received: from RELAY.CS.NET by SAIL.Stanford.EDU with TCP; 12 Apr 90 08:35:48 PDT
Received: from [192.41.197.3] by RELAY.CS.NET id aa02009; 12 Apr 90 7:27 EDT
Received: from ccut.cc.u-tokyo.ac.jp by relay.cc.u-tokyo.ac.jp (5.61/2.7W)
id AA05376; Thu, 12 Apr 90 20:27:40 +0900
Received: by ccut.cc.u-tokyo.ac.jp (5.61/6.4J.6-ut2.49)
id AA21166; Thu, 12 Apr 90 20:27:37 +0900
Received: from maia.ito.ecei.tohoku.ac.jp by hirose.cc.tohoku.ac.jp (3.2/6.4J.6-Apr10)
id AA16428; Thu, 12 Apr 90 19:41:10 JST
Received: by maia.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA14000; Thu, 12 Apr 90 19:44:58 JST
Received: by aoba.ito.ecei.tohoku.ac.jp (4.12/6.4J.6)
id AA17561; Thu, 12 Apr 90 19:45:35 jst
Received: by ito.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA02715; Thu, 12 Apr 90 19:32:56 JST
Date: Thu, 12 Apr 90 19:32:56 JST
From: Takayasu Ito <ito@ito.ecei.tohoku.ac.jp>
Return-Path: <ito@ito.ecei.tohoku.ac.jp>
Message-Id: <9004121032.AA02715@ito.ito.ecei.tohoku.ac.jp>
To: JMC%SAIL.STANFORD.EDU%relay.cs.net@u-tokyo.ac.jp
Subject: 'Foreword'
I hope that you have already received my letter and a package of Workshop paperson parallel Lisp.
We would like to have Foreword by you for our workshop proceedings in order to
encourage the researchers working on Parallel Lisp.
Now I have all the final drafts execpt Harrison-Ammarguellat's final version
with several typographical corrections(which has been sent by airmail last week)I appreciate very much if you would write Foreword and send it by April 16 or
17.
Please let me know by e-mail when you are unable to do so.
With best regards,
Takayasu Ito
∂12-Apr-90 1416 MRC@CAC.Washington.EDU re: Death Penalty Alternative
Received: from akbar.cac.washington.edu by SAIL.Stanford.EDU with TCP; 12 Apr 90 14:16:13 PDT
Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu
(5.61/UW-NDC Revision: 2.12 ) id AA03343; Thu, 12 Apr 90 14:17:03 -0700
Date: Thu, 12 Apr 1990 14:09:35 PDT
From: Mark Crispin <MRC@CAC.Washington.EDU>
Sender: Mark Crispin <mrc@Tomobiki-Cho.CAC.Washington.EDU>
Subject: re: Death Penalty Alternative
To: John McCarthy <JMC@sail.stanford.edu>
In-Reply-To: <14H9dY@SAIL.Stanford.EDU>
Message-Id: <MailManager.639954575.7590.mrc@Tomobiki-Cho.CAC.Washington.EDU>
In <14H9dY@SAIL.Stanford.EDU>, John McCarthy writes:
>Karish thinks that Texas should have an income tax. I suspect that he
>would think it ok if the Federal Courts were to find some way of ruling
>that Texas must have an income tax. We conservatives would find that
>abominable.
Not just conservatives! In liberal Washington State, with a gigantic budget
surplus, the Democratic governor is still trying to impose an income tax (over
the dead bodies of the citizens!). He managed to ram through a gasoline tax
increase; Washington has one of the highest gas taxes in the country. This
governor also signed into law legislation that would gut the authority of
Boundary Review Boards to prevent special-interest incorporations (read: a
Pacific Northwest version of the same NIMBY's and anti-growth advocates that
ruined the SF Bay Area).
Then again, Washington has a history of clowns for governor.
-------
∂12-Apr-90 1503 Mailer Re: housing policy for unmarried cohabitants
Received: from fanaraaken.STANFORD.EDU by SAIL.Stanford.EDU with TCP; 12 Apr 90 14:58:23 PDT
Received: from localhost by fanaraaken.STANFORD.EDU (4.0/4.7); Thu, 12 Apr 90 14:57:54 PDT
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: su-etc@SAIL.Stanford.EDU
Subject: Re: housing policy for unmarried cohabitants
In-Reply-To: Your message of 12 Apr 90 14:32:00 -0700.
<10Hraj@SAIL.Stanford.EDU>
Date: Thu, 12 Apr 90 14:57:52 -0700
From: James S. Vera <vera@fanaraaken.STANFORD.EDU>
In some article, John McCarthy <JMC@SAIL.Stanford.EDU> writes:
>Many of the benefits offered to married couples by common law, by
>statute law at various levels and by Stanford University policy are
>motivated by the idea that married couples often produce children
>and that it is a public benefit to help this - at least not to
>increase its costs. In so far as benefits are to be offered to
>couples that are presumed not to produce children, they need to
>be disentangled from what is supposed to benefit genuine
>families. I use the word ``genuine'' intentionally.
>
>What thought have the advocates of recognizing homosexual
>combinations given to that? Or are they merely out to grab
>everything in sight that isn't nailed down?
If such were the case, we would forbid these benefits to couples which
are unable or unwilling to bear children. I do not believe Stanford
restricts infertile, childless, married couples from receiving the
benefits that more prolific married couples receive.
Further, homosexual couples are capable of raising children. Adoption
is certainly an option, and in the case of Lesbian couples artificial
insemination is another possibility.
Thus your objection on the basis of the productiveness of a given
couple does not seem "genuine".
∂12-Apr-90 1739 jones@Sunburn.Stanford.EDU undergrad colloquium
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 Apr 90 17:39:01 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA00459; Thu, 12 Apr 90 17:39:46 -0700
Date: Thu, 12 Apr 1990 17:39:45 PDT
From: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: undergrad colloquium
Message-Id: <CMM.0.88.639967185.jones@Sunburn.Stanford.EDU>
John,
Thanks for taking the time to come to the undergrad colloquium. I think the
students' interest was quite evident.
Have you picked a sail replacement yet?
Roy
∂12-Apr-90 2220 Mailer re: housing policy for unmarried cohabitants
Received: from akbar.cac.washington.edu by SAIL.Stanford.EDU with TCP; 12 Apr 90 22:20:43 PDT
Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu
(5.61/UW-NDC Revision: 2.12 ) id AA05178; Thu, 12 Apr 90 22:21:31 -0700
Date: Thu, 12 Apr 1990 22:04:07 PDT
From: Mark Crispin <MRC@CAC.Washington.EDU>
Sender: Mark Crispin <mrc@Tomobiki-Cho.CAC.Washington.EDU>
Subject: re: housing policy for unmarried cohabitants
To: John McCarthy <JMC@sail.stanford.edu>
Cc: su-etc@sail.stanford.edu
In-Reply-To: <14Hugl@SAIL.Stanford.EDU>
Message-Id: <MailManager.639983047.4084.mrc@Tomobiki-Cho.CAC.Washington.EDU>
JMC, if you and fellow conservatives are really concerned about smart people
not having enough children, you may consider doing something about the strong
negative inducements against marriage and children:
1) The income tax "marriage penalty", which puts a higher tax burden upon a
dual-income couple if they are married than one which is unmarried. Note
that Reagan's "tax reform" eliminated Schedule W which partially abated the
marriage penalty.
2) The present structure of alimony and child-support schedules, which are
oriented at preserving the recipient's standard of living without any
consideration for the donor's standard of living. In many states, the
income of a new spouse of an alimony/child-support donor is included when
setting payments!
-------
∂12-Apr-90 2257 vera@fanaraaken.STANFORD.EDU Re: housing policy for unmarried cohabitants
Received: from fanaraaken.STANFORD.EDU by SAIL.Stanford.EDU with TCP; 12 Apr 90 22:57:38 PDT
Received: from localhost by fanaraaken.STANFORD.EDU (4.0/4.7); Thu, 12 Apr 90 22:57:08 PDT
To: JMC@SAIL.STANFORD.EDU
Subject: Re: housing policy for unmarried cohabitants
Date: Thu, 12 Apr 90 22:57:05 -0700
From: James S. Vera <vera@fanaraaken.STANFORD.EDU>
In article <rHup#@SAIL.Stanford.EDU> JMC@SAIL.STANFORD.EDU (John McCarthy) writes:
>[In reply to message from vera@fanaraaken.STANFORD.EDU sent Thu, 12 Apr 90 14:57:52 -0700.]
>
>There are great institutional costs to micromanagement, so the
>laws whose proposers explicitly wished to encourage families have
>no provisions for checking whether a couple intends to have
>children. However, many laws and other institutional arrangements
>provide additional benefits for families with children.
>
>While homosexuals might adopt children, they hardly ever do, and
>society is ambivalent about whether they should even be allowed to.
If a law wished to encourage childbirth, micromanagement is not a
problem. It need only allocate benefits to couples which have
children.
To imply that Stanford provides married student housing in order to
encourage childbirth is farfetched. It is far more likely that such
housing merely reflects the University's attempt to meet student
needs. That is clearly implied in the Daily article which started
this thread. Providing housing for unmarried couples is exactly in
line with the motivation of the proposers of married housing.
As for homosexuals adopting, few do adopt; but the same is true of
heterosexuals. The special difficulties that homosexuals have in
adopting are declining as the recent case in Mass. shows.
James S. Vera | Internet |Standard Disclaimers
Stanford University|vera@fanaraaken.stanford.edu|Blah Blah Blah Blah
Bellcore |vera2@rigel.cc.bellcore.com |vvv My WARNING vvv
I'm running for ASSU Grad. Senate, Eng. & Earth Sci. Be Suspicious!
∂13-Apr-90 0800 JMC
Call Boston Minsky Grosz Putnam Fredkin
∂13-Apr-90 1140 bthomas@Neon.Stanford.EDU Re: housing policy for unmarried cohabitants
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Apr 90 11:39:53 PDT
Received: from LOCALHOST by Neon.Stanford.EDU with SMTP (5.61/25-eef) id AA25984; Fri, 13 Apr 90 11:40:40 -0700
Message-Id: <9004131840.AA25984@Neon.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: housing policy for unmarried cohabitants
In-Reply-To: Your message of 12 Apr 90 18:03:00 -0700.
<14Hugl@SAIL.Stanford.EDU>
Date: Fri, 13 Apr 90 11:40:37 -0700
From: bthomas@Neon.Stanford.EDU
Assuming you're right about smart people not having enough children
(I'm willing to entertain the thought) I'm not sure what our society
can do about it. I would obviously fight hard against any policy
which coerces intelligent women into having children and staying home
with them. On the other hand, I don't think intelligence is so strongly
determined by genetics that it would suffice to have smart people
procreate and then turn the babies over to others to raise them.
The only reasonable way I can see to encourage an increase in the birth
rate among bright people is to provide good child care and good
parental leave for fathers and mothers. I personally plan to have
one child, but I don't see how one can have several children without
sacrificing one's career.
What's your solution?
Becky Thomas
∂13-Apr-90 1231 bthomas@Neon.Stanford.EDU Re: housing policy for unmarried cohabitants
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Apr 90 12:31:26 PDT
Received: from LOCALHOST by Neon.Stanford.EDU with SMTP (5.61/25-eef) id AA29767; Fri, 13 Apr 90 12:32:16 -0700
Message-Id: <9004131932.AA29767@Neon.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: Re: housing policy for unmarried cohabitants
In-Reply-To: Your message of 13 Apr 90 12:18:00 -0700.
<hHZdx@SAIL.Stanford.EDU>
Date: Fri, 13 Apr 90 12:32:15 -0700
From: bthomas@Neon.Stanford.EDU
I agree that objective study is both necessary and unlikely, given the
political climate. Although I consider myself politically (mostly)
liberal, I have nothing against the pursuit of knowledge, even if the
knowledge we gain makes people uncomfortable.
I think your idea of providing infant-care subsidies is an interesting
one. I would apply, no question, despite the inevitable charges
of elitism.
I grant that there is still enough land to support a much larger US
population. (I come from West Virginia, where the population has been
steadily shrinking for at least a decade and maybe two.) It's less
clear that there is enough gasoline, coal, heating oil, etc., but I
feel confident that those are problems we will solve eventually.
The real shortage may be in jobs. But perhaps that is instead a
shortage of educated workers - I haven't really thought carefully
about it.
Back to the chold-care matter - you've given a fairly short-term
solution. What about the longer term?
Becky
∂13-Apr-90 1445 VAL re: phone call on my line
[In reply to message rcvd 13-Apr-90 14:27-PT.]
Thanks. He invited me for an interview on April 27.
∂13-Apr-90 1708 Mailer re: housing policy for unmarried cohabitants
To: JMC@SAIL.Stanford.EDU, vera@FANARAAKEN.STANFORD.EDU,
su-etc@SAIL.Stanford.EDU
From: Robert W Floyd <RWF@SAIL.Stanford.EDU>
[In reply to message from JMC rcvd 12-Apr-90 14:32-PT.]
Whether production of children is a public service is not clear,
but there is one clear reason to provide services encouraging
heterosexual cohabitation: it mitigates the notorious effects
of testosterone poisoning. Anyone who has passed downwind of
an all-male frat will know what I mean. Male homosexual unions
exacerbate the symptoms, while female unions diminish the supply
of women willing to sacrifice themselves to try to help a TP
victim reconcile his affliction with membership in a civilized
society. So far as I can see, marital status is largely
irrelevant to the matter, except that requiring marriage
reduces the number of phoney cohabitations.
∂13-Apr-90 1738 mogul@wrl.dec.com Re: Death Penalty Alternative
Received: from decpa.pa.dec.com by SAIL.Stanford.EDU with TCP; 13 Apr 90 17:38:13 PDT
Received: by decpa.pa.dec.com; id AA16794; Fri, 13 Apr 90 17:39:05 -0700
Received: by jove.pa.dec.com; id AA18298; Fri, 13 Apr 90 17:39:01 -0700
Date: Fri, 13 Apr 90 17:39:01 -0700
From: mogul@wrl.dec.com (Jeffrey Mogul)
Message-Id: <9004140039.AA18298@jove.pa.dec.com>
To: JMC@SAIL.STANFORD.EDU
Subject: Re: Death Penalty Alternative
Newsgroups: su.etc
In-Reply-To: <$HrYo@SAIL.Stanford.EDU>
Organization: DEC Western Research
Cc:
What I worry
about is that the Federal judiciary should be able to force Texas
to have an income tax, whether it be directly or by imposing
costs that make it necessary, e.g. on grounds that otherwise
Texas is inflicting cruel and unusual punishments.
Are you really saying that a state government can avoid its obligations
under the 14th amendment (as currently construed to apply the Bill
of Rights to state government actions) simply by saying "we don't
care to pay for it"?
Maybe I don't want to pay my mortgage this month, either.
What would you say about a judge who issued an injunction against
a state that said it didn't want to spend enough money to print
enough ballots to allow all its citizens to vote?
-Jeff
Jeff Mogul writes:
Are you really saying that a state government can avoid
its obligations under the 14th amendment (as currently
construed to apply the Bill of Rights to state
government actions) simply by saying "we don't care to
pay for it"?
What would you say about a judge who issued an
injunction against a state that said it didn't want to
spend enough money to print enough ballots to allow all
its citizens to vote?
∂13-Apr-90 2307 beeson@ucscd.UCSC.EDU a mathematical problem for your amusement
Received: from ucscd.UCSC.EDU by SAIL.Stanford.EDU with TCP; 13 Apr 90 23:07:01 PDT
Received: by ucscd.UCSC.EDU (5.61/1.34)
id AA00522; Fri, 13 Apr 90 23:09:31 -0700
Date: Fri, 13 Apr 90 23:09:31 -0700
From: beeson@ucscd.UCSC.EDU (20012000)
Message-Id: <9004140609.AA00522@ucscd.UCSC.EDU>
To: jmc@sail.stanford.edu
Subject: a mathematical problem for your amusement
Is there a point in the unit square at rational distance from all
four corners?
(This problem came to me from Kushner; neither he nor I knows the
answer. I worked on it about an hour using a C program hoping to show
the relevant equations unsolvable mod k for some k, but no soap.)
∂14-Apr-90 0049 beeson@ucscd.UCSC.EDU yes indeed
Received: from ucscd.UCSC.EDU by SAIL.Stanford.EDU with TCP; 14 Apr 90 00:49:37 PDT
Received: by ucscd.UCSC.EDU (5.61/1.34)
id AA01454; Sat, 14 Apr 90 00:52:10 -0700
Date: Sat, 14 Apr 90 00:52:10 -0700
From: beeson@ucscd.UCSC.EDU (20012000)
Message-Id: <9004140752.AA01454@ucscd.UCSC.EDU>
To: jmc@sail.stanford.edu
Subject: yes indeed
the coordinates of the point itself must be rational, so the areas
will be rational too.
∂14-Apr-90 0141 LES re: Spider contest
[In reply to message rcvd 14-Apr-90 01:26-PT.]
Sounds good to me. I assume that there should be just one play per
position.
∂15-Apr-90 0427 ito@ito.ecei.tohoku.ac.jp R
Received: from RELAY.CS.NET by SAIL.Stanford.EDU with TCP; 15 Apr 90 04:27:04 PDT
Received: from [192.41.197.3] by RELAY.CS.NET id aa03100; 15 Apr 90 7:27 EDT
Received: from ccut.cc.u-tokyo.ac.jp by relay.cc.u-tokyo.ac.jp (5.61/2.7W)
id AA10700; Sun, 15 Apr 90 20:27:33 +0900
Received: by ccut.cc.u-tokyo.ac.jp (5.61/6.4J.6-ut2.50)
id AA27610; Sun, 15 Apr 90 20:27:29 +0900
Received: from maia.ito.ecei.tohoku.ac.jp by hirose.cc.tohoku.ac.jp (3.2/6.4J.6-Apr10)
id AA03379; Sun, 15 Apr 90 19:10:46 JST
Received: by maia.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA03280; Sun, 15 Apr 90 19:14:33 JST
Received: by aoba.ito.ecei.tohoku.ac.jp (4.12/6.4J.6)
id AA28906; Sun, 15 Apr 90 19:15:32 jst
Received: by ito.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA04064; Sun, 15 Apr 90 18:49:33 JST
Date: Sun, 15 Apr 90 18:49:33 JST
From: Takayasu Ito <ito@ito.ecei.tohoku.ac.jp>
Return-Path: <ito@ito.ecei.tohoku.ac.jp>
Message-Id: <9004150949.AA04064@ito.ito.ecei.tohoku.ac.jp>
To: JMC%sail.stanford.edu%relay.cs.net@u-tokyo.ac.jp
Subject: R
Thank you very much for accepting our invitation to write 'Foreword'.
I look forward to receiving it soon.
Takayasu Ito
∂16-Apr-90 0958 weening@Gang-of-Four.Stanford.EDU Re: Am I truthful?
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 16 Apr 90 09:58:08 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12710; Mon, 16 Apr 90 09:59:13 -0700
Message-Id: <9004161659.AA12710@Gang-of-Four.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: RPG@SAIL.Stanford.EDU
Subject: Re: Am I truthful?
In-Reply-To: Your message of "15 Apr 90 14:14:00 PDT."
<TIxQ4@SAIL.Stanford.EDU>
Date: Mon, 16 Apr 90 09:59:11 PDT
From: Joe Weening <weening@Gang-of-Four.Stanford.EDU>
You write:
The first project to build a parallel computer was
probably Illiac 4 proposed by the early 1960s. It was
over-elaborate, the cellular automaton influenced design made it
almost immune to programming, and by the time it was working, it
had been over-run by the Cray I, and ordinary serial computer
with added vector facilities.
I would change the last part to:
had been overrun by the Cray-1, a pipelined serial computer
with added vector facilities.
I don't know enough about the Illiac 4 to verify what you said about
it. Your next paragraph says:
The largest number of arithmetic operations per second
is obtained by designs that offer very limited communication
These days, there seem to be fewer problems for which this is
important. Because the speed of memory access has not kept pace with
the speed of arithmetic logic, we now see problems, even of the
numerical type, where memory access not arithmetic is the bottleneck.
Depending on the problem, either memory latency or total bandwidth is
the limiting factor. So when you say:
Designs offering
the best communication, e.g. fully shared full-speed memory,
cannot compute as fast other designs and don't scale easily to
very large numbers of processors.
two interpretations come to my mind:
a. A tradeoff between hardware for arithmetic and hardware for memory
access reduces the amount of computation that can be performed.
b. The fully shared memory cannot run fast enough to support as much
arithmetic processing as a local-memory design can.
Of these, I only agree with the second. But since memory access, and
not arithmetic, is almost certainly the dominant cost in symbolic
computation, this does not argue against shared-memory systems. So
I'm agreeing with your basic points.
∂16-Apr-90 1302 rpg@lucid.com Truth
Received: from lucid.com by SAIL.Stanford.EDU with TCP; 16 Apr 90 13:02:49 PDT
Received: from rose ([192.31.212.83]) by heavens-gate.lucid.com id AA21179g; Mon, 16 Apr 90 13:01:08 PDT
Received: by rose id AA18879g; Mon, 16 Apr 90 13:02:26 PDT
Date: Mon, 16 Apr 90 13:02:26 PDT
From: Richard P. Gabriel <rpg@lucid.com>
Message-Id: <9004162002.AA18879@rose>
To: jmc@sail.stanford.edu, weening@gang-of-four.stanford.edu
Subject: Truth
I think you were truthful enough. I also read Joe's remarks and
I have two counter-remarks:
1. I think other serial computers with vector processors overran
Illiac-4-like machines along with the Cray 1, so I rewrote that
sentence differently.
2. I did not take Joe's first reading of the sentence:
Designs offering
the best communication, e.g. fully shared full-speed memory,
cannot compute as fast other designs and don't scale easily to
very large numbers of processors.
(His reading was this:
a. A tradeoff between hardware for arithmetic and hardware for memory
access reduces the amount of computation that can be performed.)
So I did not alter it. The description of queue-based multiprocessing
confused me so I propose a rewrite. Here is the full statement, which
has trivial modifications:
Since computers were first invented, it has been known
that serial computation has limits that can be far exceeded by
using parallel computation. Even very early computers used
parallelism in carrying out arithmetic operations, and improved
hardware has expanded this kind of parallelism.
The first project to build a parallel computer was probably
Illiac 4 proposed by the early 1960s. It was over-elaborate, the
cellular automaton influenced design made it almost immune to
programming, and by the time it was working, it had been over-run by
the Cray I, and other ordinary serial computers with added vector
facilities and pipelining.
Parallel computing poses a harsh dilemma for the system
designer. The largest number of arithmetic operations per second is
obtained by designs that offer very limited communication among the
processors. If the problem fits such a design, it can run very fast,
but for many kinds of problem, effective parallelism cannnot be obtained
without good communication. Designs offering the best communication,
e.g. fully shared full-speed memory, cannot compute as fast as other
designs and don't scale easily to very large numbers of processors.
Ingenuity sometimes provides unexpected solutions, but sometimes it
seems that no amount of ingenuity will substitute for shared memory.
The largest numerical computations are those involving partial
differential equations. When these are replaced by difference
equations in the most obvious ways, they seem to lend themselves to
regular arrays of processors. However, as soon as shock waves require
concentrating the computation on dynamically selected parts of space,
and radiation propagates influences at the speed of light, the most
obvious grids waste computation.
The idea of queue-based multiprocessing arose in the early
1960s, but support was not offered for actually implementing it. The
idea is that processes can dynamically generate subprocesses that can
be done in parallel, and these subtasks are put in a queue structure
from which processors take tasks when they become free. On the one
hand, queue based multiprocessing seems to require a shared memory,
which is expensive. On the other hand, it offers straightforward ways
of programming almost any kind of problem using techniques that aren't
far from those used in programming for serial computers. Moreover,
the programs produced don't depend on the number of processors, which
can even change dynamically. The languages needed are just the usual
serial languages augmented by a few constructions for declaring
parallelism.
Queue-based multiprocessing is particularly well suited
for symbolic computation, where the same recursive process may
involve data structures of similar structure but of enormously
varied size, and where the data structures are dynamically
determined. Lisp can be made into a parallel language in a
variety of ways without distorting its character. Moreover, many
Lisp programs written for serial machines can be made to take
advantage of parallelism of this kind. Putting Lisp programs on
parallel machines based on the idea of a cellular automaton is
problematical, and if a solution is found for a particular
program, it is likely to be strongly configuration dependent.
Projects to build parallel Lisp systems in the form of
compilers and interpreters for existing or announced shared
memory multiprocessors began in the middle 1980s and have
proceeded uneventfully. It seems to be a straightforward task
whenever the necessary resources can be assembled and maintained.
The initial proposals for parallel constructs were similar to
each other. In fact my original idea in proposing the workshop
reported in these papers was that it would be a standardization
conference, and on the basis of some experience with the parallel
constructs, a proposal could be made for the incorporation of
parallelism into Common Lisp. Unfortunately, it seems that the
field of parallel Lisp is not quite ready for standardization. I
hope standardization will be pursued in a future meeting.
The present workshop is about the first in which extensive
experience in actually implementing and using the parallel constructs
is extensively reported. The approaches taken are adequately
introduced in the Preface.
It seems to me that both queue-based multi-processing and
systems with weaker communication are destined to survive and
will be suitable for different kinds of application. Queue-based
multi-procesing will provide general and straightforward
facilities of all kinds of work, but some kinds of program will
compute faster on more specialized systems.
∂16-Apr-90 1445 gini@cs.umn.edu your visit to Minneapolis
Received: from cs.umn.edu by SAIL.Stanford.EDU with TCP; 16 Apr 90 14:45:51 PDT
Received: by cs.umn.edu (5.59/1.14)
id AA08555; Mon, 16 Apr 90 16:46:34 CDT
Date: Mon, 16 Apr 90 16:46:34 CDT
From: "Maria Gini" <gini@cs.umn.edu>
Message-Id: <9004162146.AA08555@cs.umn.edu>
To: jmc@sail.stanford.edu
Subject: your visit to Minneapolis
John,
I will be very happy to pick you up at the Minneapolis airport on
Wednesday.
What is your flight and when does it arrive?
Maria Gini
gini@cs.umn.edu
(612) 625-5582
∂16-Apr-90 1625 mps@Gang-of-Four.Stanford.EDU Trip to Minn
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 16 Apr 90 16:25:39 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA16215; Mon, 16 Apr 90 16:26:46 -0700
Date: Mon, 16 Apr 90 16:26:46 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9004162326.AA16215@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Trip to Minn
Maria responded to my email. She will meet you at the gate
∂16-Apr-90 1749 rick@hanauma.stanford.edu automobile hydrogen fuel
Received: from hanauma.stanford.edu by SAIL.Stanford.EDU with TCP; 16 Apr 90 17:49:21 PDT
Received: by hanauma.stanford.edu (5.51/7.0)
id AA23685; Mon, 16 Apr 90 17:49:32 PDT
Date: Mon, 16 Apr 90 17:49:32 PDT
From: rick@hanauma.STANFORD.EDU (Richard Ottolini)
Message-Id: <9004170049.AA23685@hanauma.stanford.edu>
To: JMC@sail
Subject: automobile hydrogen fuel
(In reference to your bookstore talk:)
Some materials scientists have studied storing hydrogen in metal-hydrides
with the advantage of greater density than liquid hydrogen and lesser
flammibility.
∂17-Apr-90 0527 cross@vax.darpa.mil Re: appointments
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 17 Apr 90 05:27:17 PDT
Received: from sun46.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA08446>; Tue, 17 Apr 90 08:27:49 -0400
Posted-Date: Tue 17 Apr 90 08:27:28-EDT
Received: by sun46.darpa.mil (4.1/5.51)
id AA10907; Tue, 17 Apr 90 08:27:29 EDT
Date: Tue 17 Apr 90 08:27:28-EDT
From: Steve Cross <CROSS@DARPA.MIL>
Subject: Re: appointments
To: JMC@sail.stanford.edu
Cc: CROSS@vax.darpa.mil, cross@vax.darpa.mil
Message-Id: <640358848.0.CROSS@SUN46.DARPA.MIL>
In-Reply-To: <rIdzl@SAIL.Stanford.EDU>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@SUN46.DARPA.MIL>
John: I'll be on travel 23-25 Apr. Barry's secretary can be
reached at 202/694-5921. As you know, I'm not planning to
continue funding the fomral reasoning work. Steve
-------
∂17-Apr-90 0601 rapaport@cs.Buffalo.EDU UB Center for Cognitive Science Workshop
Received: from ub.d.umn.edu ([131.212.32.6]) by SAIL.Stanford.EDU with TCP; 17 Apr 90 06:01:02 PDT
Received: from adara.cs.Buffalo.EDU by ub.d.umn.edu (5.59/UMD-891211)
id AA07507; Tue, 17 Apr 90 08:01:05 CDT
Received: by adara.cs.Buffalo.EDU (4.1/1.01)
id AA03838; Tue, 17 Apr 90 08:58:08 EDT
Date: Tue, 17 Apr 90 08:58:08 EDT
From: rapaport@cs.Buffalo.EDU (William J. Rapaport)
Message-Id: <9004171258.AA03838@adara.cs.Buffalo.EDU>
To: AVELLINO%UGA.BITNET@electra.cs.Buffalo.EDU,
DNUTE%UGA.BITNET@electra.cs.Buffalo.EDU,
FABRIZIO%ICNUCEVM.BITNET@electra.cs.Buffalo.EDU,
M_AND_M_EB@ub.d.umn.edu, NILSSON@score.stanford.edu,
Nous%iubacs.BITNET@electra.cs.Buffalo.EDU,
Pacss@bingvaxu.cc.binghamton.edu,
RFB1%LEHIGH.BITNET@electra.cs.Buffalo.EDU,
SCHAGRIN%SNYFREBA.BITNET@electra.cs.Buffalo.EDU, WLODZ@ibm.com,
acl@cs.rochester.edu, appelt@kl.sri.com, asher@sygmund.cgs.utexas.edu,
barwise@csli.stanford.edu, bonnie@CENTRAL.CIS.UPENN.EDU,
caroldon%nature.Berkeley.EDU@jade.berkeley.edu,
cooper%epistemi.edinburgh@nss.cs.ucl.ac.uk, covey@andrew.cmu.edu,
danorman@ucsd.edu, dietrich@bingvaxu.cc.binghamton.edu,
dipert@cs.Buffalo.EDU, dunn@iuvax.cs.indiana.edu,
hobbs@warbucks.ai.sri.com, jbarnden%nmsu.CSNET@electra.cs.Buffalo.EDU,
jtn@nutter.cs.vt.edu, ihnp4!osu-eddie!bgsuvax!ray@kitty.UUCP,
lb0q@andrew.cmu.edu, lucey%SNYFREBA.BITNET@electra.cs.Buffalo.EDU,
maida@shire.cs.psu.edu, orilia@ico.olivetti.com, phil@ub.d.umn.edu,
phlpwb%gsuvm1.BITNET@electra.cs.Buffalo.EDU, pollard@RI.CMU.EDU,
r0731%csuohio.BITNET@electra.cs.Buffalo.EDU, rapaport@cs.Buffalo.EDU,
rc2z@andrew.cmu.edu, rcohen%watdragon.waterloo.edu@relay.cs.net,
rcooper@csli.stanford.edu, rcummins@rvax.ccit.arizona.edu,
rkm@arp.anu.oz.au, rlc@emx.utexas.edu,
rochester!seismo!caip!princeton!mind!harnad@rocksanne.UUCP,
rperrault@kl.sri.com, rs2l+@andrew.cmu.edu,
rwyatt%wcu.BITNET@electra.cs.Buffalo.EDU, sag@csli.stanford.edu,
sasha%colgate.CSNET@relay.cs.net, searle@cogsci.berkeley.edu,
surf404@kub.nl, talmy@cogsci.berkeley.edu,
vardi%almvma.BITNET@electra.cs.Buffalo.EDU, wiebe@ai.toronto.edu,
zenon%uwovax.BITNET@electra.cs.Buffalo.EDU
Subject: UB Center for Cognitive Science Workshop
WHERE DOES I COME FROM?
Subjectivity and the Debate over Computational Cognitive Science
=================================================================
A two-day workshop for Social Scientists, Cognitive Scientists
and all other interested parties.
May 21-22, 1990
STATE UNIVERSITY OF NEW YORK AT BUFFALO
280 Park Hall
Amherst Campus
Buffalo, NY 14260
Presented by
the Center for Cognitive Science and
Conferences in the Disciplines
Invited Speakers:
JEFF COULTER
Department of Sociology
Boston University
Author of "Rethinking Cognitive Theory and the Social Construction of Mind"
EUGENE GENDLIN
Department of Behavioral Sciences
University of Chicago
Author of "Experiencing and the Creation of Meaning"
MARK JOHNSON
Department of Philosophy
Southern Illinois University
Author of "The Body in the Mind"
and "Metaphors We Live By" (with George Lakoff)
STUART SHANKER
Department of Philosophy
York University
Author of "Wittgenstein and the Turning Point in the Philosophy of
Mathematics" and editor of "Philosophy in Britain Today"
(includes Dr. Shanker's article ``Computer Vision or Mechanist Myopia?'')
--------------------------------------------
A videotape of a talk given at the University at Buffalo
on April 16, 1990, by Roger Penrose, author of "The Emperor's New Mind",
will also be shown (presuming all goes well), followed by a discussion.
====================================================================
WHERE DOES I COME FROM?
Subjectivity and the Debate over Computational Cognitive Science
For centuries, philosophers studying the great mysteries of human sub-
jectivity have focused on the mind/body problem and the difference
between human beings and animals. Now a new ontological question takes
center stage: to what extent can a manufactured object (a computer)
exhibit qualities of mind? There have been passionate exchanges between
those who believe that a ``manufactured mind'' is possible and those who
believe that mind cannot exist except as a living, socially situated,
embodied person.
As with earlier arguments, this one shows no sign of being resolved.
But the fight over computationalism (the belief that all mental
processes can be generated by computer programs) has immediate, ``hard''
consequences for technological research and development, social and cog-
nitive science methodology, and for our everyday experience of the world
and ourselves.
Workshop speakers and panelists come from a variety of disciplinary
backgrounds (Computer Science, Linguistics, Psychology, Sociology, and
others) and will bring a wide variety of perspectives to our topic. All
presentations will be accessible to educated non-specialists.
==========================================================================
SCHEDULE FOR Monday, May 21, 1990
DAY 1: ``Self, Language, and Body''
9:00-9:30 Registration and coffee/muffins
9:30-9:40 Welcome/Introductions
9:40-10:20 Overview of Computationalism and the ``Verstehen''
tradition
10:30-noon Mark Johnson, Philosophy, Southern Illinois University,
12:00-1:30 LUNCH
1:30-3:00 Eugene, Gendlin, Behavioral Sciences, University of
Chicago, ``Crossing and Dipping: Some Terms for
Approaching the Interface Between Natural Under-
Standing and Logical formation.''
3:15-3:30 BREAK
3:30-5:00 Panel #1, ``What Does it Mean to Understand Language?''
William Rapaport, coordinator.
===========================================================================
SCHEDULE for Tues., May 22, 1990
DAY 2: ``Minds and Machines''
9:00-10:30 Stuart Shanker, Philosophy, York University,
``Models of Discovery: Computational Theories of
`Insight'.''
10:30-10:45 BREAK - coffee/muffins
10:45-noon Viewing and discussion of a videotaped talk by
Roger Penrose (Mathematics, Oxford University):
``Computers, Minds, and the Law of Physics.''
12:00-1:30 LUNCH
1:30-3:00 Jeff Coulter, Sociology, Boston University,
Informed Neuron: Reflections on the Use of Information
Theory in the Behavioral Sciences.''
3:00-3:15 BREAK
3:15-5:00 Panel #2: ``Getting to the Heart of the Mind: Methods
and Models.''
==========================================================================
HOTELS
Hampton Inn, 10 Flint Rd., Amherst, NY (716) 689-4414. Ask for ``Where
Does I Come From'' University Conference Rate, single $53 , double $60.
You must make reservation by May 6, 1990, to be guaranteed a room.
Super 8 Motel, 1 Flint Rd., Amherst, NY (716) 688-0811. Ask for ``Where
Does I Come From'' University Conference Rate, single $35.88, double
$40.88. You must make reservation by May 6, 1990 to be guaranteed a
room.
---------------------------------------------------------------------
For more information contact:
Mary Galbraith, Center for Cognitive Science, State University of New York
at Buffalo, c/o Department of Psychology, 385 Park Hall, Buffalo, NY 14260.
(716) 636-2694; maryg@cs.buffalo.edu, maryg@sunybcs.bitnet
MEAL RESERVATION & PAYMENT
To reserve places for Monday and Tuesday lunches, send a non-refundable
$10.00 check (payable to Center for Cognitive Science) to the Center for
Cognitive Science, c/o Department of Psychology, 385 Park Hall, State
University of New York at Buffalo, Buffalo, NY 14260. Please include the
attached registration coupon.
Those registering before May 10 will receive updated schedules, direc-
tions to workshop locations, and other information. There is no charge
for registration.
============================================================================
Name:
Address:
Phone:
Institution:
Registration by email: send the above information to
maryg@cs.buffalo.edu or maryg@sunybcs.bitnet. Registration at the door:
9:00 a.m., May 21, 1990 at 280 Park Hall, Amherst Campus, SUNY Buffalo. Late
registrants may have to forage for lunch.
∂17-Apr-90 0620 rapaport@cs.Buffalo.EDU Alonzo Church Symposium
Received: from ub.d.umn.edu ([131.212.32.6]) by SAIL.Stanford.EDU with TCP; 17 Apr 90 06:20:09 PDT
Received: from adara.cs.Buffalo.EDU by ub.d.umn.edu (5.59/UMD-891211)
id AA10877; Tue, 17 Apr 90 08:20:06 CDT
Received: by adara.cs.Buffalo.EDU (4.1/1.01)
id AA03982; Tue, 17 Apr 90 09:17:16 EDT
Date: Tue, 17 Apr 90 09:17:16 EDT
From: rapaport@cs.Buffalo.EDU (William J. Rapaport)
Message-Id: <9004171317.AA03982@adara.cs.Buffalo.EDU>
To: AVELLINO%UGA.BITNET@electra.cs.Buffalo.EDU,
DNUTE%UGA.BITNET@electra.cs.Buffalo.EDU,
FABRIZIO%ICNUCEVM.BITNET@electra.cs.Buffalo.EDU,
M_AND_M_EB@ub.d.umn.edu, NILSSON@score.stanford.edu,
Nous%iubacs.BITNET@electra.cs.Buffalo.EDU,
Pacss@bingvaxu.cc.binghamton.edu,
RFB1%LEHIGH.BITNET@electra.cs.Buffalo.EDU,
SCHAGRIN%SNYFREBA.BITNET@electra.cs.Buffalo.EDU, WLODZ@ibm.com,
acl@cs.rochester.edu, appelt@kl.sri.com, asher@sygmund.cgs.utexas.edu,
barwise@csli.stanford.edu, bonnie@CENTRAL.CIS.UPENN.EDU,
caroldon%nature.Berkeley.EDU@jade.berkeley.edu,
cooper%epistemi.edinburgh@nss.cs.ucl.ac.uk, covey@andrew.cmu.edu,
danorman@ucsd.edu, dietrich@bingvaxu.cc.binghamton.edu,
dipert@cs.Buffalo.EDU, dunn@iuvax.cs.indiana.edu,
hobbs@warbucks.ai.sri.com, jbarnden%nmsu.CSNET@electra.cs.Buffalo.EDU,
jtn@nutter.cs.vt.edu, ihnp4!osu-eddie!bgsuvax!ray@kitty.UUCP,
lb0q@andrew.cmu.edu, lucey%SNYFREBA.BITNET@electra.cs.Buffalo.EDU,
maida@shire.cs.psu.edu, orilia@ico.olivetti.com, phil@ub.d.umn.edu,
phlpwb%gsuvm1.BITNET@electra.cs.Buffalo.EDU, pollard@RI.CMU.EDU,
r0731%csuohio.BITNET@electra.cs.Buffalo.EDU, rapaport@cs.Buffalo.EDU,
rc2z@andrew.cmu.edu, rcohen%watdragon.waterloo.edu@relay.cs.net,
rcooper@csli.stanford.edu, rcummins@rvax.ccit.arizona.edu,
rkm@arp.anu.oz.au, rlc@emx.utexas.edu,
rochester!seismo!caip!princeton!mind!harnad@rocksanne.UUCP,
rperrault@kl.sri.com, rs2l+@andrew.cmu.edu,
rwyatt%wcu.BITNET@electra.cs.Buffalo.EDU, sag@csli.stanford.edu,
sasha%colgate.CSNET@relay.cs.net, searle@cogsci.berkeley.edu,
surf404@kub.nl, talmy@cogsci.berkeley.edu,
vardi%almvma.BITNET@electra.cs.Buffalo.EDU, wiebe@ai.toronto.edu,
zenon%uwovax.BITNET@electra.cs.Buffalo.EDU
Subject: Alonzo Church Symposium
UNIVERSITY AT BUFFALO SYMPOSIUM IN HONOR OF ALONZO CHURCH
Friday 18 May - Sunday 20 May 1990
On Sunday, May 20, 1990, the State University of New York at Buffalo will
confer an honorary doctoral degree on Alonzo Church, the distinguished
mathematical logician. On the preceding day, Professor Church will
deliver the main address at a symposium in his honor. The other princi-
pal speakers are Martin Davis, Professor of Mathematics at the Courant
Institute of New York University, and Leon Henkin, Professor of
Mathematics at the University of California at Berkeley. Hartley
Rogers, Professor of Mathematics at the Massachusetts Institute of Tech-
nology, will introduce the principal speakers and moderate discussion
periods following the lectures. Professors Davis, Henkin, and Rogers are
former doctoral students of Professor Church.
SCHEDULE (all events take place on the South Campus (also called Main Street
Campus), except where indicated)
Friday 18 May: Two preconference presentations:
6:30 PM: Stewart Shapiro (Ohio State University)
``Effectiveness''
111 Kimball Hall.
7:30 PM: Michael Scanlan (Oregon State University)
``Church's Teachers''
125 Kimball Hall.
8:30 PM: Reception in Goodyear 10.
Saturday 19 May:
9:00 AM: Registration and Coffee. All lectures in Diefendorf Hall,
South Campus.
10:00 AM: Opening Ceremonies.
10:30 AM: Alonzo Church.
12:00 noon: Luncheon in Goodyear 10.
2:00 PM: Leon Henkin.
3:30 PM: Coffee.
4:00 PM: Martin Davis.
5:30 PM: Reception.
7:00 PM: Dinner in Goodyear 10.
Sunday 20 May: Events on North Campus.
9:00 AM: Program for Gifted High School Mathematics Students,
Stewart Shapiro (Ohio State University)
``Church and Automated Theorem Proving''
112 O'Brian Hall.
10:00 AM: General Commencement, Alumni Arena.
conferring of honorary degree on Alonzo Church,
HOTELS
Hotel Lenox, 140 North Street, Buffalo, NY 14201. (716) 884-1700 and
(800) 82LENOX. Ask for ``University Rate'', $38 single, $8 each addi-
tional person. Quiet hotel in a residential neighborhood of restored
mansions at the edge of downtown. Five minute walk to subway plus ten
minute subway ride to South Campus.
Hyatt Regency, Two Fountain Plaza, Buffalo, NY 14202. (716) 856-1234
and (800) 228-9000. Ask for ``University at Buffalo Symposium Rate'',
$109. Luxury hotel in restored downtown area near theater district.
Subway stop at the door, 12 minute subway ride to South Campus.
Holiday Inn, 1880 Niagara Falls Blvd., Amherst, NY 14150. (716) 691-
8181 and (800) 465-4329. Ask for ``University at Buffalo Symposium
Rate'', $75 single or double. Ten minute drive to South Campus.
NOTE: Deadline for guaranteed special rates is April 18, 1990.
DIRECTIONS
The State University at Buffalo South Campus (also called Main St.
Campus) is on Main Street (NY 5) at Bailey Ave. (US 62) in the northeast
corner of the City of Buffalo. Public transportation connects city
hotels and the airport with the campus. The subway stop for the sympo-
sium is called ``South Campus''. The Buffalo subway is clean,
graffiti-free and safe, with handicapped access.
MEAL RESERVATION DEPOSIT
To reserve places at the Saturday luncheon and dinner, send a nonrefund-
able $25 check payable to ``Buffalo Church Symposium SUNYRF''.
INFORMATION
Buffalo Church Symposium, Department of Philosophy, State University of
New York at Buffalo, Buffalo, NY 14260. Telephone (716) 636-2438 week-
days 10 AM to 4 PM. Ask for ``Symposium Information Desk''.
John Corcoran, Symposium Organizer
Sponsored by the Office of the President, the Buffalo Logic Colloquium,
the Department of Philosophy, the Department of Mathematics, the Depart-
ment of Computer Science, the Center for Cognitive Science, and the
Mathematics Education Program (Graduate School of Education.
∂17-Apr-90 1122 scherlis@vax.darpa.mil Request for help
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 17 Apr 90 11:22:09 PDT
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA10232>; Tue, 17 Apr 90 14:04:44 -0400
Posted-Date: Tue 17 Apr 90 14:02:07-EDT
Received: by sun45.darpa.mil (4.1/5.51)
id AA13193; Tue, 17 Apr 90 14:02:09 EDT
Date: Tue 17 Apr 90 14:02:07-EDT
From: William L. Scherlis <SCHERLIS@DARPA.MIL>
Subject: Request for help
To: SW-PI@vax.darpa.mil
Cc: boehm@vax.darpa.mil, mettala@vax.darpa.mil, jkramer@vax.darpa.mil
Message-Id: <640378927.0.SCHERLIS@VAX.DARPA.MIL>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@VAX.DARPA.MIL>
To the Software PIs:
We are making progress here developing an overall software program
that links STARS, SEI, the system software activity, and the other
software efforts in the office. As part of this process, we need to
request some additional information from you (and others who you may
contact), concerning events of the now-distant past. We are seeking
specific examples of impacts that the DARPA-sponsored software-related
efforts of the 70s and 80s have had on practice. Hence this informal
request for some help. Could you please send to me in the next few
days (i.e., before Monday evening the 23rd) a few sets of specific
bullets which indicate (1) the idea, (2) who was involved, and where
it was documented (approximately), (3) the impact on practice (where,
who), (4) how the extent of the impact can be assessed or measured.
Replies can be very short and informal, This is important to
demonstrating the impact of our work. (We don't need material that
has already been provided in annual reports of the past two years,
unless there are new impacts in industry or practice to report.) If
you have suggestions of other sources (e.g., specific historical
papers that analyze impacts in software areas), please send them along
also. Your help at this time on this will enable us build a stronger
program in ISTO.
Bill
-------
∂17-Apr-90 1152 S.SALUT@Macbeth.Stanford.EDU [liza@ernie.Berkeley.EDU (Liza C. Gabato): CS Division, U.C.B. Seminar for the week of 4/16/90 (long).]
Received: from Macbeth.Stanford.EDU by SAIL.Stanford.EDU with TCP; 17 Apr 90 11:52:48 PDT
Date: Tue 17 Apr 90 11:54:09-PDT
From: Alex Bronstein <S.SALUT@Macbeth.Stanford.EDU>
Subject: [liza@ernie.Berkeley.EDU (Liza C. Gabato): CS Division, U.C.B. Seminar for the week of 4/16/90 (long).]
To: jmc@sail.Stanford.EDU, clt@sail.Stanford.EDU
Message-ID: <12582594179.23.S.SALUT@Macbeth.Stanford.EDU>
John, Carolyn,
This fellow defends an opinion exactly opposite of the CS wisdom of the past
20 years, it might be interesting to hear what he has to say... (Or bring
him to Stanford to hear what he has to say...)
Alex
---------------
Return-Path: <@Score.Stanford.EDU:liza@ernie.Berkeley.EDU>
Received: from shelby.Stanford.EDU by Macbeth.Stanford.EDU with TCP; Fri 13 Apr 90 16:30:56-PDT
Received: from Score.Stanford.EDU by shelby.Stanford.EDU (5.61+/inc-1.0)
id AA09926; Fri, 13 Apr 90 16:29:50 -0700
Received: from ernie.Berkeley.EDU by SCORE.STANFORD.EDU with TCP; Fri 13 Apr 90 16:28:28-PDT
Received: by ernie.Berkeley.EDU (5.61/1.36)
id AA15132; Fri, 13 Apr 90 16:22:25 -0700
Date: Fri, 13 Apr 90 16:22:25 -0700
From: liza@ernie.Berkeley.EDU (Liza C. Gabato)
Message-Id: <9004132322.AA15132@ernie.Berkeley.EDU>
To: seminar@ernie.Berkeley.EDU
Subject: CS Division, U.C.B. Seminar for the week of 4/16/90 (long).
CS 298-3
Computer Science Colloquium
Professor William Kahan
Computer Science Division, U.C. Berkeley
Better To Prescribe Arithmetic Than Describe It
Wednesday, April 18, 1990
4:00 p.m. - 5:00 p.m.
60 Evans Hall
The Turing Award Lecture presented to the ACM in Wash-
ington, D.C. on February 21, 1990.
We do have standards for computing: languages, commun-
ication, interfaces, magnetic media, hardware components,
..., and now floating-point arithmetic too. They must be
good things, or we should not have so many. Some are better
than others; the IEEE standards for floating-point arith-
metic are among the better ones, we shall argue, because
they prescribe so tightly what arithmetic must do. Earlier
attempts to make sense of approximate arithmetic have been
descriptive instead, relying upon sets of axioms or upon
linguistic rules from which a programmer might hope to
deduce something about what computers will do with his pro-
gram. Counter-examples reveal that, though well-intentioned
and even ingenious, the descriptive approach is not categor-
ical enought to sustain the kind of program verification and
portability we need.
Users of hardware that conforms to IEEE 754 benefit
principally in their enjoyment of numerical software that my
originally have been developed for non-conforming hardware
but also runs on theirs (after recompilation), and runs at
least about as well, probably better. But many valuable
features of the standard remain unused for lack of access
through existing compilers; these features run the risk of
atrophy if that lack of access persists.
The IEEE standards have some flaws; they leave too much
about the handling of floating-point exceptions to the ima-
gination. We need something more than waving an unnamed
flag when O/O occurs, yet something far less brutal than
abortion. What is proposed herein requires no precise
interrupts; it insinuates no new or invisible spaghetti-like
control structures; yet it helps programmer avoid hordes of
precautionary tests for events that hardly every happen.
-------
∂17-Apr-90 1427 ito@ito.ecei.tohoku.ac.jp R
Received: from RELAY.CS.NET by SAIL.Stanford.EDU with TCP; 17 Apr 90 14:27:30 PDT
Received: from [192.41.197.3] by RELAY.CS.NET id aa19552; 17 Apr 90 17:28 EDT
Received: from ccut.cc.u-tokyo.ac.jp by relay.cc.u-tokyo.ac.jp (5.61/2.7W)
id AA15816; Wed, 18 Apr 90 06:27:40 +0900
Received: by ccut.cc.u-tokyo.ac.jp (5.61/6.4J.6-ut2.53)
id AA22754; Wed, 18 Apr 90 06:27:37 +0900
Received: from maia.ito.ecei.tohoku.ac.jp by hirose.cc.tohoku.ac.jp (3.2/6.4J.6-Apr10)
id AA13055; Tue, 17 Apr 90 23:41:01 JST
Received: by maia.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA01792; Tue, 17 Apr 90 23:44:46 JST
Received: by aoba.ito.ecei.tohoku.ac.jp (4.12/6.4J.6)
id AA05756; Tue, 17 Apr 90 23:46:00 jst
Received: by ito.ito.ecei.tohoku.ac.jp (4.0/6.4J.6)
id AA05044; Tue, 17 Apr 90 23:10:22 JST
Date: Tue, 17 Apr 90 23:10:22 JST
From: Takayasu Ito <ito@ito.ecei.tohoku.ac.jp>
Return-Path: <ito@ito.ecei.tohoku.ac.jp>
Message-Id: <9004171410.AA05044@ito.ito.ecei.tohoku.ac.jp>
To: jmc%sail.stanford.edu%relay.cs.net@u-tokyo.ac.jp
Subject: R
Thank you very much for sending Foreword. I will send all the papers
to Springer tomorrow morning.
Takayasu Ito
∂17-Apr-90 1552 VAL Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
AGENT-ORIENTED PROGRAMMING
Yoav Shoham
Stanford University
Monday, April 23, 2:30pm
MJH 252
I will discuss our work on a computational framework called
Agent-Oriented Programming.
As a programming paradigm, AOP can be viewed as an extention of
Object-Oriented Programming. It extends OOP by having modules
not only communicate with one another, but also possess knowledge
and beliefs, choices and abilities, and possibly other notions.
As a logical theory, it extends standard epistemic logics. Beside
temporalizing the K (knowledge) and B (belief) operators, it
introduces operators for choice (C) and ability (A).
In either case the intuition about these mentalistic-sounding
notions is guided by intuition about the commonsense, everyday
concepts, though the actual formal definitions come nowhere close
to capturing the full linguisitic meanings.
∂18-Apr-90 0607 CLT qlisp
To: weening@Gang-of-Four.Stanford.EDU
CC: JMC@SAIL.Stanford.EDU
We need to get qlisp affairs finished before you leave.
What is needed is:
The reference manual -- we should confer with arg on this.
A reasonably final draft of the primer
A final report (which can include the above as appendices).
Lets talk soon to see how we can arrange to get all this done
in the time available.
∂18-Apr-90 1124 korf@CS.UCLA.EDU Your visit
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 18 Apr 90 11:24:49 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA21402;
Wed, 18 Apr 90 11:25:37 -0700
Message-Id: <9004181825.AA21402@shemp.cs.ucla.edu>
Received: by denali; Wed, 18 Apr 90 11:39:00 pdt
Date: Wed, 18 Apr 90 11:39:00 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: Your visit
John,
I've been trying to find a large lecture hall for you to give a talk when you
visit on the 15th and 16th of next month. Unfortunately, such rooms are booked
pretty solid during the quarter. The best I could do is either Tuesday or
Thursday afternoon from 4 to 6, since these are our regular seminar slots and we
have a room permanently reserved. I realize that neither of these are ideal for
you, but if you could give a talk then, that would be great. If not, lets get
together anyway, and you can assume you have an open invitation to come give a
talk at a future date.
-rich
∂18-Apr-90 1404 shoham@Hudson.Stanford.EDU next meeting
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 18 Apr 90 14:04:12 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA02082; Wed, 18 Apr 90 14:03:23 -0700
Date: Wed, 18 Apr 1990 14:03:22 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Cc: val@sail.Stanford.EDU
Subject: next meeting
Message-Id: <CMM.0.88.640472602.shoham@Hudson.Stanford.EDU>
I recall a tentative arrangement we made to talk about elephants on May 2. I
think you pointed out it was undergraduate lunch. I may have suggested 10 and
you may have said ok. If so I was wrong; I have to leave promptly at 10:15,
again. Would you consider 9:00 again, or have another suggestion?
Yoav
∂18-Apr-90 1422 VAL re: next meeting
To: shoham@HUDSON.STANFORD.EDU, JMC@SAIL.Stanford.EDU
[In reply to message from shoham@Hudson.Stanford.EDU sent Wed, 18 Apr 1990 14:03:22 PDT.]
9am on May 2 is fine with me.
--Vladimir
∂19-Apr-90 0812 shekhar@cs.umn.edu request for a paper
Received: from cs.umn.edu by SAIL.Stanford.EDU with TCP; 19 Apr 90 08:12:03 PDT
Received: by cs.umn.edu (5.59/1.14)
id AA10396; Thu, 19 Apr 90 10:12:53 CDT
Date: Thu, 19 Apr 90 10:12:53 CDT
From: "Shashi Shekhar" <shekhar@cs.umn.edu>
Message-Id: <9004191512.AA10396@cs.umn.edu>
To: jmc@sail.stanford.edu
Subject: request for a paper
Prof. McCarthy,
I enjoyed the discussion of 'knows-about' predicate and
cooperation between experts in Minneapolis. I would appreciate
receiving a copy of your paper on "measure of the value of information"
modelling the interaction between a weatherman and a businessman.
Shashi
∂19-Apr-90 0848 jes@cs.brown.edu Request for help
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 19 Apr 90 08:48:12 PDT
Posted-Date: Thu, 19 Apr 90 11:26:56 EDT
Received: from [128.148.128.2] by vax.darpa.mil (5.61/5.61+local)
id <AA00999>; Thu, 19 Apr 90 11:27:52 -0400
Received: from daedalus.cs.brown.edu by cs.brown.edu (4.1/SMI-4.0)
id AA20570; Thu, 19 Apr 90 11:28:04 EDT
Date: Thu, 19 Apr 90 11:26:56 EDT
From: jes@cs.brown.edu
Message-Id: <9004191526.AA06530@daedalus.cs.brown.edu>
To: SCHERLIS@vax.darpa.mil
Cc: SW-PI@vax.darpa.mil, boehm@vax.darpa.mil, mettala@vax.darpa.mil,
jkramer@vax.darpa.mil, darpa@cs.brown.edu
In-Reply-To: William L. Scherlis's message of Tue 17 Apr 90 14:02:07-EDT <640378927.0.SCHERLIS@VAX.DARPA.MIL>
Subject: Request for help
SUBJECT: Impact of 1970s and 1980s DARPA Sponsored Research
Our first ISTO DARPA contract in the 1980s, entitled ``Ideographics,'' began
on January 15, 1983 and ran for two years. It was renamed ``A graphical
approach to software development'' and was extended to December 31, 1986. It
was renewed on September 1, 1988 and retitled ``Multiparadigm Design
Environments.'' It is scheduled to run through June 30, 1990. The following
is work supported by these projects that has had an important impact.
Research supported by DARPA ISTO that has had an impact is described below.
SOFTWARE DEVELOPMENT ENVIRONMENTS:
FIELD is an open, UNIX-based, software development environment which
integrates, using a message server, an annotation editor, debugger, data
structure display facility, cross-referencing facility, profiler, and other
tools [14,15,16] . It provides support for C++ and configuration management.
It is the product of Steve Reiss' research.
o FIELD will be ported to Motif and sold by DEC. Expected announce date
is 6/30 and expected ship date is 9/30.
THIS IS CONFIDENTIAL UNTIL ANNOUNCED BY DIGITAL.
o HP has used the FIELD selective broadcasting message server concept in
their SoftBench product.
o FIELD has been distributed to approximately 30 sites.
THE GARDEN SOFTWARE DEVELOPMENT ENVIRONMENT:
GARDEN is a closed, UNIX-based, object-oriented software development
environment in which each object is executable [10,11,13,17]. It provides
visual and textual editors for objects to support conceptual programming
through the graphical expression of languages which are co-executable. It is
the product of Steve Reiss' research.
o GARDEN has been proposed as a basis for experimentation with parallel
languages for parallel processing by Intermetrics to Rome AFB.
o GARDEN has been distributed to approximately 100 sites.
THE BROWN WORKSTATION ENVIRONMENT:
BWE contains graph drawing, editing, input and output handling packages. It is
primarily the work of Steve Reiss.
o BWE is a toolset used in GARDEN and FIELD. It has been distributed to
approximately 100 sites [12].
PROGRAMMING LANGUAGES:
o Papers and edited books by Peter Wegner on object-oriented
language design and type theory have influenced the work of designers
and implementers of object-oriented languages and environments
[25,26,27].
o ML is one of the most successful vehicles for studying types. Its
type inferencing algorithm is one of its major features. Paris
Kanellakis and John Mitchell have shown that type inferencing in ML is
PSPACE-hard and in DEXPTIME. Mairson of Brandeis has recently
extended their work to show that it is actually DEXPTIME-hard.
COMPILERS:
o The static single assignment form (SSA) [29] and techniques that use
it [28,30,31] were developed by Ken Zadeck and co-authors as a method
of optimizing redundant compiled code. It is being used in the
compiler for PL.8 for the IBM System 6000 workstations and in Fran
Allen's PTRAN compiler research project at IBM Yorktown Heights. The
latter work should lead to IBM compilers for parallel Fortran. These
techniques are also used by Johnson and Heintz in a Smalltalk compiler
at U. Illinois and as crucial part of a compiler for a dataflow
machine under development at Los Alamos.
DATABASES:
ObServer [4,5] is an object server that provides object clustering,
prefetching, and sophisticated locking and notification for accessing data
without imposing serializability constraints. ENCORE is a full
object-oriented database with a sophisticated query algebra [6,7,8] that
supports query optimization over sets, type extensions in multiple programming
languages, and non-serializable data access. ENCORE and ObServer are products
of Stan Zdonik's research
o ObServer is being used as the object manager for EIS
(EngineeringInformation System), a \$19M contract managed by Honeywell
andsubcontracted to Xerox, which is funded by Air Force to provide a
framework for integrating engineering tools.
o The data model used by EIS (called FUGUE) was heavily influenced
by the ENCORE database designed by Stan Zdonik and his students, as
isEIS's query language.
o ObServer is being used in the ``Intermedia'' hypermedia system
developed by the IRIS research group at Brown.
o ObServer is being used by Prof. Sriram of MIT's Civil Engineering
Department to build an integrated design environment for
civilengineers.
o ObServer is being used by Tom Cheatham at Harvard and Software
Options, Inc. to support a software engineering environment.
o ObServer has been distributed to approximately 40 sites.
o Research by Paris Kanellakis on object-identity and constraints may
well affect the design of the next generation of databasesystems.
o Query evaluations for several types of one-sided and two sided queries
in knowledge bases was applied by Adam Buchsbaum, Paris Kanellakis
andJeff Vitter [39].
OPERATING SYSTEMS:
o The Threads lightweight process package is considered one of the better
packages of its kind [3]. It is being offered by Encore as a
product. It is the result of research of Tom Doeppner.
ARTIFICIAL INTELLIGENCE:
o Working closely with Honeywell Systems and Research Center to develop
a production version of our temporal database system.
o Expanded the group of temporal database users to include several
European companies and academic institutions collaborating on ESPRIT
projects.
GRAPHICS:
o The book by Jim Foley and Andy van Dam [21] is often cited as the
standard reference work in the field. Andy founded SIGGRAPH, led the
SIGGRAPH core graphics standards effort in the seventies [22], and
organized and led the PHIGS+ graphics standards efforts, the ANSI and
ISO standard for graphics workstations [23]. He also spawned the AVS
(Application Visualization System) project at Stellar [20] and
recruited a team of former Brown students to implement it.
o BAGS, the research testbed for modeling rendering and animation
developed by Andy's graphics group [24], has been usedat DEC, UNC, U
Penn. Running on Sun, DEC and HP. It will probably be ported to IBM
System 6000.
ARCHITECTURE:
o For his PhD thesis Dan Loprestideveloped an algorithm and a chip for
string comparison that may influence the Human Genome Project. His new
programmable systolic array, B-SYS, designed with a student, is faster
than a Cray for string comparison and will be fabricated shortly.
BioRad, a leading biotechnology firm, has expressed interest in B-SYS
for their workstation products.
o Dan will receive Honorable Mention in the 1990 Gordon Bell Prize
Competition with a person from SRC for the parallel speed-up
obtainedin a scientific application on the SPLASH highly parallel
programmable logic array developed by SRC.
o Jeff Vitter is co-author of an IBM patent with Eugene Lindstrom of IBM
on external sorting on bubble memory secondary storage. It givesa
method to efficiently implement the complete repertoire of relational
database operations, giving an order-of-magnitude improvement over
current state-of-the-art algorithms.
ALGORITHMS:
o Graph partitioning is an important problem arising in processor
allocation and VLSI placement and routing. It can be used
forallocation of tasks to processors on a massively parallel machine
so as to minimize interprocessor communication. John Savage and
Markus Wloka have demonstrated that simulated annealing at low
temperatures and the famous Kernighan-Lin heuristic for this problem
cannot be parallelized (they are log-space hard for P) and have
produced new parallelizable heuristics that have been efficiently
implemented on the Connection Machine.
o John Savage and Markus Wloka have developed optimal parallel
algorithms which have been efficiently implemented on the Connection
Machine for the left-edge channel routing heuristic [18] and
constraint-graph generation [19] , two important problems in VLSI
synthesis.
o Jeff Vitter and Elizabeth Shriver have developed disk sorting
algorithms to exploit parallelism in I/O that are likely to
havecommercial value [33]. Recent Improvements have been made by
Mark Nodine and Jeff Vitter [32].
o Jeff Vitter's work on data compression with Paul Howard will be
transferred to the NASA effort in large-scale computing
concerningtheir new space telescope and earth observatory system.
o Jeff Vitter's optimization analysis of coalesced hashing is routinely
used for tuning the algorithm [9].
o Teamwork is a CASE tool offered by Cadre Technologies that automates
structured methodologies using interactive computer graphics.
Researchdone by Roberto Tamassia and co-authors [34,35,36,37,38] has
led to good graph-drawing algorithms that will be used in a new
version of Teamwork for laying out Entity-Relationship diagrams and
Data-Flow diagrams. Later Cadre will offer layout algorithms for Buhr
diagrams used for object oriented system design in Ada and for PERT
diagrams used in project management.
o Roberto Tamassia has also been a consultant for the DEC database group
in Colorado Springs on graph drawing algorithms. DEC is interested
inusing such algorithms in interactive graphic tools for database
design and graphic query processing. Interest in these algorithms has
also been expressed by GTE and Texas Instruments.
o Roberto Tamassia has developed dynamic algorithms for graph and
geometric problems [40,41,42,43,44].
o Phillip Klein, Roberto Tamassia, and Jeff Vitter have developed
parallel algorithms for problems in planar graphs such as,
transitiveclosure, point location and graph drawings [1,2].
HYBRID PROJECT:
Brown University and Digital Equipment Corporation have proposed to DARPA
under BAA 90-01 to design, build and test a hybrid supercomputer consisting of
tightly couple massively parallel and high-performance serial machines. Many
scientific and engineering problems are well-suited this computational model,
problems that fit the data generation/data analysis paradigm, a paridigm in
which data can be generated in parallel but analysis is primarily a serial
task.
More than a dozen Brown faculty members outside Computer Science are
co-Principal Investigators on this project. The six computer science faculty
members have proposed to help with language support, operating systems,
databases, profiling, debugging and scientific visualization by extending our
current work. We are encouraging Digital to incorporate our ideas in systems
they will offer for this platform.
The effort to solicit support for this research topic was organized by
a handful of Brown faculty members (van Dam, Doeppner, Reiss, Savage,
Zadeck, and Zdonik). John Savage coordinated the preparation and
submission of the hybrid supercomputing proposal.
REFERENCES:
1. M.-Y. Kao and P.N. Klein, "Towards Overcoming the Transitive-Closure
Bottleneck: Efficient Parallel Algorithms for Planar Digraphs," Proc. 22nd ACM
Symp. on Theory of Computing, (To appear), 1990.
2. R. Tamassia and J.S. Vitter, "Optimal Parallel Algorithms for Transitive
Closure and Point Location in Planar Structures," Proc. ACM Symp. on Parallel
Algorithms and Architectures, pp. 399-408, 1989.
3. T.W. Doeppner, "Threads: A System for the Support of Concurrent
Programming," Technical Report CS-87-11, Dept. of Computer Science, Brown
University, 1987.
4. M. Hornick and S. Zdonik, "The Implementation of a Shared, Clustered
Memory System for an Object-Oriented Database System," ACM Transactions on
Office Information Systems, 1987.
5. S. Reiss, A.H. Skarra, and S.B. Zdonik, "An Object Server for an
Object-Oriented Database System," Springer Verlag, 1989.
6. G. Shaw and S. Zdonik, "A Query Algebra for Object-Oriented Databases,"
Proc. Sixth Int. IEEE Conference on Data Engineering, Los Angeles, CA, 1990.
7. G. Shaw and S. Zdonik, "Object-Oriented Queries: Equivalence and
Optimization," Proceedings of the First International Conference on Deductive
and Object-Oriented Databases, Kyoto, Japan, December, 1989.
8. G. Shaw and S. Zdonik, "An Object-Oriented Query Algebra," Proc. Second
Int. Workshop on Database Programming Languages, Morgan Kaufmann, June 1989.
9. J.S. Vitter and W.C. Chen, "Design and Analysis of Coalesced Hashing,"
Oxford University Press, New York, 1987.
10. S. Reiss, "Conceptual Programming," Proc. 5th International Software
Process Workshop, 1989.
11. S. Reiss and S. Meyers, "Representing Programs in Multiparadigm Software
Development Environments," Proc. COMPSAC '89, 1989.
12. S. Reiss and J. Stasko, "The Brown Workstation Environment: A User
Interface Design Toolkit," Proc. IFIP Working Conference on Engineering for
Human-Computer Interaction, (to appear).
13. S.P. Reiss, "An Object-Oriented Framework for Graphical Programming," in
Research Directions in Object-Oriented Programming, Edited by B. Shriver, P.
Wegner, MIT Press, pp. 189-218, 1987.
14. S.P. Reiss, "On the Use of Annotations for Integrating the Source in a
Program Development Environment," Technical Report, Department of Computer
Science (submitted for publication), Brown University, 1989.
15. S.P. Reiss, "Interacting with the FIELD Environment," Technical Report,
Department of Computer Science (submitted for publication), Brown University,
1990.
16. S.P. Reiss, "Connecting Tools Using Message Passing in the FIELD Program
Development Environment," IEEE Software, 1990 (To appear).
17. S.P. Reiss, "Working in the Garden Environment for Conceptual
Programming," IEEE Software, Vol. 4, pp. 16-27, November 1987.
18. J.E. Savage and M.G. Wloka, "A Parallel Algorithm for Channel Routing,"
Graph-Theoretic Concepts in Computer Science, in Lecture Notes in Computer
Science, Edited by J. van Leeuwen, Vol. 344, Springer-Verlag, Amsterdam, pp.
288-301, 1988.
19. J.E. Savage and M.G. Wloka, "Parallel Constraint Graph Generation," Proc.
Decennial Caltech Conf. on VLSI, MIT Press, Cambridge, MA, pp. 241-259, March
1989.
20. A. van Dam, "The Application Visualization System: A Computational
Environment for Scientific Visualization," IEEE Computer Graphics and
Application, July 1989.
21. J.S. Foley, A. van Dam, S. Feiner, and J. Hughes, "Computer Graphics:
Principles and Practice," Addison-Wesley, April 1990 (To appear).
22. ANSI (American National Standards Institute), "American National Standard
for Information Processing Systems --- Computer Graphics --- Graphical Kernel
System (GKS) Functional Description," ANSI X3.124--1985, ANSI, New York, 1985.
23. ANSI (American National Standards Institute), "American National Standard
for Information Processing Systems --- Programmer' s Hierarchical Interactive
Graphics System (PHIGS) Functional Description, Archive File Format,
Clear-Text Encoding of Archive File," ANSI X3.144--1988, ANSI, New York, 1988.
24. Paul S. Strauss "BAGS: The Brown Animation Generation System," Technical
Report CS-88-22 (Ph.D. Thesis), Dept. of Computer Science, Brown University,
1988.
25. L. Cardelli and P. Wegner, "On Understanding Types, Data Abstraction, and
Polymorphism," Computing Surveys, December 1985.
26. P. Wegner, "The Object-Oriented Classification Paradigm," in Research
Directions in Object-Oriented Programming, Edited by B. Shriver and P. Wegner,
MIT Press, 1987.
27. Computing Surveys Special Issue on Programming Language Paradigms, %E
Peter Wegner September 1989.
28. D.R. Chase, W. Wegman, and F.K. Zadeck, "Analysis of Pointers and
Structures," Proc. SIGPLAN'90 Symp. on Compiler Construction, 1990.
29. R. Cytron, J. Ferrante, B.K. Rosen, M.N. Wegman, and F.K. Zadeck, "An
efficient method of computing static single assignment form," Conf. Rec.
Sixteenth ACM Symp. on Principles of Programming Languages, pp. 25-35, 1989.
30. B.K. Rosen, M.N. Wegman, and F.K. Zadeck, "Global value numbers and
redundant computations," Conf. Rec. Fifteenth ACM Symp. on Principles of
Programming Languages, pp. 12-27, 1988.
31. M.N. Wegman and F.K. Zadeck, "Constant propagation with conditional
branches," Technical Report CS-88-02, Department of Computer Science, Brown
University, 1988.
32. M.H. Nodine and J.S. Vitter, "Greed Sort: An Optimal Sorting Algorithm
for Multiple Disks," Technical Report CS-90-04, Dept. of Computer Science,
Brown University, February 1990.
33. J.S. Vitter and E.A. Shriver, "Optimal Algorithms for Parallel Block
Transfer," Proceedings of the 22nd Annual ACM Symposium on Theory of
Computing, Baltimore, MD, May 1990.
34. C. Batini, E. Nardelli, and R. Tamassia, "A Layout Algorithm for
Data-Flow Diagrams," IEEE Transactions on Software Engineering, Vol. SE-12,
No. 4, pp. 538-546, 1986.
35. G. Di Battista and R. Tamassia, "Algorithms for Plane Representations of
Acyclic Digraphs," Theoretical Computer Science, Vol. 61, pp. 175-198, 1988.
36. G. Di Battista, R. Tamassia, and I.G. Tollis, "Area Requirement and
Symmetry Display in Drawing Graphs," Proc. ACM Symp. on Computational
Geometry, pp. 51-60, 1989.
37. R. Tamassia, "On Embedding a Graph in the Grid with the Minimum Number of
Bends," SIAM J. Computing, Vol. 16, No. 3, pp. 421-444, 1987.
38. R. Tamassia, G. Di Battista, and C. Batini, "Automatic Graph Drawing and
Readability of Diagrams," IEEE Transactions on Systems, Man and Cybernetics,
Vol. SMC-18, No. 1, pp. 61-79, 1988.
39. A. Buchsbaum, P.C. Kanellakis, and J.S. Vitter, "A Data Structure for Arc
Insertion and Regular Path Finding," Proceedings 1st Symposium on Discrete
Algorithms, pp. 22-31, January 1990.
40. G. Di Battista and R. Tamassia, "Incremental Planarity Testing," Proc.
30th IEEE Symp. on Foundations of Computer Science, pp. 436-441, 1989.
41. G. Di Battista and R. Tamassia, "On-Line Graph Algorithms with
SPQR-Trees," Proc. 17th International Colloquium on Automata, Languages and
Programming (ICALP 90), 1990, to appear.
42. D. Eppstein, G.F. Italiano, R. Tamassia, R.E. Tarjan, J. Westbrook, and
M. Yung, "Maintenance of a Minimum Spanning Forest in a Dynamic Planar Graph,"
Proc. ACM-SIAM Symp. on Discrete Algorithms, 1990.
43. F.P. Preparata and R. Tamassia, "Fully Dynamic Point Location in a
Monotone Subdivision," SIAM J. Computing, Vol. 18, No. 4, pp. 811-830, 1989.
44. R. Tamassia and F.P. Preparata, "Dynamic Maintenance of Planar Digraphs,
with Applications," Algorithmica, 1990, (To appear).
∂19-Apr-90 2142 harnad@clarity.Princeton.EDU Dynamic Behavioral Modeling: BBS Call for Commentators
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 19 Apr 90 21:42:43 PDT
Received: from Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA09372; Fri, 20 Apr 90 00:39:57 -0400
Received: from cognito.Princeton.EDU by Princeton.EDU (5.58+++/2.32/mailrelay)
id AA05606; Fri, 20 Apr 90 00:38:52 EDT
Received: by cognito.Princeton.EDU (4.1/1.98)
id AA19517; Fri, 20 Apr 90 00:36:15 EDT
Date: Fri, 20 Apr 90 00:36:15 EDT
From: harnad@clarity.Princeton.EDU (Stevan Harnad)
Message-Id: <9004200436.AA19517@cognito.Princeton.EDU>
To: harnad@clarity.Princeton.EDU
Subject: Dynamic Behavioral Modeling: BBS Call for Commentators
To BBS Associates:
Below is the abstract of a forthcoming target article to appear in
Behavioral and Brain Sciences (BBS), an international,
interdisciplinary journal providing Open Peer Commentary on important
and controversial current research in the biobehavioral and cognitive
sciences. To be considered as a commentator or to suggest other appropriate
commentators, please send email to:
harnad@confidence.princeton.edu
____________________________________________________________________
Modeling Behavioral Adaptations
Colin W. Clark
Institute of Applied Mathematics
University of British Columbia
Vancouver BC V6T 1Y4
Canada
ABSTRACT: The behavioral landscape for any individual organism is a
complex dynamical system consisting of the individual's own
physiological and mental states and the state of the physical and
biological environment in which it lives. To understand the adaptive
significance of behavioral traits one must formulate, analyse and test
simplified models of this complex landscape. The target article
describes a technique of dynamic behavioral modeling with many
desirable characteristics. There is an explicit treatment of state
variables and their dynamics. Darwinian fitness is represented
directly in terms of survival and reproduction. Behavioral decisions
are modeled simultaneously and sequentially with biologically
meaningful parameters and variables, generating empirically testable
predictions. The technique has been applied to field and laboratory
data in a wide variety of species and behaviors. Some limitations
result from the unwieldiness of large-scale dynamic models in
parameter estimation and numerical computation. (This article is a
follow-up to a previous BBS paper by Houston & Macnamara, but it can
be read independently.)
Keywords: Dynamic programming; optimization; control theory; game
theory; behavioral ecology; evolution; adaptation; fitness.
∂20-Apr-90 0904 shoham@Hudson.Stanford.EDU
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 20 Apr 90 09:04:02 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA12217; Fri, 20 Apr 90 09:03:12 -0700
Date: Fri, 20 Apr 1990 9:03:10 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
In-Reply-To: Your message of 20 Apr 90 0858 PDT
Message-Id: <CMM.0.88.640627390.shoham@Hudson.Stanford.EDU>
Ok, so it's set. I'm afraid I will again have to leave right at 10:15.
Yoav
∂20-Apr-90 0905 shoham@Hudson.Stanford.EDU
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 20 Apr 90 09:05:02 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA12228; Fri, 20 Apr 90 09:04:12 -0700
Date: Fri, 20 Apr 1990 9:04:12 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>, val@sail.Stanford.EDU
In-Reply-To: Your message of 20 Apr 90 0858 PDT
Message-Id: <CMM.0.88.640627452.shoham@Hudson.Stanford.EDU>
Sorry, Vladimir, I forgot my mailer doesn't automatically cc you; ok for May
2 at 9. Yoav
∂20-Apr-90 1012 jutta@coyote.stanford.edu Spring Quarter AI Division Lunch
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 20 Apr 90 10:12:21 PDT
Received: by coyote.stanford.edu; Fri, 20 Apr 90 09:10:26 PST
Date: 20 Apr 1990 0910-PST (Friday)
From: Jutta McCormick <jutta@coyote.stanford.edu>
To: latombe@coyote.stanford.edu, binford@coyote.stanford.edu,
eaf@SUMEX-AIM.Stanford.EDU, mrg@sunburn.stanford.edu,
ok@coyote.stanford.edu, jmc@sail.stanford.edu, nilsson@cs.stanford.edu,
shoham@hudson.stanford.edu, winograd@csli.stanford.edu,
ginsberg@cs.stanford.edu, lifschitz@cs.stanford.edu
Cc: jutta@coyote.stanford.edu
Subject: Spring Quarter AI Division Lunch
The Spring Quarter AI Division lunch is planned for Wednesday, May 30,
at noon, at the Faculty Club. Topic: Intelligent Offices.
Please let me know right away if you will be able to attend.
-Jutta
(jutta@coyote)
∂20-Apr-90 1042 VAL Bondarenko on Yale shooting
To: JMC@SAIL.Stanford.EDU, reiter@cs.toronto.edu,
morris@INTELLICORP.COM, ginsberg@SUNBURN.Stanford.EDU,
"cv00@utep.bitnet"@FORSYTHE.STANFORD.EDU
CC: andrei@CSLI.STANFORD.EDU
Dear John, Ray, Paul, Matt and Michael,
Andrei Bondarenko, our visitor from the Soviet Union, made an interesting
observation about the Yale Shooting Problem. He proposes to formalize the
law of inertia using normal defaults with prerequisites:
holds(f,s) : holds(f,result(a,s)) / holds(f,result(a,s)),
not holds(f,s) : not holds(f,result(a,s)) / not holds(f,result(a,s)).
This doesn't look very original; indeed, this is essentially what Ray
suggested in Sec. 1.1.4 of his default logic paper. But what was new
to me is that this formulation, unlike most straightforward encodings
of inertia, doesn't lead to unwanted models. The fact that holds(f,s)
is a prerequisite, rather than the antecedent of the conclusion, forces
us to reason "forward in time." As far as temporal projection is
concerned, this seems to settle the Yale shooting difficulty.
What do you think?
--Vladimir
∂20-Apr-90 1112 jutta@coyote.stanford.edu URGENT AI Division business
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 20 Apr 90 11:12:33 PDT
Received: by coyote.stanford.edu; Fri, 20 Apr 90 10:10:31 PST
Date: 20 Apr 1990 1010-PST (Friday)
From: Jutta McCormick <jutta@coyote.stanford.edu>
To: latombe@coyote.stanford.edu, binford@coyote.stanford.edu,
eaf@SUMEX-AIM.Stanford.EDU, mrg@sunburn.stanford.edu,
ok@coyote.stanford.edu, jmc@sail.stanford.edu, nilsson@cs.stanford.edu,
shoham@hudson.stanford.edu, winograd@csli.stanford.edu,
ginsberg@sunburn.stanford.edu, lifschitz@cs.stanford.edu,
jones@cs.stanford.edu
Cc: jutta@coyote.stanford.edu
Subject: URGENT AI Division business
Jean-Claude would like to have a meeting of the AI Division very soon to
decide who is going to teach which courses next year. Please let me
know right away if you can make it on Wednesday, May 2, 1:30 p.m. The
location will most likely be the Cedar Hall conference room.
--Jutta
∂20-Apr-90 1527 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
AGENT-ORIENTED PROGRAMMING
Yoav Shoham
Stanford University
Monday, April 23, 2:30pm
MJH 252
I will discuss our work on a computational framework called
Agent-Oriented Programming.
As a programming paradigm, AOP can be viewed as an extention of
Object-Oriented Programming. It extends OOP by having modules
not only communicate with one another, but also possess knowledge
and beliefs, choices and abilities, and possibly other notions.
As a logical theory, it extends standard epistemic logics. Beside
temporalizing the K (knowledge) and B (belief) operators, it
introduces operators for choice (C) and ability (A).
In either case the intuition about these mentalistic-sounding
notions is guided by intuition about the commonsense, everyday
concepts, though the actual formal definitions come nowhere close
to capturing the full linguisitic meanings.
∂20-Apr-90 1648 korf@CS.UCLA.EDU Re: May
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 20 Apr 90 16:48:05 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA05886;
Fri, 20 Apr 90 16:48:56 -0700
Message-Id: <9004202348.AA05886@shemp.cs.ucla.edu>
Received: by denali; Fri, 20 Apr 90 17:02:19 pdt
Date: Fri, 20 Apr 90 17:02:19 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 20 Apr 90 0854 PDT <8L7eB@SAIL.Stanford.EDU>
Subject: Re: May
John,
Tuesday evening would be fine. Sorry I couldn't work out the scheduling
problem for a talk.
-rich
∂22-Apr-90 1524 harnad@clarity.Princeton.EDU SPP Symposium/Workshop
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 22 Apr 90 15:21:50 PDT
Received: from Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA21641; Sun, 22 Apr 90 18:04:14 -0400
Received: from cognito.Princeton.EDU by Princeton.EDU (5.58+++/2.32/mailrelay)
id AA13209; Sun, 22 Apr 90 18:03:24 EDT
Received: by cognito.Princeton.EDU (4.1/1.98)
id AA25707; Sun, 22 Apr 90 18:04:17 EDT
Date: Sun, 22 Apr 90 18:04:17 EDT
From: harnad@clarity.Princeton.EDU (Stevan Harnad)
Message-Id: <9004222204.AA25707@cognito.Princeton.EDU>
To: AMR@ibm.com (Alexis Manaster-Ramer), PSYKIMP@vms2.uni-c.dk (Kim Plunkett),
dietrich@bingvaxu.cc.binghamton.edu,
dietrich@bingvaxu.cc.binghamton.edu
(For: Howard Pattee),
granger@uci.BITNET, hayes.pa@xerox.com, jmc@sail.stanford.edu,
ltm@xp.psych.nyu.edu (Larry Maloney), movshon@CMCL2.NYU.EDU,
port@iuvax.cs.indiana.edu, powers@informatik.uni-kl.de (David Powers),
psyirv@umnacvx.BITNET, rey@cs.umd.edu (Georges Rey),
treisman@vax.oxford.ac.uk
Subject: SPP Symposium/Workshop
Here is another Abstract. Please those participants who have
not sent one, do send one so I can circulate it to all
participants in advance. -- Stevan
---
From: TREISMAN@vax.oxford.ac.uk Michel Treisman
The Title (provisional) and informal Abstract on my contribution to the
Symbol Grounding Workshop are:
Categorization as a psychophysical problem.
If the same observer has to categorize the same stimulus on two
different occasions he may make different decisions each time. Why is
this? In some situations the observer will show an excessive tendency
to repeat previous responses to similar stimuli; this is sometimes
referred to as 'assimilation'. At other times he or she may avoid
previous responses: 'contrast'. Why should categorization be so
unreliable? Or what does this observation tell us about the process of
making a judgment? A psychophysical model will be outlined which
provides an explanation for these phenomena in terms of mechanisms
which tend to optimize uncertain judgments, and the relations between
different types of categorization, at different levels of complexity,
will be considered.
∂22-Apr-90 1819 harnad@clarity.Princeton.EDU Neuropsychology of Schizophrenia: BBS Call for Commentators
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 22 Apr 90 18:18:41 PDT
Received: from Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA23234; Sun, 22 Apr 90 21:07:55 -0400
Received: from reason.Princeton.EDU by Princeton.EDU (5.58+++/2.32/mailrelay)
id AA13647; Sun, 22 Apr 90 21:06:49 EDT
Received: by reason.Princeton.EDU (4.1/1.98)
id AA04787; Sun, 22 Apr 90 21:02:35 EDT
Date: Sun, 22 Apr 90 21:02:35 EDT
From: harnad@clarity.Princeton.EDU (Stevan Harnad)
Message-Id: <9004230102.AA04787@reason.Princeton.EDU>
To: harnad@elbereth.rutgers.edu
Subject: Neuropsychology of Schizophrenia: BBS Call for Commentators
Below is the abstract of a forthcoming target article to appear in
Behavioral and Brain Sciences (BBS), an international,
interdisciplinary journal providing Open Peer Commentary on important
and controversial current research in the biobehavioral and cognitive
sciences. To be considered as a commentator or to suggest other appropriate
commentators, please send email to:
harnad@clarity.princeton.edu or write to:
BBS, 20 Nassau Street, #240, Princeton NJ 08542 [tel: 609-921-7771]
Please specify the aspect of the article that you are qualified and
interested to comment upon. If you are not a current BBS Associate,
please send your CV and/or the name of a current Associate who would be
prepared to nominate you.
____________________________________________________________________
The Neuropsychology of Schizophrenia
J Gray,* J. Feldon,** JNP Rawlins,*** DR Helmsley* & AD Smith****
*Institute of Psychiatary, London
** Psychology, Tel Aviv University
*** Psychology, University of Oxford
**** Pharmacology, University of Oxford
A model is proposed for integrating the neural and cognitive aspects
of the positive symptoms of acute schizophrenia using evidence from:
postmortem neuropathology and neurochemistry, clinical and preclinical
studies of dopaminergic neurotransmission, anatomical connections
between the limbic system and the basal ganglia, attentional and other
cognitive abnormalities underlying the positive symptoms of
schizophrenia, specific animal models of some of these abnormalities,
and previous attempts to model the cognitive functions of the
spetohippocampal system and the motor functions of the basal ganglia.
Anatomically, the model emphasises the projections from the
septohippocampal system, via the subiculum and the amygdala to nucleus
accumbens and their interaction with the ascending dopaminergic
projection to the accumbens. Psychologically, the model emphasizes a
failure in schizophrenia to integrate stored memories of past
regularieties of perceptual input with ongoing motor programs in the
control of current perception. A number of recent experiments that
offer support for the model are fully described, including anatomical
studies of limbic-striatal connections, studies in the rat of the
effects of damage to these connections and of the effects of
amphetamine and neuroleptics on the partial reinforcement extinction
effect, latent inhibition and the Kamin blocking effect, and studies
of the latter two phenomena in acuate and chronic schizophrenics.
∂22-Apr-90 2011 skitodie@med.stanford.edu PDP
Received: from med-isg.stanford.edu by SAIL.Stanford.EDU with TCP; 22 Apr 90 20:10:54 PDT
Received: from med.stanford.edu by med-isg.stanford.edu (4.0/inc-1.0)
id AA14074; Sun, 22 Apr 90 00:53:16 PDT
Received: by med.stanford.edu (5.57/Ultrix3.0-C)
id AA20382; Sun, 22 Apr 90 20:11:43 PDT
Message-Id: <9004230311.AA20382@med.stanford.edu>
To: jmc@sail.stanford.edu
Cc: skitodie@med.stanford.edu
Subject: PDP
Date: Sun, 22 Apr 90 20:11:42 -0700
From: skitodie@med.stanford.edu
In speaking in Roy Jones' CS200 seminar a week and a half ago you put
forth two separate arguments against the feasibility of Parallel
Distributed Processing as a neural modelling technique, at least in
its current formulation.
One argument, which I found somewhat convincing, involved one's
ability to immediately learn to pronounce a word which conflicts with
all known rules of pronunciation (e.g. Xiang). However, couldn't this
just suggest that we first check for exceptions, and if none exist
then do a "lookup" in our weight matrix for the given pronunciation?
The second argument dealt with Rumelhart's paper on one's recognition
of rooms based upon the items contained in the room. You argued that
this is inadaquate because some rooms (e.g. a butler's pantry) are so
named due merely to their location. Are you assuming that the
connections needed to add location into our PDP schema would be so
great that the schema would become unfeasible (and thus an incorrect
model of the brain)?
I would appreciate you clarifying these points! Thanks for your time.
Scott Forstall
SkiToDie@med.stanford.edu
∂23-Apr-90 1419 VAL bug in Bondarenko's solution
To: JMC@SAIL.Stanford.EDU, reiter@CS.TORONTO.EDU,
ginsberg@Sunburn.Stanford.EDU,
"cv00@utep.bitnet"@FORSYTHE.STANFORD.EDU
CC: morris@INTELLICORP.COM, andrei@CSLI.STANFORD.EDU
The following message from Paul Morris shows that Bondarenko's solution
doesn't quite work. It would be interesting to know whether this can be
fixed without resorting to nonnormal defaults.
--Vladimir
∂20-Apr-90 1135 MORRIS@IntelliCorp.COM Re: Bondarenko on Yale shooting
Received: from IntelliCorp.COM (INTELLICORP.COM.INTELLICORP.COM) by SAIL.Stanford.EDU with TCP; 20 Apr 90 11:35:09 PDT
Date: Fri 20 Apr 90 11:35:59-PDT
From: Paul Morris <MORRIS@INTELLICORP.COM>
Subject: Re: Bondarenko on Yale shooting
To: VAL@SAIL.Stanford.EDU
Cc: MORRIS@INTELLICORP.COM
In-Reply-To: <8L9JJ@SAIL.Stanford.EDU>
Vladimir,
It seems to me the unwanted model is still there.
Let s1 = result(wait,s0), and s2 = result(shoot,s1).
Suppose we have
(A1) holds(loaded,s0).
(A2) holds(alive,s1).
(A3) holds(loaded,s0) : holds(loaded,s1) / holds(loaded,s1).
(A4) holds(alive,s1) : holds(alive,s2) / holds(alive,s2).
(A5) not holds(loaded,s0) : not holds(loaded,s1) / not holds(loaded,s1).
(A6) not holds(alive,s1) : not holds(alive,s2) / not holds(alive,s2).
(A7) holds(loaded,s1) -> not holds(alive,s2).
Here "->" is material implication.
Note that the unwanted model, with holds(alive,s2), supports
not holds(loaded,s1), by (A7). This blocks (A3). Now there is no
way of concluding not holds(alive,s2), so (A4) is enabled.
Should (A7) be a default instead of an implication?
Paul
-------
∂23-Apr-90 1646 casley@Neon.Stanford.EDU Origins of Lisp
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 23 Apr 90 16:46:27 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA18619; Mon, 23 Apr 90 16:47:16 -0700
Date: Mon, 23 Apr 90 16:47:16 -0700
From: Ross Casley <casley@Neon.Stanford.EDU>
Message-Id: <9004232347.AA18619@Neon.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Origins of Lisp
In a recent exchange I wrote:
I understand that pure lisp is what originally interested McCarthy. It
was only later that somebody suggested programming an interpreter for it,
and with that came the motivation for the non-applicative parts of lisp.
I seem to remember that you wrote this somewhere, but couldn't remember
where.
Anyway, this caused some disagreement, so I wondered if you would tell me
what the real situation was.
I won't quote anything you write without your permission.
-Ross
∂23-Apr-90 1646 guha@sumex-aim.stanford.edu meeting
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 23 Apr 90 16:46:27 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA17960; Mon, 23 Apr 90 16:49:32 PDT
Date: Mon, 23 Apr 1990 16:49:29 PDT
From: "Ramanathan V. Guha" <guha@sumex-aim.stanford.edu>
To: jmc@sail.stanford.edu
Subject: meeting
Message-Id: <CMM.0.88.640914569.guha@sumex-aim.stanford.edu>
I am going to be in town from about the 3rd to the 10th of May. Do you have
any free time when we could meet and talk about some of the stuff I have been
doing on contexts recently?
Thanks
Guha
∂24-Apr-90 1027 @Score.Stanford.EDU:james@cs.rochester.edu a question
Received: from Score.Stanford.EDU by SAIL.Stanford.EDU with TCP; 24 Apr 90 10:27:21 PDT
Received: from cayuga.cs.rochester.edu by SCORE.STANFORD.EDU with TCP; Tue 24 Apr 90 10:29:10-PDT
Received: from slate.cs.rochester.edu by cayuga.cs.rochester.edu (5.61/r) id AA12160; Mon, 23 Apr 90 18:27:35 -0400
Received: by slate.cs.rochester.edu (5.61/s) id AA05693; Mon, 23 Apr 90 18:27:31 -0400
Date: Mon, 23 Apr 90 18:27:31 -0400
From: james@cs.rochester.edu
Message-Id: <9004232227.AA05693@slate.cs.rochester.edu>
To: mccarthy@score.stanford.edu
Subject: a question
John
I just got a call from Morgan Kaufman about the paper
(Mccarthy & Hayes) that they are reprinting in the Readings
in PLanning volume. The new copy
you gave them is the right paper but doesn't have Hayes
name on it. I assumed this was an oversight and had them
add it. Is that right? Let me know if there's any problem
with this
thanks
james
∂24-Apr-90 1118 korf@CS.UCLA.EDU Oops
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 24 Apr 90 11:18:16 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA27168;
Tue, 24 Apr 90 11:19:09 -0700
Message-Id: <9004241819.AA27168@shemp.cs.ucla.edu>
Received: by denali; Tue, 24 Apr 90 11:32:36 pdt
Date: Tue, 24 Apr 90 11:32:36 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: Oops
John,
I forgot that I'm taking a Japanese Class that meets from 7 to 9:30 on Tuesday
nights. Would it be possible to meet early on Tuesday evening? I also don't mind
getting together later, as I recall from IJCAI that you are a bit of a nightowl.
I'll be free most of the rest of Tuesday and Wednesday, though. Sorry about
that, but it completely slipped my mind.
-rich
∂24-Apr-90 1212 VAL Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
THE FRAME PROBLEM:
FROM TOY WORLDS TO A GENERAL THEORY
Vladimir Lifschitz
Stanford University
Monday, April 30, 2:30pm
MJH 252
Ideas related to formalizing action and change are usually tested
on "toy domains," such as the blocks world or the "Yale shooting"
example. We will discuss what appears to be the most promising
available formalization of the blocks world and investigate how
general it is. We will see that that solution is applicable to a
whole class of domains involving situations and actions, of which
the blocks world is only one example. The class is described in
terms of purely formal, mostly syntactic, properties of axioms.
∂24-Apr-90 1920 harnad@clarity.Princeton.EDU Movshon Abstract for SPP
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 24 Apr 90 19:20:29 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA23417; Tue, 24 Apr 90 22:17:46 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA00714; Tue, 24 Apr 90 22:17:47 EDT
Date: Tue, 24 Apr 90 22:17:47 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9004250217.AA00714@cognito.Princeton.EDU>
To: AMR@ibm.com (Alexis Manaster-Ramer), PSYKIMP@vms2.uni-c.dk (Kim Plunkett),
dietrich@bingvaxu.cc.binghamton.edu,
dietrich@bingvaxu.cc.binghamton.edu
(For: Howard Pattee),
granger@uci.BITNET, hayes.pa@xerox.com, jmc@sail.stanford.edu,
ltm@xp.psych.nyu.edu (Larry Maloney), movshon@CMCL2.NYU.EDU,
port@iuvax.cs.indiana.edu, powers@informatik.uni-kl.de (David Powers),
psyirv@umnacvx.BITNET, rey@cs.umd.edu (Georges Rey),
treisman@vax.oxford.ac.uk
Subject: Movshon Abstract for SPP
Date: Tue, 24 Apr 90 10:48:59 EDT
From: tony@cortex.psych.nyu.edu (Tony Movshon)
Invariant stimulus features and the cortical representation of visual
information.
J. Anthony Movshon, Center for Neural Science, New York University,
New York, NY 10003.
The responses of visual cortical neurons depend upon a number of
different features of the visual stimuli that fall within their
receptive fields. In most cases there is no difference in the nature
of the response produced by varying the stimulus along different
dimensions. For this reason, it is commonly recognized that the
firing of an individual neuron cannot be used unambiguously to infer
the character of a particular visual stimulus. Rather, it is necessary
to examine the distribution of activity across a population of
neurons. In considering how the multi-dimensional nature of visual
neural signals might most readily be disambiguated, it seems that
special significance might be attached to those stimulus dimensions
for which particular groups of neurons show an invariant selectivity.
An invariant selectivity is a selectivity for the value of a stimulus
along some dimension that is independent of the value of the stimulus
along other dimensions. For example, the selectivity of neurons in
the primary visual cortex (V1) for such stimulus variables as
orientation and spatial frequency is largely independent of the
precise stimulus conditions used to measure them. On the other hand,
their selectivity for the direction of motion of targets depends in a
complex way on the spatial and temporal composition of the target, and
is therefore not invariant.
In the visual cortex of the macaque monkey, many distinct visual
areas have been identified with electrophysiological and anatomical
techniques. A number of ``lower-order'' cortical areas seem to
contain neurons whose activity is primarily controlled by signals of
retinal origin - prominent among these are areas V2, V3, V4 and MT, as
well as the primary visual cortex, V1. The responses of neurons in all
these areas seem to depend on the same collection of visual stimulus
dimensions, including spatial location and size, contour orientation,
spatial frequency, chromatic composition, drift rate, direction of
motion, and binocular disparity. Neurons in different areas can have
more or less sensitivity to variations in one or another of these
parameters, so that in quantitative terms it may be argued that
signals from one area carry more information than signals from another
about particular stimulus features. It is largely on the basis of
quantitative arguments of this kind that a particular role for one or
another area in a particular aspect of visual processing has been
asserted - qualitative differences in the way that visual signals are
represented in different cortical areas have not received much
attention.
In this paper I will argue that a special significance is, in
fact, attached to the particular stimulus dimensions that are the
subjects of invariant representation within an area. For example,
despite the fact that neurons in V1 carry signals about the direction
and speed of motion of objects, the fact that their invariant
selectivity is for orientation, spatial and temporal frequency makes
it impossible for them to carry invariant information about speed and
direction. Neurons in MT, on the other hand, carry invariant
information about motion, at the expense of losing the invariant
representation of spatial and temporal parameters. Analogous
reorganization of signals about color, stereoscopic depth, and other
stimulus features may explain the existence of other representations
of the visual image in the visual cortex.
∂24-Apr-90 1952 harnad@clarity.Princeton.EDU Granger Abstract for SPP
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 24 Apr 90 19:52:50 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA24032; Tue, 24 Apr 90 22:50:08 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA00800; Tue, 24 Apr 90 22:50:06 EDT
Date: Tue, 24 Apr 90 22:50:06 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9004250250.AA00800@cognito.Princeton.EDU>
To: harnad@learning.siemens.com
Subject: Granger Abstract for SPP
Another Abstract: Those who have not sent theirs, please do! -- SH
Subject: Re: SPP Symposium/Workshop
Date: Tue, 24 Apr 90 18:44:30 -0700
From: granger@ICS.UCI.EDU
tentative title and abstract for the symbol-grounding workshop:
Perceptual memory categorization in primary sensory cortex.
Richard Granger
Center for the Neurobiology of Learning and Memory
University of California, Irvine
Recent results from neurobiological simulation work have led to a novel
hypothesis: that the physiological operation of a primary sensory cortical
area (olfactory (piriform) cortex) automatically organizes learned
perceptual cues into a hierarchical memory (Ambros-Ingerson, Granger and
Lynch, Science, 1990). In the simulation, repetitive perceptual samples
("sniffs", or "glances") of learned cues traverse the constructed
hierarchy, such that initial samples yield relatively coarse-grained
category responses whereas later samples yield increasingly finer-grained
information about the cue. The resulting iterative recognition of cues
shares many characteristics with the robust psychological phenomenon of
"basic levels": within a hierarchically nested set of categories such as
"animal-bird-robin", there is a specific level of abstraction that is more
readily processed (e.g., recognized faster) than the others; "bird" in
this example (Mervis and Rosch, Ann.Rev.Psych., 1981). The correspondence
raises the possibility that aspects of this psychological phenomenon may
arise from fundamental physiological mechanisms in primary sensory cortex.
∂25-Apr-90 1425 lin@Neon.Stanford.EDU Reading committee
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 25 Apr 90 14:25:22 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA15535; Wed, 25 Apr 90 14:26:13 -0700
Date: Wed, 25 Apr 1990 14:26:12 PDT
From: Fangzhen Lin <lin@Neon.Stanford.EDU>
Reply-To: lin@cs.stanford.edu
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: Reading committee
Message-Id: <CMM.0.88.641078772.lin@Neon.Stanford.EDU>
Hi. I wonder whether you are willing to be the other two members of my
thesis reading committee (besides Yoav Shoham). I'll put something (two
pages) in your mail slots for you to read.
Thanks for your time,
-Fangzhen
∂25-Apr-90 1442 VAL re: Reading committee
To: lin@CS.Stanford.EDU
CC: JMC@SAIL.Stanford.EDU
[In reply to message from lin@cs.stanford.edu sent Wed, 25 Apr 1990 14:26:12 PDT.]
Fangzhen,
I'll be happy to be on your committee.
--Vladimir
∂25-Apr-90 1618 cdp!kpeters@labrea.stanford.edu RISKS contribution
Received: from labrea.stanford.edu by SAIL.Stanford.EDU with TCP; 25 Apr 90 16:18:30 PDT
Received: by labrea.stanford.edu; Wed, 25 Apr 90 15:20:12 PST
Date: Wed, 25 Apr 90 15:20:12 PST
From: cdp!kpeters@labrea.stanford.edu
Message-Id: <9004252320.AA26025@labrea.stanford.edu>
To: labrea!jmc@sail.stanford.edu
Subject: RISKS contribution
Cc: kpeters@labrea.stanford.edu
Dear Dr. McCarthy:
Charles Dunlop and Rob Kling are preparing an edited collection on the
social implications of computing, to be published by Academic Press
under the tentative title COMPUTERIZATION AND CONTROVERSY: Value Conflicts
and Social Choices. The collection will contain over 50 articles on such
topics as technological utopianism, privacy, computerization and schooling,
economic dimensions, and gender issues, to name a few. The book will
contain substantial introductions to each of sections (each section
contains 5-8 papers) which will frame the important debates and discuss
additional readings.
As part of the section on "Relationships in Electronic Communities", the
editors would like to include a few selections from the RISKS bboard
which discuss issues surrounding the "rec.humor.funny" bboard
censorship case at Stanford.
We would like to request permission to reproduce your contribution to
Volume 8, Issue 31 (February 27, 1989) of RISKS. We
will also be requesting permission to reproduce the original article
from the San Jose Mercury Times posted in Volume 8, Issue 30 of the
RISKS ("Computer Users Worry That Stanford Sets Precedent"), and
the RISKS responses to the article from Les Earnest and Jerry
Hollombe. May we please have non-exclusive world rights to the
material, in all languages?
We would be pleased to give you a copy of the volume upon publication.
I will be sending you a copy of this letter via snail mail by the end of
this week. In the meantime, I would appreciate it if you would
acknowledge receipt of this e-mail message and indicate your willingness
to give us permission to reproduce your contribution. Thank you.
Sincerely,
Sari Kalin
Associate Editor
Academic Press
124 Mt. Auburn Street
Cambridge, MA 02138
617/876-3901
cdp!kpeters@labrea.stanford.edu
as of May 5, our address will be 955 Massachusetts Avenue, Cambridge, MA
02139.
∂25-Apr-90 1900 rathmann@eclipse.stanford.edu new draft; reference request
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 25 Apr 90 19:00:15 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA26536; Wed, 25 Apr 90 19:01:45 PDT
Date: Wed, 25 Apr 90 19:01:45 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9004260201.AA26536@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: new draft; reference request
Dear Professor McCarthy,
Welcome back, I hope you had a good trip.
I have a new revision of my thesis which includes responses to the
recommendations you and other members of the committee made at my oral
exam. I think it is almost at the point where you can sign it (or
tell me what I need to fix before you can sign). I would greatly
appreciate it if you could take a look at it and see where it stands.
The deficiencies I know about are marked with xx's.
One other request -- can I name you as a reference? As a reading
committee member, I think you are in a good position to evaluate my
work.
Ideally, we should get together some time soon to talk about all of
this. Unfortunately, I'll be traveling starting tomorrow and not
returning until Saturday May 5th, and Pat Simmons tells me that is
just about when you will be leaving on another trip. Can we set
something up for when you return?
If you are reading this tonight, (Wednesday) I would greatly
appreciate a call at (408) 338-7823 regarding whether I can list you
as a reference. One of my first stops is a job interview at the
University of Hawaii, and it would help matters a lot if I could tell
them who my references are.
thanks a lot,
Peter
∂25-Apr-90 2138 lin@Neon.Stanford.EDU re: Reading committee
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 25 Apr 90 21:38:12 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA08200; Wed, 25 Apr 90 21:39:02 -0700
Date: Wed, 25 Apr 1990 21:39:02 PDT
From: Fangzhen Lin <lin@Neon.Stanford.EDU>
Reply-To: lin@cs.stanford.edu
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: Reading committee
In-Reply-To: Your message of 25 Apr 90 2136 PDT
Message-Id: <CMM.0.88.641104742.lin@Neon.Stanford.EDU>
Thanks a lot, -fz
∂25-Apr-90 2143 rathmann@eclipse.stanford.edu Re: new draft; reference request
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 25 Apr 90 21:43:47 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA26781; Wed, 25 Apr 90 21:45:21 PDT
Date: Wed, 25 Apr 90 21:45:21 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9004260445.AA26781@eclipse.stanford.edu>
To: JMC@SAIL.Stanford.EDU
Cc: rathmann@eclipse.stanford.edu
In-Reply-To: John McCarthy's message of 25 Apr 90 2141 PDT <HNc3#@SAIL.Stanford.EDU>
Subject: Re: new draft; reference request
I actually return on the 5th, fairly late I think. I can certainly
call you the day before though, and see if we can arrange something.
-Peter
∂25-Apr-90 2148 rathmann@eclipse.stanford.edu Re: new draft; reference request
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 25 Apr 90 21:48:13 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA26837; Wed, 25 Apr 90 21:49:48 PDT
Date: Wed, 25 Apr 90 21:49:48 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9004260449.AA26837@eclipse.stanford.edu>
To: JMC@SAIL.Stanford.EDU
Cc: rathmann@eclipse.stanford.edu
In-Reply-To: John McCarthy's message of 25 Apr 90 2146 PDT <lNc8I@SAIL.Stanford.EDU>
Subject: Re: new draft; reference request
Ok, somehow this will work out. Are you willing to be a reference
for me?
∂25-Apr-90 2202 rathmann@eclipse.stanford.edu Re: new draft; reference request
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 25 Apr 90 22:02:31 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA26882; Wed, 25 Apr 90 22:04:06 PDT
Date: Wed, 25 Apr 90 22:04:06 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9004260504.AA26882@eclipse.stanford.edu>
To: JMC@SAIL.Stanford.EDU
Cc: rathmann@eclipse.stanford.edu
In-Reply-To: John McCarthy's message of 25 Apr 90 2154 PDT <DNcvy@SAIL.Stanford.EDU>
Subject: Re: new draft; reference request
Thanks.
So, I'll try calling you in the evening of the 4th, and we can take it
from there.
-Peter
∂26-Apr-90 0623 @MCC.COM:ai.guha@MCC.COM re: meeting
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 26 Apr 90 06:23:35 PDT
Received: from pp.aca.mcc.com by MCC.COM with TCP/SMTP; Thu 26 Apr 90 08:24:09-CDT
Posted-Date: Thu, 26 Apr 90 08:23 CDT
Received: by pp.aca.mcc.com (5.51/ACAv4.1i)
id AA11310; Thu, 26 Apr 90 08:21:27 CST
Date: Thu, 26 Apr 90 08:23 CDT
From: R. V. Guha <ai.guha@MCC.COM>
Subject: re: meeting
To: JMC@sail.stanford.edu
In-Reply-To: <DNbfF@SAIL.Stanford.EDU>
Message-Id: <19900426132334.5.GUHA@GAIA.ACA.MCC.COM>
Date: 25 Apr 90 2126 PDT
From: John McCarthy <JMC@sail.stanford.edu>
[In reply to message sent Mon, 23 Apr 1990 16:49:29 PDT.]
The only possibility is the third. How about 2pm?
Ok, I'll be there at 2pm at your office.
Thanks
Guha
∂26-Apr-90 0714 james@cs.rochester.edu re: a question
Received: from cayuga.cs.rochester.edu by SAIL.Stanford.EDU with TCP; 26 Apr 90 07:14:08 PDT
Received: from slate.cs.rochester.edu by cayuga.cs.rochester.edu (5.61/r) id AA25509; Thu, 26 Apr 90 10:15:00 -0400
Received: by slate.cs.rochester.edu (5.61/s) id AA05752; Thu, 26 Apr 90 10:14:57 -0400
Date: Thu, 26 Apr 90 10:14:57 -0400
From: james@cs.rochester.edu
Message-Id: <9004261414.AA05752@slate.cs.rochester.edu>
To: JMC@SAIL.Stanford.EDU
Subject: re: a question
Hmmm - thats funny. I bet its the version from your collected
papers volume, since its not a copy of the original article.
Let me know if you think there's any other problem besides
pats name being missing!
james
∂26-Apr-90 1130 JMC
Linsky at TINLUNCH
∂26-Apr-90 1145 korf@CS.UCLA.EDU Re: Oops
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 26 Apr 90 11:45:09 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA03621;
Thu, 26 Apr 90 11:46:01 -0700
Message-Id: <9004261846.AA03621@shemp.cs.ucla.edu>
Received: by denali; Thu, 26 Apr 90 11:59:32 pdt
Date: Thu, 26 Apr 90 11:59:32 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 25 Apr 90 2133 PDT <HNbm5@SAIL.Stanford.EDU>
Subject: Re: Oops
John,
I think I rather not wait that long to eat dinner, so why don't you join your
colleagues, and we'll plan on getting together after my class ends at 9:30. We
can meet at UCLA or anyplace else that's convenient for you.
-rich
∂26-Apr-90 1255 nilsson@Tenaya.Stanford.EDU Re: renewal of courtesy and consulting faculty
Received: from Tenaya.Stanford.EDU by SAIL.Stanford.EDU with TCP; 26 Apr 90 12:55:17 PDT
Received: by Tenaya.Stanford.EDU (NeXT-1.0 (From Sendmail 5.52)/25-eef) id AA16698; Thu, 26 Apr 90 12:55:08 PDT
Date: Thu, 26 Apr 90 12:55:08 PDT
From: Nils Nilsson <nilsson@tenaya.stanford.edu>
Message-Id: <9004261955.AA16698@Tenaya.Stanford.EDU>
To: John McCarthy <JMC@sail.stanford.edu>
Subject: Re: renewal of courtesy and consulting faculty
Cc: nilsson@Tenaya.Stanford.EDU
Yes, we are updating our list to take these
promotions into acct. (Ted Shortliffe, for example.)
-Nils
∂26-Apr-90 1335 cdp!kpeters@labrea.stanford.edu re: RISKS contribution
Received: from labrea.stanford.edu by SAIL.Stanford.EDU with TCP; 26 Apr 90 13:35:13 PDT
Received: by labrea.stanford.edu; Thu, 26 Apr 90 12:36:58 PST
Date: Thu, 26 Apr 90 12:36:58 PST
From: cdp!kpeters@labrea.stanford.edu
Message-Id: <9004262036.AA01533@labrea.stanford.edu>
To: labrea!JMC@sail.stanford.edu
Subject: re: RISKS contribution
Cc: kpeters@labrea.stanford.edu
Thanks for your response. I will send a copy of your contribution
(along with a copy of the other humor bboard contributions) for you to see
where yours fits in the context of the other pieces we are running.
You can make any necessary modifications.
Thanks again,
Sari Kalin
∂26-Apr-90 1846 nilsson@Tenaya.Stanford.EDU Principia Schedule
Received: from Tenaya.Stanford.EDU by SAIL.Stanford.EDU with TCP; 26 Apr 90 18:46:20 PDT
Received: by Tenaya.Stanford.EDU (NeXT-1.0 (From Sendmail 5.52)/25-eef) id AA16907; Thu, 26 Apr 90 18:45:51 PDT
Date: Thu, 26 Apr 90 18:45:51 PDT
From: Nils Nilsson <nilsson@tenaya.stanford.edu>
Message-Id: <9004270145.AA16907@Tenaya.Stanford.EDU>
To: principia@cs.stanford.edu
Subject: Principia Schedule
Cc: jmc@sail.stanford.edu
Here is the schedule for the Principia speakers for
the rest of Spring Quarter. (Note that the last
date is still open. Note also the deviation from our usual
bi-weekly schedule.) All meetings at 2:30 p.m. in mjh 252.
Everyone welcome. -Nils
--------
May 2: Mike Genesereth: "McCarthy's Idea: The Advice Taker
and Where it Stands Today"
May 9: Jens Christensen (mini-orals)
May 16: Subbarao Kambhampati on PRIAR
May 30: open (last one of Spring Quarter)
∂26-Apr-90 2218 eaf@sumex-aim.stanford.edu Re: quote
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 26 Apr 90 22:18:16 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA25131; Thu, 26 Apr 90 22:21:25 PDT
Date: Thu, 26 Apr 1990 22:21:25 PDT
From: Edward A. Feigenbaum <eaf@sumex-aim.stanford.edu>
To: John McCarthy <JMC@sail.stanford.edu>
Subject: Re: quote
In-Reply-To: Your message of 26 Apr 90 1719 PDT
Message-Id: <CMM.0.88.641193685.eaf@sumex-aim.stanford.edu>
John, the number for that I think appeared in a paper that doug lenat
and I co-authored. I think Doug has revised his uppeer limit on the number of
facts to be 100 million. I'm willing to believe that the number is between
10 and 100 million, but it's not the kind of number I want to stake my
credibility on.
Ed
∂26-Apr-90 2219 eaf@sumex-aim.stanford.edu [John McCarthy <JMC@sail.stanford.edu> : quot]
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 26 Apr 90 22:19:26 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA25139; Thu, 26 Apr 90 22:22:36 PDT
Date: Thu, 26 Apr 1990 22:22:35 PDT
From: Edward A. Feigenbaum <eaf@sumex-aim.stanford.edu>
To: jmc@sail.stanford.edu
Subject: [John McCarthy <JMC@sail.stanford.edu> : quot]
Message-Id: <CMM.0.88.641193755.eaf@sumex-aim.stanford.edu>
Rules are not the only kinds of facts that computers can know.
Ed
---------------
Return-Path: <JMC@sail.stanford.edu>
Received: from SAIL.Stanford.EDU by sumex-aim.stanford.edu (4.0/inc-1.0)
id AB20969; Thu, 26 Apr 90 17:24:45 PDT
Message-Id: <1rOuuy@SAIL.Stanford.EDU>
Date: 26 Apr 90 1720 PDT
From: John McCarthy <JMC@sail.stanford.edu>
Subject: quote
To: eaf@sumex-aim.stanford.edu
Actually, what I wanted to know is whether you
believe that 10 million ordinary rules will do it.
∂27-Apr-90 0113 hoffman@csli.Stanford.EDU IEEE Expert Interview (reprint from last summer)
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 27 Apr 90 01:12:55 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA19011; Fri, 27 Apr 90 01:15:11 PDT
Date: Fri 27 Apr 90 01:15:10-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: IEEE Expert Interview (reprint from last summer)
To: jmc@sail.stanford.edu
Cc: hoffman@CSLI.Stanford.EDU
Message-Id: <641207710.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
As you may remember, the interview which we did last summer has generated
a lot of interest and will be republished in IEEE Expert (as you have
indicated your windicated your willingness.) They would also like to expand it
a little and, if you want to make any changes, change it. I would like to come
by tomorrow sometime (Friday) and ask the questions which I have included in
this note. If not tomorrow, then as early as suits your convenience. As
usual, if you don't like any of the questions, then there is no need to
answer them.
thanks
reid
questions:
(1) Could you give some specific examples of your recent work in
formalizing context and use of speech acts?
(2) Traditionally, AI has had serious difficulties with the so-called
"scaling" problem: many of the tools developed in AI work well for
small cases but not for large. What do you think underlies this
problem and do you think that any promising developments have been
made to account for this problem beyond the development of faster
hardware?
(3) Do you think that Artificial Intelligence needs to have its
theoretical foundations strengthened? For example, do you think that
there needs to be a theoretical reconstruction of terms like
"knowledge" or "representation"? In either case, how would you define
knowledge and representation?
(4) A frequent critique of work in AI is that, since it lacks
scientific foundations, very little of the work builds on previous
work in any direct way. Would you comment on this critique and
suggest some ways that much of the work falls under the same
theoretical rubric?
(5) As a bit of curveball, what would you say that the role of emotion
is in intelligence or symbol manipulation and how might that effect
work in AI?
(6) What role do you think that formal logic has in human thought and
successful AI? Does it actually describe the processes at some
functional level? Is it just a convenience for use in engineering? A
simplification or the real thing?
(7) As a prominent researcher in Artificial Intelligence, what would
you say that intelligence is and how do we know if some system
(biological or silicon) is intelligent? Would you suscribe to the
Turing Test and would you think that it's a variant of behaviorism?
(8) Many AI researchers take implementation of their ideas as proof of
professional respectability: in other words, talk is cheap without
actual implementation. For example, PhD theses and publications in AI
Journals require implementation rather than just focus on the idea.
How do you feel about this emerging test of professional competence?
Should the field adopt it, and if not, what professional barriers
should there be? Put another way, what skills should one expect an AI
researcher to always have at her or his fingertips, that warrant the
status of "AI Professional"?
-------
∂27-Apr-90 0915 harnad@clarity.Princeton.EDU SPP Abstract: Powers
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 27 Apr 90 09:14:58 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA09913; Fri, 27 Apr 90 12:07:33 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA04536; Fri, 27 Apr 90 12:07:33 EDT
Date: Fri, 27 Apr 90 12:07:33 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9004271607.AA04536@cognito.Princeton.EDU>
To: andrewsj@vassar.BITNET
Subject: SPP Abstract: Powers
Date: Fri, 27 Apr 90 16:25:49 MET DST
From: David Powers AG Siekmann <powers@informatik.uni-kl.de>
Metaphor and Symbol
David M.W. Powers
University of Kaiserslautern FRG
Over the last 12 years, simulation research has been undertaken in
Machine Learning of Natural Language and Ontology in in the context of
a particular theoretical framework as made explicit in certain
hypotheses.
The first decade of this research, including a broad cognitive science
review, is presented in a monograph [Powe89]. Recent developments and
consolidation of earlier work are to be found in [Powe90].
The main hypotheses are:
i) that the mechanisms responsible for language
phenomena are more general than is often credited;
ii) that the learning of language is inseparably tied
to the learning of ontology;
iii) that automatic self-organization and hierarchy
formation can gives rise to a basic conceptual framework
based on positive examples;
iv) that the mechanism responsible for language
learning phenomena have a considerable overlap with those
for the rest of our sensory-motor experience;
v) that recognition and generation activities are
logically and at least partially physically separate;
vi) that the recognition components act as critics for
the production components and provide negative
information;
vii) that interaction with the world also provides
implicit negative information;
iix) that contrast and similarity assessment of content
in contexts provide a basic learning mechanism based on
metaphor and paradigm;
ix) that cognitive restrictions not only restrict our
learning capability but the range of natural languages
with the effect that our limitations actually assist the
learning process;
x) that language should be examined from the
perspective of ontological learning in an active
environment;
xi) that the concepts learned at one level are the
symbolic building blocks for another level;
xii) that the exchange of information with the
environment as mediated by our sensory-motor system is not
inherently different in form or representation from our
higher level concepts.
Experiments fall into the following categories:
i) neuro-visual association/learning [Powe84,89]
ii) mixed-mode parsing/learning [Powe84,87,89]
iii) word-class learning [Powe84,87,89]
iv) concurrent parsing/learning [Powe84,87,89]
v) statistical formula learning [Powe84,89,90]
vi) neural net formula learning [Powe84,89,90]
vii) critical formula learning [Powe87,89]
iix) critical semantic learning [Powe87,89]
ix) ontological learning [Powe89]
x) morphological learning [Powe90]
In particular these batteries of experiments are all
performed in multiple contexts, in the sense of one or
more of: sensory vs linguistic modality, level of
hierarchy, or various natural languages.
Of particular relevance in this context are the fact that
similar results were achieved in batteries five and six,
applied to the same task. In battery six, particular
neurons could be associated with words and classes, and
that particular grammatical rules could be associated with
particular synapses. Battery nine made use of a simulated
robot world and learnt both nouns and verbs. Batteries
five, six and ten all showed the importance of boundary
conditions, suggested the extension of the linguistic and
psycholinguistic notions of closed class and pivot class,
and clearly demonstrated the importance of such classes
and conditions for segmentation and in guiding the
learning and concept formation processes in a context
having only positive examples.
These hypotheses and results suggest two possible
interpretations which resolve the symbol grounding
problem:
i) that there are effectively no symbols, as we
experience only translations of a variety of effects which
convey information about a world which we cannot more
directly experience and which may not even exist in the
way we imagine (a view with several centuries worth of
support);
ii) that neural networks are indeed symbolic, as
evidenced by our learning simulations in which neural and
synaptic components can be given a meaningful
interpretation;
iii) that some sort of interaction between the language
system and the domain of discourse is appropriate, if not
indeed necessary.
REFERENCES
[Powe84] David M. W. Powers, Experiments in
Computer Learning of Natural Language, Proc. Aust Comp.
Conf., pp489-499, 1984.
[Powe87] David M. W. Powers, L. Davila, D. M.
Meagher, D. Menyies, Further experiments in Computer
Learning of Natural Language, Proc. Aust. Joint AI
Conf., pp458-468 , 1987.
[Powe89] David M. W. Powers, C. C. R. Turk,
RMachine Learning of Natural LanguageS, Springer Verlag,
London/Berlin, 1989
[Powe90] David M. W. Powers, Automatic learning of
orthological closed classes from positive examples, to be
submitted.
∂27-Apr-90 1019 hoffman@csli.Stanford.EDU re: IEEE Expert Interview (reprint from last summer)
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 27 Apr 90 10:19:03 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA05758; Fri, 27 Apr 90 10:21:17 PDT
Date: Fri 27 Apr 90 10:21:15-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: re: IEEE Expert Interview (reprint from last summer)
To: JMC@SAIL.Stanford.EDU
Message-Id: <641240475.0.HOFFMAN@CSLI.Stanford.EDU>
In-Reply-To: <lOWzi@SAIL.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
I will arrive at 3PM at your office, thanks.
reid
-------
∂27-Apr-90 1634 jlm@lucid.com PhD program
Received: from lucid.com by SAIL.Stanford.EDU with TCP; 27 Apr 90 16:34:28 PDT
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA02142g; Fri, 27 Apr 90 16:32:28 PDT
Received: by bhopal id AA05389g; Fri, 27 Apr 90 16:35:58 PDT
Date: Fri, 27 Apr 90 16:35:58 PDT
From: Jim McDonald <jlm@lucid.com>
Message-Id: <9004272335.AA05389@bhopal>
To: mccarthy@sail.stanford.edu
Subject: PhD program
Is there some time convenient for you when we could talk about my
plans for the PhD program?
Dick said that at the review meeting you said wanted me to "work for
you", which leaves me a little confused whether you meant you wanted
to be my advisor or something else.
If you do plan to be my advisor, I'd like to clarify what's expected
of me, so I can make financial and other arrangements.
For background, I plan to pass most or all of the comprehensive exams
this fall and begin work on the dissertation itself sometime during
that academic year, depending upon how many classes I need to take for
exams I fail. Also, I discovered that I have about 5 terms of
residency credits from several years ago, so theoretically I could
finish rather quickly if I do well on the comprehensive and can focus
effectively on the dissertation.
As of today, I'm free to meet essentially anytime given a few hours
notice.
Thanks,
jlm (Jim McDonald)
∂28-Apr-90 1613 hoffman@csli.Stanford.EDU edited Interview (with a few questions embeddedin [...]
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 28 Apr 90 16:13:05 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA20122; Sat, 28 Apr 90 16:15:22 PDT
Date: Sat 28 Apr 90 16:15:21-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: edited Interview (with a few questions embeddedin [...]
To: jmc@sail.stanford.edu
Message-Id: <641348121.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
Q: Traditionally, AI has had serious difficulties with the so-called "scaling" problem: many of the tools developed in AI work well for small cases but not for large. What do you think underlies this problem?
I am not sure that the so-called "scaling problem" is really the difficulty. Instead, the problem may be that more concepts are required than have been included in the simple examples. However, there is one difficulty related to the scaling problem: we still have not been able to express reasoning heuristics in a declarative way. For example, a theorem prover in geometry cannot be guided by statements similar to what every high school teacher says such as "if you want to prove two triangles congruent and you have pairs of corresponding sides equal, then concentrate on proving that the included angles are equal." This kind of information cannot be told to a theorem proving program except by building it directly into the algorithm.
[Your choice of one of two sentences. First]
This lack of ability to express heuristics declaratively causes some of the scaling problems since___[If you want to add anything here.]
[Second]
So, some of the scaling problems arise from our current inability to express this kind of heuristic declaratively.
Q: Do you think that AI needs to have its theoretical foundations strengthened? For example, do you think that there needs to be a theoretical reconstruction of terms like "knowledge" or "representation"? In either case, how would you define knowledge and representation?
I am not sure that defining knowledge and representation is appropriate. I think that it is appropriate to have axiomatic theories of "a person knows a fact," "a person knows the value of an expression," or "a person knows about some subject." There are probably some corresponding theories related to representation: for example, some theories expressing a correspondence between a database and facts in the world. However, I am not sure that a sentence beginning "knowledge is" or "representation is" is what we need.
Q: A frequent criticism of work in AI is that, since it lacks scientific foundations, very little of the work builds on previous work in any direct way. Would you comment on this criticism?
I don't agree that AI lacks scientific foundations although it certainly needs much more. However, I do agree that often previous work is not directly usable in new work. I am hoping to make some dent in this problem with a theory of context, but that would exhaust the scope of this interview.
Q: What would you say that the role of emotion is in intelligence or symbol manipulation and how might that effect work in AI?
I don't agree with Johnson-Laird that emotion needs to play any fundamental role in Machine Intelligence [capitalization?], although computer programs that have to interact with human beings may certainly have to know something about emotions.
Q: What role do you think that formal logic has in human thought and successful AI? Does it actually describe the processes at some functional level? Is it just a convenience for use in engineering? A simplification or the real thing?
The original goal of formal logic was to systematize human thought and to create formalisms in which thought can be done better than people usually do it. The extent to which human thought can be considered formal is a problem on which I have not worked and wish the people who are doing it good luck. However, my previous discussion related to ascribing mental qualities to machines and Newell's knowledge level says that we can understand the behavior of people often by ascribing beliefs and goals to them.
Q: What would you say that intelligence is and how do we know if some system (biological or silicon) is intelligent? Would you suscribe to the Turing Test and would you think that it's a variant of behaviorism?
I woudn't ascribe to the Turing Test per se. In any case, the Turing Test would only be a sufficient rather than a necessary condition for intelligence. As our understanding of the components of intelligence grows, we will begin to describe systems in terms of which components they embody. I would suscribe to the Turing Test only in the following sense: I lowing sense: I would have very little in common with someone who could not be convinced by any empirical evidence that a system was intelligent. I should also remark that, as described in Turing's paper, the Turing Test is ambiguous because it does not specify who tests the machine and what this person knows about intelligence and Artificial Intelligence. It's certainly possible to fool people and it has already been done simply by using techniques similar to fortune-tellers' and psychics', by distracting the person in conversation with the machine into some area in which he supplies all of the information inadvertantly.
Q: Many AI researchers take implementation of their ideas as proof of professional respectability: in other words, talk is cheap without actual implementation. How do you feel about this emerging test of professional competence? Should the field adopt it, and if not, what professional barriers should there be? Put another way, what skills should one expect an AI researcher to always have at her or his fingertips, that warrant the status of "AI Professional"?
Requiring implementation for professional respectability goes too far. No such requirement is made in other sciences In physics and in chemistry, the experimenters and theorists respect each other. On the other hand, there are other fields in which the applicability of theory to experiment is much less and in which theorists often propound propositions that are not subject to experimental test. In AI, the situation is mixed. Both theoretical and experimental ideas need to be published and part of the problem of the field is to get closer connections between theory and experiment. As to the term "AI professional", this is appropriate for the engineering aspect of AI and not appropriate for the scientific aspect of AI.
-------
∂28-Apr-90 1615 hoffman@csli.Stanford.EDU unedited interview
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 28 Apr 90 16:15:24 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA20229; Sat, 28 Apr 90 16:17:39 PDT
Date: Sat 28 Apr 90 16:17:37-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: unedited interview
To: jmc@sail.stanford.edu
Message-Id: <641348257.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
(1) Could you give some specific examples of your recent work in formalizing context and use of speech acts?
I will be publish papers on these topics shortly.
(2) Traditionally, AI has had serious difficulties with the so-called "scaling" problem: many of the tools developed in AI work well for small cases but not for large. What do you think underlies this problem and do you think that any promising developments have been made to account for this problem beyond the development of faster hardware?
People refer to the scaling problem, but I am not sure that that is really the difficulty. Maybe the problem is that more concepts are required than have been included in the simple examples. However, there is one difficulty related to the scaling problem and that is that we still have not been able to express reasoning heuristics in a declarative way. For example, a theorem prover in geometry cannot be guided by the user saying what every high school teacher says. If you want to prove two triangles congruent and you have pairs of corresponding sides equal, then concentrate on proving that the included angles are equal. This cannot be told to a theorem proving program except by building it into the theorem-proving program, so I think that some of the scaling problems are due to the fact that we still don't have a way of expressing this kind of heuristic in a declarative way.
(3) Do you think that Artificial Intelligence needs to have its theoretical foundations strengthened? For example, do you think that there needs to be a theoretical reconstruction of terms like "knowledge" or "representation"? In either case, how would you define knowledge and representation?
I am not sure that defining knowledge and representation is appropriate. I think that it is appropriate to have axiomatic theories of a person knows a fact or a person knows the value of an expression or a person knows about some subject, and there are probably some corresponding theories related to representation: for example, some theories expressing a correspondence between a database and facts in the world. However, I am not sure that a sentence beginning "knowledge is" or "representation is" is what we need.
(4) A frequent criticism of work in AI is that, since it lacks scientific foundations, very little of the work builds on previous work in any direct way. Would you comment on this criticism and suggest some ways that much of the work falls under the same theoretical rubric?
I don't agree that AI lacks scientific foundations. Although, certainly it needs a lot more. However, I do agree that often previous work is not directly usable in new work. I am hoping to make some dent in this with an aid of a theory of context, but that goes too far right now or rather that would take me further than I am prepared to go right now.
(5) As a bit of curveball, what would you say that the role of emotion is in intelligence or symbol manipulation and how might that effect work in AI?
I don't agree with Johnson-Laird that emotion needs to play any fundamental role in Machine Intelligence, although computer programs that have to interact with human beings may certainly have to know something about emotions.
(6) What role do you think that formal logic has in human thought and successful AI? Does it actually describe the processes at some functional level? Is it just a convenience for use in engineering? A simplification or the real thing?
The original goal of formal logic was to systematize human thought and make formalisms in which thought can be done better than people usually do it. The extent to which ordinary human thought can be considered formal is a problem that I haven't worked much on and wish the people who are doing it good luck. However, my previous discussion related to ascribing mental qualities to machines and Newell's knowledge level show that (or say that) we can understand the behavior of people often by ascribing beliefs and goals to them.
(7) As a prominent researcher in Artificial Intelligence, what would you say that intelligence is and how do we know if some system (biological or silicon) is intelligent? Would you suscribe to the Turing Test and would you think that it's a variant of behaviorism?
I woudn't ascribe to the Turing Test per se. In any case, I think that the Turing Test would only be a sufficient condition for intelligence rather than a necessary condition. It seems to me that as we understand more and more of the components of intelligence, we will begin to describe systems in terms of which components they embody. I would suscribe to the Turing Test only in the following sense: that someone who could not be convinced by any empirical evidence that a system was intellegent is someone with whom I have very little in common. Certainly, the Turing Test is a strong version of that. It also should be remarked that, as described in Turing's paper, the Turing Test is ambiguous because it doesn't say who tests the machine and what this person knows about intelligence and artificial intelligence. It's certainly possible to fool people and it has already been done. Simply by using the techniques that fortune-tellers and psychics use of distracting the person being talked to into some area i
n which he supplies all of the information inadvertantly.
(8) Many AI researchers take implementation of their ideas as proof of professional respectability: in other words, talk is cheap without actual implementation. For example, PhD theses and publications in AI Journals require implementation rather than just focus on the idea. How do you feel about this emerging test of professional competence? Should the field adopt it, and if not, what professional barriers should there be? Put another way, what skills should one expect an AI researcher to always have at her or his fingertips, that warrant the status of "AI Professional"?
Taking the attitude that implementation is required for professional respectability is going too far. This...no such requirement is made in other sciences, in physics and in chemistry, the experimenters and theorists have a lot of respect for one another and the astronomers have respect for the theoretical cosmologists. On the other hand, there are other fields in which the applicability of theory to experiment is much less and where theorists often propound propositions in such a way that isn't subject to experimental test. In AI, the situation is mixed and both theoretical and experimental ideas need to be published and part of the problem of the field is to get closer connections between theory and experiment. As to the term "AI professional", this is appropriate for the engineering aspect of AI and not appropriate for the scientific aspect of AI.
-------
∂28-Apr-90 1637 hoffman@csli.Stanford.EDU two things
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 28 Apr 90 16:37:28 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA20645; Sat, 28 Apr 90 16:39:43 PDT
Date: Sat 28 Apr 90 16:39:42-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: two things
To: jmc@sail.stanford.edu
Message-Id: <641349582.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
(1) I was presuming that you didn't want line-feeds. If so, let me know and
I will reconvert the document.
(2) My spelling checker is out of order, so I wasn't able to spell check it
before I sent it to you, but I will before I send it down.
thanks
reid
-------
∂28-Apr-90 1959 mommyduc@med.stanford.edu Reid Hoffman and the Dinkelspiel Award
Received: from med-isg.stanford.edu by SAIL.Stanford.EDU with TCP; 28 Apr 90 19:59:12 PDT
Received: from med.stanford.edu by med-isg.stanford.edu (4.0/inc-1.0)
id AA27472; Fri, 27 Apr 90 23:09:13 PDT
Received: by med.stanford.edu (5.57/Ultrix3.0-C)
id AA25916; Sat, 28 Apr 90 20:00:05 PDT
Date: Sat, 28 Apr 90 20:00:05 PDT
From: mommyduc@med.stanford.edu (Frank Chen)
Message-Id: <9004290300.AA25916@med.stanford.edu>
To: jmc@sail.stanford.edu
Subject: Reid Hoffman and the Dinkelspiel Award
Hi, my name is Frank Chen, and I am an undergraduate majoring in Symbolic
Systems. I'm writing you this note to let you that I am nominating Reid
Hoffman for the Dinkelspiel Award. This annual award recognizes undergraduates
who have made outstanding contributions to undergraduate education.
The more I get involved with SSP, the more Reid's dedication is evident.
The Forum, the new Senior Honors Seminar, and TAing SSP 20 are a brief
sample of his work.
I am writing you because your name comes up often during his impassioned
dinner conversations about representation or "what does it mean to be
digital?" or Saussure . . . If you have some time, would you consider
writing a note for the selection committee?
Though the official deadline has long since passed, the Committee is
still accepting recommendations. Their address is:
Committee on Academic Appraisal and Achievement
Subcommittee on Departmental Honors
Old Union, Room 139
Stanford, CA 94305
I can be reached at this account or at 328-9252 if you have any questions.
Thanks for your time,
Frank Chen
∂29-Apr-90 1730 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
THE FRAME PROBLEM:
FROM TOY WORLDS TO A GENERAL THEORY
Vladimir Lifschitz
Stanford University
Monday, April 30, 2:30pm
MJH 252
Ideas related to formalizing action and change are usually tested
on "toy domains," such as the blocks world or the "Yale shooting"
example. We will discuss what appears to be the most promising
available formalization of the blocks world and investigate how
general it is. We will see that that solution is applicable to a
whole class of domains involving situations and actions, of which
the blocks world is only one example. The class is described in
terms of purely formal, mostly syntactic, properties of axioms.
∂29-Apr-90 2157 hoffman@csli.Stanford.EDU previous messages
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 29 Apr 90 21:57:20 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA17649; Sun, 29 Apr 90 21:59:36 PDT
Date: Sun 29 Apr 90 21:59:35-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: previous messages
To: jmc@sail.stanford.edu
Cc: hoffman@CSLI.Stanford.EDU
Message-Id: <641451575.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
Unfortunately, I have to send out those proofs fairly soon, which is why
I sent them to you. I hope that everything was fine with you, although there
was one point in which I thought you might expand your answer. (I thought
that I understood, but it wasn't my place to put any words in your mouth.)
I will also get the publisher to send you a copy when it's published.
thanks
reid
-------
∂29-Apr-90 2158 hoffman@csli.Stanford.EDU fairly soon in the previous letter means tomorrow evening.
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 29 Apr 90 21:58:29 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA17728; Sun, 29 Apr 90 22:00:47 PDT
Date: Sun 29 Apr 90 22:00:46-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: fairly soon in the previous letter means tomorrow evening.
To: jmc@sail.stanford.edu
Message-Id: <641451646.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
thanks again and I hope that the interview generates as much interest as it
did the first time.
reid
-------
∂30-Apr-90 0800 JMC
Nixon about evaluation, T visit.
∂30-Apr-90 0800 JMC
combination lock
∂30-Apr-90 0824 carlstea@sierra.Stanford.EDU re: Parking
Received: from sierra.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 Apr 90 08:24:39 PDT
Received: by sierra.Stanford.EDU (4.1/4.7); Mon, 30 Apr 90 08:25:10 PDT
Date: Mon, 30 Apr 90 08:25:10 PDT
From: carlstea@sierra.Stanford.EDU (Mary L. Carlstead)
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: Parking
In-Reply-To: Your message of 27 Apr 90 1729 PDT
Message-Id: <CMM.0.88.641489105.carlstea@>
PLEASE make your concerns known to the Parking & Transportation Committee.
∂30-Apr-90 1040 susan@nessus.stanford.edu Parking
Received: from nessus.stanford.edu by SAIL.Stanford.EDU with TCP; 30 Apr 90 10:40:33 PDT
Received: by nessus.stanford.edu (5.57/inc-1.2)
id AA12111; Mon, 30 Apr 90 10:39:23 PDT
Date: Mon, 30 Apr 90 10:39:23 PDT
From: susan@nessus.stanford.edu (Susan Gere)
Message-Id: <9004301739.AA12111@nessus.stanford.edu>
To: hk.jpf@forsythe.stanford.edu
Subject: Parking
Cc: jmc@sail.stanford.edu
(I wanted to make sure you counted Prof JMc's comment:)
> From JMC@SAIL.Stanford.EDU Fri Apr 27 17:35:21 1990
> Return-Path: <JMC@SAIL.Stanford.EDU>
> Received: by umunhum.stanford.edu (5.57/inc-1.2)
> id AA17588; Fri, 27 Apr 90 17:35:19 PDT
> Received: from SAIL.Stanford.EDU by sierra.Stanford.EDU (4.1/4.7); Fri, 27 Apr 90 17:30:53 PDT
> Message-Id: <vO#HE@SAIL.Stanford.EDU>
> Date: 27 Apr 90 1729 PDT
> From: John McCarthy <JMC@SAIL.Stanford.EDU>
> Subject: re: Parking
> To: carlstea@SIERRA.STANFORD.EDU, ee-faculty@SIERRA.STANFORD.EDU,
> ee-adminlist@SIERRA.STANFORD.EDU
> Status: R
>
> [In reply to message from carlstea@sierra.Stanford.EDU sent Fri, 27 Apr 90 15:01:12 PDT.]
>
> The University's new parking policy is probably based on the
> popular, though false, idea that the individual transportation
> provided by the automobile is on its way out. It is appropriate
> for engineering departments to try to persuade Stanford to be
> more objective about this rather than just joining a bandwagon.
For myself, I find the idea of
"c. Reducing parking supply consistent with the goal of reducing demand"
simply ludicrous; perhaps the University should check this novel idea of
reducing demand by reducing supply with some of the award-winning minds
in our own Economics department...
(and be sure to forward to the State Department the good news that we
can reduce shipments of food to starving nations, just as restricting
access to birth control reduces the number of unwanted pregnancies...)
-smg
∂30-Apr-90 1330 JMC
Lowood re tape copying
∂30-Apr-90 1403 harnad@clarity.Princeton.EDU SPP: Powers abstract (shortened)
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 30 Apr 90 13:58:33 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA16254; Mon, 30 Apr 90 16:49:05 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA06397; Mon, 30 Apr 90 16:53:34 EDT
Date: Mon, 30 Apr 90 16:53:34 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9004302053.AA06397@reason.Princeton.EDU>
To: andrewsj@vassar.BITNET
Subject: SPP: Powers abstract (shortened)
Date: Mon, 30 Apr 90 20:39:42 MET DST
From: David Powers AG Siekmann <powers@informatik.uni-kl.de>
To: Stevan Harnad <harnad@clarity>
Cc: andrewsj@VASSAR.BITNET
Metaphor and Symbol
David M.W. Powers
University of Kaiserslautern FRG
The presentation is based on the monograph [Powe89] on Machine Learning
of Natural Language and Ontology, updated with recent developments and
indication of consolidation of earlier work into a coherent approach.
Hypotheses concerning the role of contrast and similarity, and the
relationship of these mechanisms to linguistic concepts of metaphor and
paradigm, neural self-organization, psycholinguistic paradoxes
concerning negative information, and consideration of language as part
of the entire ontology have lead to a series of experiments in machine
learning of aspects of language both individually or in combination.
The work makes use of a simulated robot world as well as textual input.
Significantly, similar results have been achieved with neural and
conventional techniques applied to the same task, with simulated
neurons being clearly associated with words and classes, and with
particular grammatical rules being associated with particular
synapses.
These results suggest three possible resolutions of the symbol
grounding problem: the symbol/non-symbol distinction is not meaningful;
neural networks can exhibit 'symbolic' behaviour and structure; and, a
sensory-motor environment can provide grounding.
REFERENCES
[Powe89] David M. W. Powers, C. C. R. Turk,
Machine Learning of Natural Language, Springer Verlag,
London/Berlin, 1989.
∂30-Apr-90 1525 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Signatures
Received: from IU.AI.SRI.COM by SAIL.Stanford.EDU with TCP; 30 Apr 90 15:25:15 PDT
Received: from peabody.teleos.com by IU.AI.SRI.COM via SMTP with TCP;
Mon, 30 Apr 90 15:25:00-PST
Received: by peabody.teleos.com (3.2/4.16) id AA01114 for
jmc@sail.stanford.edu; Mon, 30 Apr 90 15:23:37 PDT
Date: Mon, 30 Apr 90 15:23:37 PDT
From: Leslie Kaelbling <leslie@teleos.com>
Message-Id: <9004302223.AA01114@peabody.teleos.com>
To: jmc@sail.stanford.edu
Subject: Signatures
Reply-To: leslie%teleos.com@ai.sri.com
When will you be available in the second half of May to sign the final
copies of my dissertation?
- L
∂30-Apr-90 2054 hoffman@csli.Stanford.EDU IEEE Interview
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 Apr 90 20:54:04 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA06548; Mon, 30 Apr 90 20:56:18 PDT
Date: Mon 30 Apr 90 20:56:17-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: IEEE Interview
To: jmc@sail.stanford.edu
Cc: hoffman@CSLI.Stanford.EDU
Message-Id: <641534177.0.HOFFMAN@CSLI.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
Professor McCarthy,
I'm sorry to rush this so, but the publisher told me that he needs the
data soon. Have you had a chance to look over the edited copy? I need to
send it to him soon (he wanted it this evening, but I guess that I might be
able to wait until tomorrow.)
thanks
reid
-------
∂30-Apr-90 2130 weening@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 Apr 90 21:30:20 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA29497; Mon, 30 Apr 90 21:31:11 -0700
Date: Mon, 30 Apr 90 21:31:11 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005010431.AA29497@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
>From: clarinews@clarinet.com (ISABELLE CLARY, UPI Business Writer)
Newsgroups: clari.biz.economy.world,clari.news.hot.east_europe,clari.tw.computers,clari.biz.top
Subject: Computer association restores ties with Soviet Union
Keywords: non-usa economy, economy, computers, manufacturing
Date: 30 Apr 90 21:28:24 GMT
Location: soviet union
NEW YORK (UPI) -- The Association for Computing Machinery said
Monday it was lifting a 13-year-old boycott of computer technology
exchanges with the Soviet Union because glasnost has helped restore some
scientific freedom and respect for human rights in the communist
country.
Bryan Kocher, president of the independent scientific society which
represents more than 80,000 computer experts worldwide, informed the
Soviets about ACM's decision in a letter to Evgely Vehilor, head of the
Computer Science Department at the Academy of Sciences in Moscow.
``It is my pleasure to inform you that ACM's Council has rescinded
its prohibition on cooperation with the Soviet Union, which was enacted
in 1977,'' Kocher said in his letter to Vehilor.
``ACM's recognition of the return of the Soviet computer scientists
to the full fellowship of the international scientific community is a
direct result of re-establishment of acceptable levels of scientific
freedom for Soviet computer scientists, improvement in Soviet respect
for the human rights of its citizens and the citizens of other countries
and reductions of world tension, which will help computer scientists
concentrate on their peaceful pursuits,'' the letter said.
ACM's members voted in favor of boycotting all computer meetings in
the Soviet Union as well as scientific exchanges with the Communist
country in 1977, following a merciless crackdown on Soviet scientists
who had dared voice opposition to the Kremlin's suppression of basic
freedoms. During the 1970s, many Soviet scientists regarded as
dissidents were sent to labor camps or confined to psychiatric asylums.
Among them was Anatoly Shcharansky who was deported to a
concentration camp but was later allowed to emigrate to Israel after the
Kremlin came under intense international pressure in the case. The
scientist later changed his name to Natan Sharansky.
``The council of the ACM voted unanimously to rescind the boycott
which had been imposed in 1977 as a response to the persecution of
Anatoly Shcharansky and other Soviet dissident scientists,'' a spokesman
for the association said.
At the time ACM issued its boycott, Nobel Prize winner Andrei
Sakharov praised the association's move and predicted it would be an
effective way to pressure the Soviet Union into giving up its repressive
policy because the Soviets lagged behind the U.S. in the computer field
and needed foreign technology exchanges.
``At the time Sakharov said `there is nothing to induce them (the
Soviets) so factually and effectively as a refusal to maintain this
cooperation,''' the ACM spokesman added.
Under the economic reforms introduced by leader Mikhail Gorbachev,
the Soviet Union drew in late 1988 a new joint-venture legislation aimed
at attracting foreign investment in the high-technology sector. The
legislation aims not only at attracting foreign capital but also at
ensuring the transfer of Western technology to the Soviet Union.
∂30-Apr-90 2209 hoffman@csli.Stanford.EDU re: IEEE Interview
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 Apr 90 22:09:33 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA08729; Mon, 30 Apr 90 22:11:52 PDT
Date: Mon 30 Apr 90 22:11:51-PDT
From: Reid Hoffman <HOFFMAN@CSLI.Stanford.EDU>
Subject: re: IEEE Interview
To: JMC@SAIL.Stanford.EDU
Message-Id: <641538711.0.HOFFMAN@CSLI.Stanford.EDU>
In-Reply-To: <8QBx3@SAIL.Stanford.EDU>
Mail-System-Version: <SUN-MM(242)+TOPSLIB(128)@CSLI.Stanford.EDU>
Thanks! I'll have them send you a copy of the final.
reid
-------
∂01-May-90 0000 JMC
Time to reconsider front door color.
∂01-May-90 0800 JMC
Nixon now.
∂01-May-90 0850 wheaton@Sunburn.Stanford.EDU
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 08:50:39 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA07525; Tue, 1 May 90 08:51:34 -0700
Date: Tue, 1 May 90 08:51:34 -0700
From: George Wheaton <wheaton@Sunburn.Stanford.EDU>
Message-Id: <9005011551.AA07525@Sunburn.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 30 Apr 90 1935 PDT <vQAsl@SAIL.Stanford.EDU>
John,
I just tried "whoi"s and found me, but the listing is archaic - I'm listed as
wheaton@polya. Thanks for letting me know, I'll change it.
gw
∂01-May-90 0900 shoham@Hudson.Stanford.EDU reimnder
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 09:00:11 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA05684; Tue, 1 May 90 08:59:01 -0700
Date: Tue, 1 May 1990 8:59:00 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU, val@sail.Stanford.EDU
Subject: reimnder
Message-Id: <CMM.0.88.641577540.shoham@Hudson.Stanford.EDU>
that we meet tomorrow at 9; I'd like to this time hear more from you, on the
details of the envisioned language and perhaps more on per vs il.
Yoav
∂01-May-90 1100 JMC
get Korf and Cubic Math back from Pehoushek
∂01-May-90 1110 RPG Lowell
Do you know his net address?
∂01-May-90 1337 jutta@Sunburn.Stanford.EDU AI Division Meeting May 2
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 13:37:29 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA17170; Tue, 1 May 90 13:34:11 -0700
Date: Tue, 1 May 1990 13:34:10 PDT
From: Jutta McCormick <jutta@Sunburn.Stanford.EDU>
To: latombe@cs.Stanford.EDU, binford@cs.Stanford.EDU,
eaf@SUMEX-AIM.Stanford.EDU, mrg@Sunburn.Stanford.EDU,
ok@coyote.Stanford.EDU, jmc@sail.Stanford.EDU, nilsson@cs.Stanford.EDU,
shoham@hudson.Stanford.EDU, winograd@csli.Stanford.EDU,
jones@cs.Stanford.EDU
Cc: jutta@Sunburn.Stanford.EDU
Subject: AI Division Meeting May 2
Message-Id: <CMM.0.88.641594050.jutta@Sunburn.Stanford.EDU>
Reminder: AI Division Meeting re next year's AI courses tomorrow, Wednesday,
May 2, in MJH 301.
(Please note that the meeting is NOT in Cedar Hall.)
∂01-May-90 1614 lum@portia.stanford.edu SURF
Received: from portia.stanford.edu by SAIL.Stanford.EDU with TCP; 1 May 90 16:14:33 PDT
Received: by portia.stanford.edu (5.59/25-eef) id AA00982; Tue, 1 May 90 16:14:20 PDT
Date: Tue, 1 May 90 16:14:20 PDT
From: Benjamin Lai <lum@portia.stanford.edu>
Message-Id: <9005012314.AA00982@portia.stanford.edu>
To: buch@sumex-aim.stanford.edu, cover@sierra.stanford.edu,
feigenbaum@sumex-aim.stanford.edu, genesereth@score.stanford.edu,
gray@sierra.stanford.edu, jlh@vsop.stanford.edu, jmc@sail.stanford.edu,
latombe@coyote.stanford.edu, luckham@sail.stanford.edu,
macovski@sierra.stanford.edu, shoham@score.stanford.edu,
shortliffe@sumex-aim.stanford.edu
Subject: SURF
Hello. My name is Benjamin Lai. I'm a graduating senior in computer science,
eventually to coterm in computer science. I just recently received something
called the SURF (Summer Undergraduate Research Fellowship) award from the
School of Engineering. What that means is that I'll be able to get $4500 to
support myself this summer from research; $1500 will come from my research
sponsor, and the School of Engineering will cover the other $3000. Yes, they
are very nice indeed. But I have to find a research sponsor this summer, by
this week or so.
I'm interested in a lot of aspects of artificial intelligence, including
representation of knowledge, learning, and machine understanding of real-world
concepts. I'd love to work in an expert systems environment. Also, I would
enjoy working in visual or auditory pattern recognition and signal processing.
I also have some interest in programming environments and programming support
tools, but as Mark Linton may not be here for the summer, I`m somewhat in the
cold in that pursuit.
If you're interested, please send e-mail to me, lum@portia, or call me at
497-0707 (I have an answering machine). I'd prefer the latter, but either is
fine.
Thanks a lot, and have a nice day!
-Ben
∂01-May-90 1617 guha@Neon.Stanford.EDU Thursdays appointment.
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 16:17:28 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA15075; Tue, 1 May 90 16:18:25 -0700
Date: Tue, 1 May 1990 16:18:24 PDT
From: "Ramanathan V. Guha" <guha@Neon.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Thursdays appointment.
Message-Id: <CMM.0.88.641603904.guha@Neon.Stanford.EDU>
I think I wont be able to make it for our appointment on Thursday. The flight
I was scheduled to leave on was overbooked (they changed planes and the new
plane had fewer seats etc.) and I got bumped off the flight (there were also
some problems with my ticket and so on). So I will be comming in Thursday
night only.
Since you told me that that was the only time you had free, I take it I wont
be able to meet you this week or next. I am writting up the stuff I was going
to tell you about and shall drop it off in your office sometime friday.
(Its a formalization of a solution for some of the issues related to context
changing, lifting, nested contexts etc.)
Thanks
Guha
∂01-May-90 1631 jutta@Sunburn.Stanford.EDU Time for AI Division Meeting
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 16:31:20 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA24160; Tue, 1 May 90 16:32:05 -0700
Date: Tue, 1 May 1990 16:32:05 PDT
From: Jutta McCormick <jutta@Sunburn.Stanford.EDU>
To: latombe@cs.Stanford.EDU, binford@cs.Stanford.EDU,
eaf@SUMEX-AIM.Stanford.EDU, mrg@Sunburn.Stanford.EDU,
ok@coyote.Stanford.EDU, jmc@sail.Stanford.EDU,
shoham@hudson.Stanford.EDU, winograd@csli.Stanford.EDU,
jones@cs.Stanford.EDU
Subject: Time for AI Division Meeting
Message-Id: <CMM.0.88.641604725.jutta@Sunburn.Stanford.EDU>
Thanks to a couple of you who pointed out that I didn't include the
meeting time in my reminder.
AI Division meeting, Wednesday, May 2, 1:30 p.m., MJH 301.
--Jutta
∂01-May-90 2158 @MCC.COM:ai.guha@MCC.COM re: Thursdays appointment.
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 1 May 90 21:58:52 PDT
Received: from pp.aca.mcc.com by MCC.COM with TCP/SMTP; Tue 1 May 90 23:59:42-CDT
Posted-Date: Tue, 1 May 90 23:41 CDT
Received: by pp.aca.mcc.com (5.51/ACAv4.1i)
id AA02964; Wed, 2 May 90 00:39:31 CDT
Date: Tue, 1 May 90 23:41 CDT
From: R. V. Guha <ai.guha@MCC.COM>
Subject: re: Thursdays appointment.
To: JMC@SAIL.Stanford.EDU
In-Reply-To: <PRBnI@SAIL.Stanford.EDU>
Message-Id: <19900502044120.7.GUHA@GAIA.ACA.MCC.COM>
Date: 01 May 90 2135 PDT
From: John McCarthy <JMC@SAIL.Stanford.EDU>
[In reply to message sent Tue, 1 May 1990 16:18:24 PDT.]
It looks like Friday at 11 would be ok.
I'll be ther at Friday 11.
Sorry for the trouble.
Guha
∂01-May-90 2321 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 1 May 90 23:21:08 PDT
Received: from Neon.Stanford.EDU by Sunburn.Stanford.EDU with SMTP (5.61+IDA/25-eef) id AA28801; Tue, 1 May 90 23:22:02 -0700
Date: Tue, 1 May 90 23:22:02 -0700
Message-Id: <9005020622.AA28801@Sunburn.Stanford.EDU>
From: The Bill Program <pony-bills@Neon.Stanford.EDU>
Subject: JMC Prancing Pony Bill
Reply-To: pony-bills@Neon.Stanford.EDU
To: John McCarthy <JMC@SAIL.STANFORD.EDU>
Prancing Pony Bill of John McCarthy (JMC) for April 1990 (5/1/1990)
Previous Balance 2.15
Monthly Interest at 1.00% 0.02
Current Charges NONE
---------
TOTAL AMOUNT DUE 2.17
PAYMENT DELIVERY LOCATION: CSD Receptionist.
Make checks payable to: STANFORD UNIVERSITY.
Please deliver payments to the Computer Science Dept receptionist, Bldg. 460.
To ensure proper crediting, please include your ACCOUNT NAME on your
check. If you pay by cash, use the small yellow envelopes provided
and write both your ACCOUNT NAME and the AMOUNT on outside.
Note: The recording of a payment may take up to three weeks after the payment
is made, but never beyond the next billing date. Please allow for this delay.
Bills are payable upon presentation. Interest of 1.00% per month will be
charged on balances remaining unpaid 25 days after bill date above.
An account with a credit balance earns interest of 0.33% per month,
based on the average daily balance.
You haven't paid your Pony bill since 10/1989.
Accounts with balances remaining unpaid for more than 55 days are
considered delinquent and are subject to reduction of credit limit.
Please pay your bill and keep your account current.
∂02-May-90 0856 CLT bing today
There are two small pots of flowers in the kitchen and
a bucket of roses behind Timothy's car seat. Please take
them when you take Timothy to school. You'll want to tip
out some of the water so it won't spill. I want the bucket back.
∂02-May-90 0927 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU Cognology
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 09:27:02 PDT
Received: by Forsythe.Stanford.EDU; Wed, 2 May 90 09:28:20 PDT
Received: from PCSITI.FB20.THD.DA.EUROPE by DDATHD21.BITNET
via GNET with RJE with RCOM ; 02 May 90 18:28:16
Date: Wed May 2 17:18:37 MEZ 1990
From: XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU
Subject: Cognology
To: jmc@sail.stanford.edu
X-Munix-To: jmc@sail.stanford.edu
John,
I am preparing an entry for the Encyclopedia for AI concerning the
term INTELLECTICS as a name for the field that studies Artificial
Intelligence. I know that you proposed the name Cognology around
1977, and wanted to give mention to that fact. Unfortunately, I have
no exact reference. I only was told that you used the name in the
title of a talk in Leningrad. Is there a written document/publication
that I could reference? If yes could you please give me details as far
as you recall them. I very much appreciate your cooperation in this
matter.
Hoping that all is fine with you,
with my best regards
=====================Wolfgang
∂02-May-90 0957 MPS
Eleanor Jadwin called. She will be in her office
this morning. 856-0833
∂02-May-90 1100 JMC
nas material to Ames, also NAS biologists
∂02-May-90 1202 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU re: Cognology
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 12:02:50 PDT
Received: by Forsythe.Stanford.EDU; Wed, 2 May 90 12:04:11 PDT
Received: from PCSITI.FB20.THD.DA.EUROPE by DDATHD21.BITNET
via GNET with RJE with RCOM ; 02 May 90 21:03:45
Date: Wed May 2 19:56:02 MEZ 1990
From: XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU
Subject: re: Cognology
To: JMC@SAIL.Stanford.EDU
X-Munix-To: JMC@SAIL.Stanford.EDU
Thanks for your quick response. May I still mention that you
did propose cognology, even though it did not happen in a publication?
Wolfgang
∂02-May-90 1433 rlm@Hudson.Stanford.EDU AIDOC ABSTRACTS
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 14:33:17 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA12141; Wed, 2 May 90 14:31:50 -0700
Date: Wed, 2 May 1990 14:31:48 PDT
From: "Robert L. Miller" <rlm@Hudson.Stanford.EDU>
To: AI-DOC:;@Hudson.Stanford.EDU
Subject: AIDOC ABSTRACTS
Message-Id: <CMM.0.88.641683908.rlm@Hudson.Stanford.EDU>
Please email me the abstract to your AI Day On Campus talk as soon as
possible. I'll need them by May 29 in order to publish a brochure for
the event.
I will create the document on a Mac, so please send straight text
only. If special equation symbols are necessary, send a hardcopy and I
will try to paste or scan it in.
Abstracts may be up to 1 page in length.
Cheers,
Robert
rlm@hudson
∂02-May-90 1506 gjohn@Neon.Stanford.EDU summer research
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 15:06:45 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA01169; Wed, 2 May 90 15:07:38 -0700
Date: Wed, 2 May 1990 15:07:37 PDT
From: "George H. John" <gjohn@Neon.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: summer research
Message-Id: <CMM.0.88.641686057.gjohn@Neon.Stanford.EDU>
Prof McCarthy,
Perhaps you remember me as the "please please PLEASE give me
an A" student in your VTSS class last quarter. My name is George
John. I'm majoring in CS, I'm roughly a junior now, and I'd very much
like to do some research this summer. At this point, I'd say I'm most
interested in AI and Artificial Life. Basically, any research that's
moving towards making computers / computer programs behave with
intelligence fascinates me.
I wanted to know if you have any room on your general problem
solver project (or other projects that I might be able to contribute
to) that an undergrad (such as myself) might be able to fill. I
remember you discussing some projects that students (undergrads, I
believe) had worked on for you, so I assumed I might have a chance...
Please save me from a boring (albeit well-compensated) job at
IBM this summer!
Thanks for your time,
(While I'm at it, thanks for taking me to lunch at the faculty club
last quarter, too. Actually, I've had no choice but to take your
advice on reading Charniak and McDermott -- it's the required text for
221.)
--George John
∂02-May-90 1522 MPS Toy
Julie, 321-4152, has the special order toy you ordered.
∂02-May-90 1602 harnad@clarity.Princeton.EDU Re: SPP and representations
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 16:02:16 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA04775; Wed, 2 May 90 18:59:51 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA07224; Wed, 2 May 90 18:59:59 EDT
Date: Wed, 2 May 90 18:59:59 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005022259.AA07224@reason.Princeton.EDU>
Subject: Re: SPP and representations
Cc: andrewsj@vassar.BITNET
Rob, sorry for the long delay in responding, I was hit with
an avalanche of deadlines when I returned from my series of talks.
It was really enjoyable talking at IU! Now, you asked:
> I would like to know more about how you plan to run the
> Chinese Room session. For example, what did you ask the
> participants to prepare? Will there be any texts available
> before the meeting from these participants - or abstracts?
> I would like to be reasonably prepared. And I would appreciate
> similar information about others in the grounding group.
I trust that you have been receiving the abstracts as they've been
arriving here. Not everyone has provided an abstract yet, and many
may not have written papers. Perhaps you want to get in touch with
some of the participants directly, to prod them.
There's been one change, though. Alexis Manaster-Ramer has dropped
out of the Searle Symposium. He was a discussant. Now here's
a tricky possibility: IF your presentation can be cast as a link
between the symbol grounding workshop in the morning and the Searle
Symposium in the evening, you may want to serve as discussant
(in the evening) rather than chairman. Thus would be a good idea only
if your talk were not cast as a nonsequitur, but really bridged the
two -- as I believe it is possible and indeed deirable to do. The
link is of course that Searle's missing "intrinsic intentionality" --
the one that makes the Chinese Room empty of meaning -- may in fact be
the nonsymbolic grounding of symbols. I don't want to force the issue
though, so tell me if making a link would be natural to what you want
to talk about. Note that ALL the other participants in the Searle
symposium will (despite some possible lip service to the contrary)
symbolic functionalists, who believe that symbol manipulation alone
already captures meaning.
> I have continued to think about the problem of your stimuli -
> the two representations for length. I dont recall what you
> called them, but one directly represents number as length,
> and the other was a binary numerical coding, where place
> represents, say, powers of 2. As a way of seeing how
> different these two codes are, we can imagine the effects of random
> noise on the code. The place-is-power code would be drastically
> damaged. Each change of one bit can have a large or small effect
> depending on which bit happened to be damaged. In the place-as-length
> code, however, noise is much easier to ignore. Most bit errors
> can be recovered due to physical continuity -
> since most neighboring bits will be the same value. So, it seems
> to me these two representations are VERY different. It
> should matter a lot which one you choose.
>
> But more fundamentally, the problem could be said to be
> that neither of these inputs is grounded. (This is why I look at
> real sounds as stimuli.) Of course, grounding in vision is
> especially hard. And size is a very abstract property of
> the visual field. But the question of the representation
> for inputs really should be determined only by the inputs themselves.
> We should supply only what a receptor surface and accompanying
> hardware provide. That is, the programmer/simulator
> should not have to `choose' a representation for input.
>
> The representation should be the OUTCOME of our simulations.
> We want our networks to discover a representation - one that
> enables the cognitive system to solve the tasks we give it.
> This representation is then grounded in stimulation.
>
> But I still don't have a suggestion for a representation
> for you to use. Maybe `length' is just too abstract a concept.
> Perhaps you need a specific set of data to model. That's what
> I'm trying to do with Chuck Watson's results.
I completely agree that the input should just be what happens at the
receptor surface. The trouble is, these neural nets (actually, these
simuulated neural nets) don't HAVE receptor surfaces: They must be
simulated. What we're looking for in the input "representation" is the
closest simulation of that receptor surface (and whatever analogs of
it are preserved in sensory transduction). So the various
representations of a line are really just trying to capture the
continuous structural properties (analog) of the receptor surface.
The place and line representations -- paired with a bit-mapped,
gaussian or lateral inhibition code -- just give us a way of fishing
around representations to pick the one(s) that behave like human
category learners. They're not to be taken too literally or seriously
quite yet...
Cheers, Stevan
∂02-May-90 1642 carlstea@sierra.Stanford.EDU re: "The Meeting"
Received: from sierra.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 16:42:23 PDT
Received: by sierra.Stanford.EDU (4.1/4.7); Wed, 2 May 90 16:42:49 PDT
Date: Wed, 2 May 90 16:42:49 PDT
From: carlstea@sierra.Stanford.EDU (Mary L. Carlstead)
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: "The Meeting"
In-Reply-To: Your message of 02 May 90 1358 PDT
Message-Id: <CMM.0.88.641691768.carlstea@>
No, I don't but Julie Fremon at 3-9362 (she's Managing Director of
Transportation Programs) would. The proposed increase is 25% - C sticker
at $50 and an A sticker at $200. There will be another hearing, something
I asked for and asked Dean Gibbons to try to arrange - Ken Down did get it
through and it will be at noon on May 10th at Terman Auditorium. Down will
send out announcements.
∂02-May-90 1746 weening@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 17:46:29 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA08660; Wed, 2 May 90 17:47:15 -0700
Date: Wed, 2 May 90 17:47:15 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005030047.AA08660@Gang-of-Four.Stanford.EDU>
To: clt@sail.Stanford.EDU, jmc@sail.Stanford.EDU
I'm sorry that I have not been very productive over the past week.
The reason is that my father is currently in the hospital, with heart
problems, and it is difficult for me to concentrate on work. Today he
was supposed to have a bypass operation, but it was postponed at the
last minute due to complications. In a few days, I hope, things will
be better.
∂02-May-90 2041 jones@Sunburn.Stanford.EDU undergraduate lunch
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 20:41:33 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA21864; Wed, 2 May 90 20:41:01 -0700
Date: Wed, 2 May 1990 20:41:00 PDT
From: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>
To: gupta@cs.Stanford.EDU, g.gorin@macbeth.Stanford.EDU, jmc@sail.Stanford.EDU,
winograd@cs.Stanford.EDU, tajnai@cs.Stanford.EDU, dill@cs.Stanford.EDU,
rwf@sail.Stanford.EDU
Subject: undergraduate lunch
Message-Id: <CMM.0.88.641706060.jones@Sunburn.Stanford.EDU>
Thanks for coming to the lunch today. It's great to see such good support
for the undergraduate program.
Now, if we could only get more undergrads involved!!!
Roy
∂02-May-90 2058 jones@Sunburn.Stanford.EDU re: undergraduate lunch
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 20:58:22 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA21957; Wed, 2 May 90 20:59:19 -0700
Date: Wed, 2 May 1990 20:59:18 PDT
From: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: undergraduate lunch
In-Reply-To: Your message of 02 May 90 2044 PDT
Message-Id: <CMM.0.88.641707158.jones@Sunburn.Stanford.EDU>
Thanks. Actually, I'd already double-checked and sent him a message.
Roy
∂02-May-90 2335 shoham@Hudson.Stanford.EDU teaching ai
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 2 May 90 23:35:02 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA14270; Wed, 2 May 90 23:33:38 -0700
Date: Wed, 2 May 1990 23:33:37 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: latombe@coyote.Stanford.EDU, genesereth@cs.Stanford.EDU,
nilsson@cs.Stanford.EDU, jmc@sail.Stanford.EDU,
feigenbaum@SUMEX-AIM.Stanford.EDU, ok@coyote.Stanford.EDU,
val@sail.Stanford.EDU, ginsberg@cs.Stanford.EDU, jones@cs.Stanford.EDU
Subject: teaching ai
Message-Id: <CMM.0.88.641716417.shoham@Hudson.Stanford.EDU>
In the course of the past few years I've reached a conclusion about teaching
AI, which I felt was too complex to raise at today's brief meeting (which
most of you attended). I thought I'd put them on record now. I'm not sure if
this should lead to a discussion or just die - I guess we'll find out.
When I first got here and taught intro AI, my goals were to:
1. be broad, so that the students get a true picture of the field
2. be intellectually profound; in particular, I didn't want smart
undergrads who were challenged to think deeply in other courses feel
this was shallow by comparison (and one shouldn't confuse intellectual
challenge with either fun or work load)
3. not teach things I thought were garbage; I'm not referring here to work
that's not my cup of tea, like say natural language, but work that I actually
thought was poor (example: most expositions of search are completely ignorant
of the relevant results in graph algorithms)
4. impart the sense of adventure which, even today, sets AI apart from other
areas.
Although that wasn't the only reason my first course didn't work too well, I
now think this is an impossible conjunction. And when I look around, I don't
see "intro to theory", despite the fact that theory is much narrower than AI.
My conclusion is that in order to not trivialize AI one needs several, say
three, courses, that between them will cover most of the territory in a
substantial way. One could in addition have a shallower survey course such as
CS123.
Straw-man candidates for the courses:
- formal methods
- reasoning (including applications such as NL, planning and learning), and
- knowledge-based systems.
In my small way I followed these conclusions, and have developed two courses.
The one concentrates on reasoning techniques and applications, which I happen
to do through Prolog (logic is not mentioned even once in that course, which
may be going overboard). One student came up to me and said: this is like 221,
except you're giving us more details! That's exactly what I'd like. The other
course is knowledge rep, and rather heavy on logic.
I realize that changing a course structure in such a radical way is difficult,
and that right now we may well be better off optimizing 221. But that doesn't
change my opinion that the structure is fundamentally flawed.
Comments?
Yoav
∂03-May-90 0748 HF.SXM@Forsythe.Stanford.EDU re: There is more to it
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 3 May 90 07:48:37 PDT
Date: Thu, 3 May 90 07:49:59 PDT
To: jmc@sail.stanford.edu
From: "Sara MacGilvra" <HF.SXM@Forsythe.Stanford.EDU>
Subject: re: There is more to it
REPLY TO 05/03/90 00:17 FROM JMC@SAIL.STANFORD.EDU "John McCarthy": re: There
is more to it
I'm sure that is true
To: JMC@SAIL.STANFORD.EDU
∂03-May-90 1258 jim@kaos.Stanford.EDU Re: Bikes and Bikes
Received: from kaos.Stanford.EDU by SAIL.Stanford.EDU with TCP; 3 May 90 12:58:36 PDT
Received: by kaos.Stanford.EDU (4.1/inc-1.0)
id AA17635; Thu, 3 May 90 12:56:43 PDT
Message-Id: <9005031956.AA17635@kaos.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: jim@kaos.Stanford.EDU
Subject: Re: Bikes and Bikes
In-Reply-To: Your message of 03 May 90 11:36:00 PDT.
<1rSprU@SAIL.Stanford.EDU>
Date: Thu, 03 May 90 12:56:42 PDT
From: Jim Helman <jim@kaos.Stanford.EDU>
High risk riders, such as young men who seek speed and death defying
thrills (and not always successfully so), are disproportionately
represented among motorcyclists. Many greenhorn riders with speed
fetishes have met their ends on Skyline road. So these statistics do
not necessarily reflect the relative safety of the vehicles
themselves. I suspect that most of the motorcycle deaths occur in
high speed situations which are hopefully not common on campus.
In the context of the discussion (low speed traffic on campuses), a
more relevant statistic would be the number of injuries and deaths
among bicyclists and motorcyclists occurring on this or other suburban
campuses. The numbers should also be normalized to reflect the actual
usage of the vehicles in on campus miles per day.
Actually, the safety comparison is probably not that relevant to
actual decision making. If you live near campus, bicycles are much
convenient, since they can be used on the inner campus. If you live a
long distance away, you may not have the time or stamina to bicycle
in.
-jim
∂03-May-90 1300 LES re: Parking
To: JMC@SAIL.Stanford.EDU, su-etc@SHELBY.STANFORD.EDU
John McCarthy says:
> The cost estimate of $30K per car for a parking structure strikes me
> as exaggerated, e.g. some creative bookkeeper loaded every possible
> cost on it. Do you have a breakdown?
I recall seeing such a figure for some recent parking structure but don't
remember where it was. I called the Stanford Planning Office this morning
to find out what they have been paying and was told it is $12-13K, though
that dosn't include the value of the land.
After discussing amortization costs, JMC says:
> If this is the real cost, they should eventually charge it, and expect
> eventually to pay employees $1000 more per year than they otherwise would.
>
> Of course, doing it this way rather than making parking free and
> paying less, gives about 30 percent to the Feds and California as
> well as generating quite substantial bookkeeping costs.
John overlooks the fact that subsidizing car drivers by providing free
parking would put the incentives in the wrong place -- bicycle commuters,
users of public transportation, and car poolers would be treated the
same as those who drive individual smog generators and use up the limited
parking areas.
> However, the current dominant ideology would not like this, because
> it regards imposing inconvenience on people for so-called environmental
> reasons as a virtue.
Damn straight.
> Finally, I'm willing to bet money that when the oil crunch finally
> comes, we will still manage to have cars as good as those we have
> now, powered by electric batteries or liquid hydrogen.
I share John's hope and belief that alternative technologies will appear
that can sustain some kind of reasonably efficient transportation network,
but I am afraid that the economics will turn out to be not so good as what
we have now, requiring substantial adjustments in the way we do things.
> The environmental ideologists will try to prevent this, but unless they
> succeed all over the world, progress will win again. I can't imagine the
> U.S. public agreeing to be austere when even one other country actually
> solves the problem.
Tsk, tsk. There are those pesky "environmental ideologists" again, on the
wrong side of every issue. I have noticed that they always seem to be
made of straw, an appropriately non-contaminating substance.
-Les Earnest
∂03-May-90 1456 rabin%humus.huji.ac.il@Forsythe.Stanford.EDU fall term visit
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 3 May 90 14:56:44 PDT
Received: by Forsythe.Stanford.EDU; Thu, 3 May 90 14:58:06 PDT
Received: by HUJIVMS (HUyMail-4j4); Fri, 04 May 90 00:27:10 +0300
Received: by humus.huji.ac.il (HU-2.1)
id AA02733; Thu, 3 May 90 19:13:48 +0300
From: michael Rabin <rabin%humus.huji.ac.il@Forsythe.Stanford.EDU>
Date: Thu, 3 May 90 19:13:48 +0300
Message-Id: <9005031613.AA02733@humus.huji.ac.il>
To: jmc@sail.stanford.edu
Subject: fall term visit
Cc: rabin
Hi John,
E-mail to Jerusalem (via Harvard) is not reliable
and I am not sure that you got my messages since
I had no ack.
Here are a few points concerning your visit.
1. Housing. There is a woman at Harvard Real Estate,
Joan Canzanelli, with whom I spoke. She can be
reached 1-800-252-5020 and will help you to find
a small place if you come alone ,or a house/ house
swap if you come with family.
2. CV. the deans office needs a cv+list of publications
to process your appointment. This is one of those
formalities that bereaaucrcies(?) thrive on.
If they have already contacted you please ignore
this, otherwise have your secretary send the material
to Martha Mooney, DAS Pierce Hall, Harvard University,
Cambridge MA,02138.
3. There is still time to arrange and announce the
Vinton Hays Lectures. I thought about one lecture
per week for about 4-5 weeks. To get people into
a routine it is probably best to have it in
consecutive weeks. What do you think about such
an arrangement?
Please ack. message. Warm regards Michael
∂03-May-90 1638 Mailer Motorcycle safety
Received: from akbar.cac.washington.edu by SAIL.Stanford.EDU with TCP; 3 May 90 16:38:00 PDT
Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu
(5.61/UW-NDC Revision: 2.12 ) id AA25830; Thu, 3 May 90 16:38:52 -0700
Date: Thu, 3 May 1990 15:43:59 PDT
From: Mark Crispin <MRC@CAC.Washington.EDU>
Sender: Mark Crispin <mrc@Tomobiki-Cho.CAC.Washington.EDU>
Subject: Motorcycle safety
To: John McCarthy <JMC@sail.stanford.edu>
Cc: su-etc@SAIL.Stanford.EDU
In-Reply-To: <1rSprU@SAIL.Stanford.EDU>
Message-Id: <MailManager.641774639.7590.mrc@Tomobiki-Cho.CAC.Washington.EDU>
Here are some more facts and figures about motorcycle accidents, to lay to
rest several of the inaccurate or downright wrong things posted here:
. 92% of all motorcycle accidents involve riders who are inexperienced (have
ridden less than a year) and who have had no formal training (e.g. taken the
Motorcycle Safety Foundation course).
. Most motorcycle accidents involve unlicensed riders.
. In most cases of serious injury, the motorcyclist was not wearing adequate
protective clothing, eye protection, or a helmet.
. Nationally, 50% of all motorcycle fatalities involve alcohol. In the cities
I am personally aware of, *all* (100%) of the recent motorcycle fatalities
in the past few years have involved alcohol.
. The usual pre-crash speed is about 31 MPH. Crash speed is about 22 MPH.
These speeds are within the capability of bicycles and mopeds. The notion
that slow-speed traffic is safer is quite wrong; the safest place for any
vehicle is on the freeway at freeway speeds. In other words, lowering the
speed limit does not do much to solve accidents.
. At the other end of the speed spectrum, there are a number of fatal
accidents in excess of 90 MPH, which is an illegal speed on any public road.
Such accidents can not be blamed on the mode of transportation -- the blame
rests squarely with the operator.
. Over 2/3 of all accidents involve a car violating the motorcycle's right-of-
way. The single most common accident is a car illegally turning left in
front of a motorcyclist.
. In most car/motorcycle accidents, the car driver claims that he didn't see
the motorcycle. This is why daytime use of headlights is mandatory in most
states, and also why many bikers use their high-beams during the day.
. Another violation of a motorcyclist's right of way is a car forcibly trying
to share the lane with the motorcycle.
. Helmets will abate head injuries in low-speed crashes. They do nothing for
neck injuries. Helmets can interfere with a motorcyclist's ability to see
or hear nearby hazards. Mandatory helmet laws are not a solution (although
I agree that bikers who don't wear helmets are good organ donors).
. Motorcyclists wearing helmets and adequate protective clothing tend to walk
away crashes, even those at high speed.
What this all means is that the bikers' are right; the overwhelming majority
of accidents happen to an identifiable class of riders (often calls "squids")
who take completely unnecessary risks, and that furthermore inattentive (to be
charitable) automobile drivers are the biggest external risk.
There is nothing intrinsically dangerous about motorcycling compared to other
forms of transportation. Motorcycles are safe, unless the bike hits something
or something hits the bike.
Many of the same things can be said about other forms of transportation.
Bicyclists tend to have more of an attitude problem than motorcyclists,
particularly in their belief that traffic controls don't apply to them. Car
drivers tend to let themselves get wrapped up in their cocoons, paying
inadequate attention to what is going on around them (I've noticed this
particularly with Volvo drivers! Apparently, just as Japanese "rocket" bikes
tend to attract a certain mindset, so do Volvos...).
Personally, I would like to see a change in our licensing laws so that, as in
several European countries, individuals under 21 can not get automobile
driving licenses but can get licenses for mopeds (first step) and smaller
motorcycles (second step). It is insane that we allow 15 year olds to drive!
I think too that some of these automobile drivers who are hostile to
motorcyclists (and make their hostility clear) would behave differently if
their son or daughter was on the road on two wheels.
In conclusion, yes, there are too many motorcycling accidents. As I showed
above, most of these accidents happen to a certain class of riders who have an
attitude problem. A secondary cause is car drivers who have an attitude
problem. An attitude readjustment on all sides would greatly reduce the
number of accidents.
None of this alters the fact that with the correct attitude, motorcycling *is*
a viable option for the commuter.
Respectfully yours,
-- Mark --
-------
∂03-May-90 1722 LES re: Bikes and Bikes
John McCarthy says:
> There's a discrepancy here. If there are 78.1 million bicyclists
> and 2.3 million bicycles are sold per year, this suggests that
> the average bicycle is 34 years old, which seems to me extremely
> doubtful. Maybe the 2.3 million refers to American manufactured
> bicycles.
Upon further investigation, I see a deeper inconsistency. According
to the Bicycle Manufacturers' Association, total unit sales in 1985
was 11.4 million, composed of 5.8 million domestic and 5.6 million
imported. The bicycle market has not crashed, so I cannot reconcile
these numbers with John's later report of 2.3 million.
-Les Earnest
∂03-May-90 1848 Mailer re: Motorcycle safety
Received: from akbar.cac.washington.edu by SAIL.Stanford.EDU with TCP; 3 May 90 18:48:37 PDT
Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu
(5.61/UW-NDC Revision: 2.12 ) id AA28012; Thu, 3 May 90 18:49:34 -0700
Date: Thu, 3 May 1990 18:39:45 PDT
From: Mark Crispin <MRC@CAC.Washington.EDU>
Sender: Mark Crispin <mrc@Tomobiki-Cho.CAC.Washington.EDU>
Subject: re: Motorcycle safety
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: su-etc@SAIL.Stanford.EDU
In-Reply-To: <rStjv@SAIL.Stanford.EDU>
Message-Id: <MailManager.641785185.7590.mrc@Tomobiki-Cho.CAC.Washington.EDU>
In <rStjv@SAIL.Stanford.EDU>, John McCarthy writes:
>Therefore, I consider it not prudent for me to drive
>motorcycles, because I don't see how I can avoid occasional inattention.
That is perhaps true. Motorcycling does require continuous attention. On the
other hand, I don't believe that operating any motor vehicle inattentively is
OK under any circumstances.
>Mark's statistic that 92 percent of accidents occur in people's first
>year of motorcycle riding has two interpretations. Perhaps within
>a year people learn HOW to ride motorcycles, or perhaps within a year
>people learn NOT to ride motorcycles.
The 92% figure is not only the first year of riding, but also *without any
formal instruction*. Not either-or. Both-and. A damn good argument for
formal training.
>My acquaintances who gave up motorcycles after injury did not seem to
>me to be reckless people.
Motorcycling itself is fun. The problem is that you do get quite frustrated
after a while having to dodge idiots on four wheels who neither know, nor
care, that you're there.
-------
∂04-May-90 0700 JMC
Harvard
∂04-May-90 0800 MPS
Bring in Phil. of B. Russell to be returned to Tanner
∂04-May-90 0800 JMC
cleaning
∂04-May-90 0800 JMC
t to bkfst?
∂04-May-90 1003 ginsberg@Sunburn.Stanford.EDU Re: teaching ai
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 May 90 10:01:24 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA26740; Fri, 4 May 90 10:02:05 -0700
Date: Fri, 4 May 90 10:02:05 -0700
From: Matthew L. Ginsberg <ginsberg@Sunburn.Stanford.EDU>
Message-Id: <9005041702.AA26740@Sunburn.Stanford.EDU>
To: feigenbaum@SUMEX-AIM.Stanford.EDU, genesereth@cs.Stanford.EDU,
ginsberg@cs.Stanford.EDU, jmc@sail.Stanford.EDU, jones@cs.Stanford.EDU,
latombe@coyote.Stanford.EDU, nilsson@cs.Stanford.EDU,
ok@coyote.Stanford.EDU, shoham@Hudson.Stanford.EDU,
val@sail.Stanford.EDU
Subject: Re: teaching ai
Yoav's suggestion that 221 be split into three courses strikes me as a
terrible idea for two reasons. The important one is that I disagree
*completely* with the idea that we shouldn't teach an intro AI course;
the second is that I don't think Yoav's four goals were an impossible
conjunct at all.
The reason there is no intro to theory course is exactly that theory *is*
narrow; there isn't a large community that would be interested in a survey
course. There are intro courses in physics, chemistry, psychology, etc,
because these disciplines have created enough interest that there are people
who are interested in learning the material but aren't prepared to make a
3-quarter commitment to it. Some of the people interested in AI will be
at the 123 level; others will be at the 221 level. To ignore either group
is to ignore our responsibilities to the school of engineering as a whole.
As far as Yoav's four goals, he lists:
1. Broad coverage
2. Intellectual depth
3. Don't teach garbage
4. Impart a sense of adventure
Needless to say, those were my goals as well when I taught 221. It is
also the case that the *first* time I taught it, I screwed it up. I
worked my tail off on the course, but there was just too much stuff I
had to learn about teaching in general and teaching AI in particular.
The second time I taught 221, I worked my butt off again. There were
still problems, but the areas in which my tau beta scores were the highest
were on things like how much the course helped the students learn to think,
and I think they had a good time and got a very good sense of the adventure
that makes AI special. I was especially glad to see that they were having
fun in light of the fact that the course was a lot of work.
Rich Korf, incidentally, has had experiences similar to mine in
teaching intro to AI at UCLA. The course is apparently viewed
(however they "score" them) as one of the better CS courses there.
The bottom line is that I think it's possible -- but hard work -- to
teach intro to AI well, and that it's one of the principal responsibilities
of the faculty here to make sure that we do it.
Matt
∂04-May-90 1502 eaf@sumex-aim.stanford.edu Re: teaching ai
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 4 May 90 15:02:40 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA03226; Fri, 4 May 90 15:05:32 PDT
Date: Fri, 4 May 1990 15:05:31 PDT
From: Edward A. Feigenbaum <eaf@sumex-aim.stanford.edu>
To: Matthew L. Ginsberg <ginsberg@sunburn.stanford.edu>
Cc: feigenbaum@sumex-aim.stanford.edu, genesereth@cs.stanford.edu,
ginsberg@cs.stanford.edu, jmc@sail.stanford.edu, jones@cs.stanford.edu,
latombe@coyote.stanford.edu, nilsson@cs.stanford.edu,
ok@coyote.stanford.edu, shoham@hudson.stanford.edu,
val@sail.stanford.edu
Subject: Re: teaching ai
In-Reply-To: Your message of Fri, 4 May 90 10:02:05 -0700
Message-Id: <CMM.0.88.641858731.eaf@sumex-aim.stanford.edu>
I think CS123, as I teach it, satisfies Yoav's criteria as well as
possible in a very short (ten week) course (ten weeks = twenty sessions=
twenty five hours; that's not even the amount of time of a one-week
industrial seminar/training-course!). The "intellectual depth" in my course
I regard as "reasonable". I would prefer to cover more of the multitudinous,
disparate topics of AI, then go into a lot of detail about a few of them.
I feel you have to trust the students' ability to learn on their own on ce
you open some doors for them.
Ed
p.s. one of these days I will get around to circulating the course prospectus
for CS123 to you folks so you can make assessments for yourselves. I think
CS123 is a much-maligned course among the AI faculty. But the students seem
to like it a lot. I get good feedback from the students.
∂04-May-90 1546 VAL Commonsense and Nonmonotonic Reasoning Seminar: No meeting
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
There will be no meeting of the nonmonotonic seminar on May 7.
--Vladimir
∂04-May-90 1633 shoham@Hudson.Stanford.EDU courses, round 2
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 May 90 16:33:23 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA24253; Fri, 4 May 90 16:27:51 -0700
Date: Fri, 4 May 1990 16:27:50 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: nilsson@cs.Stanford.EDU, latombe@coyote.Stanford.EDU,
JMC@sail.Stanford.EDU, val@sail.Stanford.EDU, ginsberg@cs.Stanford.EDU,
genesereth@cs.Stanford.EDU, winograd@cs.Stanford.EDU,
ok@coyote.Stanford.EDU, feigenbaum@SUMEX-AIM.Stanford.EDU,
jones@cs.Stanford.EDU
Subject: courses, round 2
Message-Id: <CMM.0.88.641863670.shoham@Hudson.Stanford.EDU>
There seems to be a beginning of a discussion, so let me continue it a little
bit. There have been three reactions:
Roy: My suggestion may have some merit in principle, but 1. it is possible to
do a better job of 221, and 2. practically speaking, there is a class of
students who cannot be required to take more than one AI course.
Matt: My suggestion has no merit. We need to teach an intro course as does e.g.
psych, and with a lot of hard work it is possible to achieve all the goals
I listed.
Ed: 123 is a more substantial course than is sometimes perceived.
These ideas make me modify my position slightly. I now agree that there is a
need for some intro course. However, this should not be a 200-level course.
Indeed, if you look at the other departments mentioned, you'll note that all
the intro courses are 100-level. And so it should be. If the idea is that
this might be the only AI course some students take, then they should have the
information to decide that it be the only course early on. I see no reason why
123 shouldn't serve this function. In particular, I see no reason why it
shouldn't be taken simultaneously by people who have no interest in continuing
in AI, and by people who are open to that possibility. The latter will then
continue to take the more detailed courses, for which my straw-man proposal
still stands: formal methods (KR emphasis), reasoning (including applications)
and knowledge-based systems. These will all be 200-level courses.
This would also address my concern that for some students 221 is too shallow in
relation to comparable courses. In my original message I didn't make a
distinction that I should have. I have no doubt that there are students,
perhaps a large number of them, for which 221 is currently adequate and
enjoyable when well taught. These are the people that would enjoy it also as
123. However, there are others that are currently turned off. I'm most
concerned with PhD students who think they want to do AI although they've
never seen it before, and find 221 insulting. Matt, I think you're a gifted
public speaker, very knowledgeable, and have taught the best 221 in recent
history (which includes my teaching it). Nonetheless also after your course
at least one PhD student told me that he found it too trivial, and decided to
do theory. I also heard second hand of others who were turned off by the
excessive logic, and the lack of philosophical discussion.
Again, this is anything but your fault. In the meeting Vladimir said that he
wasn't sure why, but he too felt the special difficulty of teaching 221. I
don't think it's merely the usual difficulty of teaching a new course; it has
to do with trying to cover a whole lot of ground and maintain depth. It's no
accident that all existing Intro AI books are lousy, and that better books,
like Mike and Nils's, are more specialized. Why do most faculty
here end up teaching somewhat more specialized courses? I'm sure you didn't
mean to suggest that this is just because the faculty are lazy and want to
teach only about their own area of expertise: Ed's 123, JC's
325 and my 227 are not the only counterexamples.
So, to summarize, I think the right structure would be 123 or equivalent as
intro, three or so 2xx as core AI for people who'll actually do AI, and then
specializations within the 3xx courses. The practical problem of introducing
major changes still stands, however.
Yoav
∂04-May-90 1717 ginsberg@Sunburn.Stanford.EDU Re: courses, round 2
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 May 90 17:17:31 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA12331; Fri, 4 May 90 17:18:26 -0700
Date: Fri, 4 May 90 17:18:26 -0700
From: Matthew L. Ginsberg <ginsberg@Sunburn.Stanford.EDU>
Message-Id: <9005050018.AA12331@Sunburn.Stanford.EDU>
To: JMC@sail.Stanford.EDU, feigenbaum@SUMEX-AIM.Stanford.EDU,
genesereth@cs.Stanford.EDU, ginsberg@cs.Stanford.EDU,
jones@cs.Stanford.EDU, latombe@coyote.Stanford.EDU,
nilsson@cs.Stanford.EDU, ok@coyote.Stanford.EDU,
shoham@Hudson.Stanford.EDU, val@sail.Stanford.EDU,
winograd@cs.Stanford.EDU
Subject: Re: courses, round 2
It sounds as if Yoav feels that 221, as I taught it, should be called 123,
and the "serious" AI students should be spared the 1-course broad brush
intro. I see two problems with this:
1. We need 123, which I guess Yoav would call 023. We have a responsibility
to nonmajors to tell that what AI is about. A course that has "no programming"
in the course description cannot be our basic intro to AI course.
2. There will always be a large group of people who are interested in AI
but don't want to commit 3 quarters to see if they really like it. They
will presumably take the highest-level 1 quarter course they can find. Some
of them will inevitably be turned off by the course, and others won't be.
But these people will include almost all the master's students, some Ph.D.'s
and a bunch of bright undergrads as well.
However you slice it, we need two intro AI courses, one for serious (and
potentially undecided) computer scientists, and the other for non-CS types.
It seems to me that the first course will be primarily directed at grad
students, and the other at undergrads, which is why the numbering is like
it is. But as long as both courses are offered, I can't imagine it matters
what the numbers really are.
As far as the 3-quarter intro, I don't think we need it. My guess is that
the enrollment would be comparable to the enrollment in the specialty
courses we already have.
Matt
∂04-May-90 1757 winograd@loire.stanford.edu Re: courses, round 2
Received: from loire.stanford.edu by SAIL.Stanford.EDU with TCP; 4 May 90 17:56:55 PDT
Received: by loire.stanford.edu (5.59/25-eef) id AA00485; Fri, 4 May 90 17:59:19 PDT
Date: Fri, 4 May 90 17:59:19 PDT
Message-Id: <9005050059.AA00485@loire.stanford.edu>
From: Terry Winograd <Winograd@csli.stanford.edu>
To: JMC@sail.Stanford.EDU, feigenbaum@SUMEX-AIM.Stanford.EDU,
genesereth@cs.Stanford.EDU, ginsberg@Sunburn.Stanford.EDU,
ginsberg@cs.Stanford.EDU, jones@cs.Stanford.EDU,
latombe@coyote.Stanford.EDU, nilsson@cs.Stanford.EDU,
ok@coyote.Stanford.EDU, shoham@Hudson.Stanford.EDU,
val@sail.Stanford.EDU, winograd@cs.Stanford.EDU
Subject: Re: courses, round 2
Matt says "We have a responsibility to nonmajors to tell that what AI
is about". Are the people in systems, numerical analysis, theory, etc.
shirking their respective responsibilities or is AI some kind of
nontechnical area that deserves different treatment, a la intro psych?
If there is an "AI for poets" course it shouldn't be a show and tell
promotional venture, but a serious discussion of the intellectual
issues of cognition, logic, thought, automation etc, which is based on
readings, discussions and extensive writing. Students who aren't willing
to put in that effort can certainly learn something about AI by picking
up one of the popular books on the field rather than taking a course.
There should also be a beginning technical course, which requires
appropriate mathematical and computer science prerequisites and could
have more depth and rigor than the current 221. I don't see the need
to cater to students who want to dabble in AI without having the
appropriate background, any more than the physics department should
provide technical courses that don't assume an appropriate amount of
calculus and differential equations. This will reduce the profitable
large enrollments some, but it will make the course much more teachable
by letting the instructor build on a pre-existing base of knowledge
and students who want it will get more out of it.
--t
∂05-May-90 0602 qphysics-owner@neat.cs.toronto.edu in search of a moderator
Received: from neat.cs.toronto.edu by SAIL.Stanford.EDU with TCP; 5 May 90 06:01:57 PDT
Received: by neat.cs.toronto.edu id <12772>; Sat, 5 May 1990 09:02:44 -0400
Received: from smoke.cs.toronto.edu ([128.100.1.68]) by neat.cs.toronto.edu with SMTP id <8643>; Sat, 5 May 1990 08:55:06 -0400
Received: from localhost by smoke.cs.toronto.edu with SMTP id 1136; Sat, 5 May 90 08:54:39 EDT
To: qphysics@cs.toronto.edu
Subject: in search of a moderator
Date: Sat, 5 May 1990 08:54:32 -0400
Sender: Jean-Francois Lamy <lamy@cs.toronto.edu>
From: Jean-Francois Lamy <lamy@cs.utoronto.ca>
Message-Id: <90May5.085439edt.1136@smoke.cs.toronto.edu>
Resent-From: qphysics-owner@cs.toronto.edu
Resent-To: qphysics@cs.toronto.edu
Resent-Reply-To: Jean-Francois Lamy <lamy@cs.utoronto.ca>
Resent-Message-Id: <90May5.090244edt.12772@neat.cs.toronto.edu>
Resent-Date: Sat, 5 May 1990 08:57:02 -0400
As of May 30th, I will no longer be in a position to act as moderator for the
qphysics@ai.utoronto.ca mailing list. The list has never been heavily used,
and I have received no update requests in the last few months. The alias can
remain active until such time as a new moderator is found or the plug is
pulled, but obviously updates and maintenance would not be performed. The
current membership is appended, should anyone feel like setting things up.
JFL
addanki@ibm.com (Sanjaya Addanki)
franz@cs.washington.edu (Franz Amador)
bambi@dist.dist.unige.it (Giovanni Adorni)
farlabosse@framentec.fr (Francois Arlabosse)
barrett@cs.washington.edu (Tony Barrett)
berleant@cs.utexas.edu (Dan Berleant)
biswas@vuse.vanderbilt.edu (Gautam Biswas)
bhaskar@ibm.com (R. Bhaskar)
bobrow.pa@xerox.com (Daniel Bobrow)
borchers@bbchd.uucp (Hans Borchers)
olivier@pasteur.bitnet (Olivier Boulot)
bert@swivax.uucp (Bert Bredeweg)
dave@mcc.com (David Murray Bridgeland)
byland@cis.ohio-state.edu (Tom Bylander)
caloud@heron.stanford.edu (Philippe Caloud)
chiu@cs.utexas.edu (Charles Chiu)
ric@cheops.eecs.unsw.oz.au (Enrico Coiera)
agc@cs.warwick.ac.uk (Anthony Cohn)
jcollins@cs.uiuc.edu (John Collins)
daniels@htikub5.bitnet (Henry Daniels)
dangelm@cs.ubc.ca (Heidi Dangelmaier)
adf@ztivax.siemens.com (Angela Dappert)
decoste@cs.uiuc.edu (Dennis DeCoste)
doshita@jpnkyoto.bitnet (Shuji Doshita)
davise@acf1.nyu.edu (Ernie Davis)
dekleer.pa@xerox.com (Johan DeKleer)
jo@arti6.vub.ac.be (Jo Dekuyper)
ddt@dist.dist.unige.it (Mauro di Manzo)
downing@cs.uoregon.edu (Keith Downing)
rdoyle@ai.jpl.nasa.gov (Richard Doyle)
dressler@ztivax.siemens.com (Oskar Dressler)
dvorak@cs.utexas.edu (Dan Dvorak)
falkenhainer.pa@xerox.com (Brian Falkenhainer)
faltings@elma.epfl.ch (Boi Faltings)
farquhar@cs.utexas.edu (Adam Farquhar)
sergio@rvax.ccit.arizona.edu (Reuben O. Fernandez)
forbus@cs.uiuc.edu (Kenneth Forbus)
freitag@ztivax.siemens.com (Hartmut Freitag)
gelsey@cs.yale.edu (Andrew Gelsey)
mg@cidam.oz.au (Mike Gigante)
stone@wheaties.ai.mit.edu (David Goldstone)
hardt@sunybcs.bitnet (Shoshana Hardt)
hayes.pa@xerox.com (Patrick Hayes)
hamilton@cs.sfu.ca (Howard Hamilton)
hogge@csc.ti.com (John Hogge)
dhubbell@cs.washington.edu (David Hubbel)
huberman.pa@xerox.com (Bernardo Huberman)
jhunter@cs.abdn.ac.uk (Jim Hunter)
iro10@ipvian.bitnet (Liliane Ironi)
iwasaki@sumex-aim.stanford.edu (Yumi Iwasaki)
josko@ibm.com (Leo Joskowicz)
kahn@sumex-aim.stanford.edu (Michael Kahn)
didier@arti6.vub.ac.be (Didier Keymeulen)
hkim@cs.uiuc.edu (Hyun-Kyung Kim)
kuipers@cs.utexas.edu (Ben Kuipers)
laasri@cs.umass.edu (Hassan Laasri)
wylee@cs.utexas.edu (Wan Lee)
woodwlee@cs.utexas.edu (Wood Wai Lee)
rrl@ee.hw.ac.uk (Roy Leitch)
liu@fog.cs.uoregon.edu (Zheng-Yang Liu)
macfarlane@hplabs.hp.com (Jane MacFarlane)
maitre@cs.umass.edu (Brigitte Maitre)
makarovic@henut5.bitnet (Andrej Makarovic)
em@pcsbst.uucp (Ewgeni Martschew)
jmc@sail.stanford.edu (John McCarthy)
mcilrait@ai.utoronto.ca (Sheila McIlraith)
murthy@ibm.com (Sesh Murthy)
narain@rand.org (Sanjay Narain)
neville@cs.washington.edu (Dorothy Neville)
nielsen@crd.ge.com (Paul Nielsen)
anil@ibm.com (Anil Nigam)
nishida@jpnkyoto.bitnet (Toyoaki Nishida)
noekel@uklirb.uucp (Klaus Noekel)
jscott@cs.washington.edu (J. Scott Penberthy)
peyralbe@crcge1.cge.fr (Catherine Peyralbe)
nporte@framentec.fr (Nathalie Porte)
cjp@cs.aber.ac.uk (Chris Price)
pu@carcvax.uconn.edu (Pearl Pu)
mkp@cs.utexas.edu (Martin Purvis)
raiman@fribm11.bitnet (Olivier Raiman)
rajamone@cse.usc.edu (Shankar Rajamoney)
dr@cs.warwick.ac.uk (David Randell)
praulefs@sc.intel.com (Peter Raulefs)
rehbold@uklirb.uucp (Robert Rehbold)
roschelle.pa@xerox.com (Jeremy Roschelle)
eps@princeton.edu (Elisha Sacks)
ejs@ida.liu.se (Erik Sandewall)
saraswat.pa@xerox.com (Vijay Saraswat)
sqi@ee.hw.ac.uk (Qiang Shen)
shirley.pa@xerox.com (Mark Shirley)
Reid.Simmons@cs.cmu.edu (Reid Simmons)
shrager.pa@xerox.com (Jeff Shrager)
skorstad@cs.uiuc.edu (Gordon Skorstad)
georg@tumki.informatik.tu-muenchen.de (Georg Strobl)
struss@ztivax.siemens.com (Peter Struss)
cs3@research.abb.arcom.ch (Hiro Sugaya)
sutschet@iitb.fhg.uucp (Gerd Sutschet)
ddt@dist.dist.unige.it (Daniele Tezza)
paddy@cs.qmc.ac.uk (Andrew Toal)
fellow@cs.utexas.edu (Heng Wei Osbert Ton)
gt@cise.it (Giorgio Tornielli)
weld@cs.washington.edu (Daniel Weld)
mpw@zermatt.lcs.mit.edu (Michael Wellman)
bwilliams.pa@xerox.com (Brian Williams)
ulysses!witt@pcsbst.uucp (Jan Witt)
lem@nrcvm01.bitnet (Rob Wylie)
yip@ai.ai.mit.edu (Ken Yip)
∂06-May-90 0707 harnad@clarity.Princeton.EDU SPP Abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 6 May 90 07:07:52 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA12895; Sun, 6 May 90 10:08:43 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA17865; Sun, 6 May 90 10:08:49 EDT
Date: Sun, 6 May 90 10:08:49 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005061408.AA17865@cognito.Princeton.EDU>
To: jmc@sail.stanford.edu
Subject: SPP Abstract
John, could you send a 1-200 word abstract of your talk at the
Searle workshop in June so the SPP can put it in the on-site
program with the others, please. You can send it to me and I
will circulate it to everyone.
chrs, Stevan
∂06-May-90 0723 harnad@clarity.Princeton.EDU SPP Abstracts needed
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 6 May 90 07:23:23 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA13016; Sun, 6 May 90 10:18:38 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA17881; Sun, 6 May 90 10:18:45 EDT
Date: Sun, 6 May 90 10:18:45 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005061418.AA17881@cognito.Princeton.EDU>
To: PSYKIMP@vms2.uni-c.dk (Kim Plunkett), dietrich@bingvaxu.cc.binghamton.edu,
dietrich@bingvaxu.cc.binghamton.edu (For: Howard Pattee),
granger@uci.BITNET, hayes.pa@xerox.com, jmc@sail.stanford.edu,
ltm@xp.psych.nyu.edu
(Larry Maloney), movshon@CMCL2.NYU.EDU,
port@iuvax.cs.indiana.edu, powers@informatik.uni-kl.de (David Powers),
psyirv@umnacvx.BITNET, rey@cs.umd.edu (Georges Rey),
treisman@vax.oxford.ac.uk
Subject: SPP Abstracts needed
Cc: andrewsj@vassar.BITNET
To all contributors to the SPP Searle Symposium
and Symbol Grounding Workshop
If you have not done so already, please send me a 1-200 word abstract
of your presentation. These are needed (urgently) for the on-site
program that will be distributed to all participants. According to my
records, I so far have abstracts from R. Granger, L. Maloney, A.
Movhon, D. Powers, and M. Treisman.
Also, if you have a written version of your paper in advance, or prior
reprints that are relevant, please send them to your discussants
(Symposium) or fellow participants (Workshop) -- the sooner the better,
so that everyone's contribution will be as informated and informative
as possible.
Best wishes,
Stevan Harnad
∂06-May-90 2156 jones@Sunburn.Stanford.EDU Re: courses, round 2
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 May 90 21:56:53 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA25869; Sun, 6 May 90 21:57:41 -0700
Date: Sun, 6 May 1990 21:57:41 PDT
From: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>
To: Yoav Shoham <shoham@Hudson.Stanford.EDU>
Cc: "H. Roy Jones" <jones@Sunburn.Stanford.EDU>, latombe@cs.Stanford.EDU,
nilsson@cs.Stanford.EDU, jmc@sail.Stanford.EDU, val@sail.Stanford.EDU,
ginsberg@cs.Stanford.EDU, genesereth@cs.Stanford.EDU,
winograd@cs.Stanford.EDU, eaf@SUMEX-AIM.Stanford.EDU
Subject: Re: courses, round 2
In-Reply-To: Your message of Fri, 4 May 1990 16:27:50 PDT
Message-Id: <CMM.0.88.642056261.jones@Sunburn.Stanford.EDU>
Yoav,
Now I really do disagree. We definitely need two introductory ai courses,
one for our students and one for interested parties ou;ide of CS and related
technical fields. In fact, physics has 3 introductory courses, one for
'poets', a second for technical mortals, and an honors one.
Roy
∂07-May-90 0020 ok@coyote.stanford.edu AI Qual Schedule - About to be made official
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 7 May 90 00:20:53 PDT
Received: by coyote.stanford.edu; Mon, 7 May 90 00:03:36 PDT
Date: 7 May 1990 0003-PDT (Monday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: ginsberg@sunburn.stanford.edu, latombe@coyote.stanford.edu,
jmc@sail.stanford.edu, nilsson@tenaya.stanford.edu,
rosenschein@teleos.com, shoham@hudson.stanford.edu
Cc: ok@coyote.stanford.edu
Subject: AI Qual Schedule - About to be made official
Here is my attempt for the qual schedule. There will be two exam times:
10:30--12noon and 1:45--3:15. The committees will meet at 3:30 to discuss
grading, and results will be made available by the end of the day.
Please let me know by next Tuesday if there is a problem with this schedule,
and then I will send it to the students.
On each committee there is a (C)hair, a (D)epth examiner, and one (O)ther.
AI Qualification Exam Schedule
Wednesday, June 6, 1990
(Preliminary Version for Approval)
10:30 -- 12noon
Adnan Darwiche: Reasoning
C: Latombe D: Nilsson O: Rosenschein
1:45 -- 3:15
David Ash: Planning
C: Nilsson D: Ginsberg O: Shoham
Michael Young: Formal Models of Inter-Agent Communication
C: Latombe D: McCarthy O: Rosenschein
3:30 -- 4:30
Meeting of Committees
∂07-May-90 1004 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Dissertation
Received: from IU.AI.SRI.COM by SAIL.Stanford.EDU with TCP; 7 May 90 10:04:49 PDT
Received: from peabody.teleos.com by IU.AI.SRI.COM via SMTP with TCP;
Mon, 7 May 90 10:04:35-PST
Received: by peabody.teleos.com (3.2/4.16) id AA04290 for
nilsson@tenaya.stanford.edu; Mon, 7 May 90 10:02:03 PDT
Date: Mon, 7 May 90 10:02:03 PDT
From: Leslie Kaelbling <leslie@teleos.com>
Message-Id: <9005071702.AA04290@peabody.teleos.com>
To: stan@teleos.com, nilsson@tenaya.stanford.edu, jmc@sail.stanford.edu
Subject: Dissertation
Reply-To: leslie%teleos.com@ai.sri.com
Rich Sutton has some problems with my related work chapter and will be out
of town for a month, so it is very unlikely that I will be able to file my
dissertation in time for Spring conferral. I'd still like to do it as
soon as possible, though, so I'd appreciate comments on the draft soon.
Thanks,
Leslie
∂07-May-90 1305 larson@unix.sri.com Re: Bikes and Bikes
Received: from unix.sri.com by SAIL.Stanford.EDU with TCP; 7 May 90 13:05:37 PDT
Received: by unix.sri.com (4.1/SMI-4.0)
id AA24190; Mon, 7 May 90 13:06:25 PDT
Date: Mon, 7 May 90 13:06:25 PDT
From: larson@unix.sri.com (Alan Larson)
Message-Id: <9005072006.AA24190@unix.sri.com>
To: JMC@SAIL.STANFORD.EDU
Subject: Re: Bikes and Bikes
Newsgroups: su.etc
In-Reply-To: <1rSprU@SAIL.Stanford.EDU>
Reply-To: larson@unix.sri.com
Organization: SRI International, Menlo Park, CA
Cc:
In article <1rSprU@SAIL.Stanford.EDU> you write:
>[In reply to message from jim@baroque.Stanford.EDU sent 3 May 90 17:18:07 GMT.]
>
>Jim Helman gives many reasons why motorcycles should be safer
>than bicycles. I can't answer them and decided to cheat by
>looking it up. The Statistical Abstract of the United States
>says that in 1987 4,000 motorcyclists were killed in accidents
>and 900 bicyclists. There were 5.1 million registered motorcycles
>in 1987. 2.3 million bicycles were sold in 1987.
I am a bit disappointed in your statistics. They don't make any
reference to miles traveled, number of trips, or number of units/users.
Since motorcycles travel at much greater speeds for greater distances,
it seems that a simple count is inappropriate.
Alan
∂07-May-90 1616 VAL Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
NEURAL NET REPRESENTATION OF A TRUTH MAINTENANCE SYSTEM
Paul Morris
Intellicorp
Monday, May 14, 2:30pm
MJH 252
We show how label updating for a justification-based TMS may be
accomplished within a neural net model. In this approach, the IN status
of a TMS node or justification is represented dynamically by a
continuously firing neuron, rather than statically as a memory trace.
This novel perspective leads to a way of defining well-foundedness in
terms of local constraints. It also suggests an epistemic
interpretation of a neural net as a kind of "fuzzy" TMS.
∂07-May-90 2105 GLB
To: JMC@SAIL.Stanford.EDU, jussi%hpljak@HPLABS.HP.COM,
sf@CSLI.STANFORD.EDU
NEWS QUIZ: WHAT COUNTRY DOES THIS AP NEWS REFER TO?
----------------------
a281 2101 06 May 90
AM-*****-Elections,0260
Local Elections Test Government, Communist Party
**** (AP) - After a campaign marred by violence, ******** voted in
nationwide local elections Sunday that are seen as an important
indication of support for the governing coalition and the opposition
Communist Party.
At least nine politicians in southern ***** have been killed by
underworld gangs during the campaign. The candidates were either
linked to local scandals or had denounced attempts by the underworld
to obtain government contracts.
-----------------------
(Hint: It's a country of strategic importance for the control
of the Holy Land, where NATO has brought 42 years of political
stability and economic prosperity.)
∂07-May-90 2110 harnad@clarity.Princeton.EDU Hayes's Abstract for SPP
Received: from Princeton.EDU ([128.112.129.115]) by SAIL.Stanford.EDU with TCP; 7 May 90 21:10:14 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA22426; Tue, 8 May 90 00:08:12 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA20038; Tue, 8 May 90 00:08:16 EDT
Date: Tue, 8 May 90 00:08:16 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005080408.AA20038@cognito.Princeton.EDU>
To: harnad@clarity.Princeton.EDU
Subject: Hayes's Abstract for SPP
Cc: andrewsj@vassar.BITNET
To: SPP Symbol Grounding Workshop and Searle Symposium Participants:
Date: Mon, 7 May 90 16:37:11 PDT
From: <hayes@parc.xerox.com>
Searle's Chinese Room
Pat Hayes
The basic flaw in Searle's argument is a widely accepted misunderstanding
about the nature of computers and computation: the idea that a computer
is a mechanical slave that obeys orders. This popular metaphor suggests
a major division between physical, causal hardware which acts, and
formal symbolic software, which gets read. This distinction runs
through much computing terminology, but one of the main conceptual
insights of computer science is that it is of little real scientific
importance. Computers running programs just aren't like the Chinese
room.
Software is a series of patterns which, when placed in the proper
places inside the machine, cause it to become a causally different
device. Computer hardware is by itself an incomplete specification of a
machine, which is completed - i.e. caused to quickly reshape its
electronic functionality - by having electrical patterns moved within
it. The hardware and the patterns together become a mechanism which
behaves in the way specified by the program.
This is not at all like the relationship between a reader obeying some
instructions or following some rules. Unless, that is, he has somehow
absorbed these instructions so completely that they have become part of
him, become one of his skills. The man in Searle's room who has done
this to his program now understands Chinese.
∂08-May-90 1730 hayes@roo.parc.xerox.com LISP syntax
Received: from arisia.Xerox.COM by SAIL.Stanford.EDU with TCP; 8 May 90 17:29:59 PDT
Received: from roo.parc.Xerox.COM by arisia.Xerox.COM with SMTP
(5.61+/IDA-1.2.8/gandalf) id AA27012; Tue, 8 May 90 17:30:57 -0700
Received: by roo.parc.xerox.com
(5.61+/IDA-1.2.8/gandalf) id AA00629; Tue, 8 May 90 17:30:53 PDT
Message-Id: <9005090030.AA00629@roo.parc.xerox.com>
Date: Tue, 8 May 90 17:30:53 PDT
From: <hayes@parc.xerox.com>
Reply-To: hayes@parc.xerox.com
To: jmc@sail.stanford.edu
Subject: LISP syntax
John, greetings. After your fascinating talk at lunch about the origin of
LISP, I found this and thought yo umight like it if its new to you.
PS Lets talk about CYC some time. And - perhaps more urgently - about the
Chinese-room session in a few weeks: do you want us to synchronise our
talks in any way? WE should probably at least be aware of one another.
Pat
--------------------
From: jasmerb@mist.cs.orst.edu (Bryce Jasmer)
Newsgroups: rec.humor.funny
Subject: The Strategic Defense Initiative (SDI/Star Wars)
Keywords: computer, funny
Date: 23 Apr 90 10:30:08 GMT
Through some clever security hole manipulation, I have been able to
break into all of the government's computers and acquire the Lisp code
to SDI. Here is the last page (tail -10) of it to prove that I actually
have the code:
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
----------
∂09-May-90 0905 CLT TTM ss#
624-34-5246
∂09-May-90 1354 @IBM.COM:LEORA@YKTVMH response to workshop proposal
Received: from IBM.COM by SAIL.Stanford.EDU with TCP; 9 May 90 13:54:26 PDT
Received: from YKTVMH by IBM.COM (IBM VM SMTP R1.2.1MX) with BSMTP id 7657; Wed, 09 May 90 13:55:11 PDT
Date: Wed, 9 May 90 16:55:02 EDT
From: LEORA@IBM.COM
To: jmc@sail.stanford.edu, val@sail.stanford.edu
Subject: response to workshop proposal
Dear John and Vladimir,
I sent Peter Patel-Schneider the proposal for the commonsense reasoning
symposium just before I left for vacation. He sent back mail, which
I am forwarding, saying that while the idea is good in principle, it
is just too general, especially point 1. Do you have any ideas on
how to modify this? I will try to phone you tomorrow, so we can
speak about this in person. (Sorry for the delay; Peter responded
a few weeks ago, but I just came back from vacation last week and have
been swamped). I am also enclosing the proposal which I sent him.
Speak to you soon,
Leora
==============================================================================
Received: from research.att.com by IBM.COM (IBM VM SMTP R1.2.1MX) with TCP; Wed, 25 Apr 90 12:16:57 PDT
Received: by research; Wed Apr 25 15:15:19 1990
Date: Wed, 25 Apr 90 15:15:04 EDT
From: pfps@allegra.tempo.nj.att.com (Peter F. Patel-Schneider)
To: LEORA@IBM.COM
In-Reply-To: LEORA@IBM.COM's message of Fri, 6 Apr 90 16:54:20 EDT
Subject: AAAI Symposium proposal
I read your proposal, and also gave it to the other members of the Symposia
Committee.
We like the idea in principle. The set of organizers looks just about
right. However, the proposal, as it now stands, appears to be too general.
Even though you qualify the topic in the last paragraph, your point 1/
could be used to justify anyone working on topics related to commensense
reasoning (including everyone working on nonmontonic logics).
Unfortunately, I don't have any really good ideas on what to replace this
point by. You really need something to replace it, as the last three
points don't make a complete set. Perhaps you should take the last
paragraph, and incorporate it into the beginning of the description.
Look at the call for participation for 1990 to see how long your
description should be. (Actually you have created more of a call for
participation than a proposal, but, since we know both you and the area,
there is really no need to change into more of a proposal.) It appears
that you could add another paragraph without any problems.
You should also think of what sort of information you want. You currently
ask for papers; you should probably indicate that you really don't need
publishable papers, but instead want working papers, extended abstracts, or
whatever of some specific length. Remember that you will have to read them
ALL!
If you have any questions, please send me mail.
Peter F. Patel-Schneider
pfps@research.att.com
==========================================================================
Peter,
Enclosed please find our proposal for the 1991 AAAI Spring Symposium Series.
Please let us know if you have any comments. Thank you very much,
John McCarthy
Vladimir Lifschitz
Leora Morgenstern
===========================================================================
Proposal for the 1991 AAAI Spring Symposium Series
Logical Formalizations of Commonsense Reasoning:
Evaluating Existing Formalisms and Moving on to New Domains
We solicit papers describing original ideas and new results on
expressing commonsense knowledge in formal declarative languages
and on formalizing commonsense reasoning. We are particularly
interested in:
1. Formalizations aimed at better solutions of known difficulties
or raising new ones.
Possible topics include:
- causality,
- relations between knowledge and action,
- knowledge and communication,
- illuminating small domains like the blocks world and extensions,
- common sense physics,
- common sense psychology and
- applications of nonmonotonic reasoning to the above.
2. Extending formalization to new domains.
3. Evaluation and mathematical analysis of the soundness and generality of
formalizations.
4. Mathematical results and counterexamples clarifying relations
between different formalisms.
Relevance to artificial intelligence will be the primary consideration.
New logics advanced for their own sakes will be regarded with suspicion.
Theorems only marginally related to the problem of formalizing common
sense, as well as papers investigating primarily computational problems
(algorithms, complexity, implementation), would not be appropriate for
this symposium either. People considering submissions may wish to
contact one of the organizers by telephone, e-mail or mail to discuss
appropriateness of the topic.
Organizers:
John McCarthy (Chairman)
Computer Science Department
Stanford University
Stanford, CA 94305
Phone: (415)723-4430
E-mail: JMC@CS.STANFORD.EDU
Vladimir Lifschitz
Computer Science Department
Stanford University
Stanford, CA 94305
Phone: (415)723-3334
E-mail: VAL@CS.STANFORD.EDU
Leora Morgenstern
IBM Watson Research Center
P.O.Box 704
Yorktown Heights, NY 10598
Phone: (914)784-7151
E-mail: LEORA@IBM.COM
∂09-May-90 1518 harnad@clarity.Princeton.EDU Optimality: BBS Call for Commentators
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 9 May 90 15:18:31 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA10084; Wed, 9 May 90 18:09:47 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA02403; Wed, 9 May 90 16:06:51 EDT
Date: Wed, 9 May 90 16:06:51 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005092006.AA02403@reason.Princeton.EDU>
To: srh@flash.bellcore.com
Subject: Optimality: BBS Call for Commentators
Cc: connectionists@cs.cmu.edu, jcha@csugreen.BITNET (James Ha)
Below is the abstract of a forthcoming target article to appear in
Behavioral and Brain Sciences (BBS), an international,
interdisciplinary journal providing Open Peer Commentary on important
and controversial current research in the biobehavioral and cognitive
sciences. To be considered as a commentator or to suggest other appropriate
commentators, please send email to:
harnad@clarity.princeton.edu or write to:
BBS, 20 Nassau Street, #240, Princeton NJ 08542 [tel: 609-921-7771]
Please specify the aspect of the article that you are qualified and
interested to comment upon. If you are not a current BBS Associate,
please send your CV and/or the name of a current Associate who would be
prepared to nominate you.
____________________________________________________________________
The Quest for Optimality: A Positive Heuristic of Science?
Paul J. H. Schoemaker
Center for Decision Research
Graduate School of Business
University of Chicago
Chicago, IL 6063
Abstract
This paper examines the strengths and weaknesses of one of science's
most pervasive and flexible metaprinciples: Optimality
is used to explain utility maximization in economics, least effort
principles in physics, entropy in chemistry, and survival of the
fittest in biology. Fermat's principle of least time involves both
teleological and causal considerations, two distinct modes of
explanation resting on poorly understood psychological primitives.
The rationality heuristic in economics provides an example from social
science of the potential biases arising from the extreme flexibility of
optimality considerations, including selective search for confirming
evidence, ex post rationalization, and the confusion of prediction with
explanation. Commentators are asked to reflect on the extent to which
optimality is (1) an organizing priniciple of nature, (2) a set of
relatively unconnected techniques of science, (3) a normative principle
for rational choice and social organization, (4) a metaphysical way of
looking at the world, or (5) something else still.
Key Words:
Optimization, Variational Principles, Rationality,
Explanation, Evolution, Economics, Adaptation, Causality, Heuristics,
Biases, Sociobiology, Control Theory, Homeostasis, Entropy, Regulation.
∂09-May-90 1805 weening@Gang-of-Four.Stanford.EDU names for parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 May 90 18:04:59 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA09337; Wed, 9 May 90 18:05:30 -0700
Date: Wed, 9 May 90 18:05:30 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005100105.AA09337@Gang-of-Four.Stanford.EDU>
To: qlisp@Gang-of-Four.Stanford.EDU
Subject: names for parallel iteration functions
In the paper "Parallel List Mapping Operations" by Ian, Dan, Carolyn
and myself (still unpublished), we defined parallel versions of Common
Lisp's mapping functions and called them QMAPC, QMAPCAR, etc.
The technique is also applicable to any function that iterates over a
sequence, such as DOLIST, DOTIMES, and many of the sequence functions.
We will end up having a lot of parallel functions but I think they
will be a useful programming tool and easy to describe. The question
is what to call them.
My proposal is to uniformly make use of a "Q" prefix to name functions
or macros of this type. I.e., QDOTIMES will be a macro that uses the
technique described in our paper.
The benefits of this proposal are:
- the "Q" prefix is already used for QLET, QLAMBDA, etc., so it fits
in with an existing naming scheme.
- there is only one Common Lisp form beginning with "Q" ("QUOTE"), so
any other name beginning with "Q" is clearly a Qlisp extension.
- the meaning of a "Q" form will always be the same as the
corresponding Common Lisp form, except for the parallelism.
All of these forms should take an argument that decides whether to
create processes, just like QLET. Aside from that, their syntax will
be the same as the corresponding Common Lisp forms to avoid confusion.
Example: QDOTIMES with dynamic process creation would be:
(qdotimes (dynamic) (i 100) (foo i j k))
DYNAMIC is another new name that I would like to propose using -- it
implements the dynamic partitioning method that Dan and I have been
working on. (DYNAMIC) with no arguments creates a process only when
the local processor's queue is empty, which has worked best as a
default on most of the programs we tested. (DYNAMIC n) creates a
process if there are n or fewer processes in the queue. (Only a few
programs have needed a non-zero value of n to get good performance.)
Please comment on this proposal if you have an opinion. I'd like to
make these names "official" so that I can describe them in the Qlisp
primer as tools available to programmers.
∂09-May-90 1912 arg@lucid.com names for parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 May 90 19:12:13 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA09626; Wed, 9 May 90 19:12:50 -0700
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA29757g; Wed, 9 May 90 19:12:21 PDT
Received: by bhopal id AA07665g; Wed, 9 May 90 19:13:05 PDT
Date: Wed, 9 May 90 19:13:05 PDT
From: Ron Goldman <arg@lucid.com>
Message-Id: <9005100213.AA07665@bhopal>
To: weening@Gang-of-Four.Stanford.EDU
Cc: qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Joe Weening's message of Wed, 9 May 90 18:05:30 -0700 <9005100105.AA09337@Gang-of-Four.Stanford.EDU>
Subject: names for parallel iteration functions
Before we add tons of new parallel versions of standard Lisp functions maybe
we should first see how good/bad it is to just use an explicit SPAWN, e.g.
(QMAPCAR (pred) #'baz list)
vs.
(MAPCAR #'(lambda (l) (spawn (pred) (baz l))) list)
seems to be clearer using QMAPCAR, while
(QDOTIMES (pred) (i n) (baz i))
vs.
(DOTIMES (i n) (spawn (pred) (baz i)))
seems about the same. (Note we plan to fix DOTIMES, etc. so that the
above will work as is.) The advantage of using explicit SPAWN's is that
the programmer doesn't have to remember which Lisp constructs have a
parallel version and which (if any?) don't.
Given that we do want to introduce lots of new parallel versions of Lisp
functions then the "Q" prefix idea seems fine, if not inevitable.
Rather than adding a simple form to specify when to execute things in
parallel (e.g. as currently done in QLET), we might want to use an
argument list (e.g. as is now done with SPAWN) so that we can specify
other information (e.g. for-effect, priority, resources, etc) as needed.
Ron
∂09-May-90 1951 weening@Gang-of-Four.Stanford.EDU Re: names for parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 May 90 19:50:56 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA09776; Wed, 9 May 90 19:51:22 -0700
Message-Id: <9005100251.AA09776@Gang-of-Four.Stanford.EDU>
To: Ron Goldman <arg@lucid.com>
Cc: qlisp@Gang-of-Four.Stanford.EDU
Subject: Re: names for parallel iteration functions
In-Reply-To: Your message of "Wed, 09 May 90 19:13:05 PDT."
<9005100213.AA07665@bhopal>
Date: Wed, 09 May 90 19:51:20 PDT
From: Joe Weening <weening@Gang-of-Four.Stanford.EDU>
Date: Wed, 9 May 90 19:13:05 PDT
From: Ron Goldman <arg@lucid.com>
Before we add tons of new parallel versions of standard Lisp functions maybe
we should first see how good/bad it is to just use an explicit SPAWN, e.g.
(QMAPCAR (pred) #'baz list)
vs.
(MAPCAR #'(lambda (l) (spawn (pred) (baz l))) list)
seems to be clearer using QMAPCAR, while
(QDOTIMES (pred) (i n) (baz i))
vs.
(DOTIMES (i n) (spawn (pred) (baz i)))
seems about the same.
I agree with your statements on which is clearer, but these pairs of
expressions will not have the same performance. In most cases, the
explicit SPAWN calls will create many more processes than the Q
functions, even when dynamic spawning is used.
The parallel list mapping paper goes into detail on why this is so,
and shows how to implement QMAPCAR very efficiently. (It doesn't
mention QDOTIMES, but this case is easier since you know right from
the start how many iterations need to be done.)
So the point of the Q functions is to insulate the programmer as much
as possible from the details of when processes are created. Someday
we may discover an even better way to do parallel iteration, or on a
new parallel machine there may be a different best way.
∂09-May-90 2034 weening@Gang-of-Four.Stanford.EDU parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 9 May 90 20:34:01 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA09873; Wed, 9 May 90 20:34:36 -0700
Date: Wed, 9 May 90 20:34:36 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005100334.AA09873@Gang-of-Four.Stanford.EDU>
To: qlisp@Gang-of-Four.Stanford.EDU
Subject: parallel iteration functions
Here's a brief summary of how the parallel iteration functions work.
In the general case, we have a sequence of N "tasks" to do, and they
can all be done in parallel. (I.e., the programmer has said that it
is safe to do so.) In the case of DOTIMES, and sequence functions on
vectors, the number N is known at the start. In the case of MAPCAR,
DOLIST, or sequence functions on lists, N is ususlly not known.
The first strategy that comes to mind is to make each task a potential
process, and use a runtime test to decide which processes to actually
create. Dynamic partitioning is an obvious candidate for this test,
since it has worked well in other programs. It does not work well
here, though. What happens is that the first few tasks get run in
separate processes, then the queues fill up; later they start to empty
again, so more processes get created, etc. With N large enough
(compared to P, the number of processors), the behavior is eventually
cyclic so the number of processes created is linear in N. This means
a lot of overhead. We would like something that is more like linear
in P, since this is all that is needed to keep the processors busy.
Other (non-dynamic) partitioning tests do not seem any more useful
here, at least we haven't found any that work well.
What helps is to structure the tasks into larger groups so that large
chunks of work can be done by most processes. The best case is when
the tasks all take the same time, we know N, and we can index into a
vector to find the data for each task. Then we can just create P
processes, each doing N/P tasks, and they will all finish at the same
time. However:
- this may not be best if the iteration is a subproblem of a larger
program that has other parallelism
- tasks may not all take the same amount of time, and we may not
know ahead of time how long they will take
- we often don't know N or can't index into the data
The first two problems are solved by imposing a recursive structure on
the data even if it doesn't have any particular structure to start
with (other than being a sequence). So far we've only looked at a
binary divide-and-conquer, because it is simple and subsumes a lot of
other possible structures. By this, I mean we divide the tasks into
two sets, then subdivide each of these into two, etc. Each set of
tasks is a potential process. The result is a balanced computation
tree, and dynamic partitioning now works well, creating processes of
the appropriate granularity and interacting well with other parallel
code that is running at the same time.
That takes care of DOTIMES and most functions on vectors. For lists,
we have the problem that we don't know N, or even if we do, it takes
O(N) time to get to most of the data. If we know N and try to divide
the tasks into two sets as before, the O(N) time to cdr through the
list results in a lot of idle time before the processors all get busy.
(Also if we don't know N, we could call LENGTH on the list to find it
before doing the computation, but this has the same problem.)
Sometimes the basic task time is large enough so that this is OK. But
we also wanted to handle cases where the individual tasks are quite
small, and found this can be done by a neat trick. To avoid the long
idle time at the beginning, we start by creating a process for just
the first task. Then we create a process for the next two tasks, then
one for the next four tasks, etc. This gets a lot of processors
working quickly, but doesn't create a huge number of processes, since
the size of the processes is growing very quickly. At some point, we
run off the end of the list and this means we have created a process
of roughly N/2 tasks, and a have created total of just log(N)
processes in getting there. All of these processes can split their
sets of tasks in half whenever the dynamic partitioning method needs
to create more processes, so we are in pretty much the same situation
as above.
There are some more details (including how to pass back return values
and construct the returned list for functions like MAPCAR), but this
is the basic idea.
∂10-May-90 1121 pehoushe@Gang-of-Four.Stanford.EDU parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 May 90 11:21:07 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12085; Thu, 10 May 90 11:21:41 -0700
Date: Thu, 10 May 90 11:21:41 -0700
From: pehoushe@Gang-of-Four.Stanford.EDU (Dan Pehoushek)
Message-Id: <9005101821.AA12085@Gang-of-Four.Stanford.EDU>
To: weening@Gang-of-Four.Stanford.EDU
Cc: qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Joe Weening's message of Wed, 9 May 90 20:34:36 -0700 <9005100334.AA09873@Gang-of-Four.Stanford.EDU>
Subject: parallel iteration functions
Ron wrote:
>Before we add tons of new parallel versions of standard Lisp functions maybe
>we should first see how good/bad it is to just use an explicit SPAWN, e.g.
> (QMAPCAR (pred) #'baz list)
> vs.
> (MAPCAR #'(lambda (l) (spawn (pred) (baz l))) list)
Using SPAWN ranges from good on large grains to mediocre on medium grains
to very bad on small grains. Using a simple fork/join version of mapcar
is also impotent because it would not be tail recursive.
Our divide-and-conquer tricks on list mapping are sufficiently effective
that we get speedup even on (qmapc #'identity list), when compared to
(mapc #'identity list). The list mapping paper explains the Amdahl's
Law aspects of qmapc; the limit on speedup is determined by the ratio
of time spent on a list element, divided by the time of the cdr
operation.
In the current implementation, there is a Qdotimes and a Qdolist, and
the Qmaplist functions, but no predicate argument. The use of
predicates in these cases would be difficult to explain. Isn't it
adequate to just say these iterative qfunctions and qforms are
parallel, to the best of the implementer's ability?
We've been using Q informally as a prefix for a long time. It seems
right, and when it goes in the Primer, it'll be formal.
When Joe suggested DYNAMIC as a name for the dynamic spawn/partition
predicate, I liked it immediately. The only potential problem is that
it hides low-level details from the programmer, which may or may not
be a good thing.
It also appears rather difficult to convince some people that the
DYNAMIC predicate is really really great, even after they've seen nice
theoretical and experimental results. This continues to be a major
problem.
-Dan
∂10-May-90 1138 rpg@lucid.com parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 May 90 11:38:15 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12188; Thu, 10 May 90 11:38:48 -0700
Received: from rose ([192.31.212.83]) by heavens-gate.lucid.com id AA01193g; Thu, 10 May 90 11:38:15 PDT
Received: by rose id AA01446g; Thu, 10 May 90 11:37:32 PDT
Date: Thu, 10 May 90 11:37:32 PDT
From: Richard P. Gabriel <rpg@lucid.com>
Message-Id: <9005101837.AA01446@rose>
To: pehoushe@Gang-of-Four.Stanford.EDU
Cc: weening@Gang-of-Four.Stanford.EDU, qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Dan Pehoushek's message of Thu, 10 May 90 11:21:41 -0700 <9005101821.AA12085@Gang-of-Four.Stanford.EDU>
Subject: parallel iteration functions
> (QMAPCAR (pred) #'baz list)
> vs.
> (MAPCAR #'(lambda (l) (spawn (pred) (baz l))) list)
Using SPAWN ranges from good on large grains to mediocre on medium grains
to very bad on small grains. Using a simple fork/join version of mapcar
is also impotent because it would not be tail recursive.
It's easy to overlook the fact that even dumb compilers can be made to
understand that the second form can be transformed to the first one,
where ``QMAPCAR'' is taken as an internal compiler form. The question
becomes one of how smart does the compiler need to be versus the
complexity of introducing a lot of constructs.
(mapc #'identity list). The list mapping paper explains the Amdahl's
Law aspects of qmapc; the limit on speedup is determined by the ratio
of time spent on a list element, divided by the time of the cdr
operation.
And CDR is very fast (a list of length 100,000 takes 80 milliseconds,
and it is a very rare list), a fact that limits my appreciation of the
*practical* import of the O(n) arguments.
We've been using Q informally as a prefix for a long time. It seems
right, and when it goes in the Primer, it'll be formal.
When Joe suggested DYNAMIC as a name for the dynamic spawn/partition
predicate, I liked it immediately. The only potential problem is that
it hides low-level details from the programmer, which may or may not
be a good thing.
``Q'' is a good standard prefix. Why drop it for DYNAMIC when many
other things could be called dynamic (such as binding, extent, etc)?
-rpg-
∂10-May-90 1201 pehoushe@Gang-of-Four.Stanford.EDU parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 May 90 12:01:30 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12298; Thu, 10 May 90 12:01:57 -0700
Date: Thu, 10 May 90 12:01:57 -0700
From: pehoushe@Gang-of-Four.Stanford.EDU (Dan Pehoushek)
Message-Id: <9005101901.AA12298@Gang-of-Four.Stanford.EDU>
To: rpg@lucid.com
Cc: weening@Gang-of-Four.Stanford.EDU, qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Richard P. Gabriel's message of Thu, 10 May 90 11:37:32 PDT <9005101837.AA01446@rose>
Subject: parallel iteration functions
3 Clarifications:
> (QMAPCAR (pred) #'baz list)
> vs.
> (MAPCAR #'(lambda (l) (spawn (pred) (baz l))) list)
In the current implementation, there is no (pred). Qmapcar is
implemented with (dynamic) and the tricks Joe mentioned, and so looks
exactly like MAPCAR, except with a Q -- (QMAPCAR #'baz list). In the
current implementation, all the parallel iteraters differ from their
serial counterparts by just a "Q" at the beginning. There is no
predicate, and adding a predicate might be wrong.
>And CDR is very fast (a list of length 100,000 takes 80 milliseconds,
>and it is a very rare list), a fact that limits my appreciation of the
>*practical* import of the O(n) arguments.
CDR is very fast, and that's why we get good speedups, even on fine
grained lists. But we still needed the doubling trick to get the best
result on long fine grained lists.
DYNAMIC is only being considered as the name of the dynamic spawning
predicate, as in:
(qlet (dynamic)
((a (fib (- n 1)))
(b (fib (- n 2))))
(+ a b))
-Dan
∂10-May-90 1447 bergman@Sunburn.Stanford.EDU US-Japan travel grant
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 10 May 90 14:47:08 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA15181; Thu, 10 May 90 14:48:04 -0700
Date: Thu, 10 May 1990 14:48:01 PDT
From: Sharon Bergman <bergman@Sunburn.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Cc: clt@sail.Stanford.EDU, mps@sail.Stanford.EDU, davis@Sunburn.Stanford.EDU,
bergman@sunburn.Stanford.EDU
Subject: US-Japan travel grant
Message-Id: <CMM.0.88.642376081.bergman@Sunburn.Stanford.EDU>
John, I don't believe I given you the account number for your
new NSF grant (Joint US-Japanese Collaboration on Research in New
Foundations of Computer Science). Here it is:
Account number: 2DMA539
Fund number: 163D065
Title: NSF INT 89-20626
Performance period: 4/1/90-3/31/92 (no 6-mo. flexibility period)
Total amount: $28,730
-Sharon Bergman
∂10-May-90 1613 ok@coyote.stanford.edu AI Qual Schedule
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 10 May 90 16:13:27 PDT
Received: by coyote.stanford.edu; Thu, 10 May 90 16:08:40 PDT
Date: 10 May 1990 1608-PDT (Thursday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: ginsberg@sunburn.stanford.edu, latombe@coyote.stanford.edu,
jmc@sail.stanford.edu, nilsson@tenaya.stanford.edu,
rosenschein@teleos.com, shoham@hudson.stanford.edu,
ash@sumex-aim.stanford.edu, darwiche@neon.stanford.edu,
young@neon.stanford.edu
Cc: ok@coyote.stanford.edu
Subject: AI Qual Schedule
This is the official schedule for the qualification exams on Wednesday
June 6, 1990. There will be two exam times: 10:30--12noon and 1:45--3:15.
On each committee there is a (C)hair, a (D)epth examiner, and one (O)ther.
AI Qualification Exam Schedule
Wednesday, June 6, 1990
10:30 -- 12noon
Adnan Darwiche: Reasoning
C: Latombe D: Nilsson O: Rosenschein
1:45 -- 3:15
David Ash: Planning
C: Nilsson D: Ginsberg O: Shoham
Michael Young: Formal Models of Inter-Agent Communication
C: Latombe D: McCarthy O: Rosenschein
∂11-May-90 0957 korf@CS.UCLA.EDU visit
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 11 May 90 09:57:17 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA23307;
Fri, 11 May 90 09:58:13 -0700
Message-Id: <9005111658.AA23307@shemp.cs.ucla.edu>
Received: by denali; Fri, 11 May 90 10:11:59 pdt
Date: Fri, 11 May 90 10:11:59 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: visit
John,
Where and when would you like to meet when you visit next week? Where are you
staying in L.A.?
-rich
∂11-May-90 1201 jfetzer@ub.d.umn.edu Current Developments
Received: from ub.d.umn.edu ([131.212.32.6]) by SAIL.Stanford.EDU with TCP; 11 May 90 12:01:04 PDT
Received: by ub.d.umn.edu (5.59/UMD-891211)
id AA30988; Fri, 11 May 90 14:02:23 CDT
From: jfetzer@ub.d.umn.edu (james fetzer)
Message-Id: <9005111902.AA30988@ub.d.umn.edu>
Subject: Current Developments
To: M_and_M_EB@ub.d.umn.edu
Date: Fri, 11 May 90 14:02:21 CDT
Cc: jfetzer@ub.d.umn.edu
X-Mailer: Elm [version 2.1 PL1]
MINDS AND MACHINES
Journal for Artificial Intelligence, Philosophy, and Cognitive Science
(ISSN 0924-6495)
We have an addition to the Editorial Board and we need more submissions:
(1) Brian Smith has joined the board, which now includes Jon Barwise,
Andy Clark, Robert Cummins, Fred Dretske, Jerry Fodor, Clark Glymour,
Stevan Harnad, John Haugeland, Jaakko Hintikka, David Israel, Philip
Johnson-Laird, Frank Keil, Henry Kyburg, John McCarthy, Donald Nute,
Zenon Pylyshyn, Barry Richards, David Rumelhart, Roger Schank, John
Searle, Brian Smith, Paul Smolensky, Stephen Stich, and Terry Winograd.
(2) The initiation of a new journal encounters two crucial difficulties.
The first is establishing an appropriate editorial board. The second is
securing suitable submissions in sufficient number. Our success in cop-
ing with the first of these problems is obvious. What may be less evi-
dent is that we need more submissions. I am therefore asking for your
assistance in locating manuscripts that may be suitable for publication.
If each of the members of the board would be willing to assume responsi-
bility for generating just ONE APPROPRIATE SUBMISSION over the next month
or so (by encouraging a colleague, a student, or some other associate to
submit a paper to the journal for review), then I believe that we should
be able to cope with the second of these problems as well. Once issues
of MINDS AND MACHINES begin to appear, this difficulty should disappear.
Jim
∂11-May-90 1305 jlm@lucid.com names for parallel iteration functions
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 May 90 13:05:30 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA17721; Fri, 11 May 90 13:05:59 -0700
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA11934g; Fri, 11 May 90 13:05:24 PDT
Received: by bhopal id AA04451g; Fri, 11 May 90 13:06:04 PDT
Date: Fri, 11 May 90 13:06:04 PDT
From: Jim McDonald <jlm@lucid.com>
Message-Id: <9005112006.AA04451@bhopal>
To: arg@lucid.com
Cc: weening@Gang-of-Four.Stanford.EDU, qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Ron Goldman's message of Wed, 9 May 90 19:13:05 PDT <9005100213.AA07665@bhopal>
Subject: names for parallel iteration functions
Date: Wed, 9 May 90 19:13:05 PDT
From: Ron Goldman <arg>
...
(QDOTIMES (pred) (i n) (baz i))
vs.
(DOTIMES (i n) (spawn (pred) (baz i)))
seems about the same. ...
Ron
I don't think so.
For many problems, I think QDOTIMES is the right level of abstraction.
E.g., assume n is 100000000. Then (QDOTIMES (pred) (i n) (baz i))
can be smart about farming out subsequences to various processes, but
(DOTIMES (i n) (spawn (pred) (baz i))) is going to thrash madly.
jlm
∂11-May-90 1431 MPS
Professor Lipset - 668-5982 between 5-7pm tonight
John Lloyd, Briston, England, Program chairman for the symposium on
computer logic to be held Nov 13-14, 1990, Brussels would like you
to present a paper
0272 303 913 --- 0272 855 329
∂11-May-90 1507 pehoushe@Gang-of-Four.Stanford.EDU Iteration funcations, mainly Qdotimes
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 May 90 15:07:11 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA18509; Fri, 11 May 90 15:07:26 -0700
Date: Fri, 11 May 90 15:07:26 -0700
From: pehoushe@Gang-of-Four.Stanford.EDU (Dan Pehoushek)
Message-Id: <9005112207.AA18509@Gang-of-Four.Stanford.EDU>
To: jlm@lucid.com
Cc: qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Jim McDonald's message of Fri, 11 May 90 13:06:04 PDT <9005112006.AA04451@bhopal>
Subject: Iteration funcations, mainly Qdotimes
Given:
(defun foo (n) (dotimes (i n)))
(defun qfoo (n) (qdotimes (i n)))
QFOO is about 1.9 times as slow serially as FOO, but gets nearly
perfect speedup. Of course, when the dotimes loop is non-empty, the
slowdown ratio can get close to 1, and the real speedup close to 8.
Please Note: There is no control expression for limiting parallelism.
A control expression does not seem particularly necessary, so we are
describing the iterators in the primer without using control
expressions.
In the current implementation, QDOTIMES implicitly uses DYNAMIC as the
partition predicate, behaving quite a bit differently from:
(DOTIMES (i n) (spawn (pred) (baz i)))
-Dan
***********************************************************************
Three timing experiments:
> (time (foo 1000000)) ;; serial code in serial
User cpu time = 1533 milliseconds
> (time (qfoo 1000000)) ;; parallel code in serial
User cpu time = 2894 milliseconds
> (qtime (qfoo 1000000)) ;; parallel code in parallel
Parallel Time: 380 milliseconds on 8 processors
NumProcesses: 690
Overhead: 24.6 milliseconds, 0.8%
Idle: 21.2 milliseconds, 0.7%
***********************************************************************
Source Code For QDOTIMES. A similar idea applies to qmapcar...
;;; For Lucid folks, #! is roughly equivalent to qlet& t, and
;;; (=& (self-task-stack-height) 0) == (DYNAMIC)
;;; is a test to see if the local task stack is empty. Please note that
;;; qdotimes only works for a fixnum number of iterations. The &'s tell
;;; the compiler that everything is made of fixnums.
(defmacro qdotimes ((var howmany
&optional (result-form nil))
&body do-form)
(let ((right (gensym)) ;; Gensyms avoid name conflicts.
(qdo-fun (gensym)))
`(labels ((,qdo-fun (,var ,right)
(declare (fixnum ,right ,var)
(function ,qdo-fun (fixnum fixnum) nil))
(loop
(when (and (=& (self-task-stack-height) 0)
(<& ,var ,right))
#!(progn
(,qdo-fun ,var (+& ,var (ash& (-& ,right ,var) -1)))
(,qdo-fun (+& ,var (+& 1 (ash& (-& ,right ,var) -1))) ,right))
(return-from ,qdo-fun nil))
(when (>& ,var ,right) (return nil))
;; The body of the dotimes
,@do-form
(incf& ,var))
(values)))
(,qdo-fun 0 (1- ,howmany))
,result-form)))
∂11-May-90 1521 iam@Gang-of-Four.Stanford.EDU typep
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 May 90 15:21:14 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA18612; Fri, 11 May 90 15:21:46 -0700
Date: Fri, 11 May 90 15:21:46 -0700
From: iam@Gang-of-Four.Stanford.EDU (Ian Mason)
Message-Id: <9005112221.AA18612@Gang-of-Four.Stanford.EDU>
To: qlisp@go4.Stanford.EDU
Subject: typep
carolyn and i revised the boyer benchmark so as to
incorporate structures. in doing so we came
across an interesting phenomenon. typep is 50%
slower (when it fails) in parallel mode than
it is in sequential mode. the following example is typical.
(defstruct (term (:print-function term-print)))
(defstruct (atomic-term (:include term)) name)
(defstruct (composite-term (:include term)) op args)
(defun is-atomic-term (term) (typep term 'atomic-term))
(defun is-term (term) (typep term 'term))
(setq *i* nil)
(progn (dotimes (i 100000)
(setq *i* (cons (make-atomic-term :name i)
(cons (make-composite-term :op i) *i*))))
nil)
> (compile 'is-atomic-term)
> (compile 'is-term)
> (progn (gc) (time (mapcar #'is-atomic-term *i*)) t)
User cpu time = 63321 milliseconds
T
> (progn (gc) (qtime (mapcar #'is-atomic-term *i*)) t)
User cpu time = 94136 milliseconds
T
> (progn (gc) (time (mapcar #'is-term *i*)) t)
User cpu time = 14554 milliseconds
T
> (progn (gc) (qtime (mapcar #'is-term *i*)) t)
User cpu time = 15561 milliseconds
T
∂11-May-90 1521 franz@cs.washington.edu A new home for the qualitative physics list
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 11 May 90 15:21:31 PDT
Received: by june.cs.washington.edu (5.61/7.0jh)
id AA11205; Fri, 11 May 90 15:06:11 -0700
Date: Fri, 11 May 90 15:06:11 -0700
From: franz@cs.washington.edu (Franz G. Amador)
Return-Path: <franz@cs.washington.edu>
Message-Id: <9005112206.AA11205@june.cs.washington.edu>
To: qphysics@cs.washington.edu
Subject: A new home for the qualitative physics list
Although the qualitative physics mailing list perished at its old
home, like the phoenix it arises from its ashes in a new nest, namely
qphysics@cs.washington.edu
with administrative matters to be directed to
qphysics-request@cs.washington.edu
where it will no doubt act as a potent pollinator for intellectual
cross-fertilization. So, have at it, gang.
Franz Amador
franz@cs.washington.edu
{rutgers,cornell,ucsd,ubc-cs,tektronix}!uw-beaver!june!franz
∂11-May-90 1526 rpg@lucid.com DYNAMIC
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 May 90 15:26:51 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA18630; Fri, 11 May 90 15:27:25 -0700
Received: from rose ([192.31.212.83]) by heavens-gate.lucid.com id AA13268g; Fri, 11 May 90 15:27:00 PDT
Received: by rose id AA03185g; Fri, 11 May 90 15:26:20 PDT
Date: Fri, 11 May 90 15:26:20 PDT
From: Richard P. Gabriel <rpg@lucid.com>
Message-Id: <9005112226.AA03185@rose>
To: qlisp@Gang-of-Four.Stanford.EDU
In-Reply-To: Dan Pehoushek's message of Fri, 11 May 90 15:07:26 -0700 <9005112207.AA18509@Gang-of-Four.Stanford.EDU>
Subject: DYNAMIC
I think this is a bad name for this predicate. First, it doesn't have
P at the end, and even if it did, it would not be descriptive. Anyone
who saw the name would not even know it had anything to do with
parallelism. At least with names like QDOTIMES, the naive user knows
it must have an obscure meaning.
DYNAMIC is a name that users probably want to use for their own purposes.
Therefore, I propose we use DYNAMIC-SPAWN-P, and if someone really
wants to use DYNAMIC, a macro can be written.
-rpg-
∂11-May-90 1724 VAL Reminder: Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
NEURAL NET REPRESENTATION OF A TRUTH MAINTENANCE SYSTEM
Paul Morris
Intellicorp
Monday, May 14, 2:30pm
MJH 252
We show how label updating for a justification-based TMS may be
accomplished within a neural net model. In this approach, the IN status
of a TMS node or justification is represented dynamically by a
continuously firing neuron, rather than statically as a memory trace.
This novel perspective leads to a way of defining well-foundedness in
terms of local constraints. It also suggests an epistemic
interpretation of a neural net as a kind of "fuzzy" TMS.
∂11-May-90 2018 arg@lucid.com typep
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 May 90 20:18:26 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA19894; Fri, 11 May 90 20:18:57 -0700
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA15304g; Fri, 11 May 90 20:18:30 PDT
Received: by bhopal id AA05245g; Fri, 11 May 90 20:19:14 PDT
Date: Fri, 11 May 90 20:19:14 PDT
From: Ron Goldman <arg@lucid.com>
Message-Id: <9005120319.AA05245@bhopal>
To: qlisp@go4.Stanford.EDU
In-Reply-To: Ian Mason's message of Fri, 11 May 90 15:21:46 -0700 <9005112221.AA18612@Gang-of-Four.Stanford.EDU>
Subject: typep
That TYPEP would be slower in parallel mode seemed strange, so I took a look
at things and now have a good explanation of what is happening. When TYPEP
is called with a structure as its first argument, it first checks if the
structure is a subtype of the second argument. If it is then it returns T
and, as Carolyn & Ian's results showed, the parallel time is the same as the
serial time. (Remember that when running in parallel the Alliant hardware
runs about 5% slower than in serial operation.)
Now if the structure is not a subtype of the type argument, then a check is
done to see if the type argument is also a structure, in which case TYPEP
will return NIL. To do this an internal hash table is checked (via
GETHASH), and that's where the extra time comes from. GETHASH normally
needs to get a reader lock for the hash table before it can do the lookup,
however there's an optimization in the code so that when it's running in
serial mode it doesn't bother to get/release the lock. I timed how long the
code to get/release the lock took and that seemed enough to account for
almost all the difference. I hope this clears up the mystery.
Note that in a few weeks I plan on bringing up a new version of Qlisp that
is always in parallel mode, i.e. Qlisp at toplevel, so timing differences
between TIME & QTIME will go away. Actually we'll probably change the
definition of (QTIME form) => (TIME (QWAIT form)).
Ron
∂12-May-90 0730 weening@Gang-of-Four.Stanford.EDU Measuring parallel processor performance
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 May 90 07:30:49 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA21037; Sat, 12 May 90 07:31:20 -0700
Date: Sat, 12 May 90 07:31:20 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005121431.AA21037@Gang-of-Four.Stanford.EDU>
To: qlisp@Gang-of-Four.Stanford.EDU
Subject: Measuring parallel processor performance
The May 1990 issue of Communications of the ACM has an interesting
article by Alan Karp and Horace Flatt, "Measuring Parallel Processor
Performance". It suggests that in addition to measuring speedup and
efficiency, we should compute a new statistic related to Amdahl's law.
Amdahl's law says that if a certain fraction f of the code must be run
serially, then speedup on p processors is limited by
1/s = f + (1-f) / p
Given a program, f is not always easy to determine by analyzing the
code. However, the speedup s can be measured, and then
f = (1/s - 1/p) / (1 - 1/p)
If this is done on varying values of p, then in a "good" parallel
system f will remain constant. If we observe that f increases with p,
then something in the system, not the program, needs to be fixed. An
example is some kind of synchronization whose time increases with the
number of processors.
If f is high, but does not increase with p, then the program is
showing a lack of sufficient parallelism and it needs to be examined.
[At the end of the article they note that one should compute f by
f = 1 - (1 - 1/s) / (1 - 1/p)
to avoid loss of precision in the subtraction.]
∂12-May-90 1636 ok@coyote.stanford.edu Young's Depth Area
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 12 May 90 16:36:31 PDT
Received: by coyote.stanford.edu; Sat, 12 May 90 16:30:44 PDT
Date: 12 May 1990 1630-PDT (Saturday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: latombe@coyote.stanford.edu, jmc@sail.stanford.edu, rosenschein@teleos.com
Cc: ok@coyote.stanford.edu
Subject: Young's Depth Area
Michael Young has sent me a reading list that describes the extent
of the scope of his depth area for the AI qual. Here's the LaTeX source
file of this list:
------------
∂13-May-90 1054 harnad@clarity.Princeton.EDU Re: abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 13 May 90 10:54:05 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.32/relay)
id AA13556; Sun, 13 May 90 13:54:16 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA25046; Sun, 13 May 90 13:54:47 EDT
Date: Sun, 13 May 90 13:54:47 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005131754.AA25046@cognito.Princeton.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: Re: abstract
Cc: hayes.pa@xerox.com
John, here's Pat's abstract. But note that we've got you down
as two separate speakers. It's going to shrink the symposium
if you become one...
Cheers, Stevan
Searle's Chinese Room
Pat Hayes
The basic flaw in Searle's argument is a widely accepted misunderstanding
about the nature of computers and computation: the idea that a computer
is a mechanical slave that obeys orders. This popular metaphor suggests
a major division between physical, causal hardware which acts, and
formal symbolic software, which gets read. This distinction runs
through much computing terminology, but one of the main conceptual
insights of computer science is that it is of little real scientific
importance. Computers running programs just aren't like the Chinese
room.
Software is a series of patterns which, when placed in the proper
places inside the machine, cause it to become a causally different
device. Computer hardware is by itself an incomplete specification of a
machine, which is completed - i.e. caused to quickly reshape its
electronic functionality - by having electrical patterns moved within
it. The hardware and the patterns together become a mechanism which
behaves in the way specified by the program.
This is not at all like the relationship between a reader obeying some
instructions or following some rules. Unless, that is, he has somehow
absorbed these instructions so completely that they have become part of
him, become one of his skills. The man in Searle's room who has done
this to his program now understands Chinese.
∂13-May-90 1055 shoham@Hudson.Stanford.EDU michael young quals
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 May 90 10:55:30 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA03214; Sun, 13 May 90 10:54:09 -0700
Date: Sun, 13 May 1990 10:54:08 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: michael young quals
Message-Id: <CMM.0.88.642621248.shoham@Hudson.Stanford.EDU>
Oussama called me to say that you thought Young's qual reading list was too
limited. When young showed it to me, I told him that the list was fairly
narrow, and that he could therefore expect detailed questions. I specifically
mentioned nonmonotonic logics. Furthermore, nonmonotonic temporal reaosning is
particularly relevant to communication. So I think you'd be perfectly within
the guidelines to ask questions in this area. Does this address you concern?
Oussama asked me clarify the matter to Young; I'd like to do it before I leave
for Amsterdam on Tuesday.
Yoav
∂14-May-90 0911 MPS
Your visa is ready. I am having the courier go and pick it up now.
Pat
∂14-May-90 0930 JMC
Hersch about freq. flyer, lack thereof
∂14-May-90 1000 JMC
devlin
∂14-May-90 1025 VAL
Are you free for lunch today, or any other day this week?
∂14-May-90 1039 @IU.AI.SRI.COM,@peabody.teleos.com:leslie@teleos.com Signatures
Received: from IU.AI.SRI.COM by SAIL.Stanford.EDU with TCP; 14 May 90 10:39:42 PDT
Received: from peabody.teleos.com by IU.AI.SRI.COM via SMTP with TCP;
Mon, 14 May 90 10:40:32-PST
Received: by peabody.teleos.com (3.2/4.16) id AA01879 for
jmc@sail.stanford.edu; Mon, 14 May 90 10:38:35 PDT
Date: Mon, 14 May 90 10:38:35 PDT
From: Leslie Kaelbling <leslie@teleos.com>
Message-Id: <9005141738.AA01879@peabody.teleos.com>
To: jmc@sail.stanford.edu
Subject: Signatures
Reply-To: leslie%teleos.com@ai.sri.com
John,
I got the comments back from Sutton, and they're mostly trivial requests for
more references. So, I still have hopes of finishing this quarter. When,
in the next few weeks, can I get you to sign the signature pages?
- Leslie
∂14-May-90 1100 JMC
call Keith Devlin to check.
∂14-May-90 1100 JMC
Hennessey nomination
∂14-May-90 1208 MPS Mileage miles
Franklin called about your mileage plus. We can correct AA. I have
the new number in your file, but he was crediting the old number.
As far as Uniteds connection with the foreign carriers. When you
get to the airport you have to fill out the United mileage plus form
and present it to the foreign carrier when you give them your ticket.
Franklin also said to keep your boarding passes. Sometimes the airlines
goof and that is your way of proving you flew and are due the miles.
∂14-May-90 1211 MPS Mileage
You must fill in the form each time you board a foreign carrier
which has reciprocal agreement with United. Sabena does, and
you are coming back on them.
∂14-May-90 1347 shoham@Hudson.Stanford.EDU qual depth area
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 14 May 90 13:47:28 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA07505; Mon, 14 May 90 13:45:59 -0700
Date: Mon, 14 May 1990 13:45:58 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: young@cs.Stanford.EDU
Cc: jmc@sail.Stanford.EDU, latombe@coyote.Stanford.EDU, rosenschein@teleos.com,
ok@coyote.Stanford.EDU
Subject: qual depth area
Message-Id: <CMM.0.88.642717958.shoham@Hudson.Stanford.EDU>
Michael,
Oussama, who is away, asked me to clarify with you the following. As
discussed between us when you presented your reading list, you will
be expected to know material on nonmonotonic logics in general, and
reasoning about change in particular, including various recent ideas.
Good luck in the exam.
Yoav
∂14-May-90 1353 MPS
Leora Morgenstern called
∂14-May-90 1445 RWF history of LISP
In Kleene's Intro to Metamath, he introduces a form of
notation representing f(a,b,c) by (f,a,b,c). Was this
the source of Cambridge Polish notation? Ref. pg. 248.
∂14-May-90 1747 young@Neon.Stanford.EDU re: qual depth area
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 14 May 90 17:47:12 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA12860; Mon, 14 May 90 17:48:07 -0700
Date: Mon, 14 May 1990 17:48:05 PDT
From: "R. Michael Young" <young@Neon.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: young@CS.Stanford.EDU, shoham@Hudson.Stanford.EDU,
latombe@COYOTE.STANFORD.EDU, rosenschein@TELEOS.COM,
ok@COYOTE.STANFORD.EDU, young@Neon.Stanford.EDU
Subject: re: qual depth area
In-Reply-To: Your message of 14 May 90 1704 PDT
Message-Id: <CMM.0.88.642732485.young@Neon.Stanford.EDU>
That's sounds fine. I was just concerned that the focus of the depth
area portion of the exam would be on nonmonotonic reasoning rather
than on communication. Having "some familiarity" with applications of
circumscription to action, knowledge, etc, is certainly to be
expected. As long as we have a common understanding of the focus of
the area I have no concerns.
Please let me know if our ideas seem to differ. I can stop by your
office for a very brief talk if that would help. Thanks.
-Michael
∂14-May-90 1918 jwl%compsci.bristol.ac.uk@NSFnet-Relay.AC.UK Symposium on Computational Logic
Received: from NSFnet-Relay.AC.UK by SAIL.Stanford.EDU with TCP; 14 May 90 19:18:18 PDT
Received: from sun.nsfnet-relay.ac.uk by vax.NSFnet-Relay.AC.UK
via Janet with NIFTP id aa00666; 14 May 90 20:05 BST
Received: from lihue by kukini.CompSci.Bristol.ac.uk id aa29423;
14 May 90 19:38 BST
Date: Mon, 14 May 90 19:38:00 BST
From: John Lloyd <jwl%compsci.bristol.ac.uk@NSFnet-Relay.AC.UK>
To: jmc@sail.stanford.edu
cc: jwl%compsci.bristol.ac.uk@NSFnet-Relay.AC.UK
Subject: Symposium on Computational Logic
Message-ID: <9005141938.aa07455@lihue.CompSci.Bristol.ac.uk>
Dear Professor McCarthy,
As chairman of the programme committee for the Symposium on Computational
Logic to be held at the Esprit'90 Conference, I invite you to contribute
a paper to the Symposium.
Esprit'90 is the annual conference of the ESPRIT programme, attracting
around 3000 participants over 4 days. This year for the first time
a Symposium will be held as a special event with its own proceedings
separate from the main proceedings. This Symposium is intended to be
a prestigious occasion with wide circulation of the proceedings.
All papers at the Symposium will be invited. There will be around 10
papers, plus a panel session. We intend to invite K. Apt, A. Bundy,
A. Colmerauer, C.A.R. Hoare, R. Kowlaski, R. Reiter, D. Scott, J. Siekmann,
and F. Turini, in addition to you. The European speakers are mostly from
the ESPRIT project "Compulog" on which the Symposium is based.
So far, Apt, Kowalski, Siekmann and Turini are confirmed, and Colmerauer,
Bundy, and Reiter are very likely. I'm not certain about Hoare, but he
is already committed to coming to the main conference and we hope to get
him for the Symposium. Also I had a garbled message this morning which I
believe said that Scott had agreed to come, but I will have to get back
to Brussels to confirm this.
The Symposium will be held in Brussels at the same site as ESPRIT'90.
The dates for the Symposium are 13th and 14th November, 1990, and it
will begin at 2pm on the 13th and go all day on the 14th. Speakers will
have 45 minutes for their presentations. The final event on the programme
will be a panel session lasting for one and a half hours. A social event
(probably a buffet supper) is planned for the evening of the 13th.
We expect approximately 200 delegates, although since this is the first
time such a Symposium has been held, the numbers are hard to estimate.
The timescale is that we would need your paper (which I am afraid is
essential!) around the middle of August. Authors retain copyright of
their papers and we expect the proceeding to be published by Springer.
The intention is that the proceedings be widely disseminated.
We have left a few weeks spare at the end, so that all papers
can be read before the final version is required (probably, around the
5th of September.) The committee may have some suggestions to make and
it will also be a chance to make any minor changes so that the papers will
fit together better overall.
The committee is looking for papers of the highest possible quality.
These could either be conventional research papers or detailed
overview/future directions papers on some aspect of computational logic.
Computational logic is to be understood in the general sense of the use
of logic in Computer Science, although we would obviously prefer, if
possible, for speakers to establish a connection with the rather wide
range of topics studied in the "Compulog" project. In any case, I will
send you information about the research topics in the "Compulog" project
to give you some idea what topics the internal speakers will be addressing.
All usual expenses associated with coming to the Symposium will be
paid for by ESPRIT. This includes travel, accommodation and meals.
Also your registration fee for the Symposium will be waived.
I hope very much you will be able to contribute to the Symposium.
I am sure you will find it an interesting and stimulating occasion.
In case you want to contact me by phone, my number at work is 272-303913
and my number at home is 272-855329.
yours sincerely,
John Lloyd
∂15-May-90 0021 ME re: interesting finger "bug"
To: action@CS.Stanford.EDU
CC: jcm@ISWIM.STANFORD.EDU, JMC@SAIL.Stanford.EDU, LES@SAIL.Stanford.EDU,
JJW@SAIL.Stanford.EDU
- - - - -
Date: Mon, 14 May 90 22:22:45 -0700
From: John C. Mitchell <jcm@Iswim.Stanford.EDU>
To: action@cs.Stanford.EDU
Subject: interesting finger "bug"
Try doing finger mccarthy@cs
- - - - -
I'll save you all some trouble. Here's what you get:
Csalias name: mccarthy In real life: John McCarthy
[doing a finger on jmc-lists@sail.stanford.edu]
[sail.stanford.edu]
Undefined switch: LISTS
The problem is that mccarthy@cs forwards not to a person but to a special
SAIL forwarding entry called jmc-lists. Since a person on SAIL can't have
a "-" in a user name, and you can't finger a forwarding address, the SAIL
FINGER program takes the "-" as indicating that a "switch" (an option)
follows. And of course, "-lists" is not a defined option.
Now, a little kludge in FINGER to make it continue despite undefined
switches would actually have gotten the finger info for JMC....
∂15-May-90 0319 nsuzuki%jpntscvm.bitnet@Forsythe.Stanford.EDU my resume
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 15 May 90 03:19:20 PDT
Received: by Forsythe.Stanford.EDU; Tue, 15 May 90 03:20:59 PDT
Received: by JPNTSCVM (Mailer R2.03B) id 9256; Tue, 15 May 90 18:58:33 JST
Date: Tue, 15 May 1990 18:57 JST
From: Norihisa Suzuki <nsuzuki%jpntscvm.bitnet@Forsythe.Stanford.EDU>
Subject: my resume
To: Professor John McCarthy <JMC@SAIL.STANFORD.EDU>
Dear John,
I sent you my resume by Fax and Airmail. Please let me know if
anything happens.
Regards, Nori
∂15-May-90 0621 rapaport@cs.Buffalo.EDU CALL FOR PAPERS
Received: from ub.d.umn.edu ([131.212.32.6]) by SAIL.Stanford.EDU with TCP; 15 May 90 06:21:32 PDT
Received: from adara.cs.Buffalo.EDU by ub.d.umn.edu (5.59/UMD-891211)
id AA11965; Tue, 15 May 90 08:22:45 CDT
Received: by adara.cs.Buffalo.EDU (4.1/1.01)
id AA14037; Tue, 15 May 90 09:20:58 EDT
Date: Tue, 15 May 90 09:20:58 EDT
From: rapaport@cs.Buffalo.EDU (William J. Rapaport)
Message-Id: <9005151320.AA14037@adara.cs.Buffalo.EDU>
To: M_and_M_EB@ub.d.umn.edu
Subject: CALL FOR PAPERS
========================================================================
CALL FOR PAPERS---CALL FOR PAPERS---CALL FOR PAPERS---CALL FOR PAPERS
========================================================================
Kluwer Academic Publishers announces
MINDS AND MACHINES
Journal for Artificial Intelligence, Philosophy, and Cognitive Science
ISSN 0924-6495
EDITORIAL FOCUS:
Machines and Mentality
Knowledge and its Representation
Epistemic Aspects of Computer Programming
Connectionist Conceptions
Artificial Intelligence and Epistemology
Computer Methodology
Computational Approaches to Philosophical Issues
Philosophy of Computer Science
Simulation and Modeling
Ethical Aspects of Artificial Intelligence
EDITOR:
James H. Fetzer, Philosophy, University of Minnesota, Duluth, MN, USA
BOOK REVIEW EDITOR:
William J. Rapaport, Computer Science, SUNY Buffalo, Buffalo, NY, USA
EDITORIAL BOARD (as of 7 May 1990)
Jon Barwise Philosophy and Mathematics, Indiana University, USA
Andy Clark Cognitive Studies, University of Sussex, UK
Robert Cummins Philosophy, University of Arizona, USA
Fred Dretske Philosophy, University of Wisconsin at Madison, USA
Jerry Fodor Philosophy, Rutgers University, USA
Clark Glymour Philosophy, Carnegie-Mellon University, USA
Stevan Harnad Psychology, Princeton University, USA
John Haugeland Philosophy, University of Pittsburgh, USA
Jaakko Hintikka Philosophy, Boston University, USA
David Israel SRI International, USA
Philip Johnson-Laird Psychology, Princeton University, USA
Frank Keil Psychology, Cornell University, USA
Henry Kyburg Philosophy, University of Rochester, USA
John McCarthy Computer Science, Stanford University, USA
Donald Nute Philosophy, University of Georgia, USA
Zenon Pylyshyn Psychology, University of Western Ontario, Canada
Barry Richards Computing, Imperial College, London, UK
David Rumelhart Psychology, Stanford University, USA
Roger C. Schank Learning Sciences, Northwestern University, USA
John Searle Philosophy, University of California at Berkeley, USA
Brian Cantwell Smith Artificial Intelligence, Xerox PARC, USA
Paul Smolensky Computer Science, University of Colorado, USA
Stephen Stich Philosophy, Rutgers University, USA
Terry Winograd Computer Science, Stanford University, USA
MINDS AND MACHINES affords an international forum for discussion and
debate of important and controversial issues concerning significant
developments within its areas of editorial focus. Well-reasoned
contributions from diverse theoretical perspectives are welcome, and
every effort will be made to insure their prompt publication. Among the
features that are intended to make this journal distinctive within the
field are these:
o Strong stands on controversial issues are specifically encouraged;
o Important articles exceeding normal journal length may appear;
o Special issues devoted to specific topics will be a regular feature;
o Review essays discussing current problem situations will appear;
o Critical responses to previously published pieces are also invited.
This journal is intended to foster a tradition of criticism within the
AI and philosophical communities on problems and issues of common
concern. Its scope explicitly encompasses philosophical aspects of
computer science. All submissions will be subject to review.
Publication will begin with a single volume of four issues per year.
The first issue will appear in January 1991.
Contributors should send 4 copies of their manuscript to:
James H. Fetzer, Editor
MINDS AND MACHINES
Department of Philosophy
University of Minnesota
Duluth, MN 55812
USA
jfetzer@ub.d.umn.edu
AI_and_PHIL@ub.d.umn.edu
Correspondence concerning books for review should be sent to:
William J. Rapaport, Book Review Editor
MINDS AND MACHINES
Center for Cognitive Science
Department of Computer Science
SUNY Buffalo
Buffalo, NY 14260
USA
rapaport@cs.buffalo.edu
rapaport@sunybcs.bitnet
Subscription information and sample copies will be available from:
Kluwer Academic Publishers Group
P.O. Box 322
3300 AH Dordrecht
The Netherlands
or
Kluwer Academic Publishers
101 Philip Drive
Norwell, MA 02061
USA
========================================================================
CALL FOR PAPERS---CALL FOR PAPERS---CALL FOR PAPERS---CALL FOR PAPERS
========================================================================
∂15-May-90 0822 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 15 May 90 08:22:49 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03466; Tue, 15 May 90 08:23:25 -0700
Date: Tue, 15 May 90 08:23:25 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005151523.AA03466@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
did you read my message to you yesterday regarding your mileage plus
miles?
∂15-May-90 0928 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 15 May 90 09:27:58 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03606; Tue, 15 May 90 09:28:34 -0700
Date: Tue, 15 May 90 09:28:34 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005151628.AA03606@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
are you going to go to the AAAI-90 workshop, July 29 thru Aug 3 in
Boston? If so I will have to send the registration in today. I can
send it in by your Am Exp card. Please let me know before you leave.
Thanks. Pat
∂15-May-90 0929 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 15 May 90 09:29:03 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03610; Tue, 15 May 90 09:29:37 -0700
Date: Tue, 15 May 90 09:29:37 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005151629.AA03610@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
The book Constructive Cominatoric has been missing from the library
since last November. No place on campus has the book. What do you
want me to do? Pat
∂15-May-90 0945 damon@Sunburn.Stanford.EDU re: interesting finger "bug"
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 15 May 90 09:45:34 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA12531; Tue, 15 May 90 09:46:32 -0700
Date: Tue, 15 May 1990 9:46:31 PDT
From: Damon Koronakos <damon@Sunburn.Stanford.EDU>
To: Martin Frost <ME@SAIL.Stanford.EDU>
Cc: action@CS.Stanford.EDU, jcm@ISWIM.STANFORD.EDU, JMC@SAIL.Stanford.EDU,
LES@SAIL.Stanford.EDU, JJW@SAIL.Stanford.EDU
Subject: re: interesting finger "bug"
In-Reply-To: Your message of 15 May 90 0021 PDT
Message-Id: <CMM.0.88.642789991.damon@Sunburn.Stanford.EDU>
Thanks to Marty for explaining finger jmc@cs.
Someone else asked about "lookup jmc" returning John McCarthy's information.
When you do a "lookup <token>", lookup tries to match <token> to several
pieces of information in a particular order:
1. last name
2. (first portion of) last name, first name
3. cs-alias name
4. account name (from which the person can run PEDIT)
[do a man lookup for more information]
John McCarthy has account "jmc@labrea", so "lookup jmc" eventually finds
his record. "Lookup jcm" will return John Mitchell's record (because he
has cs-alias jcm).
Chers,
Damon
∂15-May-90 0946 rathmann@eclipse.stanford.edu did you get a chance to look at the section?
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 15 May 90 09:46:43 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA29225; Tue, 15 May 90 09:47:42 PDT
Date: Tue, 15 May 90 09:47:42 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9005151647.AA29225@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: did you get a chance to look at the section?
Did you get a chance to look at the section on authoritative
partitions? It was not complete, but that is easily fixed.
On the other hand, if you want something completely different, that is
more difficult, and I would like to get started as soon as possible.
-Peter
∂15-May-90 1149 Mailer re: delay in aid to Nicaragua and Panama
Received: from akbar.cac.washington.edu by SAIL.Stanford.EDU with TCP; 15 May 90 11:49:12 PDT
Received: from tomobiki-cho.cac.washington.edu by akbar.cac.washington.edu
(5.61/UW-NDC Revision: 2.12 ) id AA20755; Tue, 15 May 90 11:50:11 -0700
Date: Tue, 15 May 1990 11:46:26 PDT
From: Mark Crispin <mrc@SUMEX-AIM.Stanford.EDU>
Subject: re: delay in aid to Nicaragua and Panama
To: John McCarthy <JMC@sail.stanford.edu>
Cc: su-etc@sail.stanford.edu
In-Reply-To: <4Y8sD@SAIL.Stanford.EDU>
Message-Id: <MS-C.642797186.1103527590.mrc@SUMEX-AIM.Stanford.EDU>
No, the Republicans do the same stupid trick of attaching unrelated amendments
to urgently needed bills to try to ram something through.
I don't see any particular necessity for public funds to be used to pay for
abortions; on the other hand consider that Republicans have attached anti-
abortion language to a lot of totally unrelated bills to push their ideology.
-------
∂15-May-90 1413 PHY Knuth
Don was in this morning and wanted to know if you had sent a reply
to his message of last week. Did he contact you directly?
-Phyllis
∂16-May-90 1423 @IBM.COM:LEORA@YKTVMH Patel-Schneider
Received: from IBM.COM by SAIL.Stanford.EDU with TCP; 16 May 90 14:23:15 PDT
Received: from YKTVMH by IBM.COM (IBM VM SMTP R1.2.1MX) with BSMTP id 0221; Wed, 16 May 90 13:51:32 PDT
Date: Wed, 16 May 90 16:41:11 EDT
From: LEORA@IBM.COM
To: jmc@sail.stanford.edu, val@sail.stanford.edu
Subject: Patel-Schneider
Dear John and Vladimir,
On Monday, I sent you some mail responding to your (John's) response
to Patel-Schneider's reaction to our workshop proposal. I suspect that
the mail didn't get through, since I got some strange messages from the
mailer, so I am resending it. Since then, I have also asked Ernie Davis
for his advice on this issue (hope you don't mind). Ernie suggested that
you (John) keep sitting on P-S's head. In particular, he suggested that
he may think that you are serving on the committee merely as a figurehead
(since the email containing the proposal came from my account), and that
you should disabuse him of this notion. In addition, he suggested that
you point out to Patel-Schneider that there are not more than a dozen
papers in AAAI 1988, IJCAI 1989, and KR 1989 conforming to the specifications
outlined in the proposal. So it certainly does not seem that our criteria
are too broad.
John, I tried to reach you before you left for Russia, but was unsuccessful.
Let's all speak together about this when you get back.
Leora
Date: 14 May 1990, 12:59:31 EDT
From: LEORA at YKTVMH
To: jmc at sail.stanford.edu, val at sail.stanford.edu
Subject: patel-schneider
My initial reaction is that if you think browbeating Patel-Schneider or
calling Bobrow will help, it makes sense to do it. It really will be easier
for us if the structure of the conference is already in place, and AAAI does
handle all of the arrangements. On the other hand, I am wondering if the AAAI
Symposium format is too restrictive for us. If AAAI is really that rigid
about what the topic of a symposium must be, we might be better off doing
something on our own.
If you do think that it makes sense for us to go it alone, we probably
will have to start working on arrangements, call for papers, etc. right away.
So we should probably come to a decision on this in the very near future.
At any rate, I must say that I am also amazed at Patel-Schneider's reaction,
particularly since AAAI is always claiming that they are trying to elevate
the quality of the Spring Symposium Series.
Leora
∂18-May-90 0227 ME your mail
To: JMC
CC: CLT, ME
I reset an E job of yours (on line 46) just now because it had your
mail file open and was idle for 3 days. That means that queued mail
couldn't be delivered to you for 3 days and would soon be returned to
sender (possibly some already might have been).
∂17-May-90 2234 linvill@sierra.Stanford.EDU Hennessy Nomination
Received: from sierra.Stanford.EDU by SAIL.Stanford.EDU with TCP; 17 May 90 22:34:12 PDT
Received: by sierra.Stanford.EDU (4.0/4.7); Thu, 17 May 90 22:34:27 PDT
Date: Thu, 17 May 90 22:34:27 PDT
From: linvill@sierra.Stanford.EDU (John G. Linvill)
To: Gibbons@sierra.Stanford.EDU, Horowitz@sierra.Stanford.EDU, Ullman@NIMBIN,
JMC@SAIL
Cc: Linvill@sierra.Stanford.EDU, Peterson@sierra.Stanford.EDU,
Cloutier@sierra.Stanford.EDU
Subject: Hennessy Nomination
Message-Id: <CMM.0.88.643008866.linvill@>
Colleagues,
Here is a draft of the Hennessy NAE nomination. Would you edit it and
make suggestions? In TIMES 12 it fits into the limited required space.
Joel Birnbaum, Gordon Bell and Dave Hodges have all agreed to serve as
references, though CS people, he says, have more credibility in the CS Peer
Group. Dave will serve, but recommends a computer scientist. John, Jeff says
you would serve as a reference. Would you?
How about Jim or Jeff serving as Nominator, the others as references?
NAE NOMINATION FOR JOHN HENNESSY
1. Name: John L. Hennessy
2. Primary Position and Affiliation:
Professor of Electrical Engineering and Computer Science
Center for Integrated Systems, 208
Stanford University
Stanford, CA 94305-4070
(415)725-3712
3. Date and Place of Birth: September 22, 1952; (Please find place from EE Office)
4. Citizen of the United Sates
5. Education
BE, Elec. Eng., Villanova University, 1973
MS, Computer Science, State University New York at Stony Brook, 1975
PhD, Computer Science, State University New York at Stony Brook, 1977
6. Registered Engineer (blank)
7. Engineering Category Computer Science and Engineering
8. Specific Outstanding Technical and Professional Engineering Accomplishments
Meriting Election to the NAE
John Hennessy first recognized that performance of Reduced Instruction Set Computers
involves both hardware and software issues and that these must be addressed together. The research
he led at Stanford with his students and colleagues produced innovations in computer architecture
(load-store architectures, simple instruction sets, user-visible pipelining, and large register files) and in
software techniques (Uopt optimizing compiler system and machine code scheduling). From the
research emerged a better understanding of the factors that affect computer performance and better
tools to measure it. All computers designed subsequently have benefitted from this work.
Hennessy, with colleagues, founded MIPS Computer Systems to implement the ideas
developed in the Stanford research group, and he remains as Chief Scientist at MIPS. He spent a
sabbatical year working at MIPS contributing to the present dominance of RISC architecture in high-
speed computing.
Hennessy's research led to higher performance computers. But it also led to a set of salient
publications illuminating hardware and software issues in computer architecture. He and David
Patterson (of Univ. of Calif. at Berkeley, leader of the RISC research team there and the first
university researcher to work in RISC computers) promptly wrote a book, "Computer Architecture: A
Quantitative Approach" which has just been published.
Hennessy's career, centered in university research and teaching, is coupled to a vigorous
industry which he stimulates in a powerful way. His overall impact on American technology,
innovation and competitiveness is unique.
9. Record of Professional Experience
1987 - present Dir., Computer Systems Laboratory, Stanford University
1986 - present Professor of Electrical Engineering and Computer Science,
Stanford University
1984 - present Chief Scientist and Co-Founder, MIPS Computer Systems
1983 - 1986 Associate Professor of Electrical Engineering and
ComputerScience (by courtesy)
1977 - 1983 Assistrant Professor of Electrical Engineering and Computer
Science (by courtesy)
10. Contributions of Record- Publications [no. (get from Margaret Rowland) ],
Patents [no. (from Margaret)], one book and contribution to a book.
The following publications are especially relevant to Hennessy's contributions.
J. L. Hennessy and D. A. Patterson, "Computer Architecture: A Quantitative Approach,"
Morgan Kaufman, San Mateo, CA 1990 A definitive text for computer architecture and
design.
C. Rowen, S. Przybylski, N. Jouppi, T. Gross, J. Shott and J. L. Hennessy, "MIPS: A
High Performance 32-Bit NMOS Microprocessor," in Digest of Int. Solid-State Circuits
Conf., IEEE, San Francisco, CA Feb. 1984. A paper by Hennessy's team on the
Stanford MIPS microprocessor featuring a high performance compiler scheduled pipeline.
T. Gross, J. L. Hennessy, S. Przybylski and C. Rowen, "Measurement and Evaluation of
the MIPS Architecture and Processor, " ACM Trans. Comput. Sys. 6(3):229-257, Aug.
1988 The first detailed evaluation of a RISC processor to be published.
11. Principal Technical Society Memberships ---
IEEE, Senior Member
Editor, IEEE Design and Test, 1984, 1985, 1986.
Assoc. Editor, IEEE MICRO, 1981, 1982.
Member, Computer Science and Technology Board, National Research Council, 1989-92.
Member, Committee to Study International Developments in Computer Science and
Technology, National Academy of Sciences, 1988.
12. Professional Recognition
Willard and Inez Bell Endowed Professor of Electrical Engineering and Computer Science,
Stanford University, 1987 - .
Presidential Young Investigator, National Science Foundation, 1984.
John J. Gallen Memorial Award, Villanova University, 1983; awarded to the most
outstanding young engineering alumnus.
13. Proposed Citation
For pioneering contributions in hardware and software for RISC Architecture, and for
conception of test measures verifying its performance advantages.
Nominator and References
Gordon Bell David Hodges
Joel Birnbaum James Gibbons
Jeffrey Ullman John McCarthy
Regards, John
∂15-May-90 1425 korf@CS.UCLA.EDU Re: Let's postpone getting together
Received: from shemp.cs.ucla.edu by SAIL.Stanford.EDU with TCP; 15 May 90 14:25:17 PDT
Received: from Denali.CS.UCLA.EDU by shemp.cs.ucla.edu
(Sendmail 5.61a+YP/2.31) id AA07579;
Tue, 15 May 90 14:26:10 -0700
Message-Id: <9005152126.AA07579@shemp.cs.ucla.edu>
Received: by denali; Tue, 15 May 90 14:40:05 pdt
Date: Tue, 15 May 90 14:40:05 pdt
From: Richard E Korf <korf@CS.UCLA.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 15 May 90 0803 PDT <4Y75G@SAIL.Stanford.EDU>
Subject: Re: Let's postpone getting together
John,
That's ok with me. I'm baically happy with the program I now have, and will
wait until we can get together to discuss it before thinking about writing
anything. I'd like to get together before the end of July, however, since I'll
be travelling for most of the rest of the summer. I plan to be in the bay area
fairly regularly, and will let you know when I'm up in your neighborhood.
-rich
∂16-May-90 0935 MPS
Eric Dietrich - dietrich@bingvaxu.cc.binghamton.edu
He needs your abstract or paper for the symposium at
U. of Md., June 8th
∂16-May-90 1108 weening@Gang-of-Four.Stanford.EDU Top Level Common Lisp
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 16 May 90 11:08:43 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA12552; Wed, 16 May 90 11:09:10 -0700
Date: Wed, 16 May 90 11:09:10 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005161809.AA12552@Gang-of-Four.Stanford.EDU>
To: qlisp@Gang-of-Four.Stanford.EDU
Subject: Top Level Common Lisp
"Parallel Computing News", from the Northeast Parallel Architectures
Center at Syracuse University, volume 3 number 4 (April 1990) has an
article about a new product called "Top Level Common Lisp", from Top
Level Inc. Has anyone heard anything about this?
The article is not very technical; it mostly lists features of the
system such as:
"future objects, which provide implicit synchronization"
"a choice of four different fork operators, each associated with
different amounts of overhead"
"external debugger ... eliminates the deadlock problems that occur
in internal debuggers"
"parallel compiler ... features a three-process pipeline"
It is designed for shared-memory MIMD machines, but they don't say
which ones it runs on.
∂16-May-90 2312 @MCC.COM:ai.guha@MCC.COM between the devil and the deep sea
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 16 May 90 23:12:48 PDT
Received: from pp.aca.mcc.com by MCC.COM with TCP/SMTP; Thu 17 May 90 01:13:44-CDT
Posted-Date: Thu, 17 May 90 01:13 CDT
Received: by pp.aca.mcc.com (5.51/ACAv4.1i)
id AA01149; Thu, 17 May 90 02:11:15 CDT
Date: Thu, 17 May 90 01:13 CDT
From: R. V. Guha <ai.guha@MCC.COM>
Subject: between the devil and the deep sea
To: hayes.pa@xerox.com, jmc@sail.stanford.edu
Message-Id: <19900517061301.5.GUHA@GAIA.ACA.MCC.COM>
Hi,
I was wondering if you had gotten a chance to read
the document I gave you last week (on contexts and discourses).
If you have I have a couple of questions.
They pertain to a lot of flak I have been getting very
recently from two groups of people. Both of them are
related the notion that utterances in nl may be converted
to wffs that are supposed to be true in some context.
The two sets of comments I have had to defend against are
(a) From the hard-core-conservative linguists. Their
basic criticisms are as follows.
- given an utterance, it has a well defined interpretation (which
is a wff that is universally true) and there are a set
of implications from this wff. Saying that the interpretation
of an utterance is a wff holds
in a context allows for the possibility for interpreting the
utterance from different contexts which allows for multiple
interpretations.
- Having a context argument in the interpretation allows for
potentially anything to affect the meaning of an utterance.
This I believe goes against one of the basic assumptions of
traditional linguistics that the meaning of an utterance
depends on a small, well-defined number of parameters (the
preceeding one or two utterances etc.)
- Since different contexts can have different wffs hold
in them, the set of wffs mentioning a particular predicate
can be very different from one context to another. If these
wffs mentioning a predicate can be thought of as defining the "intentional
meaning" of the predicate, it now becomes possible to have
different contexts using a particular predicate with the
meaning of the predicate differing slightly from one context
to another.
All these I am told are complete no-no's as far as the linguists
(and for any theory that has anything to say about discourses etc.).
(b) From some of the situated semantics folks - mostly their
criticism is that it is just not possible to build this whole
thing from logic and that my construction (using the predicate
ist, the notion of fluents etc.) is really subverting logic
completely.
I am really quite confused as to how I should react to both
these criticisms. I should probably ignore the criticism from
the situated folks, but am not so sure about the linguists.
Do you have any suggestions or advice?
Hoping to hear from you soon.
Guha
∂17-May-90 1402 harnad@clarity.Princeton.EDU Re: spp abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 17 May 90 14:02:30 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA05414; Thu, 17 May 90 16:59:46 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA02876; Thu, 17 May 90 16:59:48 EDT
Date: Thu, 17 May 90 16:59:48 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005172059.AA02876@reason.Princeton.EDU>
Subject: Re: spp abstract
Cc: andrewsj@vassar.BITNET, harnad@clarity.Princeton.EDU
> Date: Thu, 17 May 90 15:07:28 -0500
> From: Robert Port <port@iuvax.cs.indiana.edu>
> To: harnad@Princeton.EDU
> Subject: spp abstract
>
> Sorry not to have gotten back to you. Im afraid slipped my very porous
> mind. I will give you a call in a couple minutes. But, to give you an
> idea what I can contribute to the conference , here is an abstract -
> for the talk that I may or may not be providing.
>
> I will call this afternoon to discuss what to do about the conference.
> I agree that my ideas could be tied to the the Chinese Room problem but
> Im not at sure I am the right person to make that case. I would rather
> make my primary contribution to the grounding discussion and just be a
> chair for the Searle group. At least if that will work for you. I will
> call.
>
> Icons and Temporal Patterns: a Dynamic Connectionist Solution
> to Symbol Grounding
>
> Harnad (1990) proposes that categories can be grounded by their direct
> relationship to a physical icon of the input stimulation. The notion of
> an icon has a clear meaning in the case of a visual display: it is a
> pattern of activity in a field of neurons that is physically isomorphic
> with the pattern of light in 2D. The standard proposal for a physical
> icon of TIME is physical distance. Such a model is naturally
> implemented by delay lines in a network (see Lang, 1990, NN).
>
> But delay lines are not a good model for human behavior. When subjects
> are trained on a complex temporal pattern, like a random sequence of
> tones, they can develop a detailed perceptual representation (Spiegel &
> Watson 1981, Watson and Foyle, 1985, JASA). Some skills that should be
> easy are very difficult -- eg, recognizing an absolute time interval in
> the face of randomly varying intervening sounds. This should be easy
> since absolute time differences are represented by weights on specific
> delay lines. So edges from inputs that have a random relation to the
> categorical identification should learn random weights. On the other
> hand, skills that should be very difficult turn out to be easy, such as
> detecting serial order of familiar patterns in the face of changes in
> rate of presentation. This should be difficult since a pattern that
> appears at different rates will be distributed differently across the
> range of delays, and should thus be learned only slowly.
>
> On other hand, we have been developing dynamic network models that
> represent learned temporal patterns of tones as stable equilibria in
> the activation space of a group of fully recurrent nodes
> (Port-Anderson, 1990, Anderson-Port, 1990). These systems were trained
> (with real-time recurrent learning) to recognize particular tone
> sequences. They are highly resistant to noise and continue to recognize
> patterns even when the rate of presentation is varied by a factor of 2
> faster or slower. Watson's results and our simulations suggest that
> brains do not produce an icon of auditory patterns in time. It implies
> that direct contact with stimulation that is distributed in time is not
> possible. Although I do not disagree with Harnad on the importance of
> symbol grounding for an account of perception, apparently, the
> grounding of categories does not require an `icon' in the sense that
> Harnad has proposed.
Bob, your abstract is very interesting and very relevant to the symbol
grounding workshop, but unless someone drops out, my hands are tied on
that score because there simply isn't enough time for another formal
presentation to the workshop. I promise that if anyone drops out, you
can present it there. The Searle Symposium is related, as you know,
but unfortunately, your own contribution does not bear directly on the
relationship. Let me try to sketch what a relevant contribution might
do:
The speakers in the Searle Symposium (all anti-Searle) will have to
claim either that the symbols in a purely symbolic program that speaks
indistinguashably from a human speaker ARE grounded, just as human
symbols are, or that they need not be grounded (and perhaps neither do
human symbols). "Grounding" can have two (related) meanings. For
Searle, it means having INTRINSIC meaning, the way our symbols do, as
opposed to DERIVED or PARASITIC meaning, as the symbols of a book (or,
he would argue, a computer program) do. The second sense of grounding,
mine, is that a symbol is grounded if and only if the device that
generates the symbol can not only use the symbol just as people do in
discourse, but can also pick out the objects and states of affairs the
symbols and symbol strings refer to in the world. If it can't, if it
only crunches symbols (be they ever so systematically interpretable
by us) then the symbols are as ungrounded as those in the
Chinese/Chinese dictionary I mention in my paper.
Now, in order to be grounded in my sense -- to be able to discriminate,
categorize, name and describe the objects and states of affairs the
symbols and symbol strings stand for -- the device must clearly begin
with the sensory PROJECTION (notice I'm not saying the sensory ICON
yet). That is simply the effect of the energetic input from the object
in question on the device's transducer surfaces. This effect includes
both spatial and temporal dimensions. Icons, for me, come in only when
it comes to RELATIVE DISCRIMINATION -- judgments of same different,
more, less -- when the stimuli being compared are both present, or
presented in rapid succession. For that kind of task, comparing their
sensory ICONs (analogs of the sensory projection) seems sufficient. For
ABSOLUTE IDENTIFICATION (categorization), however (which you need in
order to reliably name any object whose sensory projection varies in any
nontrivial way from presentation to presentation among CONFUSABLY
SIMILAR ALTERNATIVES), icons are insufficient, and INVARIANT FEATURES
of the sensory projection must be found and learned. That's what I use
nets for, and it's not clear you disagree with me: Time invariance is a
perfectly legitimate invariance!
In any case, my objective is not to resolve our apparent disagreement,
but to point out that, to address the Searle issue, you would have to
address the question of whether or not sensory grounding -- not just
in icons, but in invariant features of the sensory projection,
including learned and context dependent ones -- is necessary to give
symbols meaning: I.e., whether or not you can have meaning without it.
The anti-Searleans, who say meaning is just the right pattern of
symbol manipulations, will argue that grounding is not necessary, the
symbol crunching is enough. What bearing, if any, does your work have
on THAT?
If it does, and you can make it explicit, than you can replace Alexis
Manaster-Ramer in the Searle Symposium (and be the only
non-anti-Searlean in the Symposium). If not, all I can do is leave you
chairing the Symbol Grounding Symposium, participating actively in the
discussion, and pinch-hitting if anyone drops out.
Please let me know.
Best wishes,
stevan
∂17-May-90 1649 pehoushe@Gang-of-Four.Stanford.EDU A good way to implement qlambda
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 17 May 90 16:48:54 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA19767; Thu, 17 May 90 16:49:20 -0700
Date: Thu, 17 May 90 16:49:20 -0700
From: pehoushe@Gang-of-Four.Stanford.EDU (Dan Pehoushek)
Message-Id: <9005172349.AA19767@Gang-of-Four.Stanford.EDU>
To: qlisp@Gang-of-Four.Stanford.EDU
Subject: A good way to implement qlambda
A quick review: Qlambda is basically a serializing closure with a lock
to control access. It is currently implemented with a sleep-lock.
The following discussion applies to the FIFO variety of sleep-lock,
and is probably only interesting to people interested in implementing
a Queue of Queues.
In Lucid Qlisp, evaluating a Qlambda T form dedicates system resources
(a Process shell) to running all calls to the qlambda. This causes a
few resource allocation problems, such as GCing qlambdas, among others
(tieing up the process shell even when the qlambda is idle). The
streams of Erasthones (computing primes via qlambdas) is quite useful
as a sample qlambda program. It also thoroughly embarasses the Lucid
implementation of Qlambda. A call to a Qlambda takes between 1 and 3
milliseconds, plus GC and general resource problems. In my version, a
call to qlambda takes <100 Microseconds, and with no serious resource
problems.
A way around some of these resource problems is to implement qlambdas
via closures with individual queues; the closures go away when nothing
points to them (including the scheduling system, which doesn't point
to empty qlambdas). And an inactive but callable qlambda would not
take up precious Process shell resources; even qlambdas with calls
waiting to be run do not take up Process resources, but only require
memory to store the currently queued calls.
A few parts of the implementation are:
;; A single Qlambda structure
(defstruct qlambda
(lock 0 :type fixnum) ;; slot for spin lock.
(size 0 :type fixnum) ;; Current number of calls waiting to run.
(head nil) ;; local head of queue
(tail nil) ;; local tail of queue
(status nil)
(next-in-queue nil)) ;; Ptr to next Qlambda structure. Important!
There is a global queue of qlambda structures which idle processors
check for work. This is essentially a "Queue of Qlambda Queues."
Currently implemented as a macro, each evaluation of a qlambda form
allocates a qlambda data structure (for maintaing its fifo queue of
callers) and creates two closures; The spawner closure queues up a
call to the spawnee closure. Some logical details are hidden in
qlambda-spawn, and can be elucidated upon request. But the idea is
that when nothing points to the Spawner closure, the empty data
structure goes away. And the only time a qlambda takes up precious
process resources is when it computes.
(defmacro qlambda (prop lambda-list &rest body)
(let ((qlambda (gensym))
(spawner (gensym))
(spawnee (gensym)))
`(let ((,qlambda (make-qlambda)))
(labels ((,spawnee ,lambda-list
,@body)
(,spawner ,lambda-list
(qlambda-spawn ,qlambda ,prop (,spawnee ,@lambda-list))))
#',spawner))))
Dan
PS. All of the above has absolutely no relationship to N stack scheduling or
dynamic spawning.
∂18-May-90 0550 XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 18 May 90 05:50:42 PDT
Received: by Forsythe.Stanford.EDU; Fri, 18 May 90 05:52:25 PDT
Received: from PCSITI.FB20.THD.DA.EUROPE by DDATHD21.BITNET
via GNET with RJE with RCOM ; 18 May 90 14:46:27
Date: Fri May 18 13:41:35 MEZ 1990
From: XIISWBIB%DDATHD21.BITNET@Forsythe.Stanford.EDU
To: jmc@sail.stanford.edu,
nilsson@Tenaya.stanford.edu
X-Munix-To: amarel@aramis.rutgers.edu, bledsoe@cs.utexas.edu, bobrow@parc.xerox.
mylopoulos@ai.toronto.edu, nilsson@Tenaya.stanford.edu, reiter@ai.toronto.ed
Subject: Intellectics
For the next edition of the Encyclopedia for AI I was asked to write
an entry on the term "Intellectics" which I wanted to let you know in
advance. Any comments are welcome.
Wolfgang
%*****************************************************************************/
%* */
%* Version: 1.00 Date: --/--/-- File: .tex */
%* Last Version: File: */
%* Changes: */
%* --/--/-- */
%* */
%* Title: */
%* Author: */
%* */
%* Usage: */
%* */
%* */
%*****************************************************************************/
\documentstyle[12pt]{article}
\setlength{\textwidth}{145mm} \setlength{\textheight}{220mm}
\setlength{\parindent}{1Em} \setlength{\parskip}{1ex}
\setlength{\topmargin}{0mm} \setlength{\headsep}{0mm}
\setlength{\headheight}{3mm} \setlength{\headsep}{8mm}
\setlength{\footheight}{5mm} \setlength{\footskip}{10mm}
\global\evensidemargin=15mm \global\oddsidemargin=15mm
\begin{document} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\title{Intellectics}
\author{W.\ Bibel \\ Technical University Darmstadt}
\date{}
\maketitle
The term ``Artificial Intelligence'' first appeared in the project proposal by
J.\ McCarthy, M.\ Minsky, and C.\ Shannon to the Rockefeller Foundation in
1956 where they proposed to hold the (by now famous) Dartmouth Conference
\cite{McCor:1}. McCarthy had in mind to denote the {\em object\/} of studies
with this term. For the scientific field itself concerned with these studies
he later proposed the name ``Cognology''.\footnote{This name appears in the
title of a talk McCarthy gave at a conference held near Leningrad (Repino) in
1977.} Whatever the reasons are, the term ``Artificial Intelligence'' got into
use for the object {\em and\/} as a name for the field alike. The term has
since been the cause of much controversy.
Part of this debate is concerned with the underlying anthropomorphism that
attributes to machines the name of a capacity, namely intelligence, thought to
be genuinely {\em human\/}. This concern rationally is not well justified
since a proper distinction is made in the term by the adjective ``Artificial''
(but there remains an irrational component in the argument).
People with a good sense for language are more concerned with the confusion
made by the term of the studied object with the field itself. Imagine if
botany, the field studying plants, were just called ``plants'' as well. Or how
would ``man'' sound instead of anthropology? It is for exactly this good
reason that McCarthy has chosen different names for field and for object. He
may not have made the best choice of a good name for the field.
The community is not unaware of these problems and therefore tends to
substitute ``Artificial Intelligence'' by ``AI''. Many people, especially
again those with some sense left for a good use of natural language, agree
that this exciting area deserves a better name than ``AI'', indeed.
For these reasons, in 1980, {\em Intellectics\/} came up as a new name for the
field that studies artificial and natural intelligence \cite{bi:E10}. This
name seems like a rather perfect construction. It contains the object of
studies, intelligence, in the root of the word. It reminds of the intellect
which may be thought of as the ``machinery'' in man that produces
intelligence, thus quite appropriately covering what is done in the field.
Finally, it implicitly emphasizes both aspects, the artificial and the human
one.
For the latter reason, Intellectics covers the {\em union\/} of the fields of
Artificial Intelligence and Cognitive Science, a recent friction that is
unfortunate for both communities anyway (and to some extent may even have its
cause in the lack of an appropriate name for the entire field).
There are few things in the world that are absolutely perfect, a rule which
also applies to the name ``Intellectics''. For one, the word's radical is of
Latin origin while the suffix derives from Greek. Such a combination has
precedences and should not be the cause of much worry, however.
``Intellectics'' also does not explicitly include a connotation that accounts
for the importance of the senses for intelligence. But this is only because
the original meaning of the Latin word ``intellegere'' has shifted somewhat
over times. It certainly does not {\em ex\/}clude a role for the senses in
this context.
In any case, no term, however good, will ever become perfect unless it is used
by the community, while a bad term like ``AI'' will not become better by
continuously using it. At the point of writing there are a number of
institutions in Europe that have adopted the term ``Intellectics'', among them
the French {\em Association pour la recherche cognitive\/} which carries the
term in its statutes, the Universities {\em Technische Universit{\"a}t
M{\"u}nchen\/} and {\em Technische Hochschule Darmstadt\/} that use it
officially for naming our field with it, to mention but a few.
\bibliographystyle{alpha}
\bibliography{lit,my-publ,/usr2/wiss/bibel/arbeiten/vollendet/eder/Hab}
\end{document} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
∂18-May-90 0729 boyer@CLI.COM Bledsoe Symposium
Received: from CLI.COM by SAIL.Stanford.EDU with TCP; 18 May 90 07:29:43 PDT
Received: by CLI.COM (4.0/1); Fri, 18 May 90 09:29:41 CDT
Date: Fri, 18 May 90 09:29:41 CDT
From: Robert S. Boyer <boyer@CLI.COM>
Message-Id: <9005181429.AA22911@CLI.COM>
To: jmc@sail.stanford.edu
Subject: Bledsoe Symposium
Reply-To: boyer@CLI.COM
I have spoken to you about a festschrift that Kluwer is publishing in
Bledsoe's honor, and you kindly agreed to contribute an article. This
note is a reminder that the article is due December 15, 1990.
Detailed instructions will follow shortly.
I am soliciting funds for a Symposium for Woody Bledsoe, which is to
be held November 11 and 12, 1991, shortly after his 70th birthday.
(The Computer Sciences Department has agreed to put up some funds, and
I think that I have got some elsewhere lined up.) My records do not
show that you have indicated whether you would come speak. I would
like to list you as one of those who will or might come. So could you
please indicate your position on this? I certainly think that listing
your name would increase my chances on getting funding.
Thanks,
Bob
∂18-May-90 1004 VAL Commonsense and Nonmonotonic Reasoning Seminar
To: "@CS.DIS[1,VAL]"@SAIL.Stanford.EDU
From: Vladimir Lifschitz <VAL@SAIL.Stanford.EDU>
There will be no more meetings of the nonmonotonic seminar this quarter.
See you in October.
--Vladimir
∂18-May-90 1559 visikka@cmx.npac.syr.edu
Received: from cmx.npac.syr.edu by SAIL.Stanford.EDU with TCP; 18 May 90 15:59:26 PDT
Date: Fri, 18 May 90 18:58:33 EDT
From: visikka@cmx.npac.syr.edu (Vishal Sikka)
Received: by cmx.npac.syr.edu (5.57/2.1-cmx at NPAC)
id AA07701; Fri, 18 May 90 18:58:33 EDT
Message-Id: <9005182258.AA07701@cmx.npac.syr.edu>
To: jmc@sail.stanford.edu
To:
Prof. John McCarthy.
CSD, Stanford.
Respected sir,
Hello! My name is Vishal Sikka, and if you recall I had a talk
with you last summer about my interest in AI and my desire to study at
Stanford. This was upon the inspiration of Prof. Minsky. I am glad to
tell you that I have been accepted at Stanford, and I shall be continuing
my education there beginning with this autumn quarter.
I am in the San Francisco bay area these days, and I was wondering
if I could meet with you sometime on monday to learn more about your
current research in AI. You had given me a copy of your (then) most recent
paper, when I visited you the last time. I have a substantiial research
experience in AI already, more of a superficial kind, touching various
aspects, and now I want to concentrate on something that appears interesting
as well as promising. So kindly suggest a convenient time, sir.
Thanx,
Vishal.
∂18-May-90 2128 hamilton@cs.sfu.ca Abstracts for AAAI?
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 18 May 90 21:28:10 PDT
Received: from relay.cdnnet.ca by june.cs.washington.edu (5.61/7.0jh)
id AA27146; Fri, 18 May 90 21:19:57 -0700
From: hamilton@cs.sfu.ca
Return-Path: <hamilton@cs.sfu.ca>
Received: by relay.CDNnet.CA (4.1/1.14)
id AA12186; Fri, 18 May 90 21:20:35 PDT
Date: 18 May 90 21:17 -0700
To: qphysics@cs.washington.edu
Message-Id: <9005190417.AA25565@cs.sfu.ca>
Subject: Abstracts for AAAI?
Why is there no activity on this mail list?
What I'd like to see is the abstracts for papers to be presented
soon at AAAI-90 or other conferences that are related to
qualitative physics. So, how about if everyone who has such
a paper mails the abstract to
qphysics@cs.washington.edu
? You could put "ABS: <your title> as the subject.
What other kinds of uses can we put this mail list to?
Howard Hamilton
∂19-May-90 2034 weld@cs.washington.edu Abstracts for AAAI?
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 19 May 90 20:34:40 PDT
Received: by june.cs.washington.edu (5.61/7.0jh)
id AA18230; Sat, 19 May 90 20:26:19 -0700
Date: Sat, 19 May 90 20:26:19 -0700
From: weld@cs.washington.edu (Dan Weld)
Return-Path: <weld@cs.washington.edu>
Message-Id: <9005200326.AA18230@june.cs.washington.edu>
To: hamilton@cs.sfu.ca
Cc: qphysics@cs.washington.edu
In-Reply-To: hamilton@cs.sfu.ca's message of 18 May 90 21:17 -0700 <9005190417.AA25565@cs.sfu.ca>
Subject: Abstracts for AAAI?
Well I agree - this list is much too quiet.
I feel slightly guilty posting my abstract since it describes work that is
very similar to what I presented at QPW-89, but here goes anyway.
APPROXIMATION REFORMULATIONS
Although computers are widely used to simulate complex physical systems,
crafting the underlying models that enable computer analysis remains
difficult. When a model is created for one task, it is often impossible to
reuse the model for another purpose because each task requires a different
set of simplifying assumptions. By representing modeling assumptions
explicitly as approximation reformulations, we have developed qualitative
techniques for switching between models. We assume that automated
reasoning proceeds in three phases: 1) model selection, 2) quantitative
analysis using the model, and 3) validation that the assumptions underlying
the model were appropriate for the task at hand. If validation discovers a
serious discrepancy between predicted and observed behavior, a new model
must be chosen. We present a domain independent method for performing this
model shift when the models are related by an approximation reformulation
and describe a Common Lisp implementation of the theory.
Let's hear some more!
Dan
∂21-May-90 0740 hogge@m2.csc.ti.com Qphysics Applications
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 21 May 90 07:40:46 PDT
Received: from TI.COM by june.cs.washington.edu (5.61/7.0jh)
id AA28915; Mon, 21 May 90 07:29:27 -0700
Return-Path: <hogge@m2.csc.ti.com>
Received: by ti.com id AA19133; Mon, 21 May 90 09:32:16 -0500
Received: from m2.csc.ti.com by tilde id AA20887; Mon, 21 May 90 09:27:19 CDT
Received: by m2.csc.ti.com (5.52/4.7)
id AA27765; Mon, 21 May 90 09:26:47 CDT
Date: Mon, 21 May 90 09:26:47 CDT
From: hogge@m2.csc.ti.com (John Hogge)
Message-Id: <9005211426.AA27765@m2.csc.ti.com>
To: qphysics@cs.washington.edu
Subject: Qphysics Applications
I've been out of Qphysics work for several years now and have several
questions:
1. How well is it being applied? Does anyone have a list of successful
qphysics applications--perhaps a list you use when seeking research funds?
2. Are there specific companies concentrating significant resources in
qphysics applications?
Regards,
--John
∂21-May-90 0844 jc@cs.utexas.edu ABS: QPC: A Compiler from Physical Models into Qualitative Differential Equations
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 21 May 90 08:44:54 PDT
Received: from cs.utexas.edu by june.cs.washington.edu (5.61/7.0jh)
id AA01768; Mon, 21 May 90 08:38:25 -0700
Return-Path: <jc@cs.utexas.edu>
Received: from ai.cs.utexas.edu by cs.utexas.edu (5.61/1.62)
id AA23653; Mon, 21 May 90 10:39:11 -0500
Posted-Date: Mon, 21 May 90 10:38:47 CDT
Message-Id: <9005211538.AA29457@ai.cs.utexas.edu>
Received: by ai.cs.utexas.edu (5.59/1.4-Client)
id AA29457; Mon, 21 May 90 10:38:50 CDT
From: jc@cs.utexas.edu (James M. Crawford)
Date: Mon, 21 May 90 10:38:47 CDT
X-Mailer: Mail User's Shell (6.5.6 6/30/89)
To: qphysics@cs.washington.edu
Subject: ABS: QPC: A Compiler from Physical Models into Qualitative Differential Equations
Cc: kuipers@cs.utexas.edu
To appear at AAAI-90:
QPC: A Compiler from Physical Models into Qualitative Differential Equations
James Crawford, Adam Farquhar, and Ben Kuipers
Department of Computer Sciences
University of Texas at Austin
Austin, Texas 78712
jc@cs.utexas.edu farquhar@cs.utexas.edu kuipers@cs.utexas.edu
Abastract:
Qualitative reasoning can, and should, be decomposed into a
model-building task, which creates a qualitative differential equation
(QDE) as a model of a physical situation, and a qualitative
simulation task, which starts with a QDE, and predicts the possible
behaviors following from the model.
In support of this claim, we present QPC, a model builder that takes the
general approach of Qualitative Process Theory [Forbus, 1984],
describing a scenario in terms of views, processes, and influences.
However, QPC builds QDEs for simulation by QSIM, which gives it access
to a variety of mathematical advances in qualitative simulation incorporated
in QSIM
We present QPC and its approach to Qualitative Process Theory, provide
an example of building and simulating a model of a non-trivial
mechanism, and compare the representation and implementation decisions
underlying QPC with those of QPE [Falkenhainer and Forbus, 1988; Forbus, 1990].
∂21-May-90 1741 arg@lucid.com Qlisp at toplevel
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 21 May 90 17:38:45 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA09655; Mon, 21 May 90 17:38:53 -0700
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA16525g; Mon, 21 May 90 17:38:16 PDT
Received: by bhopal id AA25567g; Mon, 21 May 90 17:38:50 PDT
Date: Mon, 21 May 90 17:38:50 PDT
From: Ron Goldman <arg@lucid.com>
Message-Id: <9005220038.AA25567@bhopal>
To: qlisp@go4.stanford.edu
Subject: Qlisp at toplevel
The time has finally come to have Qlisp go into parallel mode when it starts
up and stay there during the top level read-eval-print loop. This would
eliminate the need for QEVAL and, if in evaluating a form futures are created
whose values are not immediately needed, then the associated processes could
continue to run after the top level loop has printed out the "value" of the
form. Below is a description of a number of new functions that will be
needed/available:
1) The function:
(change-number-of-processors n) ; 1 <= n <= max-number-of-processors
will allow the user to dynamically change the number of processors being
used by Qlisp. The global variable *number-of-processors* will now be a
read-only variable. (When only using one processor it's now an error to
try to acquire a locked spin lock.)
If for some reason it's necessary to run Qlisp in serial mode this
can be done by calling CHANGE-NUMBER-OF-PROCESSORS with an argument
of :serial instead of a number. This can only be done from top level
and will result in killing all currently running user processes. To
re-enter parallel mode just call CHANGE-NUMBER-OF-PROCESSORS again
with a valid numeric argument.
2) To kill all current user processes:
(kill-all-processes)
which will force all user processes to run their unwind-protect cleanup
forms and go away. Note that qlambda processes don't get destroyed, but
rather they stop what they're doing, their queues are flushed and they
revert to an idle state, waiting to be called.
Note that this function should only be called from top level.
3) To put the system into a "clean" state so timings can be done repeatably:
(reset-world)
which will call KILL-ALL-PROCESSES and then do a GC to flush old futures
and qlambda process closures that are no longer pointed to.
Note that it's up to the user to reset any user locks that may have been
left set along with any other user specific reinitialization.
4) To let the user know about running processes the toplevel prompt will be
changed to include the number of actively running processes. There will
also be a way to disable this verbose prompt. Likewise the debugger
prompt will include both actively running processes and processes waiting
to enter the debugger. A more full report on the status of user processes
will be possible using:
(qstatus)
which will report the number of processes:
a) created / scheduled
b) active : ready to run / waiting for (1) futures
(2) locks
(3) work (qlambda's)
(4) being resumed (suspended)
5) When the user commands the debugger with :a to return to toplevel, or to a
higher debugger level, all of the processes created during the evaluation
of the current toplevel/debugger will be automatically killed.
6) Disksave can only be done from toplevel. Any running user processes will
first be killed (by KILL-ALL-PROCESSES) before the image is saved. Qlambda
process closures will be saved and when the new image starts up new
processes for them will automatically be spawned. If the user wants the
new image to have any other running user processes then a user function
with code to spawn them needs to be called when the new image is started up.
(C.f. documentation on the :restart-function argument to disksave).
7) Good timing information will become more of a problem on the Alliant.
Time values obtained from the get_rusage system call on the Alliant are
really only accurate to 10 milliseconds, the tick rate of the system
time slice clock. When making such calls in concurrent mode, the
breakdown of runtime into user and system components is bogus --- user
time is often zero or negative! --- though the sum or total runtime seems
accurate.
Now that a form may return before all of the processes it has spawned
have finished (time (form)) may report quite different times then
(time (qwait (form))). So I propose that we change QTIME so that
(qtime form) will be equivalent to (time (qwait form)).
8) Another problem with the Alliant OS is that when the toplevel
read-eval-print loop does the read, the entire Qlisp task is suspended,
so no progress can be made by any other still running processes. The
new function:
(qwait-all)
will not return until all the running processes finish or are waiting
for a lock/future/etc.
∂22-May-90 0608 @grenada:wellman@wrdc.af.mil Abstracts for AAAI?
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 22 May 90 06:08:09 PDT
Received: from [134.131.28.1] by june.cs.washington.edu (5.61/7.0jh)
id AA12079; Tue, 22 May 90 05:56:44 -0700
Return-Path: <@grenada:wellman@wrdc.af.mil>
Received: by wrdc.af.mil (5.57/Ultrix 3.0)
id AA06784; Tue, 22 May 90 08:54:27 EDT
Date: Tue, 22 May 90 08:54 EDT
From: Mike Wellman <wellman@wrdc.af.mil>
Subject: Abstracts for AAAI?
To: hamilton@cs.sfu.ca
Cc: qphysics@cs.washington.edu, wellman@wrdc.af.mil
In-Reply-To: <9005190417.AA25565@cs.sfu.ca>
Message-Id: <19900522125431.8.WELLMAN@[134.131.24.6]>
This paper is to appear in the Proceedings of the Sixth Conference on
Uncertainty in AI:
Exploiting Functional Dependencies in Qualitative Probabilistic Reasoning
Michael P. Wellman
Functional dependencies restrict the potential interactions among variables
otherwise connected in a probabilistic network. This restriction can be
exploited in qualitative probabilistic reasoning by introducing deterministic
variables and modifying the inference rules to produce stronger conclusions
in the presence of functional relations. I describe how to accomplish these
modifications in qualitative probabilistic networks by exhibiting the update
procedures for graphical transformations involving probabilistic and
deterministic variables and combinations. A simple example demonstrates
that the augmented scheme can reduce qualitative ambiguity that would
arise without the special treatment of functional dependency. Analysis of
qualitative synergy reveals that new higher-order relations are required to
reason effectively about synergistic interactions among deterministic
variables.
∂22-May-90 0851 mkatz@garlic.Stanford.EDU Qlisp at toplevel
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 22 May 90 08:51:08 PDT
Received: from garlic.Stanford.EDU by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA11985; Tue, 22 May 90 08:51:32 -0700
Received: by garlic.Stanford.EDU (5.57/Ultrix3.0-C)
id AA04142; Tue, 22 May 90 08:56:32 PDT
Date: Tue, 22 May 90 08:56:32 PDT
From: mkatz@garlic.stanford.edu (Morris Katz)
Message-Id: <9005221556.AA04142@garlic.Stanford.EDU>
To: arg@lucid.com
Cc: qlisp@gang-of-four.stanford.edu
In-Reply-To: Ron Goldman's message of Mon, 21 May 90 17:38:50 PDT <9005220038.AA25567@bhopal>
Subject: Qlisp at toplevel
Date: Mon, 21 May 90 17:38:50 PDT
From: Ron Goldman <arg@lucid.com>
The time has finally come to have Qlisp go into parallel mode when it starts
up and stay there during the top level read-eval-print loop. This would
eliminate the need for QEVAL and, if in evaluating a form futures are created
whose values are not immediately needed, then the associated processes could
continue to run after the top level loop has printed out the "value" of the
form. Below is a description of a number of new functions that will be
needed/available:
6) Disksave can only be done from toplevel. Any running user processes will
first be killed (by KILL-ALL-PROCESSES) before the image is saved. Qlambda
process closures will be saved and when the new image starts up new
processes for them will automatically be spawned. If the user wants the
new image to have any other running user processes then a user function
with code to spawn them needs to be called when the new image is started up.
(C.f. documentation on the :restart-function argument to disksave).
Are you really sure that you do not want DISK-SAVE to do a QWAIT-ALL before
saving, rather than a KILL-ALL-PROCESSES. I would think that this would be
the more intuitive default. Of course, either default can be overridden by
simply calling either KILL-ALL-PROCESSES or QWAIT-ALL prior to DISK-SAVE.
-------------------------------------------------------------------------------
Morry Katz
katz@cs.stanford.edu
-------------------------------------------------------------------------------
∂22-May-90 1432 arg@lucid.com Qlisp at toplevel
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 22 May 90 14:32:00 PDT
Received: from LUCID.COM by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA13501; Tue, 22 May 90 14:32:15 -0700
Received: from bhopal ([192.31.212.13]) by heavens-gate.lucid.com id AA27451g; Tue, 22 May 90 14:31:39 PDT
Received: by bhopal id AA26837g; Tue, 22 May 90 14:32:13 PDT
Date: Tue, 22 May 90 14:32:13 PDT
From: Ron Goldman <arg@lucid.com>
Message-Id: <9005222132.AA26837@bhopal>
To: mkatz@garlic.stanford.edu
Cc: qlisp@go4.stanford.edu
In-Reply-To: Morris Katz's message of Tue, 22 May 90 08:56:32 PDT <9005221556.AA04142@garlic.Stanford.EDU>
Subject: Qlisp at toplevel
Morry - I can see your point, but note that qwait-all merely waits for all
actively running processes to finish or to block by waiting for a
lock/future/etc. When starting a disksave there shouldn't be any user
processes around in a running or blocked state, and the only way to
guarantee that is to call kill-all-processes. In the best of all
possible Qlisp worlds it might be desirable to allow user processes to
be preserved by disksave so they would continue execution when the
new image starts up. However for now that's just too much work to
make happen.
Ron
∂22-May-90 1835 @UWAVM.ACS.WASHINGTON.EDU:nishida@kuis.kyoto-u.ac.jp ABSTRACT: Reasoning with Model Lattices
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 22 May 90 18:35:15 PDT
Received: from uwavm.acs.washington.edu by june.cs.washington.edu (5.61/7.0jh)
id AA01659; Tue, 22 May 90 18:26:24 -0700
Received: from UWAVM.ACS.WASHINGTON.EDU by UWAVM.ACS.WASHINGTON.EDU (IBM VM SMTP R1.2) with BSMTP id 6929; Tue, 22 May 90 18:24:59 PDT
Received: from JPNKYOTO.BITNET by UWAVM.ACS.WASHINGTON.EDU (Mailer R2.04) with
BSMTP id 7959; Tue, 22 May 90 18:24:56 PDT
Received: from JPNKYOTO by JPNKYOTO.BITNET (Mailer X1.25) with BSMTP id 9680;
Wed, 23 May 90 10:26:01 JST
Received: from kuis.kuis.kyoto-u.ac.jp by jpnkyoto.kyoto-u.ac.jp (IBM VM SMTP
R1.2.1) with TCP; Wed, 23 May 90 10:26:00 JST
Received: by kuis.kuis.kyoto-u.ac.jp (4.0/6.4J.5) with SMTP
id AA18079; Wed, 23 May 90 10:25:50 JST
Received: by pathos.kuis.kyoto-u.ac.jp (4.0/6.4J.5) with TCP
id AA00262; Wed, 23 May 90 10:25:39 JST
Return-Path: <nishida%kuis.kyoto-u.ac.jp@UWAVM.ACS.WASHINGTON.EDU>
Message-Id: <9005230125.AA00262@pathos.kuis.kyoto-u.ac.jp>
To: qphysics%cs.washington.edu@bitnet-relay
Subject: ABSTRACT: Reasoning with Model Lattices
Date: Wed, 23 May 90 10:25:34 +0900
From: nishida%kuis.kyoto-u.ac.jp@UWAVM.ACS.WASHINGTON.EDU
Reasoning with Model Lattices
Toyoaki Nishida and Shuji Doshita
Model inference is an activity of constructing a domain model
from observations. Model inference enables knowledge based systems
to cope with novel situations by autonomously collecting data,
making hypothesis, and even designing experiments.
A crucial issue in model inference is to distinguish relevant features
from lots of irrelevant ones. It is critical especially
when a large amount of general knowledge is available for the domain.
This paper presents a computational model of model inference for
physical devices.
The contribution of this paper is twofold:
(1) we have introduced a notion of model lattice to
organize a collection of alternative models
and have presented a theoretical characterization of model inference
(2) we demonstrate that causal analysis helps contrain the
search for relevant features in modeling physical devices.
reference:
@incollection{Nishida89b,
author = "Toyoaki Nishida and Shuji Doshita",
title = "Reasoning with Model Lattices",
editor = "M. Tokoro and Y. Anzai and A. Yonezawa",
booktitle = "Concepts and Characteristics of Knowledge-Based Systems",
publisher = "Elsevier Science Publishers B. V. (North-Holland)",
year = "1989",
pages = "325-347",
note = "Selected and Reviewed Papers from the IFIP TC 10/ WG 10.1
Workshop Mount Fuji, Japan, 9-12 November, 1987"}
Toyoaki Nishida
∂24-May-90 1033 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 24 May 90 10:33:01 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA00809; Thu, 24 May 90 10:33:22 -0700
Date: Thu, 24 May 90 10:33:22 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005241733.AA00809@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
David Mumford, Harvard, 617 495-5487
RE: Auto Theorem Proving Competition
He would like you on the committee to help select the winners for the
prizes to be awarded next January.
∂24-May-90 1625 mps@Gang-of-Four.Stanford.EDU Farewell Dinner for Joe Weening
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 24 May 90 16:25:23 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03501; Thu, 24 May 90 16:25:34 -0700
Date: Thu, 24 May 90 16:25:34 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005242325.AA03501@Gang-of-Four.Stanford.EDU>
To: napier@cs.Stanford.EDU, yossi@theory.Stanford.EDU, farhad@cs.Stanford.EDU,
ark@sail.Stanford.EDU, zm@sail.Stanford.EDU, gangolli@cs.Stanford.EDU,
rwf@sail.Stanford.EDU, jmc@sail.Stanford.EDU, iam@sail.Stanford.EDU,
me@sail.Stanford.EDU, chandler@cs.Stanford.EDU,
pehoushek@cs.Stanford.EDU, appelt@ai.sri.com,
d.doggette@macbeth.Stanford.EDU, shankar@go4.Stanford.EDU,
kolk@jessica.Stanford.EDU, pst@ack.Stanford.EDU,
crew@neon.Stanford.EDU, rokicki@neon.Stanford.EDU, johnh@src.dec.com,
clt@go4.Stanford.EDU, rpg@sail.Stanford.EDU, bjr@sunburn.Stanford.EDU,
winkler@cs.Stanford.EDU, gotelli@cs.Stanford.EDU, rdz@go4.Stanford.EDU
Subject: Farewell Dinner for Joe Weening
SUBJECT: Farewell dinner
DATE: Wednesday, May 20
TIME: 7:00 pm
PLACE: Grand China
5100 El Camino Real (near Rengstorff)
Los Altos
COST: $12.00 per person
Because we are a large group, I will have to insist on prepayment for
your dinner.
Please make a check payable to Grand China for $12.00 per person and
bring no later than 5:00 pm Tuesday, May 29 to my office (MJH 358).
I believe this will give those of you off campus to mail your check to
my attention.
Thank you all.
Pat Simmons
∂24-May-90 1643 mps@Gang-of-Four.Stanford.EDU Farewell Dinner for Joe Weening (correction)
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 24 May 90 16:42:58 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03654; Thu, 24 May 90 16:42:56 -0700
Date: Thu, 24 May 90 16:42:56 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005242342.AA03654@Gang-of-Four.Stanford.EDU>
To: napier@cs.Stanford.EDU, yossi@theory.Stanford.EDU, farhad@cs.Stanford.EDU,
ark@sail.Stanford.EDU, zm@sail.Stanford.EDU, gangolli@cs.Stanford.EDU,
rwf@sail.Stanford.EDU, jmc@sail.Stanford.EDU, iam@sail.Stanford.EDU,
me@sail.Stanford.EDU, chandler@cs.Stanford.EDU,
pehoushek@cs.Stanford.EDU, appelt@ai.sri.com,
d.doggette@macbeth.Stanford.EDU, shankar@go4.Stanford.EDU,
kolk@jessica.Stanford.EDU, pst@ack.Stanford.EDU,
crew@neon.Stanford.EDU, rokicki@neon.Stanford.EDU, johnh@src.dec.com,
clt@go4.Stanford.EDU, rpg@sail.Stanford.EDU, bjr@sunburn.Stanford.EDU,
winkler@cs.Stanford.EDU, gotelli@cs.Stanford.EDU, rdz@go4.Stanford.EDU
Subject: Farewell Dinner for Joe Weening (correction)
Hi Folks
Marty just caught a boo-boo in the message to you. The date is the
30th and not the 20th of May. The keyboard is getting tired this late
in the day and makes lots of mistakes. Thanks.
Pat
∂25-May-90 0648 harnad@clarity.Princeton.EDU SPP Abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 25 May 90 06:48:04 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA01687; Fri, 25 May 90 09:49:00 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA08477; Fri, 25 May 90 09:49:03 EDT
Date: Fri, 25 May 90 09:49:03 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005251349.AA08477@cognito.Princeton.EDU>
To: jmc@sail.stanford.edu
Subject: SPP Abstract
John, we still don't have the abstract you promised for ths week. -- Stevan
∂25-May-90 2009 beeson@ucscd.UCSC.EDU Pure Bargaining
Received: from ucscd.UCSC.EDU by SAIL.Stanford.EDU with TCP; 25 May 90 20:09:19 PDT
Received: by ucscd.UCSC.EDU (5.61/1.34)
id AA01186; Fri, 25 May 90 20:12:42 -0700
Date: Fri, 25 May 90 20:12:42 -0700
From: beeson@ucscd.UCSC.EDU (20012000)
Message-Id: <9005260312.AA01186@ucscd.UCSC.EDU>
To: jmc@sail.stanford.edu
Subject: Pure Bargaining
A naturally-occurring example of the hypothetical bargaining
situation you described to me: My two daughters have trouble
agreeing on the book which I am to read at bedtime (since they are
of different ages). For I while I tried telling them, I will read
whatever book you want so long as you both agree, but if you can't
agree then I won't read any book. Now this was pure bargaining!
and my youngest daughter followed your tough-guy strategy, insisting
that we read the book she wanted, or else we wouldn't read any. So
after a while my older daughter would unhappily give in. (The younger
one looks so sweet and innocent, too!) Democracy fails miserably
when the size of the electorate is 2; so I took over as dictator and
said "I will choose the book, and no arguments!" But I didn't realize
until after our discussion that the breakdown of the method was
theoretically predictable.
∂28-May-90 0845 harnad@clarity.Princeton.EDU Urgent: SPP Abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 28 May 90 08:45:36 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA00547; Mon, 28 May 90 11:46:20 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA13246; Mon, 28 May 90 11:46:30 EDT
Date: Mon, 28 May 90 11:46:30 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005281546.AA13246@cognito.Princeton.EDU>
To: jmc@sail.stanford.edu
Subject: Urgent: SPP Abstract
John, we urgently need your abstract for the on-site program.
It needn't be formal, just the gist of your arugument against
Searle. Please reply!
Best wishes, Stevan
∂28-May-90 1129 rathmann@eclipse.stanford.edu Welcome back
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 28 May 90 11:29:36 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA23791; Mon, 28 May 90 11:31:44 PDT
Date: Mon, 28 May 90 11:31:44 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9005281831.AA23791@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: Welcome back
Welcome back from you trip. It seems that you are bringing Gorbachev
back with you this time!
Did you have any comments/changes for my thesis? If there is
something major, I would like to start right away.
Thanks,
Peter
∂28-May-90 1709 harnad@clarity.Princeton.EDU Confirmation
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 28 May 90 17:09:44 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA13406; Mon, 28 May 90 20:10:25 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA13648; Mon, 28 May 90 20:10:33 EDT
Date: Mon, 28 May 90 20:10:33 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005290010.AA13648@cognito.Princeton.EDU>
To: jmc@sail.stanford.edu
Subject: Confirmation
Got it, thanks. -- SH
∂28-May-90 1723 harnad@clarity.Princeton.EDU J. McCarthy's Abstract for SPP Searle Symposium
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 28 May 90 17:23:04 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA13482; Mon, 28 May 90 20:15:10 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA13661; Mon, 28 May 90 20:15:14 EDT
Date: Mon, 28 May 90 20:15:14 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005290015.AA13661@cognito.Princeton.EDU>
To: dietrich@bingvaxu.cc.binghamton.edu,
dietrich@bingvaxu.cc.binghamton.edu
(For: Howard Pattee),
granger@uci.BITNET, hayes.pa@xerox.com, jmc@sail.stanford.edu,
ltm@xp.psych.nyu.edu (Larry Maloney), movshon@CMCL2.NYU.EDU,
plunkett@amos.ucsd.edu
(Kim Plunkett, UCSD),
port@iuvax.cs.indiana.edu,
powers@informatik.uni-kl.de
(David Powers), psyirv@umnacvx.BITNET,
rey@cs.umd.edu (Georges
Rey), treisman@vax.oxford.ac.uk
Subject: J. McCarthy's Abstract for SPP Searle Symposium
Cc: andrewsj@vassar.BITNET
Abstract for SPP Symposium on Searle's Chinese Room Argument:
From: John McCarthy <JMC@SAIL.Stanford.EDU>
John Searle begins his (1990) "Consciousness, Explanatory Inversion and
Cognitive Science" with:
"Ten years ago in this journal I published an
article (Searle, 1980a and 1980b) criticising what I
call Strong AI, the view that for a system to have
mental states it is sufficient for the system to
implement the right sort of program with right inputs
and outputs. Strong AI is rather easy to refute and
the basic argument can be summarized in one sentence:
`a system, me for example, could implement a program for
understanding Chinese, for example, without
understanding any Chinese at all.' This idea, when
developed, became known as the Chinese Room Argument."
The Chinese Room Argument can be refuted in one sentence:
"Searle confuses the mental qualities of one computational process,
himself for example, with those of another process that the first
process might be interpreting, a process that understands Chinese, for
example."
That accomplished, the lecture will discuss the ascription of mental
qualities to machines with special attention to the relation between
syntax and semantics, i.e. questions suggested by the Chinese Room
Argument. I will deal explicitly with Searle's four ``axioms'',
which, although they don't have a unique interpretation, suggest
various ideas worth discussing.
∂28-May-90 2120 rathmann@eclipse.stanford.edu new "almost final" draft
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 28 May 90 21:20:07 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA24744; Mon, 28 May 90 21:22:10 PDT
Date: Mon, 28 May 90 21:22:10 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9005290422.AA24744@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: new "almost final" draft
I am leaving an updated copy of my thesis in your mailbox. Hopefully,
this is pretty close to being the final version.
-Peter
∂28-May-90 2317 young@Neon.Stanford.EDU re: qual depth area
Received: from Neon.Stanford.EDU by SAIL.Stanford.EDU with TCP; 28 May 90 23:17:23 PDT
Received: by Neon.Stanford.EDU (5.61/25-eef) id AA20007; Mon, 28 May 90 23:18:28 -0700
Date: Mon, 28 May 1990 23:18:27 PDT
From: "R. Michael Young" <young@Neon.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: young@Neon.Stanford.EDU
Subject: re: qual depth area
In-Reply-To: Your message of 14 May 90 1754 PDT
Message-Id: <CMM.0.88.643961907.young@Neon.Stanford.EDU>
Professor McCarthy,
Is there a time before you leave for the nonmon conference that I can drop in
to discussion the scope of my depth area on the AI qual? I don't think it
would take more than 5 or so minutes.
-Michael
∂29-May-90 0000 JMC
Rabin at Harvard
∂29-May-90 1100 rlm@Hudson.Stanford.EDU AIDOC ABSTRACTS NEEDED
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 29 May 90 11:00:03 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA14345; Tue, 29 May 90 10:57:53 -0700
Date: Tue, 29 May 1990 10:57:52 PDT
From: "Robert L. Miller" <rlm@Hudson.Stanford.EDU>
To: AI-DOC:;@Hudson.Stanford.EDU
Subject: AIDOC ABSTRACTS NEEDED
Message-Id: <CMM.0.88.644003872.rlm@Hudson.Stanford.EDU>
Please email me the abstract to your AI Day On Campus talk. I need
them today in order to publish a brochure for the event.
I will create the brochure on a Mac, so if possible, please send
straight text only. If special equation symbols are necessary, send a
hardcopy and I will try to paste or scan it in.
Abstracts may be up to 1 page in length.
Cheers,
Robert
rlm@hudson
∂29-May-90 1121 mps@Gang-of-Four.Stanford.EDU reply to message
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 29 May 90 11:21:24 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA07649; Tue, 29 May 90 11:21:40 -0700
Date: Tue, 29 May 90 11:21:40 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005291821.AA07649@Gang-of-Four.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
In-Reply-To: John McCarthy's message of 28 May 90 0008 PDT <d#P7R@SAIL.Stanford.EDU>
Subject: reply to message
I already sent in your registration and have received the receipt that
you will use to pick up your registration materials and badge. If you
want on-campus housing your request has to be received by July 1.
single, 35.00/night. Housing off-campus is due by June 29. You have
several choices. Pat
∂29-May-90 1124 harnad@clarity.Princeton.EDU Biederman Abstract: SPP Symbol Grounding Workshop
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 29 May 90 11:24:08 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.37/relay)
id AA05590; Tue, 29 May 90 14:19:20 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA04882; Tue, 29 May 90 14:19:18 EDT
Date: Tue, 29 May 90 14:19:18 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005291819.AA04882@reason.Princeton.EDU>
To: srh@flash.bellcore.com
Subject: Biederman Abstract: SPP Symbol Grounding Workshop
Cc: andrewsj@vassar.BITNET
From: Irv Biederman <PSYIRV@vx.acs.umn.edu>
ABSTRACT FOR SPP SYMBOL GROUNDING WORKSHOP
Grounding Mental Symbols in Object Images
Irving Biederman and John E. Hummel
University of Minnesota
We describe a neural net (NN) implementation of a theory of real time
visual shape recognition that takes as input the edges corresponding to
the occlusional and orientation discontinuities in an image. As output
the model activates a unit that is selective for a specified
arrangement of simple volumes (or geons) and thus achieves a basic (or
entry) level classification according to Biederman's (1987)
Recognition-by-Components theory of object recognition. The output unit
can qualify as a symbol of the object in that it reflects the major
invariances of visual object recognition. The model solves four
fundamental problems in object recognition that likely confront all
attempts at visual basic-level symbol grounding:
1) Translational, size, and orientation invariance: The same output
unit(s), corresponding to the object, are activated no matter where the
image falls in the visual field, the size of the image, and the
orientation in depth (up to parts occlusion),
2) Appropriate grouping (or organization) of image elements into
appropriate parts,
3) A basis of determining invariant object centered relations (such as
TOP- OF or SIDE-CONNECTED), and
4) A basis for computing the similarity (or equivalence) of object
images.
These problems all required a solution to the "binding problem"--
determining what groups with what. In the present case, for example,
how are the various segments of the parts of an object grouped
according to their appropriate parts. Most NN models have employed
enumeration, assigning a unit to each attribute combination. Such
enumerative schemes are unsatisfactory in that they require a
prohibitively large number of units to represent even modest input
domains. Moreover, they do not express the equivalence of inputs. By
employing different units to represent the different locations of an
object, for example, the information that it is the same object in the
different locations is not represented. Our model achieves binding
through phase locking of the oscillatory activity of cells that are
tuned to oriented image edges. The phase locking (or synchrony) is
established by "fast enabling links" (FELs) between pairs of a)
collinear, b) coterminating, and c) parallel adjacent edge cells. These
units then activate invariant representations of geons and relations in
intermediate layers.
The model offers some perspective on what it is that makes a category
"basic." A category such as chair will encompass a number of
distinguishably different geon models that, perceptually, may be as
distant as members of different classes.
∂29-May-90 1436 kuipers@cs.utexas.edu research job opportunity
Received: from june.cs.washington.edu by SAIL.Stanford.EDU with TCP; 29 May 90 14:36:28 PDT
Received: from cs.utexas.edu by june.cs.washington.edu (5.61/7.0jh)
id AA22899; Tue, 29 May 90 14:24:20 -0700
Return-Path: <kuipers@cs.utexas.edu>
Received: from ai.cs.utexas.edu by cs.utexas.edu (5.61/1.62)
id AA12851; Tue, 29 May 90 16:24:07 -0500
Date: Tue, 29 May 90 16:23:54 CDT
From: kuipers@cs.utexas.edu
Posted-Date: Tue, 29 May 90 16:23:54 CDT
Message-Id: <9005292123.AA13894@ai.cs.utexas.edu>
Received: by ai.cs.utexas.edu (5.59/1.4-Client)
id AA13894; Tue, 29 May 90 16:23:54 CDT
To: qphysics@cs.washington.edu
Subject: research job opportunity
Cc: kuipers@cs.utexas.edu
Research Associate in Qualitative Reasoning
Department of Computer Science
University of Texas at Austin
It appears that funds will be available to support a post-doctoral
Research Associate in the Qualitative Reasoning Group at the University
of Texas at Austin. The goal of this position will be to coordinate a
substantial research effort on the use of qualitative reasoning as an
intelligent framework for quantitative inference and problem-solving in
science and engineering.
Our group has a vigorous research program in qualitative simulation,
qualitative model-building, model-based monitoring and diagnosis, and
the combination of qualitative and quantitative inference methods. We
have a variety of ongoing relationships with the departments of
Chemical, Mechanical, and Electrical and Computer Engineering. Austin
is also a wonderful place to live.
We are looking for someone combining substantial background in
engineering and numerical mathematics with substantial background in
artificial intelligence, model-based expert systems, and qualitative
reasoning. A doctorate is required, plus evidence of depth of knowledge
spanning both fields. The ideal candidate will be capable of
self-directed, high-quality work at the interface between symbolic
qualitative reasoning and numerical engineering problem-solving. He or
she should have a clear technical vision of the relationship between
symbolic and numerical inference methods, should be capable of working
within the QSIM framework we have developed, and should be capable of
participating in the technical direction of a sizable group of
researchers.
The position is funded for the first year, with a strong possibility of
funding for a second year. There are no teaching responsibilities, but
the opportunity exists for teaching if desired.
The position could start as early as July 1, 1990. The position will
be held open until June 15, but we will continue accepting applications
until the position is filled.
Please send a vita, copies of relevant papers, and arrange to have three
letters of reference sent to:
Prof. Benjamin Kuipers
Department of Computer Sciences
University of Texas at Austin
Austin, Texas 78712 USA
email: kuipers@cs.utexas.edu
∂29-May-90 1533 VAL Going to Tahoe
Gelfond, Bondarenko and I will be in MJH on Thursday, May 31, at 1:30,
ready to leave for Tahoe whenever you are ready.
--Vladimir
∂30-May-90 0736 mps@Gang-of-Four.Stanford.EDU Joe's Dinner
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 May 90 07:35:55 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA16061; Wed, 30 May 90 07:36:06 -0700
Date: Wed, 30 May 90 07:36:06 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005301436.AA16061@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Joe's Dinner
Would you please give a little talk after dinner about Joe. We would
all appreciate hearing from you.
I would also like to leave a little early to make sure everything is
ready at the resturant. I am the one arranging this dinner for Joe.
Another item, Joyce would like to know how you are voting on LaVoy?
Is there anyone else I should ask to speak tonight? Don Knuth, Gene
Golub and Zohar are going along with a bunch of Ph.D., staff, and
friends.
Thanks.
Pat
∂30-May-90 0915 CLT calenday
wed 13-jun-90 7pm dinner at Hurds
[is there any conflict?]
∂30-May-90 0953 jutta@Sunburn.Stanford.EDU reminder
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 May 90 09:53:16 PDT
Received: by Sunburn.Stanford.EDU (5.61+IDA/25-eef) id AA09642; Wed, 30 May 90 09:54:02 -0700
Date: Wed, 30 May 1990 9:54:00 PDT
From: Jutta McCormick <jutta@Sunburn.Stanford.EDU>
To: latombe@coyote.Stanford.EDU, binford@cs.Stanford.EDU,
eaf@SUMEX-AIM.Stanford.EDU, mrg@Sunburn.Stanford.EDU,
ok@coyote.Stanford.EDU, jmc@sail.Stanford.EDU, nilsson@cs.Stanford.EDU,
shoham@cs.Stanford.EDU, winograd@csli.Stanford.EDU,
lifschitz@cs.Stanford.EDU, ginsberg@cs.Stanford.EDU
Cc: jutta@Sunburn.Stanford.EDU
Subject: reminder
Message-Id: <CMM.0.88.644086440.jutta@Sunburn.Stanford.EDU>
REMINDER: AI Division lunch today, Wednesday, May 30, noon, Faculty Club
(reservation for Latombe). Topic: Intelligent offices
∂30-May-90 1007 CLT revision/calendar item
fri 15-jun-90 6:30pm dinner at Hurds
∂30-May-90 1147 linvill@sierra.Stanford.EDU re: Hennessy Nomination
Received: from sierra.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 May 90 11:47:15 PDT
Received: by sierra.Stanford.EDU (4.0/4.7); Wed, 30 May 90 11:46:56 PDT
Date: Wed, 30 May 90 11:46:56 PDT
From: linvill@sierra.Stanford.EDU (John G. Linvill)
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: Hennessy Nomination
In-Reply-To: Your message of 28 May 90 0012 PDT
Message-Id: <CMM.0.88.644093214.linvill@>
John,
The deadline for submission of the nominations came and we had to
put down a list of four. We could not include you since there wasn't the
chance to discuss it with you.
I hope the nomination will be successful. If not, we should
resubmit next year, and will keep your interest in mind for then.
Regards, John
∂30-May-90 1210 gangolli@Theory.Stanford.EDU Rides to Grand China tonight
Received: from Thor.Stanford.EDU by SAIL.Stanford.EDU with TCP; 30 May 90 12:10:24 PDT
Received: from LOCALHOST by Thor.Stanford.EDU with SMTP (5.61/25-theory-eef) id AA10919; Wed, 30 May 90 12:09:19 -0700
Message-Id: <9005301909.AA10919@Thor.Stanford.EDU>
To: napier@cs.Stanford.EDU, yossi@theory.Stanford.EDU, farhad@cs.Stanford.EDU,
ark@sail.Stanford.EDU, zm@sail.Stanford.EDU, gangolli@cs.Stanford.EDU,
rwf@sail.Stanford.EDU, jmc@sail.Stanford.EDU, iam@sail.Stanford.EDU,
me@sail.Stanford.EDU, chandler@cs.Stanford.EDU,
pehoushek@cs.Stanford.EDU, appelt@ai.sri.com,
d.doggette@macbeth.Stanford.EDU, shankar@Gang-of-Four.Stanford.EDU,
kolk@jessica.Stanford.EDU, pst@ack.Stanford.EDU,
crew@neon.Stanford.EDU, rokicki@neon.Stanford.EDU, johnh@src.dec.com,
clt@Gang-of-Four.Stanford.EDU, rpg@sail.Stanford.EDU,
bjr@sunburn.Stanford.EDU, winkler@cs.Stanford.EDU,
gotelli@cs.Stanford.EDU, rdz@Gang-of-Four.Stanford.EDU,
mps@Gang-of-Four.Stanford.EDU
Subject: Rides to Grand China tonight
Date: Wed, 30 May 90 12:09:16 -0700
From: Anil R. Gangolli <gangolli@Theory.Stanford.EDU>
Let me suggest that those people driving and those needing rides to
Grand China for the "weaning Weening" dinner tonight converge in the
MJH lobby at 6:30pm. --anil.
∂30-May-90 1215 "JC_._POSTMAST_@_THPINYC_(JC)%THPINYC"@mcimail.com John C. Dvorak 6/26/90 Editorial
Received: from NRI.Reston.VA.US by SAIL.Stanford.EDU with TCP; 30 May 90 12:15:44 PDT
Received: from mcimail.com by NRI.NRI.Reston.VA.US id ad13498;
30 May 90 15:03 EDT
Date: Wed, 30 May 90 14:04 EST
From: JC <"JC_._POSTMAST_@_THPINYC_(JC)%THPINYC"@mcimail.com>
To: MCCARTHY <JMC@sail.stanford.edu>
Subject: John C. Dvorak 6/26/90 Editorial
Message-Id: <00900530190400/0003921119NB4EM@mcimail.com>
MHS: Source date is: 30-May-90 13:41 EST
Dear Colleagues and Correspondents to the EMAIL LEAGUE!
-A major blow has been struck in our favor by the editorial by Mr. Dvorak in
the latest PC Magazine entitled "It's Time E-Mail Standardized Worldwide."
He lists the draft "demands" of the league, mentions my name and email
address (ohmigod!) and thoughtfully dissents from the general view expressed
here that DNS is probably the defacto way to go. As this hits the stand is
clearly the time to take the next step, and have the PC magazines of the
world recommend Email addressing to their advertisers.
As always, I welcome your thoughts.
John Coonrod, Email: jc%thpinyc@mcimail.com
∂30-May-90 1433 VAL Message from Arkady
His suggestion is to buy a 386- or 486-based PC ($20,000), a few ATs
($3,000 apiece) that can be used as terminals, and a lazer printer
($2,000). There are people here who will configure the system and
check it out. Some additional programming will need to be done, and
it will be possible and less expensive to find people in Russia who can
do it, rather than here.
Another possibility is to get a minicomputer. An ARIX ($25,000-30,000)
will support 20-30 terminals. But the software would cost between $20,000
and $50,000, so that the totel may be more that what we want, and it
won't really give more power.
∂30-May-90 1519 harnad@clarity.Princeton.EDU SPP Abstract: Pattee
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 30 May 90 15:19:11 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.38/relay)
id AA04003; Wed, 30 May 90 17:40:48 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA00559; Wed, 30 May 90 17:40:45 EDT
Date: Wed, 30 May 90 17:40:45 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005302140.AA00559@reason.Princeton.EDU>
To: andrewsj@vassar.BITNET
Subject: SPP Abstract: Pattee
To: SPP Symbol-Grounding Workshop and Searle Symposium Participants
Abstract for SPP Workshop on Symbol Grounding
The Measurement Problem in Physics and Brain Theory
H.H. Pattee
Department of Systems Science
TJ Watson School of Engineering
SUNY
Binghamton NY 13901
The measurement problem in physics is a special case of the
symbol-grounding problem of brain theory, which in turn is a special
case of the epistemological problem of relating the knower and the
known. In quantum theory the measurement problem is notoriously
obscure, because the results of measuring quantum events are
nonclassically observer-dependent and yet must be expressed only in
classical language. Even the measurement of classical events cannot be
completely described by physical laws, because measurement involves
intent, i.e., the what, where, and when of the measurements must be
determined by an "observer," not by the laws. It also makes no sense
to say that a measurement has occurred unless there is a "result."
At issue in physical theory are the necessary conditions for
"observer," and "result." Working physicists evade the philosophical
issues by using language confined to formal, operational symbol
systems that restrict measurements to numerical results and
prediction to computation. Formal symbol systems do allow unambiguous
predictions, but only at the cost of generating logical antinomies and
conceptual puzzles like "Schroedinger's Cat" and "Wigner's friend."
If one relaxes the syntactic precision of formal symbols, however, and
extends the concepts of "observer" and "result" to the subsymbolic,
dynamical level of measurement-control constraints in simpler
organisms, some of these puzzles are resolved. The problem I raise for
brain theory is that formal symbolic models of the brain may likewise
produce unambiguous results only at the cost of conceptual puzzles like
Searle's "Chinese Room."
∂30-May-90 1656 ok@coyote.stanford.edu Re: Depth Examiner
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 30 May 90 16:56:25 PDT
Received: by coyote.stanford.edu; Wed, 30 May 90 16:57:04 PDT
Date: 30 May 1990 1657-PDT (Wednesday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: latombe@coyote.stanford.edu, jmc@sail.stanford.edu,
nilsson@tenaya.stanford.edu, rosenschein@teleos.com,
shoham@hudson.stanford.edu, young@neon.stanford.edu
Cc: ok@coyote.stanford.edu
Subject: Re: Depth Examiner
This is to confirm the changes of (D)epth area examiner and (O)ther
examiner in Michael Young's committee:
AI Qualification Exam Schedule
Wednesday, June 6, 1990
1:45 -- 3:15
Michael Young: Formal Models of Inter-Agent Communication
C: Latombe D: Rosenschein O: McCarthy
∂30-May-90 2336 @MCC.COM:ai.guha@MCC.COM re: between the devil and the deep sea
Received: from MCC.COM by SAIL.Stanford.EDU with TCP; 30 May 90 23:36:23 PDT
Received: from hi3.aca.mcc.com.aca.mcc.com (HI3.ACA.MCC.COM) by MCC.COM with TCP/SMTP; Thu 31 May 90 01:37:19-CDT
Posted-Date: Thu, 31 May 90 01:37 CDT
Received: from GAIA.ACA.MCC.COM by hi3.aca.mcc.com.aca.mcc.com (4.0/ACAv4.1i)
id AA17313; Thu, 31 May 90 01:33:01 CDT
Date: Thu, 31 May 90 01:37 CDT
From: R. V. Guha <ai.guha@MCC.COM>
Subject: re: between the devil and the deep sea
To: JMC@SAIL.Stanford.EDU
In-Reply-To: <d#PuV@SAIL.Stanford.EDU>
Message-Id: <19900531063709.3.GUHA@GAIA.ACA.MCC.COM>
Date: 28 May 90 0016 PDT
From: John McCarthy <JMC@SAIL.Stanford.EDU>
[In reply to message sent Thu, 17 May 90 01:13 CDT.]
You didn't hear from me soon, because I was in Moscow. I
have read your paper and have comments. Also I am willing
to defend our approach against the criticisms you cite.
I'd like to see your comments on it. I realized that there is a
set of unstated assumptions we are making about the kinds of
problems we would like to be able to handle, what we think are
reasonable and unreasonable assumptions etc. and I am trying to
draw up a coherent picture of this. This should enable us to
defend our position quite well.
There are some assumptions/tactics
the traditional linguists accept that we don't seem to be using
(eg. they make the assumption that there exists a small number
of context parameters that can be extracted from the textually
nearby sentences that can be use to "flesh out" the contextuality
in the sentence. This has severe limitations and is one of the
positions that Barwise & co. attack.) I also read completely
the Barwise book (The situation in logic) and it seems that we
can do everything they want to do with their constructs (including
things like inferencing in a context). There however seem to be
some deep differences (aside from the technical details) in that
they seem to assume the existance of objective account of the
content of an utterance (as opposed to our approach where we
just go on lifting the sentences into more and more general contexts).
However, I have not yet completely understood how to translate
their material into ours (their foundations are very different -
they don't even use the foundation axiom etc.).
I shall be comming to Palo Alto in a couple of weeks and plan
to spend most of the summer there and am looking forward to
discussing this with you. In the mean time, if you have the time,
could you send me the commented version of the paper? I am
sorry for not replying earlier - i was down with some kind
of flu.
Thanks
Guha
∂31-May-90 0900 scherlis@vax.darpa.mil ANNUAL REPORTING
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 31 May 90 09:00:28 PDT
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA21719>; Thu, 31 May 90 11:11:24 -0400
Resent-Message-Id: <9005311502.AA14320@sun45.darpa.mil>
Received: by sun45.darpa.mil (4.1/5.51)
id AA14320; Thu, 31 May 90 11:02:19 EDT
Received-Date: Wed, 30 May 90 21:46:14 -0400
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA18791>; Wed, 30 May 90 21:46:14 -0400
Posted-Date: Wed 30 May 90 21:37:13-EDT
Received: by sun45.darpa.mil (4.1/5.51)
id AA13866; Wed, 30 May 90 21:37:14 EDT
Date: Wed 30 May 90 21:37:13-EDT
From: William L. Scherlis <SCHERLIS@vax.darpa.mil>
Subject: ANNUAL REPORTING
To: SW-PI@vax.darpa.mil
Cc: njacobs@vax.darpa.mil
Message-Id: <644117833.0.SCHERLIS@VAX.DARPA.MIL>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@VAX.DARPA.MIL>
Resent-Date: Thu 31 May 90 11:02:18-EDT
Resent-From: William L. Scherlis <SCHERLIS@DARPA.MIL>
Resent-To: SW-PI@vax.darpa.mil
TO: The Software Principal Investigators
SUBJECT: Annual Reporting
We are again working to simplify the annual reporting process.
1. TECHNICAL REPORTS. Our annual technical reports will be in a
format similar to the one we used last year (i.e., soliciting
accomplishments of the past year, objectives for the forthcoming year,
technology transition progress, major program milestones, and the
like). These reports form the basis for our making an internal case
in DARPA concerning the value and impact of the software research
program. They are also useful in helping us understand what you are
doing so we can work with you more effectively (e.g., by suggesting
potential technology transition opportunities).
2. FINANCIAL REPORTS. We will also request some financial information,
primarily to help us compare actual expenditures to funds allocated.
This helps us optimize yearly allocations so there are more funds to go
around in any given year. If you are spending more slowly than you
expected, we can shift some of your funding authorization to a later
year. This will NOT have adverse impact on you, but it provides
considerable improvement in both flexibility and budget security for the
whole program. Flexibility here means that we can use the funds garnered
in a given year to start up more efforts that year, thus expanding the
program.
Budget security here means that we can keep the elapsed time between our
authorization to you to spend funds and your actual billing of the
government down to a level that does not attract the interest of
budgeteers who are looking for funds to cut: Delays in spending
authorized funds indicate to budget cutters that our program
authorizations are too high and could be reduced. Once a reduction
happens, it becomes difficult to restore programs to their full level in
the later years. This creates risk for the entire program. (Note that
you can request "no-cost extensions" to contracts to extent the time
available to you to carry out the work.)
This fiscal year, for example, ISTO experienced some major cuts due to
delays between the time contractors were authorized to spend money and
the time they actually submitted bills. So you can help in two ways: (1)
Provide honest assessments of your expenditure situation when we request
the data, and (2) Encourage your contracts/budget colleagues to bill the
government promptly when funds are expended.
ACTION:
While you are thinking about these bureaucratic things, please
ensure that the regular quarterly reports you are already
producing are being sent to me at DARPA as well as to our
colleagues at your contracting agent organization.
We will send out a specific email solicitation for the annual reports in
the next week or two. You will have several weeks to prepare the
reports. As before, Nicole Jacobs (njacobs@darpa.mil) will be the
primary contact at DARPA concerning these requests.
Thanks,
Bill
-------
∂31-May-90 0959 MPS
Prof. Peters, 32212 would like to talk to you about
Andre Bondarko
∂31-May-90 1001 weening@Gang-of-Four.Stanford.EDU SAIL -> Gang-of-Four
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 10:01:20 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA22522; Thu, 31 May 90 10:01:35 -0700
Date: Thu, 31 May 90 10:01:35 -0700
From: weening@Gang-of-Four.Stanford.EDU (Joe Weening)
Message-Id: <9005311701.AA22522@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: SAIL -> Gang-of-Four
Shall I copy your files from SAIL to Gang-of-Four, as we planned?
∂31-May-90 1013 mps@Gang-of-Four.Stanford.EDU
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 10:13:26 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA23768; Thu, 31 May 90 10:13:42 -0700
Date: Thu, 31 May 90 10:13:42 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005311713.AA23768@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
You never gave me the address where you want me to send all those
papers to. Would you please? Also, will you be in today because of
your trip to Tahoe. Thanks
∂31-May-90 1131 scherlis@vax.darpa.mil ISTO SOFTWARE TECHNOLOGY COMMUNITY MEETING
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 31 May 90 11:31:43 PDT
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA21722>; Thu, 31 May 90 11:11:31 -0400
Resent-Message-Id: <9005311502.AA14323@sun45.darpa.mil>
Received: by sun45.darpa.mil (4.1/5.51)
id AA14323; Thu, 31 May 90 11:02:25 EDT
Received-Date: Wed, 30 May 90 22:05:02 -0400
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA18873>; Wed, 30 May 90 22:05:02 -0400
Posted-Date: Wed 30 May 90 21:56:00-EDT
Received: by sun45.darpa.mil (4.1/5.51)
id AA13884; Wed, 30 May 90 21:56:01 EDT
Date: Wed 30 May 90 21:56:00-EDT
From: William L. Scherlis <SCHERLIS@vax.darpa.mil>
Subject: ISTO SOFTWARE TECHNOLOGY COMMUNITY MEETING
To: SW-PI@vax.darpa.mil
Cc: boehm@vax.darpa.mil, squires@vax.darpa.mil, mettala@vax.darpa.mil,
jkramer@vax.darpa.mil, boesch@vax.darpa.mil, njacobs@vax.darpa.mil
Message-Id: <644118960.0.SCHERLIS@VAX.DARPA.MIL>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@VAX.DARPA.MIL>
Resent-Date: Thu 31 May 90 11:02:24-EDT
Resent-From: William L. Scherlis <SCHERLIS@DARPA.MIL>
Resent-To: SW-PI@vax.darpa.mil
TO: The Software Principal Investigators
SUBJECT: ISTO Software Technology Community Meeting
There is a major DARPA/ISTO event coming up in the next month that
will involve the Software Technology PIs.
1. ISTO SOFTWARE STRATEGIC PLAN. The software picture has changed
considerably at DARPA in the last several months. Barry Boehm came on
board as Director of ISTO six months ago, and we now have a team of four
program managers directly concerned with software technology:
Jack Kramer - STARS and the Software Engineering Instutute
Erik Mettala - Domain specific software architectures, object bases
Brian Boesch - Distributed and parallel systems software
Bill Scherlis - Software environments, prototyping, computer science
In addition, Stephen Squires, who is the DARPA focal point for high
performance computing issues, has been working with us to develop
appropriate ties to the high performance computing efforts.
The six of us have been working together over the past several months to
develop a strategic plan for the software efforts at DARPA, which are now
concentrated in ISTO. Naturally enough, we are following an iterative
model in developing the plan. Over the past months, we have been
interacting with many of you concerning both specific parts of the plan
and the formulation of an overall shared technical vision, and we are now
eager to interact in a more structured way with the ISTO research
community to help refine the plan. This plan will form the basis for the
ISTO Software Technology program as of this October.
The purpose of the plan is to enable us to implement a principled
investment strategy, with the resultant benefits:
- An effective synergy and balance can be maintained among software
program elements and between long-term and short-term.
- Opportunities for research collaboration and for coordinated
technology transition efforts can be better identified and
planned for.
- Opportunities for researcher participation are made clear, including
means to opportunistically insert unexpected new ideas.
- Technology consumers and researchers both have a clear model of how
they can effectively interact.
- Consumers of results can develop realistic expectations and more
effectively participate at early stages of technology development.
- Results of the program should be usable in other areas, including
high performance computing and artifical intelligence systems.
The major challenge, of course, is to find the optimum level of technical
commitment for the plan. Too much commitment raises risk through the
elimination of alternative approaches. Too little commitment creates the
illusion of a plan, but fails to achieve the benefits. At the meeting,
we will work with you to ensure that our plan draft maintains a mixed
strategy and, in particular, does not eliminate possibilities that could
productively be pursued.
I must emphasize that the plan is designed to enable us BOTH to
continually produce results that will ultimately be usable by others AND
to be effectively receptive to unexpected new ideas as they emerge from
your research efforts. This means that we not only must achieve
technology transition out from our program, but our downstream users must
also be able to achieve a "reality transition" back into our program,
enabling us to refine our research agenda and thereby increase the real
impact of our efforts.
2. TECHNICAL MEETING. Because of the many changes in the office, and the
need to address the many opportunities presented by the various planning
efforts (including the HPC plan and the DoD-wide Software Master Plan --
see below), we are calling a Software Technology Community Meeting for
the end of June. We realize this is short notice, but we are eager to
involve you in a substantive discussion of these plans as soon as
possible. We expect to implement these plans at the start of the fiscal
year.
The agenda of the meeting will include presentations of elements of the
current draft of the ISTO Software Strategic Plan, discussion of related
efforts, and opportunity to interact concerning issues raised. Our
goal is to improve and refine the plan, so we can use the revised result
as a basis for our programs. This will affect most of our community, so
please come and participate!
Our intent is to have regular software community meetings in the future,
with well-defined goals and agendas for each. Most of the other ISTO
research communities have regular meetings (e.g., vision, speech,
networking, VLSI CAD). Some of these have published proceedings, and
considerable benefit has been realized. (But you should know that we are
definitely not planning a full ISTO-wide PI meeting.)
The Software Technology Community Meeting meeting will be in the
Washington area, at a specific location to be announced, starting at noon
on Wednesday June 27 and ending at noon June 29. There may be additional
informal sessions in the evenings and on the afternoon of the last day.
DARPA representation will include Barry Boehm, Stephen Squires, the four
program managers involved with software, and probably several other ISTO
people.
3. RELATED PLAN EFFORTS. This planning effort is related to two other
major planning efforts that we are involved with.
One is the DoD-wide effort, initiated by the Office of the Secretary of
Defense, to develop a DoD Software Master Plan. This plan, now in draft
form and available to the community, concerns software DoD-wide, and
specifically addresses the four major mechanisms available to the DoD to
address software issues:
- Software acquisition and life-cycle management. This addresses how we
contract for, develop, acquire, manage, maintain, enhance, and
otherwise deal with software through the formal acquisition
process.
- DoD software policies and standards. There are hundreds of DoD
directives, standards, and other guidance documents that relate to
software acquisition and management. Also, DoD participates in
commercial standards efforts in order to represent its interests
(e.g., interoperability, security, tool interfaces).
- Personnel. Although DoD contracts out most of its software activity,
it must maintain a base of software expertise in-house in order to
represent its own needs effectively. In addition, the volume of
DoD software business (9% of the total DoD budget, according to
industry estimates) requires DoD to attend to the personnel base
in industry and in research facilities.
- Technology and technology transition. A DoD Software *Technology* Plan
is to be developed, with appropriate additional funding. Technology
transition activities are also recommended, such as shadow
projects, information clearinghouses, and assessment mechanisms.
The DoD Software *Technology* Plan will likely be developed over the next
few months, with DARPA/ISTO participation. In addition to basic computer
science work and focused technology transition efforts, plan elements are
likely to include the following areas: Software engineering environment
frameworks, software engineering tools, prototyping technology, means to
develop and utilize reusable software assets, domain-specific software
architectures, software re-engineering for the vast inventory of existing
software, management tools including metrics and cost estimation,
ultrareliable and secure software, distributed and parallel software,
scalable AI capabilities, and systems software.
A draft of this plan was circulated in February, and a DoD-wide community
meeting was held in April in order to promote discussion and comment.
Many people from indistry, government, and research laboratories
attended, and we have since received many hundreds of written comments.
A final draft Master Plan is in production.
The other major planning effort is the High Performance Computing
program, which you have been hearing about for some time. This program
has major software elements. Copies of this plan are widely available in
the community. If you want one, send your full postal address by email
to Nicole Jacobs (njacobs@darpa.mil). Please limit requests to one or
two per institution. (It may also be possible to have some copies
distributed of the February draft DoD Software Master Plan. Let Nicole
know if you are desperate to have one.)
We encourage you to look over these plans before our meeting.
4. DETAILS AND PREPARATION. Details concerning agenda and location in
Washington will be supplied in the next week. We may also make certain
modest requests to specific projects concerning preparation for the
meeting, possibly including some short presentations.
We encourage each project to send senior technical representatives:
ACTION:
Please send a list of proposed invitees (for your project)
to Nicole Jacobs (njacobs@darpa.mil) by close of business
Wednesday June 6. We encourage early responses.
By Thursday June 7, we expect to provide details sufficient for you to
make travel arrangements to Washington.
Bill
-------
∂31-May-90 1347 jfeldman%icsib2.Berkeley.EDU@jade.berkeley.edu Chinese Room
Received: from jade.berkeley.edu by SAIL.Stanford.EDU with TCP; 31 May 90 13:47:12 PDT
Received: from icsib2.Berkeley.EDU
by jade.berkeley.edu (5.61.1/1.16.25)
id AA27417; Thu, 31 May 90 13:44:29 PDT
Received: by icsib2.berkeley.edu. (4.0/SMI-4.0)
id AA02602; Thu, 31 May 90 13:47:55 PDT
Date: Thu, 31 May 90 13:47:55 PDT
From: jfeldman%icsib2.Berkeley.EDU@jade.berkeley.edu (Jerry Feldman)
Message-Id: <9005312047.AA02602@icsib2.berkeley.edu.>
To: jmc@sail.stanford.edu
Subject: Chinese Room
John,
I would like to see a written version of your Syntax,Semantics and
Systems talk if one is available.
Jerry
∂31-May-90 1358 scherlis@vax.darpa.mil Re: telephone appointment
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 31 May 90 13:58:13 PDT
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA24035>; Thu, 31 May 90 16:59:59 -0400
Posted-Date: Thu 31 May 90 16:50:51-EDT
Received: by sun45.darpa.mil (4.1/5.51)
id AA14626; Thu, 31 May 90 16:50:52 EDT
Date: Thu 31 May 90 16:50:51-EDT
From: William L. Scherlis <SCHERLIS@DARPA.MIL>
Subject: Re: telephone appointment
To: JMC@sail.stanford.edu
Message-Id: <644187051.0.SCHERLIS@VAX.DARPA.MIL>
In-Reply-To: <1Tar1V@SAIL.Stanford.EDU>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@VAX.DARPA.MIL>
John --
I'll try to call you Monday, late morning your time. But if I
can't get to it on Monday, then I will try again on Friday. (There
are three all-day meetings Tues-Thurs.)
Bill
-------
∂31-May-90 1501 harnad@clarity.Princeton.EDU SPP Workshop: Plunkett Abstract
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 15:01:08 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.38/relay)
id AA15219; Thu, 31 May 90 17:59:21 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA01047; Thu, 31 May 90 17:59:27 EDT
Date: Thu, 31 May 90 17:59:27 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9005312159.AA01047@reason.Princeton.EDU>
To: andrewsj@vassar.BITNET
Subject: SPP Workshop: Plunkett Abstract
Symbol Grounding During the Child's Second Year
Kim Plunkett
Psychology Department
Aarhus University
Many studies of cognitive and linguistic development attest to the
substantial symbolic achievements of children during their second year
despite their limited syntactic development. These achievements include
symbolic play, conventional gesturing skills and object permanence.
Developmentalists often cite children's emergent vocabularies as
evidence of acquired symbolic capacity. This paper reviews evidence on
the degree to which children's early vocabularies are symbolically
grounded and the mechanisms that might support these developments.
Recent connectionist work suggests how vocabulary development can be
the product of a self-organising system that coordinates
representations of different domains (perceptual and linguistic). The
reorganistional processes are like a developmental partitioning of
linguistic and perceptual coordinations into functionally discrete,
causally active domains. Symbol grounding arises from a developmental
process which is a decontextualised formal system in some situations
and content driven in others.
∂31-May-90 1609 mps@Gang-of-Four.Stanford.EDU Renewed book
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 16:09:04 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA19090; Thu, 31 May 90 16:09:15 -0700
Date: Thu, 31 May 90 16:09:15 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005312309.AA19090@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Renewed book
The book by Shortliffe has been renewed until June 14
∂31-May-90 1633 mps@Gang-of-Four.Stanford.EDU Rathman Thesis
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 16:33:42 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA19290; Thu, 31 May 90 16:33:56 -0700
Date: Thu, 31 May 90 16:33:56 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9005312333.AA19290@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Rathman Thesis
Peter was in today checking on his thesis. He has to file on the 6th
of June. I told him he could probably talk to you when you get back
from Fallen Leaf at home.
Pat
∂31-May-90 1826 hanson@csli.Stanford.EDU invitation
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 31 May 90 18:26:05 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA23690; Thu, 31 May 90 18:28:36 PDT
Date: Thu, 31 May 1990 18:28:33 PDT
From: Kristin Hanson <hanson@csli.stanford.edu>
To: clt@sail.stanford.edu, jmc@sail.stanford.edu
Subject: invitation
Message-Id: <CMM.0.88.644203713.hanson@csli.stanford.edu>
I don't think we've ever met, but I would like to invite you to a
good-bye party that a friend and I are having for Gianluigi Bellin.
It's to be the evening of Friday, June 8, around 8:30 p.m. at my
house, which is at 226 Donohoe Street in East Palo Alto (near Menalto,
this side of the freeway). I hope you will be able to make it.
- Kristin Hanson
∂31-May-90 1943 rathmann@eclipse.stanford.edu Any comments on my thesis?
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 31 May 90 19:43:46 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA05079; Thu, 31 May 90 19:45:50 PDT
Date: Thu, 31 May 90 19:45:50 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9006010245.AA05079@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: Any comments on my thesis?
Professor McCarthy,
Did you have any comments on my thesis draft. Time is short.
If there are to be any changes I will have to make them soon.
-Peter
∂01-Jun-90 0103 ME SAIL printing
To: JMC, CLT, VAL, GLB
CC: ME
You can still print from SAIL, although printing has been turned
off for other users.
∂01-Jun-90 1041 MPS
Dr. Irwin Oreskes would like you to call him. He thinks he
knows you from your childhood years in LA. He will be out
here for his daughters graduation and would like to see you.
Call him during the evening (his time) at 212 873-4623
∂01-Jun-90 1500 harnad@clarity.Princeton.EDU Synopsis of SG Problem & 11 SPP Abstracts
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 1 Jun 90 15:00:25 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.38/relay)
id AA05564; Fri, 1 Jun 90 17:55:30 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA21186; Fri, 1 Jun 90 17:55:31 EDT
Date: Fri, 1 Jun 90 17:55:31 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9006012155.AA21186@cognito.Princeton.EDU>
To: andrewsj@vassar.BITNET
Subject: Synopsis of SG Problem & 11 SPP Abstracts
To: Contributors to SPP Symbol Grounding Workshop (especially,
but Searle Symposium too)
Upon re-reading all the abstracts I have the feeling that the focus of the
Workshop is a bit blurred. It's certainly not that I want all or any of
you to agree with me on the solution, but we have to at least give some
indication that we are all addressing the same problem, so that the
Workshop does not just become a pot-pourri of everyone's ongoing work.
So here is a fairly compact synopsis of the symbol grounding problem.
May I ask you, in presenting your papers, to indicate explicitly their
bearing on the problem itself, as formulated here?
Looking forward to seeing you all at the conference. -- Stevan
--------------------------------------------------------------------
Synopsis of the Symbol Grounding Problem
Stevan Harnad
Department of Psychology
Princeton University
Princeton NJ 08544
A symbol system is a set of physical tokens (e.g., scratches on paper,
holes on a tape, flip-flop states in a computer) and rules for
manipulating them (e.g., erase "0" and write "1"). The rules are purely
syntactic: They operate only on the (arbitrary) shapes of the symbols,
not their meanings. The symbols and symbol combinations can be given a
systematic semantic interpretation, for example, they can be
interpreted as meaning objects ("cat," "mat") or states of affairs
("the cat is on the mat"). The symbol grounding problem is that the
meanings of the symbols are not grounded in the symbol system itself.
They derive from the mind of the interpreter. Hence, on pain of
infinite regress, the mind cannot itself be just a symbol system,
syntactically manipulating symbols purely on the basis of their shapes.
The problem is analogous to attempting to derive meaning from a
Chinese/Chinese dictionary if one does not first know Chinese. One just
goes around in endless circles, from meaningless symbol to meaningless
symbol. The fact that the dictionary is systematically interpretable is
of no help, because its interpretation is not grounded in the
dictionary itself. Hence "Strong AI," the hypothesis that cognition is
symbol manipulation, is incorrect, as Searle has argued.
How can one ground the meanings of symbols within the symbol system
itself? This is impossible in a pure symbol system, but in a hybrid
system, one based bottom-up on nonsymbolic functions such as
transduction, analog transformations and sensory invariance extraction,
the meanings of elementary symbols can be grounded in the system's
capacity to discriminate, categorize and name the objects and states of
affairs that its symbols refer to, based on the projections of those
objects and states of affairs on its sensory surfaces. The grounded
elementary symbols -- the names of the ground-level sensory object
categories -- can then be rulefully combined and recombined into
higher-order symbols and symbol strings. But unlike in a pure symbol
system, these symbol manipulations would not be purely syntactic ones,
constrained only by the arbitrary shapes of the symbol tokens; they
would also be constrained by (indeed, grounded in) the nonarbitrary
shapes of the distal objects, their proximal sensory projections, the
analogues of the sensory projections that subserve discrimination, and
the learned and innate sensory invariants that subserve categorization
and naming.
The solution to the symbol grounding problem, then, will come from an
understanding of the mechanisms capable of accomplishing sensory
categorization and the learning of concrete and abstract categories.
Among the canidates are sensory icons and neural nets that learn
sensory invariants.
------------------------------------------------------------------
11 Abstracts for SPP Symposium On Searle's Chinese Room
and
Workshop on Symbol Grounding
Sunday, June 9, University of Maryland, College Park
For information: andrewsj@vassar.bitnet
(1) SYNTAX, SEMANTICS AND SYSTEMS
John McCarthy <JMC@SAIL.Stanford.EDU>
Artificial Intelligence
Stanford University
John Searle begins his (1990) "Consciousness, Explanatory Inversion and
Cognitive Science" with:
"Ten years ago in this journal I published an
article (Searle, 1980a and 1980b) criticising what I
call Strong AI, the view that for a system to have
mental states it is sufficient for the system to
implement the right sort of program with right inputs
and outputs. Strong AI is rather easy to refute and
the basic argument can be summarized in one sentence:
`a system, me for example, could implement a program for
understanding Chinese, for example, without
understanding any Chinese at all.' This idea, when
developed, became known as the Chinese Room Argument."
The Chinese Room Argument can be refuted in one sentence:
"Searle confuses the mental qualities of one computational process,
himself for example, with those of another process that the first
process might be interpreting, a process that understands Chinese, for
example."
That accomplished, the lecture will discuss the ascription of mental
qualities to machines with special attention to the relation between
syntax and semantics, i.e. questions suggested by the Chinese Room
Argument. I will deal explicitly with Searle's four ``axioms'',
which, although they don't have a unique interpretation, suggest
various ideas worth discussing.
-----------------------------------------------------------
(2) SEARLE'S CHINESE ROOM
Pat Hayes <hayes@parc.xerox.com>
The basic flaw in Searle's argument is a widely accepted misunderstanding
about the nature of computers and computation: the idea that a computer
is a mechanical slave that obeys orders. This popular metaphor suggests
a major division between physical, causal hardware which acts, and
formal symbolic software, which gets read. This distinction runs
through much computing terminology, but one of the main conceptual
insights of computer science is that it is of little real scientific
importance. Computers running programs just aren't like the Chinese
room.
Software is a series of patterns which, when placed in the proper
places inside the machine, cause it to become a causally different
device. Computer hardware is by itself an incomplete specification of a
machine, which is completed - i.e. caused to quickly reshape its
electronic functionality - by having electrical patterns moved within
it. The hardware and the patterns together become a mechanism which
behaves in the way specified by the program.
This is not at all like the relationship between a reader obeying some
instructions or following some rules. Unless, that is, he has somehow
absorbed these instructions so completely that they have become part of
him, become one of his skills. The man in Searle's room who has done
this to his program now understands Chinese.
--------------------------------------------------------------------
(3)E MEASUREMENT PROBLEM IN PHSYICS AND BRAIN THEORY
H.H. Pattee
Department of Systems Science
TJ Watson School of Engineering
SUNY
Binghamton NY 13901
The measurement problem in physics is a special case of the
symbol-grounding problem of brain theory, which in turn is a special
case of the epistemological problem of relating the knower and the
known. In quantum theory the measurement problem is notoriously
obscure, because the results of measuring quantum events are
nonclassically observer-dependent and yet must be expressed only in
classical language. Even the measurement of classical events cannot be
completely described by physical laws, because measurement involves
intent, i.e., the what, where, and when of the measurements must be
determined by an "observer," not by the laws. It also makes no sense
to say that a measurement has occurred unless there is a "result."
At issue in physical theory are the necessary conditions for
"observer," and "result." Working physicists evade the philosophical
issues by using language confined to formal, operational symbol
systems that restrict measurements to numerical results and
prediction to computation. Formal symbol systems do allow unambiguous
predictions, but only at the cost of generating logical antinomies and
conceptual puzzles like "Schroedinger's Cat" and "Wigner's friend."
If one relaxes the syntactic precision of formal symbols, however, and
extends the concepts of "observer" and "result" to the subsymbolic,
dynamical level of measurement-control constraints in simpler
organisms, some of these puzzles are resolved. The problem I raise for
brain theory is that formal symbolic models of the brain may likewise
produce unambiguous results only at the cost of conceptual puzzles like
Searle's "Chinese Room."
------------------------------------------------------------------
(4) METAPHOR AND SYMBOL
David Powers <powers@informatik.uni-kl.de>
Robotics
University of Kaiserslautern FRG
The presentation is based on the monograph [Powe89] on
Machine Learning of Natural Language and Ontology,
updated with recent developments and indication of
consolidation of earlier work into a coherent approach.
Hypotheses concerning the role of contrast and
similarity, and the relationship of these mechanisms to linguistic
concepts of metaphor and paradigm, neural self-organization,
psycholinguistic paradoxes concerning negative information,
and consideration of language as part of the entire ontology
have lead to a series of experiments in machine learning of
aspects of language both individually or in combination. The work
makes use of a simulated robot world as well as textual input.
Significantly, similar results have been achieved with neural and
conventional techniques applied to the same task, with simulated
neurons being clearly associated with words and classes, and
with particular grammatical rules being associated with
particular synapses.
These results suggest three possible resolutions of the symbol
grounding problem: the symbol/non-symbol distinction
is not meaningful; neural networks can exhibit 'symbolic' behaviour
and structure; and, a sensory-motor environment can provide
grounding.
REFERENCES
[Powe89] David M. W. Powers, C. C. R. Turk,
Machine Learning of Natural Language, Springer Verlag,
London/Berlin, 1989.
-------------------------------------------------------------------
(5) CATEGORIZATION AS A PSYCHOPHYSICAL PROBLEM
Michel Treisman treisman@vax.oxford.ac.uk
Psychology Department
University of Oxford
If the same observer has to categorize the same stimulus on two
different occasions he may make different decisions each time. Why is
this? In some situations the observer will show an excessive tendency
to repeat previous responses to similar stimuli; this is sometimes
referred to as 'assimilation'. At other times he or she may avoid
previous responses: 'contrast'. Why should categorization be so
unreliable? Or what does this observation tell us about the process of
making a judgment? A psychophysical model will be outlined which
provides an explanation for these phenomena in terms of mechanisms
which tend to optimize uncertain judgments, and the relations between
different types of categorization, at different levels of complexity,
will be considered.
---------------------------------------------------------------------
(6) GROUNDING MENTAL SYMBOLS IN OBJECT IMAGES
Irving Biederman <PSYIRV@vx.acs.umn.edu> and John E. Hummel
Psychology Department
University of Minnesota
We describe a neural net (NN) implementation of a theory of real time
visual shape recognition that takes as input the edges corresponding to
the occlusional and orientation discontinuities in an image. As output
the model activates a unit that is selective for a specified
arrangement of simple volumes (or geons) and thus achieves a basic (or
entry) level classification according to Biederman's (1987)
Recognition-by-Components theory of object recognition. The output unit
can qualify as a symbol of the object in that it reflects the major
invariances of visual object recognition. The model solves four
fundamental problems in object recognition that likely confront all
attempts at visual basic-level symbol grounding:
1) Translational, size, and orientation invariance: The same output
unit(s), corresponding to the object, are activated no matter where the
image falls in the visual field, the size of the image, and the
orientation in depth (up to parts occlusion),
2) Appropriate grouping (or organization) of image elements into
appropriate parts,
3) A basis of determining invariant object centered relations (such as
TOP- OF or SIDE-CONNECTED), and
4) A basis for computing the similarity (or equivalence) of object
images.
These problems all required a solution to the "binding problem"--
determining what groups with what. In the present case, for example,
how are the various segments of the parts of an object grouped
according to their appropriate parts. Most NN models have employed
enumeration, assigning a unit to each attribute combination. Such
enumerative schemes are unsatisfactory in that they require a
prohibitively large number of units to represent even modest input
domains. Moreover, they do not express the equivalence of inputs. By
employing different units to represent the different locations of an
object, for example, the information that it is the same object in the
different locations is not represented. Our model achieves binding
through phase locking of the oscillatory activity of cells that are
tuned to oriented image edges. The phase locking (or synchrony) is
established by "fast enabling links" (FELs) between pairs of a)
collinear, b) coterminating, and c) parallel adjacent edge cells. These
units then activate invariant representations of geons and relations in
intermediate layers.
The model offers some perspective on what it is that makes a category
"basic." A category such as chair will encompass a number of
distinguishably different geon models that, perceptually, may be as
distant as members of different classes.
------------------------------------------------------
(7) INVARIANT STIMULUS FEATURES AND THE
CORTICAL REPRESENTATION OF VISUAL INFORMATION
J. Anthony Movshon tony@cortex.psych.nyu.edu
Center for Neural Science
New York University,
New York, NY 10003.
The responses of visual cortical neurons depend upon a number of
different features of the visual stimuli that fall within their
receptive fields. In most cases there is no difference in the nature
of the response produced by varying the stimulus along different
dimensions. For this reason, it is commonly recognized that the
firing of an individual neuron cannot be used unambiguously to infer
the character of a particular visual stimulus. Rather, it is necessary
to examine the distribution of activity across a population of
neurons. In considering how the multi-dimensional nature of visual
neural signals might most readily be disambiguated, it seems that
special significance might be attached to those stimulus dimensions
for which particular groups of neurons show an invariant selectivity.
An invariant selectivity is a selectivity for the value of a stimulus
along some dimension that is independent of the value of the stimulus
along other dimensions. For example, the selectivity of neurons in
the primary visual cortex (V1) for such stimulus variables as
orientation and spatial frequency is largely independent of the
precise stimulus conditions used to measure them. On the other hand,
their selectivity for the direction of motion of targets depends in a
complex way on the spatial and temporal composition of the target, and
is therefore not invariant.
In the visual cortex of the macaque monkey, many distinct visual
areas have been identified with electrophysiological and anatomical
techniques. A number of ``lower-order'' cortical areas seem to
contain neurons whose activity is primarily controlled by signals of
retinal origin - prominent among these are areas V2, V3, V4 and MT, as
well as the primary visual cortex, V1. The responses of neurons in all
these areas seem to depend on the same collection of visual stimulus
dimensions, including spatial location and size, contour orientation,
spatial frequency, chromatic composition, drift rate, direction of
motion, and binocular disparity. Neurons in different areas can have
more or less sensitivity to variations in one or another of these
parameters, so that in quantitative terms it may be argued that
signals from one area carry more information than signals from another
about particular stimulus features. It is largely on the basis of
quantitative arguments of this kind that a particular role for one or
another area in a particular aspect of visual processing has been
asserted - qualitative differences in the way that visual signals are
represented in different cortical areas have not received much
attention.
In this paper I will argue that a special significance is, in
fact, attached to the particular stimulus dimensions that are the
subjects of invariant representation within an area. For example,
despite the fact that neurons in V1 carry signals about the direction
and speed of motion of objects, the fact that their invariant
selectivity is for orientation, spatial and temporal frequency makes
it impossible for them to carry invariant information about speed and
direction. Neurons in MT, on the other hand, carry invariant
information about motion, at the expense of losing the invariant
representation of spatial and temporal parameters. Analogous
reorganization of signals about color, stereoscopic depth, and other
stimulus features may explain the existence of other representations
of the visual image in the visual cortex.
-----------------------------------------------------------
(8) COLOR AND COLOR CONSTANCY
Laurence T. Maloney ltm@xp.psych.nyu.edu
Department of Psychology
Center for Neural Science
New York University
The initial visual information that determines color appearance in
human vision depends as much on the lighting in a scene as on the
spectral properties of surfaces in the scene. A visual system that
bases color appearance on the properties of the surface, discounting
the contribution of the illuminant, is termed \fIcolor constant\fP.
I describe a class of algorithms designed to allow vision systems to
estimate information (analogous to color) about surface properties
despite changes in the illuminant. These linear model algorithms
include work by Brill, Buchsbaum, Maloney and Wandell, and others.
These algorithms share strong assumptions about the range of possible
illuminants and possible surface reflectances present in a scene. I
describe evidence suggesting that many common surfaces and illuminants
satisfy the constraints required by linear model algorithms.
Hilbert (1987, Chap. 7) discusses the consequences of this work
for philosophy.
Maloney, L. T., and Wandell, B. A., Color constancy: A computational
method for recovering surface spectral reflectance. Journal of the
Optical Society of America A, 1986, \fB3\fP, 29-33.
Maloney, L. T., Evaluation of linear models of surface spectral
reflectance with small numbers of parameters. Journal of the Optical
Society of America A, 1986, \fB3\fP, 1673-1683.
Hilbert, D. R., \fIColor and Color Perception; A Study in
Anthropocentric Realism.\fP (Stanford, CA: CSLI, 1987).
---------------------------------------------------
(9) PERCEPTUAL MEMORY CATEGORIZATION IN PRIMARY SENSORY CORTEX
Richard Granger granger@ICS.UCI.EDU
Center for the Neurobiology of Learning and Memory
University of California, Irvine
Recent results from neurobiological simulation work have led to a novel
hypothesis: that the physiological operation of a primary sensory
cortical area (olfactory (piriform) cortex) automatically organizes
learned perceptual cues into a hierarchical memory (Ambros-Ingerson,
Granger and Lynch, Science, 1990). In the simulation, repetitive
perceptual samples ("sniffs", or "glances") of learned cues traverse
the constructed hierarchy, such that initial samples yield relatively
coarse-grained category responses whereas later samples yield
increasingly finer-grained information about the cue. The resulting
iterative recognition of cues shares many characteristics with the
robust psychological phenomenon of "basic levels": within a
hierarchically nested set of categories such as "animal-bird-robin",
there is a specific level of abstraction that is more readily processed
(e.g., recognized faster) than the others; "bird" in this example
(Mervis and Rosch, Ann.Rev.Psych., 1981). The correspondence raises the
possibility that aspects of this psychological phenomenon may arise
from fundamental physiological mechanisms in primary sensory cortex.
-------------------------------------------------------------------
(10) ICONS AND TEMPORAL PATTERNS:
A DYNAMIC CONNECTIONIST SOLUTION TO SYMBOL GROUNDING
Harnad (1990) proposes that categories can be grounded by their direct
relationship to a physical icon of the input stimulation. The notion of
an icon has a clear meaning in the case of a visual display: it is a
pattern of activity in a field of neurons that is physically isomorphic
with the pattern of light in 2D. The standard proposal for a physical
icon of TIME is physical distance. Such a model is naturally
implemented by delay lines in a network (see Lang, 1990, NN).
But delay lines are not a good model for human behavior. When subjects
are trained on a complex temporal pattern, like a random sequence of
tones, they can develop a detailed perceptual representation (Spiegel &
Watson 1981, Watson and Foyle, 1985, JASA). Some skills that should be
easy are very difficult -- eg, recognizing an absolute time interval in
the face of randomly varying intervening sounds. This should be easy
since absolute time differences are represented by weights on specific
delay lines. So edges from inputs that have a random relation to the
categorical identification should learn random weights. On the other
hand, skills that should be very difficult turn out to be easy, such as
detecting serial order of familiar patterns in the face of changes in
rate of presentation. This should be difficult since a pattern that
appears at different rates will be distributed differently across the
range of delays, and should thus be learned only slowly.
On other hand, we have been developing dynamic network models that
represent learned temporal patterns of tones as stable equilibria in
the activation space of a group of fully recurrent nodes
(Port-Anderson, 1990, Anderson-Port, 1990). These systems were trained
(with real-time recurrent learning) to recognize particular tone
sequences. They are highly resistant to noise and continue to recognize
patterns even when the rate of presentation is varied by a factor of 2
faster or slower. Watson's results and our simulations suggest that
brains do not produce an icon of auditory patterns in time. It implies
that direct contact with stimulation that is distributed in time is not
possible. Although I do not disagree with Harnad on the importance of
symbol grounding for an account of perception, apparently, the
grounding of categories does not require an `icon' in the sense that
Harnad has proposed.
-----------------------------------------------------------------
(11) SYMBOL GROUNDING DURING THE CHILD'S SECOND YEAR
Kim Plunkett PSYKIMP@vms2.uni-c.dk
Psychology Department
Aarhus University
Many studies of cognitive and linguistic development attest to the
substantial symbolic achievements of children during their second year
despite their limited syntactic development. These achievements include
symbolic play, conventional gesturing skills and object permanence.
Developmentalists often cite children's emergent vocabularies as
evidence of acquired symbolic capacity. This paper reviews evidence on
the degree to which children's early vocabularies are symbolically
grounded and the mechanisms that might support these developments.
Recent connectionist work suggests how vocabulary development can be
the product of a self-organising system that coordinates
representations of different domains (perceptual and linguistic). The
reorganistional processes are like a developmental partitioning of
linguistic and perceptual coordinations into functionally discrete,
causally active domains. Symbol grounding arises from a developmental
process which is a decontextualised formal system in some situations
and content driven in others.
---------------------------------------------
∂01-Jun-90 1545 pratt@coraki.stanford.edu notes for CS 350
Received: from coraki.stanford.edu by SAIL.Stanford.EDU with TCP; 1 Jun 90 15:45:49 PDT
Received: by coraki.stanford.edu (4.0/25-eef) id AA06973; Fri, 1 Jun 90 15:47:11 PDT
Date: Fri, 1 Jun 90 15:47:11 PDT
From: Vaughan Pratt <pratt@cs.stanford.edu>
Message-Id: <9006012247.AA06973@coraki.stanford.edu>
To: jmc@sail
Subject: notes for CS 350
Are there notes for CS 350?
-v
∂03-Jun-90 1247 rathmann@eclipse.stanford.edu
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 3 Jun 90 12:47:02 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA13038; Sun, 3 Jun 90 12:49:02 PDT
Date: Sun, 3 Jun 90 12:49:02 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9006031949.AA13038@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Hello, rumor has it that you are in the building.
Before you leave, if possible, I would like to meet with you to either
have you sign my thesis, or find out what needs to be done to it
before you can sign.
thanks,
Peter
∂03-Jun-90 1636 rathmann@eclipse.stanford.edu circumscribing equality (your formulation)
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 3 Jun 90 16:36:28 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA13402; Sun, 3 Jun 90 16:38:26 PDT
Date: Sun, 3 Jun 90 16:38:26 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9006032338.AA13402@eclipse.stanford.edu>
To: jmc@sail.stanford.edu
Cc: rathmann@eclipse.stanford.edu
Subject: circumscribing equality (your formulation)
Here is a telegraphic description of a homomorphism based version of
equality circumscription, which lives more or less entirely within
normal circumscription. If it is not comprehensible, I will be happy
to fill in the details (*after* Wednesday.)
It can be done, but requires some ugly things.
Basically, we define a new function symbol h, which serves the same
purpose as the original definition of <=
If we just have a finite number of constants the formulation is:
constants: a1 a2 a1' a2'
add a function symbol h, and axioms: h(a1')=a1
h(a2')=a2
This implies that if a1'=a2' then a1=a2, and is equivalent to saying
(a1', a2') <= (a1, a2) in your formulation. (Note that I haven't
given a way to say <.)
-----
Here is the formulation for ground terms. I'll stick with a simple
example with two constants a,b and a function S.
step 1:
Add primed versions of the constants and functions of the original
theory, so that the constants are now a',b',a,b and functions S'
and S.
step2: add a function symbol h, and add the homomorphism axioms:
-for every constant:
h(a')=a
h(b')=b
-for every function symbol:
forall X . h(S'(X))=S(h(X))
Again, this is the condition to say that the primed terms are "less
collapsed" than the unprimed terms. The "least collapsed" model for
this example should include two non-intersecting sets isomorphic to
the integers - the successors of a and the successors of b.
Note: this implies all the sentences like:
if S'(a')=S'(b') then S(a)=S(b)
since h(S'(a'))=S(h(a'))=S(a)
h(S'(b'))=S(h(b'))=S(b)
Again, this is only a formulation of <=, we have no way yet to say <.
-----
We have some options on how to say <. I don't think any of them are
very pretty.
Here's one. Add a new predicate gr'() which is meant to be true
whenever the argument is denoted by a primed ground term. Define the
predicate with axioms:
for primed constants:
gr'(a'), gr'(b')
for primed function symbols
forall X . gr'(X) implies gr'(S'(X))
In the overall circumscription, gr' has to be minimized (probably with
highest priority, although I haven't checked).
Given this predicate, we say that the primed "model" is < the
unprimed one if there exist ground terms which are made equal by the
homomorphism h. The axiom would be:
exist X
exist Y
gr'(X) & gr'(Y) &
NOT(X=Y) &
h(X) = h(Y)
The minimization axiom would be something like this. (I am making up
notation as I go along).
Z(S,a,b) & (NOT
(EXISTS S',a',b' h .
Z(S',a,b')
h:(S',a',b') -> (S,a,b) ; the homomorphism axioms above
(S',a',b') < (S,a,b) ; the < condition defined
; with gr'()
))
-Peter
∂03-Jun-90 1738 rathmann@eclipse.stanford.edu Re: circumscribing equality (your formulation)
Received: from eclipse.stanford.edu by SAIL.Stanford.EDU with TCP; 3 Jun 90 17:38:54 PDT
Received: by eclipse.stanford.edu (4.0/inc-1.0)
id AA13435; Sun, 3 Jun 90 17:40:52 PDT
Date: Sun, 3 Jun 90 17:40:52 PDT
From: rathmann@eclipse.stanford.edu (Peter K. Rathmann)
Message-Id: <9006040040.AA13435@eclipse.stanford.edu>
To: JMC@SAIL.Stanford.EDU
Cc: rathmann@eclipse.stanford.edu
In-Reply-To: John McCarthy's message of 03 Jun 90 1709 PDT <14b#5p@SAIL.Stanford.EDU>
Subject: Re: circumscribing equality (your formulation)
Yes.
That is how roughly how Marianne and I defined preferred model
semantics. It is a matter of taste which is worse - a predicate with
only syntactic meaning, like gr'(), or an explicitly quantified
function variable.
Because we used homomorphisms to minimize non-equality predicates, we
got fairly complicated semantics for non-ground-term universe
elements. By separating the two, and using normal circumscription to
minimize predicates, it might be possible to avoid this.
-Peter
∂04-Jun-90 0901 JMC
Dina Bolla
∂04-Jun-90 1023 rww@ibuki.com Hungary
Received: from ames.arc.nasa.gov by SAIL.Stanford.EDU with TCP; 4 Jun 90 10:23:05 PDT
Received: by ames.arc.nasa.gov (5.61/1.2); Mon, 4 Jun 90 10:23:31 -0700
Received: by ibuki.com (5.52/4.7)
id AA06520; Mon, 4 Jun 90 09:23:16 PDT
Date: Mon, 4 Jun 90 09:23:16 PDT
From: rww@ibuki.com (Richard Weyhrauch)
Message-Id: <9006041623.AA06520@ibuki.com>
To: jmc@sail.stanford.edu
Subject: Hungary
Do you remember Bolint Dolmaki from Hungary? Do you have his
address?
Richard
∂04-Jun-90 1101 mostow@cs.rutgers.edu letter from Stefanuk
Received: from edison.rutgers.edu by SAIL.Stanford.EDU with TCP; 4 Jun 90 11:01:18 PDT
Received: by edison.rutgers.edu (5.59/SMI4.0/RU1.2/3.05)
id AA00258; Mon, 4 Jun 90 14:00:12 EDT
Date: Mon, 4 Jun 90 14:00:12 EDT
From: mostow@cs.rutgers.edu
Message-Id: <9006041800.AA00258@edison.rutgers.edu>
To: jmc@sail.stanford.edu
Cc: priediti@cs.rutgers.edu
Subject: letter from Stefanuk
Reply-To: mostow@cs.rutgers.edu
Dear Professor McCarthy:
I was excited to find in today's mail an envelope apparently from you. However
upon opening it I discovered it was a letter from a V. L. Stefanuk from the
USSR Academy of Science, expressing some dismay that my IJCAI89 paper with
Armand Prieditis on automated discovery of admissible heuristics did not cite
his IJCAI73 and IJCAI75 papers on problem representation. It mentions that the
latter paper was omitted from the English version of the IJCAI75 proceedings
but that you were one of the participants who received a copy.
I surmise that this person was a visitor who borrowed one of your envelopes and
that you had nothing to do with the letter otherwise. But I thought I should
let you know in case you were interested or wanted to add anything.
Thanks. - Jack Mostow
P.S. You may remember me from 1980-81, when I was a Research Associate in HPP.
∂04-Jun-90 1107 rww@ibuki.com Soviet Commerce
Received: from ames.arc.nasa.gov by SAIL.Stanford.EDU with TCP; 4 Jun 90 11:07:12 PDT
Received: by ames.arc.nasa.gov (5.61/1.2); Mon, 4 Jun 90 10:22:23 -0700
Received: by ibuki.com (5.52/4.7)
id AA06500; Mon, 4 Jun 90 09:17:43 PDT
Date: Mon, 4 Jun 90 09:17:43 PDT
From: rww@ibuki.com (Richard Weyhrauch)
Message-Id: <9006041617.AA06500@ibuki.com>
To: jmc@sail.stanford.edu
Subject: Soviet Commerce
John,
Has Ershov said anything to you about Soviet Software Commerce? What
do you think about the russian capabilities to produce software?
Do you have Ershov's and Stefanoks addresses?
Thanks.
Richard
PS what about the strategic issues? I was actually thinking of
importing software if they could produce it, but maybe exporting
Lisp is a good idea? (By importing software produced there it
seems that all strategic and currency are made easier!!!. Surely
bits for hard currency may appeal to them.
∂04-Jun-90 1121 wood%hpmeg@LANL.GOV Preprint Request
Received: from p.lanl.gov (milnet-gw.lanl.gov) by SAIL.Stanford.EDU with TCP; 4 Jun 90 11:21:47 PDT
Received: from hpmeg.lanl.gov by p.lanl.gov (5.61/1.14)
id AA15517; Mon, 4 Jun 90 12:22:46 -0600
Message-Id: <9006041822.AA15517@p.lanl.gov>
Received: by hpmeg
(15.11/15.6) id AA09451; Mon, 4 Jun 90 12:22:00 mdt
Date: Mon, 4 Jun 90 12:22:00 mdt
From: chris wood <wood@hpmeg>
To: JMC@SAIL.Stanford.EDU
Subject: Preprint Request
Cc: wood@hpmeg
Dr. McCarthy:
Unfortunately I'll miss your talk on Searle at the upcoming SPP meeting.
Could you please send me a copy? E-mail is fine.
Thanks,
Chris Wood
Biophysics Group (P-6) M715
Los Alamos National Laboratory
Los Alamos, NM 87545
(505) 665-2546
∂04-Jun-90 1214 cphoenix@csli.Stanford.EDU Paper on agreement
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Jun 90 12:14:20 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA09320; Mon, 4 Jun 90 12:16:56 PDT
Date: Mon, 4 Jun 90 12:16:56 PDT
From: cphoenix@csli.Stanford.EDU (Chris Phoenix)
Message-Id: <9006041916.AA09320@csli.Stanford.EDU>
To: jmc@sail.stanford.edu
Subject: Paper on agreement
I'm writing a paper on the concepts of "trade" ("I'll do X if you do Y"),
"agreement", and maybe "promise", for CS324. Anton Schwartz, the TA for
that class, suggested that I might look at your Elephant 2000 paper, but
didn't know how relevant that would be. Can you give me any suggestions
on which literature to look at?
The paper will be pretty short, and will approach the problem by looking
at the cognitive process each agent goes through in making the trade, and
each agent's mental representation of the situation.
Thanks...
Chris
∂04-Jun-90 1219 cphoenix@csli.Stanford.EDU re: Paper on agreement
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Jun 90 12:19:23 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA09435; Mon, 4 Jun 90 12:21:59 PDT
Date: Mon, 4 Jun 90 12:21:59 PDT
From: cphoenix@csli.Stanford.EDU (Chris Phoenix)
Message-Id: <9006041921.AA09435@csli.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: re: Paper on agreement
OK, thanks anyway... can you give me any quick pointers on where else to
look?
∂04-Jun-90 1224 cphoenix@csli.Stanford.EDU re: Paper on agreement
Received: from csli.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Jun 90 12:24:42 PDT
Received: by csli.Stanford.EDU (4.1/inc-1.0)
id AA09680; Mon, 4 Jun 90 12:27:17 PDT
Date: Mon, 4 Jun 90 12:27:17 PDT
From: cphoenix@csli.Stanford.EDU (Chris Phoenix)
Message-Id: <9006041927.AA09680@csli.Stanford.EDU>
To: JMC@SAIL.Stanford.EDU
Subject: re: Paper on agreement
OK, thanks. Prof. Shoham is teaching the class, so I've already asked him.
I'll stop bothering you now... :-)
∂04-Jun-90 1508 ME X terminals
To: JMC, CLT
Can I come over today to install the X terminals at your house?
Also, is one of the modems already there (Telebit T2500)?
∂04-Jun-90 1534 @Sunburn.Stanford.EDU:pony-errors@neon.stanford.edu JMC Prancing Pony Bill
Received: from Sunburn.Stanford.EDU by SAIL.Stanford.EDU with TCP; 4 Jun 90 15:34:45 PDT
Received: from Neon.Stanford.EDU by Sunburn.Stanford.EDU with SMTP (5.61+IDA/25-eef) id AA23694; Mon, 4 Jun 90 15:35:48 -0700
Date: Mon, 4 Jun 90 15:35:48 -0700
Message-Id: <9006042235.AA23694@Sunburn.Stanford.EDU>
From: The Bill Program <pony-bills@Neon.Stanford.EDU>
Subject: JMC Prancing Pony Bill
Reply-To: pony-bills@Neon.Stanford.EDU
To: John McCarthy <JMC@SAIL.STANFORD.EDU>
Prancing Pony Bill of John McCarthy (JMC) for May 1990 (6/4/1990)
Previous Balance 2.17
Monthly Interest at 1.00% 0.02
Current Charges NONE
---------
TOTAL AMOUNT DUE 2.19
PAYMENT DELIVERY LOCATION: CSD Receptionist.
Make checks payable to: STANFORD UNIVERSITY.
Please deliver payments to the Computer Science Dept receptionist, Bldg. 460.
To ensure proper crediting, please include your ACCOUNT NAME on your
check. If you pay by cash, use the small yellow envelopes provided
and write both your ACCOUNT NAME and the AMOUNT on outside.
Note: The recording of a payment may take up to three weeks after the payment
is made, but never beyond the next billing date. Please allow for this delay.
Bills are payable upon presentation. Interest of 1.00% per month will be
charged on balances remaining unpaid 25 days after bill date above.
An account with a credit balance earns interest of 0.33% per month,
based on the average daily balance.
You haven't paid your Pony bill since 10/1989.
Accounts with balances remaining unpaid for more than 55 days are
considered delinquent and are subject to reduction of credit limit.
Please pay your bill and keep your account current.
∂04-Jun-90 1653 VAL
∂04-Jun-90 1651 VAL offer
To: pbw@CS.UTEXAS.EDU, aldale@CS.UTEXAS.EDU
Dear Paul and Al,
Thank you very much for arranging that I get the official letter before
we leave for Europe. I'll contact you as soon as we get back (that is,
around July 10) to discuss a few remaining questions regarding the conditions
of my employment. I don't anticipate any serious problems, and I hope that
everything will be easily resolved to our mutual satisfaction.
And thank you, of course, for all the work you have done in connection with
this matter, and for all your efforts to make UT Austin an attractive place
for both of us.
Best regards,
Vladimir
∂04-Jun-90 1807 VAL Travel plans
I'm leaving on June 6, coming back on July 7. You can contact me
from June 14 through 17 by sending a message to Fiora Pirri,
ITALIANO%IRMIASI.RM.CNR.IT@Forsythe (say "For Fiora Pirri" in
the subject line).
∂04-Jun-90 2225 ME X terminal logout
After you log out from any SAIL windows, use the right mouse key to
click on the menu item 'kill twm'. This will log you out from any
Unix hosts where you have windows open, and it will leave you in
the "Serial Session" on the X terminal, where you'll find yourself
still logged in (to Sunburn for now). Log out there. That will leave
you talking to Tip-MJHf. Type the Quit or Q command to make the
Tip hang up its end of the modem connection. You'll see a few lines
of garbage and 5 seconds later a "NO CARRIER" message (from your modem).
Now you're back where you started: To dial out you would type "atdn"
at this point.
∂05-Jun-90 0744 bundy@aipna.edinburgh.ac.uk Carolyn Talcott
Received: from NSFnet-Relay.AC.UK by SAIL.Stanford.EDU with TCP; 5 Jun 90 07:44:53 PDT
Received: from sun.nsfnet-relay.ac.uk by vax.NSFnet-Relay.AC.UK
via Janet with NIFTP id aa21345; 5 Jun 90 14:58 BST
Date: Tue, 5 Jun 90 15:11:21 BST
Message-Id: <3157.9006051411@sin.aipna.ed.ac.uk>
From: Alan Bundy <bundy@aipna.edinburgh.ac.uk>
Subject: Carolyn Talcott
To: jmc@sail.stanford.edu
Do you have an email address for Carolyn? Thanks.
Alan
∂05-Jun-90 0753 mps@Gang-of-Four.Stanford.EDU Rathman Thesis
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 5 Jun 90 07:53:06 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA08506; Tue, 5 Jun 90 07:53:16 -0700
Date: Tue, 5 Jun 90 07:53:16 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9006051453.AA08506@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Rathman Thesis
He has to file his thesis tomorrow. He came to see you last Friday
and as you were not here, I told him I would ask you today about it.
∂05-Jun-90 1006 beeson@ucscd.UCSC.EDU software law (an idle question)
Received: from ucscd.UCSC.EDU by SAIL.Stanford.EDU with TCP; 5 Jun 90 10:06:45 PDT
Received: by ucscd.UCSC.EDU (5.61/1.34)
id AA03502; Tue, 5 Jun 90 10:10:46 -0700
Date: Tue, 5 Jun 90 10:10:46 -0700
From: beeson@ucscd.UCSC.EDU (20012000)
Message-Id: <9006051710.AA03502@ucscd.UCSC.EDU>
To: jmc@sail.stanford.edu
Subject: software law (an idle question)
suppose someone wrote, independently, an exact clone of an existing
program such as Mathematica. Could they market it (at a lower price)
or could they be sued for copyright infringement?
∂05-Jun-90 1658 ok@coyote.stanford.edu AI Qual Schedule (Reminder)
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 5 Jun 90 16:57:52 PDT
Received: by coyote.stanford.edu; Tue, 5 Jun 90 16:57:44 PDT
Date: 5 Jun 1990 1657-PDT (Tuesday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: ginsberg@sunburn.stanford.edu, latombe@coyote.stanford.edu,
jmc@sail.stanford.edu, nilsson@tenaya.stanford.edu,
rosenschein@teleos.com, shoham@hudson.stanford.edu,
ash@sumex-aim.stanford.edu, darwiche@neon.stanford.edu,
young@neon.stanford.edu
Cc: merryman@sunburn.stanford.edu, ok@coyote.stanford.edu
Subject: AI Qual Schedule (Reminder)
AI Qualification Exam Schedule
Wednesday, June 6, 1990
10:30 -- 12noon (MJH, Room 222)
Adnan Darwiche: Reasoning
C: Latombe D: Nilsson O: Rosenschein
1:45 -- 3:15 (MJH, Room 222)
David Ash: Planning
C: Nilsson D: Ginsberg O: Shoham
1:45 -- 3:15 (MJH, Room 301)
Michael Young: Formal Models of Inter-Agent Communication
C: Latombe D: Rosenschein O: McCarthy
3:30 -- 4:30 Meeting of committees (MJH. Room 222)
∂05-Jun-90 2107 ash@sumex-aim.stanford.edu Re: AI Qual Schedule (Reminder)
Received: from sumex-aim.stanford.edu by SAIL.Stanford.EDU with TCP; 5 Jun 90 21:07:42 PDT
Received: by sumex-aim.stanford.edu (4.0/inc-1.0)
id AA20998; Tue, 5 Jun 90 21:10:49 PDT
Date: Tue, 5 Jun 1990 21:10:48 PDT
From: David Ash <ash@sumex-aim.stanford.edu>
To: Oussama Khatib <ok@coyote.stanford.edu>
Cc: ginsberg@sunburn.stanford.edu, latombe@coyote.stanford.edu,
jmc@sail.stanford.edu, nilsson@tenaya.stanford.edu,
rosenschein@teleos.com, shoham@hudson.stanford.edu,
ash@sumex-aim.stanford.edu, darwiche@neon.stanford.edu,
young@neon.stanford.edu, merryman@sunburn.stanford.edu,
ok@coyote.stanford.edu
Subject: Re: AI Qual Schedule (Reminder)
In-Reply-To: Your message of 5 Jun 1990 1657-PDT (Tuesday)
Message-Id: <CMM.0.88.644645448.ash@sumex-aim.stanford.edu>
> 10:30 -- 12noon (MJH, Room 222)
↑↑↑
>
> Adnan Darwiche: Reasoning
>
> C: Latombe D: Nilsson O: Rosenschein
>
>
> 1:45 -- 3:15 (MJH, Room 222)
↑↑↑
>
> David Ash: Planning
222 is a student office. Do you mean 220, Nils' conference room?
-d
∂06-Jun-90 0950 ok@coyote.stanford.edu Re: AI Qual Schedule (Reminder)
Received: from coyote.stanford.edu by SAIL.Stanford.EDU with TCP; 6 Jun 90 09:49:53 PDT
Received: by coyote.stanford.edu; Wed, 6 Jun 90 09:49:22 PDT
Date: 6 Jun 1990 0949-PDT (Wednesday)
From: Oussama Khatib <ok@coyote.stanford.edu>
To: David Ash <ash@sumex-aim.stanford.edu>
Cc: ginsberg@sunburn.stanford.edu, latombe@coyote.stanford.edu,
jmc@sail.stanford.edu, nilsson@tenaya.stanford.edu,
rosenschein@teleos.com, shoham@hudson.stanford.edu,
ash@sumex-aim.stanford.edu, darwiche@neon.stanford.edu,
young@neon.stanford.edu, merryman@sunburn.stanford.edu,
ok@coyote.stanford.edu
Subject: Re: AI Qual Schedule (Reminder)
In-Reply-To: David Ash <ash@sumex-aim.stanford.edu> /
Tue, 5 Jun 1990 21:10:48 PDT.
<CMM.0.88.644645448.ash@sumex-aim.stanford.edu>
Thank you David. I thought 222 was Nils' conference room.
The correct room number is 220.
-Oussama
∂06-Jun-90 1002 tajnai@Hudson.Stanford.EDU AI Day on Campus -- RSVP for lunch
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 6 Jun 90 10:02:00 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA22478; Wed, 6 Jun 90 09:59:28 -0700
Date: Wed, 6 Jun 1990 9:59:26 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: AI-DOC:;@Hudson.Stanford.EDU
Cc: hill@Hudson.Stanford.EDU
Subject: AI Day on Campus -- RSVP for lunch
Message-Id: <CMM.0.88.644691566.tajnai@Hudson.Stanford.EDU>
Let me know if you are planning to attend the barbeque lunch in Terman
Grove for AI Day on Campus. We must turn in a count to the caterer.
Carolyn
∂06-Jun-90 1031 ME Tip crashes
To: JMC
CC: CLT
∂06-Jun-90 0942 JMC
Is there some way to tell if the TIP is down?
Yes, if you can get into the network from somewhere else.
(Tip-MJHf indeed crashed last night and has been crashing a lot this week.
Hopefully we'll be able to do something to fix it.)
Here are the 2400-baud phone lines on Tip-MJHe (which isn't crashing).
Tip-E (2400,1200,300) 321-4761 through -65 (5 lines)
Tip-E (2400,1200,300) 321-8971 through -75 (5 lines)
Tip-E (2400 only) 321-5733 through -34 (2 lines)
Tip-E (2400 only) 327-9028 through -29 (2 lines)
To dial one of these numbers, first go to the Serial Session window
under SETUP on your X terminal (click on Serial Session). Then click
on the baud rate until it says 2400. Then dial Tip E (e.g., atdt3214761).
Then type some RETURNs until you get the Tip prompt. Then say "tip-mjhf"
to try to connect to the probably-dead tip. If you can't connect, then
it has probably crashed. You can then connect to somewhere else, but
you probably shouldn't run xinitremote at 2400 (although you can try to
see how slow it is). You won't have the keys set up specially for SAIL,
but you can type TTY ANSI on SAIL and be a no-EDIT-key display. Or type
"set term=ansi" on a Unix system to use it in the ANSI emulation window.
∂06-Jun-90 1033 ME Tip crashes cont'd
To: JMC
CC: CLT
Be sure to set the X serial speed back to 9600 before using Tip-MJHf
and xinitremote. Also, to get into SETUP mode, push the SETUP key
near the top right of the keyboard.
∂06-Jun-90 1401 MPS Free Software
Sorry, you are right. I always thought Boston and Cambridge kinda
ran together as one city. Glad you caught my error before I mailed
out the order. Thanks
∂06-Jun-90 1553 MPS
There will be no BofD meeting at MAD this month. Robin, Nafeh's
secretary called today.
∂07-Jun-90 0700 JMC
scherlis, hotels, car in Texas, Stefanyuk membership, Jeff
∂07-Jun-90 0730 mumford@sancho.harvard.edu ATP committee
Received: from sancho.harvard.edu by SAIL.Stanford.EDU with TCP; 7 Jun 90 07:30:02 PDT
Date: Thu, 7 Jun 90 10:29:44 EDT
From: mumford@sancho.harvard.edu (david mumford)
Message-Id: <9006071429.AA14482@sancho.harvard.edu>
To: jmc@sail.stanford.edu
Subject: ATP committee
Dear John,
Thank you for agreeing to serve on a committee to award the
'current' and 'milestone' prizes in Automatic Theorem Proving. So
far, we have me, Jack Schwartz, John Selfridge, Gerry Sussman, yourself
and, probably, Larry Wos (he was a bit hesitant, but will get back to me).
I talked to Jack and he thought that perhaps we should add another
person, although both of us want to avoid adding obvious candidates for
one of the awards. He suggested Robin Milnor. Can you comment: i) would
he seem to you a strong candidate so that we shouldn't ask him to be
on the committee? and ii) if not, do you have a telephone number and
e-mail address for him, so I can ask him?
At the same time, perhaps you can begin thinking of nominations.
I propose we each make a short list of nominations, which I will collate
and circulate for comments.
Regards,
David (Mumford)
∂07-Jun-90 1040 MPS
Yes, I am taking care of your letter file. Will you be in today to
sign the Gibbons letter?
∂07-Jun-90 1348 MPS
Call L. Morgenstern.
∂07-Jun-90 1500 ME SAIL font
To: JMC
CC: ME
Well, it turned out that the SAIL font was simply missing from Sunburn
for some reason. (Probably flushed somehow when new fonts were installed.)
Anyway, its on Sunburn now, so you should be able to use it normally there.
Let me know if it works from home tonight.
∂07-Jun-90 1901 H.HARPER@Macbeth.Stanford.EDU re: speaking of seccesionist states from federations
Received: from Macbeth.Stanford.EDU by SAIL.Stanford.EDU with TCP; 7 Jun 90 19:00:54 PDT
Date: Thu 7 Jun 90 19:02:42-PDT
From: copacetic fig for life <H.HARPER@Macbeth.Stanford.EDU>
Subject: re: speaking of seccesionist states from federations
To: JMC@SAIL.Stanford.EDU
In-Reply-To: <1DdYPv@SAIL.Stanford.EDU>
Message-ID: <12596041537.21.H.HARPER@Macbeth.Stanford.EDU>
oh. thank you.
-------
∂08-Jun-90 0854 njacobs@vax.darpa.mil ISTO Software Technology Community Meeting
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 8 Jun 90 08:54:20 PDT
Received: from sun45.darpa.mil by vax.darpa.mil (5.61/5.61+local)
id <AA02720>; Fri, 8 Jun 90 11:36:07 -0400
Posted-Date: Fri 8 Jun 90 11:25:12-EDT
Received: by sun45.darpa.mil (4.1/5.51)
id AA19954; Fri, 8 Jun 90 11:25:13 EDT
Date: Fri 8 Jun 90 11:25:12-EDT
From: Nicole L. Fields <NJACOBS@DARPA.MIL>
Subject: ISTO Software Technology Community Meeting
To: SW-PI@vax.darpa.mil
Message-Id: <644858712.0.NJACOBS@VAX.DARPA.MIL>
Mail-System-Version: <SUN-MM(218)+TOPSLIB(128)@VAX.DARPA.MIL>
MEMORANDUM FOR DISTRIBUTION
SUBJECT: 1990 ISTO Software Technology Community Meeting
The Information Science and Technology Office (ISTO) of the Defense
Advanced Research Projects Agency (DARPA) is sponsoring a Software
Technology Community Meeting for invited researchers and government
officials on 27-29 June in Washington, DC.
The workshop will be held at the Omni Shoreham Hotel in Washington, DC.
You are asked to reserve your room and to make arrangements for
payment yourself. A special workshop rate of $82.43 plus taxes is available.
Rooms will held until 12 June, so call early and be sure to identify
yourself as part of the ISTO Software Technology Community Meeting.
The number is (202) 234-0700. Since space at the workshop is limited,
it is necessary that everyone planning to attend notify Nicole Jacobs as
soon as possible either by phone at (202) 694-5800 or by electronic
mail at njacobs@vax.darpa.mil.
Please plan your travel so that you can attend the first session
beginning at 12:30 PM Wednesday, 27 June and the last session ending
at 3:30 PM Friday, 29 June. The workshop registration desk will be
open from 11:00 to 1:00 on that day. Additional information on the
meeting and registration will be sent to everyone by the end of next week.
If anyone has any special dietary requirements please contact
Nicole Jacobs by COB Wednesday, 13 June.
-------
∂08-Jun-90 1332 @coraki.stanford.edu:pratt@cs.stanford.edu Re: notes for CS 350
Received: from coraki.stanford.edu by SAIL.Stanford.EDU with TCP; 8 Jun 90 13:32:37 PDT
Received: by coraki.stanford.edu (4.0/25-eef) id AA15268; Fri, 8 Jun 90 13:33:42 PDT
Message-Id: <9006082033.AA15268@coraki.stanford.edu>
To: John McCarthy <JMC@sail.stanford.edu>
Subject: Re: notes for CS 350
In-Reply-To: Your message of 02 Jun 90 1721 PDT.
<10buvX@SAIL.Stanford.EDU>
Date: 08 Jun 90 13:33:41 PDT (Fri)
From: pratt@cs.stanford.edu
What is cs350?
Mathematical Theory of Computation. The catalog shows it as being
taught by you but not offered 1889-90 [sic].
-v
∂08-Jun-90 1415 betsy@russell.Stanford.EDU Information for Keith
Received: from russell.Stanford.EDU by SAIL.Stanford.EDU with TCP; 8 Jun 90 14:15:18 PDT
Received: by russell.Stanford.EDU (4.1/inc-1.0)
id AA16394; Fri, 8 Jun 90 14:18:06 PDT
Date: Fri, 8 Jun 1990 14:18:04 PDT
From: Betsy Macken <betsy@russell.stanford.edu>
To: barwise@russell.Stanford.EDU, john@russell.Stanford.EDU,
briansmith.pa@xerox.com, grosz@harvard.harvard.edu,
karttunen.pa@xerox.com, bmoore@ai.sri.com, stan@teleos.com,
bresnan@russell.Stanford.EDU, sag@russell.Stanford.EDU,
wasow@russell.Stanford.EDU, winograd@russell.Stanford.EDU,
kaplan.pa@xerox.com, jmc@sail.stanford.edu, rperrault@ai.sri.com,
suppes@russell.Stanford.EDU
Cc: peters@russell.Stanford.EDU, etch@russell.Stanford.EDU, israel@ai.sri.com,
laura@russell.Stanford.EDU
Subject: Information for Keith
Message-Id: <CMM.0.88.644879884.betsy@russell.stanford.edu>
Keith will arrive on 27 June to begin work on the CSLI history SDF has
commissioned him to write. He is going to begin with "the initial grant
acquisition phase". Do any of you have material from this period in your
own files that might be useful to Keith? Would you check to see
if you have messages or papers or whatever that might be of use to him and
may not be in our central files? Please let me know either way so
that if you don't have anything, I won't bother about it again.
Thanks,
Betsy
∂11-Jun-90 0930 rlm@Hudson.Stanford.EDU
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 Jun 90 09:30:10 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA14925; Mon, 11 Jun 90 09:27:24 -0700
From: Robert L. Miller <rlm@Hudson.Stanford.EDU>
Resent-Message-Id: <9006111627.AA14925@Hudson.Stanford.EDU>
Date: Mon, 11 Jun 1990 9:27:24 PDT
Resent-To: AI-DOC:;@Hudson.Stanford.EDU
Resent-Date: Mon, 11 Jun 1990 9:27:24 PDT
Resent-From: "Robert L. Miller" <rlm@Hudson.Stanford.EDU>
Message-Id: <CMM.0.88.645121644.rlm@Hudson.Stanford.EDU>
Apparently-To: prior@hudson.Stanford.EDU
Apparently-To: rlm@hudson.Stanford.EDU
Apparently-To: tajnai@cs.stanford.edu
Apparently-To: der@psych.stanford.edu
Apparently-To: Rindfleisch@sumex-aim.stanford.edu
Apparently-To: Shortlife@sumex-aim.stanford.edu
Apparently-To: Nii@sumex-aim.stanford.edu
Apparently-To: OK@coyote.stanford.edu
Apparently-To: Iwasaki@sumex-aim.stanford.edu
Apparently-To: Bhayes-roth@sumex-aim.stanford.edu
Apparently-To: gruber@sumex-aim.stanford.edu
Apparently-To: ginsberg@cs.stanford.edu
Apparently-To: Fagan@sumex-aim.stanford.edu
Apparently-To: rse@sumex-aim.stanford.edu
Apparently-To: musen@sumex-aim.stanford.edu
Apparently-To: Winograd@csli.stanford.edu
Apparently-To: waldinger@ai.sri.com
Apparently-To: Marty@cis.stanford.edu
Apparently-To: shoham@cs.stanford.edu
Apparently-To: nilsson@tenaya.stanford.edu
Apparently-To: jmc@sail.stanford.edu
Apparently-To: ZM@sail.stanford.edu
Apparently-To: latombe@coyote.stanford.edu
Apparently-To: Genesereth@cs.stanford.edu
Apparently-To: binford@peace.stanford.edu
Apparently-To: Binford@coyote.stanford.edu
Apparently-To: eaf@sumex-aim.stanford.edu
AI Day on Campus will start this Thursday, June 14, 8:30 am, at
Terman Auditorium. Parallel tracks 1 and 2 will be held in Terman Aud.
and Skilling Aud. respectively, beginning at 9:30 am.
Please check the schedule below for individual speaking times.
Audio/Visual Requirements: SITN will provide for slides (hardcopy,
used with an overhead camera and monitors, is preferred),
transparencies, VHS video, and 35mm slides. If you are using anything
except a hardcopy slide, please let me know and I will see that SITN
will be readily prepared for your talk.
Will post a separate message for demonstrations.
Robert Miller
Computer Forum
rlm@hudson
---------------------------------------------------------------------
The Stanford Computer Forum Presents
AI Day On Campus
Thursday, June 14, 1990, Stanford University
Schedule:
8:30 - 9:00 Registration, Terman Breezeway
9:00 - 9:25 An Overview of Artificial Intelligence
at Stanford, Terman Auditorium
Dr. Robert Engelmore
Track 1, Terman Auditorium
9:30 - 10:30 Foundations
Multivalued Logics: Making Formal Reasoning
a Practical AI Tool
Dr. Matthew Ginsberg
Formal Approaches to the Construction of
Correct Reactive Programs
Prof. Zohar Manna
10:30 - 10:45 Break
10:45 - 12:15 Agent-Centered Reasoning
Informable Agents
Prof. Michael Genesereth
Agent-Oriented Programming
Prof. Yoav Shoham
ACTNET: A Language for Programming
Embedded, Real-Time Systems
Prof. Nils Nilsson
12:15- 1:30 Barbecue Lunch, Terman Grove
1:30 - 2:30 Demonstrations (see next column)
2:45 - 4:15 Robotics and Vision
Spatial Reasoning: From Robotics
to Engineering
Prof. Jean-Claude Latombe
Sensor-Based Robot Control Systems
Prof. Oussama Khatib
Geometric and Physical Reasoning in
Inspection and Manufacturing
Prof. Thomas Binford
4:15 - 4:30 Break
4:30 - 5:30 AI, the Brain and Society
Artificial Intelligence, Human Intelligence,
and Social Intelligence: What's the Connection?
Prof. Terry Winograd
Brain Style Computation
Prof. David Rumelhart
Track 2, Skilling Auditorium
9:30 - 10:30 Knowledge-Based Systems I
Adaptive Intelligent Systems
Dr. Barbara Hayes-Roth
Intelligent Real-Time Monitoring and Control
Dr. Barbara Hayes-Roth
10:30 - 10:45 Break
10:45 - 12:15 Knowledge-Based Systems II
How Things Work: Knowledge-Based Modeling of Physical Devices
Dr. Tom Gruber and/or Dr. Yumi Iwasaki
AI in Design and Manufacturing
Prof. Jay M. Tenenbaum
Designworld
Prof. Michael Genesereth
12:15- 1:30 Barbecue Lunch, Terman Grove
1:30 - 2:30 Demonstrations (see below)
2:45 - 4:15 Knowledge-Based Systems III
Knowledge-Based CASE
Ms. Penny Nii
Generation of Knowledge-Acquisition Tools
from Explicit Task Models
Prof. Mark Musen
Experiments with Speech Input to Expert Systems
Dr. Larry Fagan
Combining Qualitative and Quantitative Modeling
Techniques in a Monitoring and Control Task
Dr. Larry Fagan
4:15 - 4:30 Break
4:30 - 5:30 AI and Technology
Artificial Intelligence Science and Technology:
Steps Toward the Millennium
Prof. Edward A. Feigenbaum
Artificial Intelligence and Electronic
Data Interchange
Prof. John McCarthy
Demonstrations, Tracks 1 & 2
VentPlan: Combining Qualitative and
Quantitative Models for Intensive Care Unit
Data Interpretation and Response
Dr. Larry Fagan
TABLEAU, An Interactive Graphic
Deductive System
Mr. R. Burback, Prof. Z. Manna,
and Prof. R. Waldinger
Multivalued Logic System
Dr. Matthew Ginsberg
Action Network Control Programs
Prof. Nils Nilsson
Guardian System for Monitoring
Intensive-Care Patients
Dr. Barbara Hayes-Roth
Designworld
Prof. Michael Genesereth
AI in Design and Manufacturing
Prof. Jay M. Tenenbaum
* Locations to be announced.
∂11-Jun-90 1035 mps@Gang-of-Four.Stanford.EDU Phone messages
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 11 Jun 90 10:35:12 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA02376; Mon, 11 Jun 90 10:35:16 -0700
Date: Mon, 11 Jun 90 10:35:16 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9006111735.AA02376@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Phone messages
There was nothing of importance on your answer machine.
∂11-Jun-90 1204 harnad@clarity.Princeton.EDU To all Spp Workshop/Symposium participants
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 11 Jun 90 12:04:19 PDT
Received: from cognito.Princeton.EDU by Princeton.EDU (5.61++/2.39/relay)
id AA05646; Mon, 11 Jun 90 15:00:26 -0400
Received: by cognito.Princeton.EDU (4.1/1.101)
id AA07396; Mon, 11 Jun 90 15:00:34 EDT
Date: Mon, 11 Jun 90 15:00:34 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9006111900.AA07396@cognito.Princeton.EDU>
To: PSYKIMP@vms2.uni-c.dk (Kim Plunkett), dietrich@bingvaxu.cc.binghamton.edu,
dietrich@bingvaxu.cc.binghamton.edu (For: Howard Pattee),
granger@uci.BITNET, hayes.pa@xerox.com, jmc@sail.stanford.edu,
ltm@xp.psych.nyu.edu
(Larry Maloney), movshon@CMCL2.NYU.EDU,
plunkett@amos.ucsd.edu (Kim
Plunkett, UCSD),
port@iuvax.cs.indiana.edu,
powers@informatik.uni-kl.de
(David Powers), psyirv@umnacvx.BITNET,
rey@cs.umd.edu (Georges
Rey), treisman@vax.oxford.ac.uk
Subject: To all Spp Workshop/Symposium participants
Cc: andrewsj@vassar.BITNET
I wanted to collectively thank all of you for your contribution. I
think both sessions were successful and everyone found them
very informative. The highly compressed format of the workshop
left too little room for discussion or even presentation, unfortunately,
but that was the price of covering so much of the relevant spectrum --
the right strategy, I think, for such a multidisciplinary audience.
I hope that it will be possible to organize a full-size conference
devoted exclusively to symbol grounding next year, perhaps at the NY
Academy of Sciences, and there all participants will have the time
available for a full technical presentation and plenty of discussion.
Best wishes to you all,
Stevan Harnad
∂11-Jun-90 1322 LISTSERV@VM1.NoDak.EDU Message
Received: from VM1.NoDak.EDU by SAIL.Stanford.EDU with TCP; 11 Jun 90 13:22:04 PDT
Received: from NDSUVM1.BITNET by VM1.NoDak.EDU (IBM VM SMTP R1.2.1MX) with BSMTP id 6970; Mon, 11 Jun 90 15:21:45 CDT
Received: by NDSUVM1 (Mailer R2.07) id 6969; Mon, 11 Jun 90 15:21:44 CDT
Date: Mon, 11 Jun 90 15:21:43 CDT
From: Revised List Processor (1.6d) <LISTSERV@VM1.NoDak.EDU>
Subject: Message
To: JMC@SAIL.STANFORD.EDU
Your mailing for THEORY-A has been submitted to the list editor:
THEORYNT@YKTVMZ.
∂11-Jun-90 1601 xanadu!peterson@uunet.UU.NET book quote(s)
Received: from uunet.uu.net by SAIL.Stanford.EDU with TCP; 11 Jun 90 16:00:57 PDT
Received: from xanadu.UUCP by uunet.uu.net (5.61/1.14) with UUCP
id AA09273; Mon, 11 Jun 90 19:02:03 -0400
Received: by xanadu (4.1/SMI-4.0.2) id AA01007; Mon, 11 Jun 90 15:52:11 PDT
Date: Mon, 11 Jun 90 15:52:11 PDT
From: xanadu!peterson@uunet.UU.NET (Chris Peterson)
Message-Id: <9006112252.AA01007@xanadu >
To: xanadu!sail.stanford.edu!jmc@uunet.UU.NET
Subject: book quote(s)
I am working on a book on nanotechnology to be published by William
Morrow; it is aimed at an intelligent but nontechnical readership. It
will cover a number of topics on which you may have an
opinion:
The technical credibility of nanotechnology, defined as 'thorough
control of
the structure of matter' via molecular manufacturing
When such an ability is likely to be achieved
Its level of importance
Milestones along the development path(s)
Potential applications and benefits
Public policy issues
If you would like to make one or more brief comments -- just
a sentence or two -- for potential inclusion in this book, please
forward them to me by mail, fax, or electronic mail by June 22. Write
informally, as if you are explaining your point to an intelligent but
nontechnical friend.
Please put no more than a few minutes into this, if you choose
to participate.
Chris Peterson
PO Box 60775
Palo Alto, CA 94306
fax 415-948-5649
email peterson@xanadu.com
∂12-Jun-90 1151 E1.I85@Forsythe.Stanford.EDU
Received: from Forsythe.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 Jun 90 11:51:09 PDT
Date: Tue, 12 Jun 90 11:53:30 PDT
To: jmc@sail
From: "Voy Wiederhold" <E1.I85@Forsythe.Stanford.EDU>
Dear John and Carolyn,
We would like to invite you to a party after the graduation
on Sunday for Peter Rathmann. It will be at our house in Palo Alto
at 3PM with Chinese food etc. Please bring your little one too.
Our address is 471 Matadero Ave. (about 1/2 mi. south of
Page Mill on the East side of EC).
Hope you can come.
Voy
∂12-Jun-90 1233 tajnai@Hudson.Stanford.EDU
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 Jun 90 12:33:47 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA22850; Tue, 12 Jun 90 12:31:41 -0700
Date: Tue, 12 Jun 1990 12:31:39 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
In-Reply-To: Your message of 12 Jun 90 0946 PDT
Message-Id: <CMM.0.88.645219099.tajnai@Hudson.Stanford.EDU>
John, my staff and I will be involved in the conference all day on
Thursday and will not be in our offices to receive the tape.
The best thing is for you to Fed Express the tape to Pat and have her
deliver it to us in Terman Auditorium as soon as it arrives. You are
scheduled for 5:00, so the tape should arrive in plenty of time.
Carolyn
∂12-Jun-90 1418 mps@Gang-of-Four.Stanford.EDU Fed Ex
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 Jun 90 14:18:15 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA07209; Tue, 12 Jun 90 14:18:15 -0700
Date: Tue, 12 Jun 90 14:18:15 -0700
From: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
Message-Id: <9006122118.AA07209@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Subject: Fed Ex
I called Fed Ex and they said to use entire address like I said. Also
include the room number and my phone number. If there is a new driver
that day it is good to have as much info on the label as possible.
Pat
∂12-Jun-90 1622 tajnai@Hudson.Stanford.EDU
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 12 Jun 90 16:22:03 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA26593; Tue, 12 Jun 90 16:19:49 -0700
Date: Tue, 12 Jun 1990 16:19:48 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: mps@Gang-of-Four.Stanford.EDU (Pat Simmons)
In-Reply-To: Your message of Tue, 12 Jun 90 15:26:36 -0700
Cc: jmc@sail.Stanford.EDU, rlm@Hudson.Stanford.EDU
Message-Id: <CMM.0.88.645232788.tajnai@Hudson.Stanford.EDU>
No, Pat. I told John that all of us will be at the AI Day on Campus
Symposium held in Terman Auditorium. I asked him to ask you to
deliver it to me at Terman Auditorium when it arrives.
There won't be anyone in our offices at all on Thursday.
Would appreciate it.
Carolyn
∂13-Jun-90 0821 tajnai@Hudson.Stanford.EDU Re: handout
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 08:21:06 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA00329; Wed, 13 Jun 90 08:18:57 -0700
Date: Wed, 13 Jun 1990 8:18:56 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: tajnai@CS.STANFORD.EDU, MPS@SAIL.Stanford.EDU
Subject: Re: handout
In-Reply-To: Your message of 13 Jun 90 0809 PDT
Message-Id: <CMM.0.88.645290336.tajnai@Hudson.Stanford.EDU>
75 copies should be enouogh.
Carolyn
∂13-Jun-90 0950 shoham@Hudson.Stanford.EDU aidocjmc
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 09:50:35 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA00982; Wed, 13 Jun 90 09:48:26 -0700
Date: Wed, 13 Jun 1990 9:48:25 PDT
From: Yoav Shoham <shoham@Hudson.Stanford.EDU>
To: nilsson@cs.Stanford.EDU, tajnai@cs.Stanford.EDU, jmc@sail.Stanford.EDU
Subject: aidocjmc
Message-Id: <CMM.0.88.645295705.shoham@Hudson.Stanford.EDU>
My suggestion that Ed represent John notwithstanding, I will be happy to
present John's work if other means don't work out. I'm sure I'd not do as
good a job as he would, but I think I understand parts of it reasonably well.
If I had (e.g. faxed) slides to work from that would make it even easier. Of
course a 2-D JMC is probably better than a 3-D YS emulation of JMC.
Yoav
∂13-Jun-90 1139 tajnai@Hudson.Stanford.EDU Re: aidocjmc
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 11:39:14 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA01979; Wed, 13 Jun 90 11:37:04 -0700
Date: Wed, 13 Jun 1990 11:37:02 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: Yoav Shoham <shoham@Hudson.Stanford.EDU>
Cc: nilsson@cs.Stanford.EDU, tajnai@cs.Stanford.EDU, jmc@sail.Stanford.EDU
Subject: Re: aidocjmc
In-Reply-To: Your message of Wed, 13 Jun 1990 9:48:25 PDT
Message-Id: <CMM.0.88.645302222.tajnai@Hudson.Stanford.EDU>
Yoav, perhaps you could answer questions after John's video.
Carolyn
∂13-Jun-90 1138 MPS Ablex
I called them and it will be out July 1990, no specific date.
The title is Formalizing Commonsense (as if you didn' know).
Their phone number is 201-767-8450 if you need to get in touch with
them.
Pat
∂13-Jun-90 1141 tajnai@Hudson.Stanford.EDU re: aidocjmc
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 11:41:17 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA02008; Wed, 13 Jun 90 11:39:10 -0700
Date: Wed, 13 Jun 1990 11:39:09 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Cc: shoham@HUDSON.STANFORD.EDU, nilsson@CS.STANFORD.EDU,
tajnai@CS.STANFORD.EDU, MPS@SAIL.Stanford.EDU
Subject: re: aidocjmc
In-Reply-To: Your message of 13 Jun 90 1005 PDT
Message-Id: <CMM.0.88.645302349.tajnai@Hudson.Stanford.EDU>
John, we will need a release from you in order to include your video
in our videotape presentation of AIDOC to Forum members.
Carolyn
∂13-Jun-90 1324 tajnai@Hudson.Stanford.EDU re: aidocjmc
Received: from Hudson.Stanford.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 13:24:32 PDT
Received: by Hudson.Stanford.EDU (5.61+IDA/25-eef) id AA02717; Wed, 13 Jun 90 13:22:26 -0700
Date: Wed, 13 Jun 1990 13:22:25 PDT
From: "Carolyn E. Tajnai" <tajnai@Hudson.Stanford.EDU>
To: John McCarthy <JMC@SAIL.Stanford.EDU>
Subject: re: aidocjmc
In-Reply-To: Your message of 13 Jun 90 1211 PDT
Message-Id: <CMM.0.88.645308545.tajnai@Hudson.Stanford.EDU>
Have Doug send me an invoice.
And yes, we are videotaping the entire proceedings.
Thanks,
Carolyn
∂13-Jun-90 1420 MPS
I got the first page. Did you get the info about Ablex? It
will be published and released July 1990. I talked to Judy about
the videotape this morning. If you have any questions, talk to her
about it. As far as I know, it will be seen around 5:00 pm up here.
That is according to C. Tajani.
Pat
∂13-Jun-90 1427 @IBM.COM:LEORA@YKTVMH patel-schneider's address
Received: from IBM.COM by SAIL.Stanford.EDU with TCP; 13 Jun 90 14:27:40 PDT
Received: from YKTVMH by IBM.COM (IBM VM SMTP R1.2.1MX) with BSMTP id 9697; Wed, 13 Jun 90 14:30:24 PDT
Date: Wed, 13 Jun 90 17:27:27 EDT
From: LEORA@IBM.COM
To: jmc@sail.stanford.edu
Subject: patel-schneider's address
John,
Sorry for not sending you Peter Patel-Schneider's address earlier.
It is pfps@research.att.com
Have a great time in Austin,
Leora
∂13-Jun-90 1643 harnad@clarity.Princeton.EDU Incest Avoidance: BBS Call for Commentators
Received: from Princeton.EDU by SAIL.Stanford.EDU with TCP; 13 Jun 90 16:42:59 PDT
Received: from reason.Princeton.EDU by Princeton.EDU (5.61++/2.39/relay)
id AA05261; Wed, 13 Jun 90 19:42:28 -0400
Received: by reason.Princeton.EDU (4.1/1.101)
id AA05148; Wed, 13 Jun 90 15:29:32 EDT
Date: Wed, 13 Jun 90 15:29:32 EDT
From: Stevan Harnad <harnad@clarity.Princeton.EDU>
Message-Id: <9006131929.AA05148@reason.Princeton.EDU>
To: harnad@elbereth.rutgers.edu
Subject: Incest Avoidance: BBS Call for Commentators
Below is the abstract of a forthcoming target article to appear in
Behavioral and Brain Sciences (BBS), an international,
interdisciplinary journal providing Open Peer Commentary on important
and controversial current research in the biobehavioral and cognitive
sciences. To be considered as a commentator or to suggest other appropriate
commentators, please send email to:
harnad@clarity.princeton.edu or write to:
BBS, 20 Nassau Street, #240, Princeton NJ 08542 [tel: 609-921-7771]
Please specify the aspect of the article that you are qualified and
interested to comment upon. If you are not a current BBS Associate,
please send your CV and/or the name of a current Associate who would be
prepared to nominate you.
____________________________________________________________________
"An evolutionary analysis of rules regulating human inbreeding and
marriage"
Nancy Wilmsen Thornhill
Department of Biology and Department of Anthrology
University of New Mexico
Albuquerque, NM 87131
ABSTRACT: Evolutionary theory predicts that humans should avoid
incest because of its severe consequences for individual reproductive
potential through production of defective offspring. Selection for
avoidance of close-kin mating has resulted in a psychological
mechanism which promotes voluntary incest avodance. All human
societies have been alleged to have rules regulating incest. If
incest is avoided, why are social rules constructed to regulate it?
The suggestion in this paper is that incest rules do not exist
primarily in order to regulate close-kin mating, but instead regulate
inbreeding between more distant kin and sexual relations between
non-kin. This suggestion inspires 3 evolutionary hypotheses which
address predictions about affinal kin mating and cousin marriage.
Rules regulating mating between affinal kin are hypothesized to be
measures of paternity protection. Cousin marriage (inbreeding) is
hypothesized to be regulated because, if it occurs, it can concentrate
wealth and power within families to the detriment of the powerful
positions of rulers in some societies and the relatively equal social
statuses of most men in others. The predictions generated by the
hypotheses were tested on a world wide sample of 129 societies taken
from the Standard Cross Cultural Sample, using the comparative method.
The hypotheses were supported. Come confounds of the method are
discussed and controlled. Several alternative anthropological
hypotheses were tested and failed in support.
KEYWORDS: incestuous inbreeding, non-incestuous inbreeding,
evolutionary theory, mental adaptations, social rules, paternity
reliability, wealth concentration, comparative method.
∂14-Jun-90 0636 jdr@vax.darpa.mil test
Received: from vax.darpa.mil (darpa.mil) by SAIL.Stanford.EDU with TCP; 14 Jun 90 06:36:23 PDT
Posted-Date: Thu 14 Jun 90 09:17:21-EDT
Received: by vax.darpa.mil (5.61/5.61+local)
id <AA01985>; Thu, 14 Jun 90 09:17:25 -0400
Date: Thu 14 Jun 90 09:17:21-EDT
From: John Reed <JDR@DARPA.MIL>
Subject: test
To: SW-PI@vax.darpa.mil
Message-Id: <645369441.0.JDR@VAX.DARPA.MIL>
Mail-System-Version: <VAX-MM(217)+TOPSLIB(128)@VAX.DARPA.MIL>
please ignore this test
-------
∂14-Jun-90 1316 me@Gang-of-Four.Stanford.EDU files moving from SAIL
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 14 Jun 90 13:16:11 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA03916; Thu, 14 Jun 90 13:17:27 -0700
Date: Thu, 14 Jun 90 13:17:27 -0700
From: me@Gang-of-Four.Stanford.EDU (Martin Frost)
Message-Id: <9006142017.AA03916@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU
Cc: clt@Gang-of-Four.Stanford.EDU, me@Gang-of-Four.Stanford.EDU
Subject: files moving from SAIL
John, I'm moving your files onto Go4 right now. They will appear in
the directory /u/jmc/sail/jmc, with one subdirectory for each SAIL
directory you have.
(We can easily change /u/jmc/sail/jmc to /u/jmc/sail at any time after
the tape has finished reading. Carolyn, we can do the same change for
your sail directory on Go4 too.)
∂14-Jun-90 1351 MPS Sarah called
She is at Whit and Marys
968-5792
∂14-Jun-90 1617 me@Gang-of-Four.Stanford.EDU JMC files moved from SAIL to Go4
Received: from Gang-of-Four.Stanford.EDU by SAIL.Stanford.EDU with TCP; 14 Jun 90 16:16:58 PDT
Received: by Gang-of-Four.Stanford.EDU (5.61/inc-1.0)
id AA05565; Thu, 14 Jun 90 16:18:13 -0700
Date: Thu, 14 Jun 90 16:18:13 -0700
From: me@Gang-of-Four.Stanford.EDU (Martin Frost)
Message-Id: <9006142318.AA05565@Gang-of-Four.Stanford.EDU>
To: jmc@sail.Stanford.EDU, jmc@Gang-of-Four.Stanford.EDU
Cc: clt@Gang-of-Four.Stanford.EDU, me@Gang-of-Four.Stanford.EDU
Subject: JMC files moved from SAIL to Go4
John, your files are now all on Go4. You'll find them in the
directory /u/jmc/sail, with one subdirectory per SAIL directory.
(Since the /f1 partition didn't hold all the files, some of them
are in the /f2 partition, but they will all appear to be under
/u/jmc/sail, thanks to the symbolic links of Unix.)
Let me know when and if you want mail forwarded automatically.