Date: Wed, 21 May 86 12:50:29 edt From: vtisr1!irlistrq To: fox Subject: IRList Digest V2 #24 Status: RO IRList Digest Wednesday, 21 May 1986 Volume 2 : Issue 24 Today's Topics: Email - IRList mail problems Query - Diff at the word level? Reply - Address for Hanspeter Giger, Question on machine sds.mfenet Software Psychology Society - V. 10 #4, Potomac Chapter Newsletter Announcement - IJCAI Site Selection and Officer Election CSLI - Knowledge, Communication, and Time COGSCI - Large Vocabulary Speech Recognition Learning to Construct Abstractions ---------------------------------------------------------------------- From: Aviezri Fraenkel Date: Thu, 15 May 86 16:54:01 -0200 Subject: IRList Hello Ed, The last IRList I got was Issue #21 of April 23. Has it stopped to appear, or did subsequent issues get lost in the mails -- I wonder... Regards, Aviezri S. Fraenkel. [Note: Thank you for alerting me to mail problems! I had sent out issues 22 and 23 in the interim but some mishap must have occurred during distribution after the message left here. I sent both out again last week, and am still getting error messages (about this second try) back about mailer failures at some sites - but I think this second attempt will make it through eventually. Please let me know, folks, if you don't have issues 1-24 of Volume 2, and I will retransmit any missing files as needed. - Ed] ------------------------------ Date: Fri, 9 May 86 19:09:19 edt From: donna@NLM-VAX.ARPA Subject: diff marks Do you know of any software that finds significant differences between text files. It is fairly easy to get nroff, etc. to find differences, but these could be changed spelling, a single new word, punctuation. We need to be able to compare two versions of a document and "mark" significant differences. Mike Lesk knew of nothing. This would make an interesting research topic for someone. Has your ir network disappeared. This would be a good question for it. Also are there any other nets that I could send this out on. Donna [Note: nlm-vax had mail problems too which have since been resolved and they should now have received the various issues they missed. Regarding the 'diff' problem, it sounds like an interesting and useful topic to study. One solution might be to take the UNIX 'diff' approach, but on a word rather than line level, with spelling correction or other inexact matching techniques added in. Does anyone have something working to do this? - Ed] ------------------------------ From: Hans-Peter Giger Date: Wed, 7 May 86 15:28:34 -0100 Subject: Re: test Hi! I got your Reply of the 5th May. Addresses seem to work now Hanspeter Giger [Note: In issues 15-18 was discussion of contacting Hanspeter Giger. We now finally have made contact and he is on the distribution list thanks to help by several people. - Ed PS - Can anyone give me an ARPANET-type address to reach maisel@sds.mfenet - he also wanted to be added to IRList but none of my attempts to find 'sds' have succeeded.] ------------------------------ Date: Wed, 14 May 86 01:09:59 edt From: ben@MIMSY.UMD.EDU Subject: Hello Here's another announcement for IRLIST... I tried to put it into a more useful form....Ben [Note: Nice Job! Thanks, Ed] SOFTWARE PSYCHOLOGY SOCIETY POTOMAC CHAPTER VOLUME 10 NUMBER 4 SUMMER 1986 June 13 Room 413-414 DIGITAL TYPOGRAPHY: ART, SCIENCE, AND TECHNOLOGY Richard Rubinstein, Digital Equipment Corporation 150 Locke Drive - LM04/H4, Marlboro, MA 01752 Digital typography is using computers to present text on screens and paper. Doing this well, producing high-quality images that people read easily, calls for an admixture of three disparate disciplines. The first, typography, is a traditional art encompassing aesthetic knowledge. Psychol- ogy offers knowledge of perception, and the means of objec- tive evaluation. And computer technology provides the vehi- cle for delivery of letterforms, economically and upon demand. The knowledge of these three specialties is expressed in three vastly different languages. Typographers know a great deal about letterforms, readability, effective layout and so on, but they say these things in terms that are not very familiar or useful to the engineers. By contrast, psycholo- gists have measurements that may be hard to relate to the real world of reading books and screens. Historically, read- ing experiments have not been typographically sophisticated. And engineers have been accused of just building systems without talking to anyone else, user or expert. I am working to convert typographic knowledge into princi- ples, data, and algorithms that engineers can use directly in the design of computer systems. I will present a number of typographical conjectures that I believe psychological research may illuminate. I will present data that I have collected that bear on these conjectures. The result of research spanning these three disciplines could be algorithms that would dramatically improve the quality of machine-made text, and a better understanding of the role of typographic quality in the effective presenta- tion of text. ____________________ Note: All meetings will be held at the George Washington University's Marvin Center (800 21st Street, N.W.) between 10:00 AM and noon. Coffee and doughnuts will be provided by the Department of Electrical Engineering and Computer Sci- ences. July 11 Room 413-414 USER MODELS AND THE AUTOMATIC SELECTION OF ARTICLES FOR AN ELECTRONIC NEWSPAPER Bob Allen, Bell Communications Research 445 South Street, Morristown, NJ 07960 The potential for automatic modelling of people's prefer- ences for news stories was examined in several experiments. One study showed that predictions were easy to make in terms of feature articles and sections of the newspaper that sub- jects read. However, a second study showed that it is rela- tively difficult to predict specific news articles that will read. Overall, the results suggest possible problems in the wide-spread application of user models. Finally, the rela- tionship betweeen computer user models and psychological models which make predictions about people (e.g., attitude and personality models) is discussed. Have a pleasant summer. No meeting in August. See you in September! ____________________ Send correspondence for this newsletter to: Software Psychology Society, c/o Skip Williamson, Knowledge Systems, Inc., 5705 Stillwell Rd., Rockville, MD 20851. ------------------------------ Date: Fri, 16 May 86 01:03:50 edt From: walker@MOUTON.ARPA Subject: IJCAI-89 Site Selection and Officer Election IJCAI-89 Site Selection and Officer Election The Trustees of the International Joint Conferences on Artificial Intelligence, Inc. are pleased to announce that IJCAI-89 will be held 20-26 August 1989 in Detroit, Michigan, USA. Wolfgang Bibel, Technical University of Munich, has been elected Conference Chair; Sri Sridharan, BBN Laboratories, has been elected Program Chair; and Sam Uthurusamy of General Motors Research Laboratories has been appointed to chair the Local Arrangements Committee. Don Walker, Bell Communications Research, the IJCAII Secretary-Treasurer, will also serve as Secretary-Treasurer for the conference. IJCAI-89 will be cosponsored by the American Association for Artificial Intelligence. All conference activities will be coordinated through the AAAI Office by Claudia Mazzetti, Executive Director of the AAAI, who will provide direct support for the IJCAI-89 Conference Committee. In accordance with customary practice for IJCAI conferences held in North America, the AAAI will also arrange the tutorial and exhibit programs at the meeting. For further information, contact one of the following: Wolfgang Bibel (IJCAI-89) Institut fuer Informatik Technische Universitaet Muenchen Postfach 202420 D-8000 Muenchen 2, West Germany Telephone: (49-89)2105-2031 bibel%germany.csnet@csnet-relay N. S. Sridharan (IJCAI-89) BBN Laboratories 10 Moulton Street Cambridge, MA 02238 Telephone: (1-617)497-3366 sridharan@bbng.arpa R. Uthurusamy (IJCAI-89) Computer Science Department General Motors Research Laboratories Warren, MI 48090, USA Telephone: (1-313)575-3177 samy%gmr.csnet@csnet-relay Donald E. Walker (IJCAI-89) Bell Communications Research 445 South Street MRE 2A379 Morristown, NJ 07960, USA Telephone: (1-201)829-4312 walker@mouton.arpa Claudia Mazzetti (IJCAI-89) AAAI Headquarters 445 Burgess Drive Menlo Park, CA 94025 Telephone: (1-415)328-3123 aaai-office@sumex-aim.arpa ------------------------------ Date: Wed, 14 May 86 01:09:40 edt From: EMMA@SU-CSLI.ARPA Subject: Van Nguyen talk From: Margaret Olender DATE: May 14, 1986 LOCATION: SRI International, Ravenswood Avenue, Building E KNOWLEDGE, COMMUNICATION, AND TIME Van Nguyen IBM Thomas J. Watson Research Center (Joint work with Kenneth J. Perry) Abstract The role that knowledge plays in distributed systems has come under much study recently. In this talk, we re-examine the commonly accepted definition of knowledge and examine how appropriate it is for distributed computing. Motivated by the draw-backs thus exposed, we propose an alternative definition that we believe to be better suited to the task. This definition handles multiple knowers and makes explicit the connection between knowledge, communication, and time. It also emphasizes the fact that knowledge is a function of one's initial knowledge, communication history and deductive abilities. The need for assuming perfect reasoning is mitigated. Having formalized these links, we then present the first proof system for programs that incorporates both knowledge and time. The proof system is compositional, sound and relatively complete, and is an extension of the Nguyen-Demers-Gries-Owicki temporal proof system for processes. Suprisingly, it does not require proofs of non-interference (as first defined by Owicki-Gries). ------------------------------ Date: Wed, 14 May 86 01:09:44 edt From: DEJONG%OZ.AI.MIT.EDU@MC.LCS.MIT.EDU Subject: Cognitive Science Calendar [Extract - Ed] From: Robert H. Kassel Subject: Speech Recognition Wednesday, 14 May 3:00pm Room: 34-401B Large Vocabulary Speech Recognition at IBM: An Expert Approach Fred Jelinek IBM T.J. Watson Research Center Yorktown Heights, NY Demonstration after the talk, Room 36-428 From: JHC at OZ.AI.MIT.EDU Thursday, 15 May 4:00pm Room: NE43- 8th floor playroom -- AI Revolving Seminar -- LEARNING TO CONSTRUCT ABSTRACTIONS Rick Lathrop MIT AI Lab One useful trait of an intelligent agent is to construct higher-level abstractions from a mass of detailed low-level information. This talk will explore one way an agent might be taught how to construct such abstractions, and why it might be a useful or interesting for an agent to do so. A main motivation is the possibility of the use of these abstractions to see similarities (between situations) that are obscured by the mass of irrelevant details at the lower level. Preliminary examples from the Rieger (causal) mechanism world, VLSI circuit analysis, and protein structure analysis will be discussed. ------------------------------ END OF IRList Digest ********************