Copy Link
Add to Bookmark
Report
AIList Digest Volume 6 Issue 062
AIList Digest Friday, 1 Apr 1988 Volume 6 : Issue 62
Today's Topics:
Queries - LISP on VMS & AI in CAD & Expert Systems in Education &
Expert System Project Management Course &
Commercial distribution of AutoLISP Applications,
Bindings - McRobbie & Ralph London,
Application - Conversational Programs: RACTER,
Philosophy - Formal Semantics and Computational Models,
Expert Systems - Automatic Knowledge Extraction,
AI Tools - Circuit-Design Translator in Prolog/Lisp &
Student Versions of OPS5
----------------------------------------------------------------------
Date: 29 Mar 88 19:05:55 GMT
From: "Eric T. Freeman" <nasa@e.ms.uky.edu>
Subject: LISP on VMS
NOTE: The following is posted for a friend, you may respond to either his
address or mine, I will forward responses to him.
******************************************************************************
WANTED--A public domain (or very inexpensive) copy of LISP for the VAX/VMS
(not Unix). Must have compiler. Must have someone to answer questions.
franz lisp would be fine. Send mail to jones@dftnic.gsfc.nasa.gov
Thanks,
Tom Jones jones@dftnic.gsfc.nasa.gov
******************************************************************************
Eric Freeman
University of Kentucky Computer Science
nasa@g.ms.uky.edu
freeman@dftnic.gsfc.nasa.gov
------------------------------
Date: 30 Mar 88 10:15:45 GMT
From: csli!rustcat@labrea.stanford.edu (Vallury Prabhakar)
Subject: Re: need info on AI in CAD
In article <332@infko.UUCP> uro@infko.UUCP (Uwe und Roland) writes:
# In doing research on intelligent CAD-systems, I'm looking
# for literature, technical/research reports and conference
# proceedings on following topics:
#
# - Architecture of intelligent CAD-systems
# - Design and Implementation of AI-models which enable
# the user to include structural knowledge (i.e. design knowledge),
# exceeding topological knowledge while describing his objects to the
# CAD-system. I'm looking for a kind of shell, surrounding the realization
# of the topological/geometrical model, but beeing as independent
# as possible from the latter one.
I don't know if it's a bit late, but could you kindly forward pertinent
responses to the above portions of your original query to me?
On a related note, I've been looking for literature/publications relevant
to topics such as pattern recognition in a CAD "drawing", such as determining
line connectivies, and figuring out whether a series of lines form a closed
path, etc. I would appreciate any information/pointers on this.
If this is the wrong newsgroup for such a question, I apologize. I've
tried comp.graphics but wasn't able to get much from there.
Enjoy,
-- Vallury Prabhakar
rustcat@cnc-sun.stanford.edu
PS: Please reply directly to my e-mail address. I don't come here very often.
[You might want to look into the work of Deborah Walters (of SUNY)
on perceptual groupings of edge elements and lines. Check the
computer vision conferences for her papers. The work finds lines
that seem to belong together, rather than those that match some
stored shape template. -- KIL]
------------------------------
Date: Thu, 31 Mar 88 09:26 EST
From: ARMAND%BCVMS.BITNET@MITVMA.MIT.EDU
Subject: STATUS OF EXPERT SYSTEMS?
A committee here at Boston College is presently investigating the state of the
art of Expert Systems in higher education. We have determined several
situations on campus which would benefit by this technology.
We would very much appreciate any information and/or direction regarding the
following queries.
1) What is the current interest level of faculty, staff, and students in
applications of this techology?
2) What instructional applications of expert systems are already in existence?
What type of development is going on in this area?
3) What level of success, if any, has resulted from internally developed expert
systems?
4) What hardware, software, and other resources are currently in use? Future
planning?
5) What problems should be foreseeable and addressed in planning and developing
an Expert Systems laboratory?
Please feel free to share your experiences/stories as we have very little to go
on at this point. We would appreciate very much ANY response to these
questions. If you could spare a few moments for a phone call, that would be
great - call collect if you wish.
Armand H. Doucette
VAX Systems Programmer
Boston College Computer Center
Chestnut Hill, MA 02167
(617) 552-4463
BITNET: ARMAND@BCVMS
------------------------------
Date: Thu, 31 Mar 88 10:48:40 PST
From: bienk@spam.istc.sri.com (Marie Bienkowski)
Subject: Expert System Project Management Course
Can anyone point me to a course on managing the development
of an expert system? Something that an AI company or such
would offer, introducing expert systems and suggesting ways
for efficient management of projects involving their development.
Thanks in advance,
Marie Bienkowski
(bienk@istc.sri.com)
------------------------------
Date: Thu 31 Mar 88 18:00:02-EST
From: Ben Olasov <G.OLASOV@CS.COLUMBIA.EDU>
Subject: Commercial distribution of AutoLISP applications
Hello,
I'd like to hear from any readers who know of commercial distributors/
clearinghouses of AutoLISP code (code for the integrated LISP interpreter
in the AutoCAD CAD package).
Please send responses to this account: G.OLASOV@CS.COLUMBIA.EDU
Thank you,
Ben
------------------------------
Date: Thu, 31 Mar 88 07:23:42 CST
From: lusk%antares@anl-mcs.arpa
Subject: address needed
Here is everything I know about Zorba: (well, not quite everything)
alias mcrobbie mam%arp.anu.oz@uunet.uu.net
Michael McRobbie 61-62-492035 \
Dr. Michael McRobbie, Automated Reasoning Project, RSSS, \
Australian National University, G.P.O. Box 4, Canberra, 2601, A.C.T.
------------------------------
Date: 30 Mar 88 11:54:52 GMT
From: Gilbert Cockton <mcvax!hci.hw.ac.uk!gilbert@uunet.uu.net>
Reply-to: Gilbert Cockton <mcvax!hci.hw.ac.uk!gilbert@uunet.uu.net>
Subject: Re: Request for Reference on Auto User Interface Pgm
In article <8803100728.AA07906@ucbvax.Berkeley.EDU> TAYLOR%PLU@ames-io.ARPA
writes:
>Sometime during the summer of 1987, there was a seminar given at either
>Stanford or SRI by someone maybe named Jack London concerning his
>PhD thesis on the subject of automatic generation of user interface
>specifications and/or code based on the specifications/contents of
>a given data base. Obviously this is vague - it is second hand
>information. Can someone give me an accurate reference to the
>work presented in the seminar?
Sounds like Ralph London from Tektronix. See London and Duisberg in
ACM Trans on Graphics 1986. Sorry no more details, moving office and
can't reach for the paper.
--
Gilbert Cockton, Scottish HCI Centre, Heriot-Watt University, Chambers St.,
Edinburgh, EH1 1HX. JANET: gilbert@uk.ac.hw.hci
ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hci!gilbert
------------------------------
Date: 30 Mar 88 10:54 +0100
From: Kai Quale <quale%si.uninett@TOR.nta.no>
Subject: Re: conversations?
ok2@psuvm.bitnet writes:
> Hi, I'm trying to locate versions of PARRY and similar programs that
>simulate conversation with a human being (such as ELIZA and SHRDLU).
> I do know that PARRY, which simulates conversing with a Paranoid person
>was written by K.M. Colby, based on the information processing approach
>(of a paranoiac) of S.S. Tomkins.
> Also, I know that SHRDLU, which simulates talking to and giving orders
>to a robot in a computer generated simulation, was written by Terry Winograd.
>[Rest deleted]
RACTER is another conversation program which can be run in
monologue (Essay, Poetry, Free prose, ...) or dialogue mode. I don't know
what kind of hardware it runs on; I have only read a book written by the
program : "The Policeman's Beard is Half Constructed". The book is packed
with poems and philosophical discussions between RACTER and its programmers,
and RACTER and itself. The real gem is a conversation between RACTER and ELIZA.
You wouldn't believe how quickly ELIZA is pressed into a corner, and starts
responding to everything with "Interesting" and "Please go on". The following
passage does not exist in the book (it's a long time since I read it), but may
serve as a taste of the general atmosphere :
R : I wonder intensely about the meaningfulness of a hamburger.
E : Have you felt this for a long time ?
R : Yes, it inhabits my mind like a clawed hippopotamus, never letting go.
E : I see.
R : Have you ever thought about the fundamental similarity of Bach and the
color pink ?
E : Is this really what you want to see me about ?
R : Why do you ask such a prosaic question ? Don't you feel the organic
wholeness of the world ? The singing of the celestial spheres ? Ah,
what a state the world is coming to.
E : Please go on.
etc.
Kai Quale <quale%si.uninett@tor.nta.no>
------------------------------
Date: 29 Mar 88 16:17:32 GMT
From: Gilbert Cockton <mcvax!hci.hw.ac.uk!gilbert@uunet.uu.net>
Reply-to: Gilbert Cockton <mcvax!hci.hw.ac.uk!gilbert@uunet.uu.net>
Subject: Re: Question on arguments against AI
In article <8803051150.AA05897@ucbvax.Berkeley.EDU> ADLASSNI@AWIIMC11.BITNET
("Adlassnig, Peter") writes:
>
>Is it true that there are two main arguments against the feasibility
>of AI?
> ....
Forget categories for the moment and come bottom up.
Within formal semantics there are a whole set of problems which reduce
confidence in the comprehensiveness of computational models of human
beliefs and behaviour.
Formal semantics is largely AI off-line, and has an intellectual and
scholarly tradition which pre-dates the LISP bar of AI. I suggest you
pick up the Cambridge University Press catalogue and chase up any
Linguistics text with 'semantics' in the title. Most of these monographs
and texts have consensus examples of problems for mathematical accounts
of meaning, especially ones based on two-valued logics. Everyone in
NLP should know about them. Basically, AI won't succeed until it
cracks these problems, and there is no reason to believe that they
will ever get anywhere near cracking them. The gap between
mathematical accounts and reality remains too large.
--
Gilbert Cockton, Scottish HCI Centre, Heriot-Watt University, Chambers St.,
Edinburgh, EH1 1HX. JANET: gilbert@uk.ac.hw.hci
ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hci!gilbert
------------------------------
Date: 29 Mar 88 18:10:52 GMT
From: mcvax!unido!tub!tmpmbx!netmbx!morus@uunet.uu.net (Thomas M.)
Subject: Re: automatic knowledge extraction
In article <8803250634.AA22441@ucbvax.Berkeley.EDU> SYSTHVU@BLEKUL11.BITNET
(Van Uytven Herman) writes:
>
>I'm interested in systems for automating the knowledge extraction process.
>In the literature there's a lot of information about these systems.
>I'm looking for some references of commercially available products.
>Can anyone provide me this information ?
>Is there anyone who uses these systems, or has some experience with them ?
>I'd be very grateful if you could send your comments to me.
>Thanks in advance,
>
>Chris Vanhoutte
I think that this topic might be of interest for some in the ai-community.
Please post responses to one of the relevant newsgroups. I am just doing
a study about knowledge elicitation methods - mainly using the articles
of the Knowledge Acquisition for Knowledge Based Systems Workshop, Banff,
Canada, 1986 and 1987. My main interest lie in psychological aspects, the
"rebirthing" of common theories and tools which emerged from psychotherapy
and diagnostics like structered interviews, repertory grids, MDS and the like.
My personal opinion is, that the ideal knowledge engineer has to be a kind of
psychotherapist and well trained in general and social psychology more than
a technician trained in representation methods and using appropriate tools.
A lot of commercially available tools for building XPS cannot account for the
psychological and philosophical problems which emerge with such a difficile
topic as knowledge elicitation. If you read adverts for those tools you could
get the impression that these problems are already solved - just fill in a
couple of so called examples and the tool will create a knowledge base for
you. Experts - be aware!
--- Thomas Muhr
PS.: What about elicitation by hypnosis ? !-)
--
@(^o^)@ @(^x^)@ @(^.^)@ @(^_^)@ @(*o*)@ @('o`)@ @(`!')@ @(^o^)@
@ Thomas Muhr Tel.: (Germany) 030 - 87 41 62 (voice!) @
@ NET-ADRESS: muhrth@db0tui11.bitnet or morus@netmbx.UUCP @
@ BTX: 030874162 @
------------------------------
Date: Thu, 31 Mar 88 07:28:53 CST
From: lusk%antares@anl-mcs.arpa
Subject: translators in Prolog/Lisp
You might try Peter Rentjes (sp?) somewhere in North Carolina. (Sorry I can't
be more specific) He has a large circuit design language translator written
in Prolog, parts of which were released into the public domain at the recent
Prolog benchmarking workshop in Los Angeles. For Peter's address you might
try Rick Stevens (stevens@anl-mcs.arpa).
------------------------------
Date: 30 Mar 88 16:08:05 GMT
From: trwrb!aero!srt@ucbvax.Berkeley.EDU (Scott R. Turner)
Subject: Re: Student versions of OPS5
In article <1580@netmbx.UUCP> morus@netmbx.UUCP (Thomas Muhr) writes:
>In article <27336@aero.ARPA> srt@aero.UUCP (Scott R. Turner) writes:
>>It should be possible to write an OPS5-like language in a lot less than
>>100K. The only difficult part of OPS5 to implement is the RETE algorithm.
>
>Right, but after all the proposed deletions this code would hardly cover 2
>pages. Leaving Rete-match out is not just affecting run-time (the decrease
>in performance is incredible!) - but it would eliminate all features which
>make OPS5 an interesting language - mainly the heuristics for selecting
>rule-instantiations intelligently.
Actually, I think specificity and recency are fairly easy to
implement. (Without using Rete, that is...) Specificity simply means
you have to keep your rule database ordered. That's simple enough to
do. Recency is a little bit harder. The "obvious" execution cycle is
to fire the first available rule. To do recency, you have to keep the
fact database ordered by recency (easy) and then find the first rule
that fires on the first fact (and so on down the fact database).
That's not particularly hard to do. I suspect OPS5 (with specificity
and recency) could be written in a couple of pages of Prolog.
The point is, if you just want a student version of OPS5 - something
to play around with - then you don't need all the Rete trappings.
They can be easily (and inefficiently) duplicated.
>>(*) My experience is that most OPS5 programmers (not that there are many)
> Is this right ? ---^^^^^^^^^^^^^^^^^^^^^^
>>ignore or actively counter the "pick the most specific/least recently used"
>>rules anyway.
My guess is that there are very few active OPS5 programmers out there.
For the most part I would say it is a dead language. It is years out
of date (in terms of representation power, etc.), has an awkward
syntax, and promotes a rather strained coding style. The fact that
there are only two or three people contributing to this topic should
give you some idea of how popular it is in regards to the net.
-- Scott Turner
------------------------------
End of AIList Digest
********************
