LAWS@SRI-AI.ARPA (03/01/85)
From: AIList Moderator Kenneth Laws <AIList-REQUEST@SRI-AI.ARPA>
AIList Digest Friday, 1 Mar 1985 Volume 3 : Issue 27
Today's Topics:
AI Tools - XLISP & Unification & Rete Networks & DUCK,
Applications - CAI & Commonsense Reasoning & Paradox,
Representation - Analytic Solid Modeling,
Bindings - General Research Corp.,
Correction - Webster service,
Linguistics - Wally & Y'alls,
Business - R&D Opportunity & Medicomp AI/DB System,
Seminars - Direct Manipulation Interfaces (CMU) &
Counterfactuals (SU)
----------------------------------------------------------------------
Date: Mon, 25 Feb 85 15:56:24 EST
From: David_Michael_Fobare%Wayne-MTS%UMich-MTS.Mailnet@MIT-MULTICS.ARPA
Subject: XLISP Questions
I am working on a version of LISP called XLISP and am wondering
if anyone else has worked with it. In particular I am having trouble
with the 'Object', 'Class', and 'Keymap' primitives. What are instance
variables? Class variables? Thanks.
David Fobare
------------------------------
Date: Tue 26 Feb 85 10:20:30-MST
From: Pete Tinker <tinker@UTAH-20.ARPA>
Subject: Production Systems/Unification
Are there any production systems (backward- or forward-chaining) which
use unification rather than simple pattern matches to bind variables?
-Pete Tinker (TINKER@UTAH-20)
------------------------------
Date: 27 Feb 85 12:49:42 PST (Wed)
From: Don Rose <drose@uci-icse>
Subject: rete networks
Hello everyone - I am currently familiarizing myself with the subject of
rete networks (esp. with regards to a possible synthesis with current
data-dependency frameworks for truth maintenance); if anyone can point the
way to relevant/interesting articles along these lines (or just on rete
networks alone), I would appreciate your help. Thanks --Donald Rose
drose@uci-icse
------------------------------
Date: Wed, 27 Feb 85 17:57:23 PST
From: Koenraad Lecot <koen@UCLA-LOCUS.ARPA>
Subject: DUCK
Can anybody provide me with references to the DUCK language? I know Smart
Systems Technologies is selling DUCK but I imagine they'll charge for
manuals (smart).
Thanks in advance,
-- Koenraad Lecot
ARPA: koen@ucla-cs.arpa
UUCP: ...!{cepu,ihnp4,randvax,sdcrdcf,trwspp,ucbvax}!ucla-cs!koen
------------------------------
Date: Wed, 27 Feb 85 15:09:44 est
From: Deepak Kumar <kumard%buffalo.csnet@csnet-relay.arpa>
Subject: Info wanted - Expert Systems for CAI
I'm taking a course on AI in Medicine where in I have decided
to present a seminar on Intelligent Computer Assisted Instruction
for Medical Diagnosis. I want to emphasize use of already-
existing Expert Systems for CAI applications. After all, when the
knowledge is there, it can be utilized for purposes other than
advising and decision making.
If there are any references or materials on the above domain
I'd like to have it. It need not be specifically to the
area of Medicine. So, if anyone has made any attempts on using
Expert Systems in Education, I'd like to get some experiences
that I can share with my audience.
Thanx.
Deepak Kumar (Dept. of CS, SUNY at Buffalo)
------------------------------
Date: Wed, 27 Feb 85 17:16:42 est
From: arora@buffalo (Kulbir S. Arora)
Subject: Request for bibliography
Is there a bibliography available on Common-sense reasoning systems
(qualitative reasoning, mental models) ?
Kulbir Arora
------------------------------
Date: 27 Feb 1985 1559-PST (Wednesday)
From: Eugene Miya <eugene@AMES-NAS.ARPA>
Subject: info requested -- paradox
Newsgroups: net.ai
Distribution: net
This is for someone else. Send replies to me.
Does anybody have pointers to structures and techniques for dealing with
paradox or apparent contradiction [e.g., light is a wave, light is a
particle]? one suggestion made was to use analogy, but we have a
copy of the "analogy considered harmful" paper by some people at parc.
thanks in advance.
--eugene miya
NASA Ames Research Center
{hplabs,ihnp4,dual,hao,vortex}!ames!aurora!eugene
emiya@ames-vmsb.ARPA
p.s. usenet responses turned up one ref to work at mit on contradiction.
------------------------------
Date: 26 February 1985 23:16-EST
From: John G. Aspinall <JGA @ MIT-MC>
Subject: Recent Article - representation of solids
In a recent AI digest, I noticed this summary of a paper from IEEE Computer
Graphics and Applications, February 1985:
From: Ken Laws <Laws@SRI-AI.ARPA>
An Overview of Analytic Solid Modeling, by M. Casale and E. Stanton
of PDA Engineering, pp. 45-56:
Describes a way of modeling complex shapes as ...
I'm very interested by these sorts of systems, and have been working with
splines for some time now, so I was quite interested in the paper. If any
reader has deeper knowledge of the system described I would appreciate some
enlightenment about a query of mine. I have read the paper thoroughly, and I
don't think it answers this.
One thing I did not understand was the following: They made a big deal out of
the necessity to represent circles (surfaces of rotation, holes, cylinders,
etc.) which I agree with -- a round object is easily produced in many ways
in the real world (lathes, drills, etc.). But there is no way to exactly
represent a circle as a polynomial of a parametric variable.
That is if x = P(u) and y = Q(u), where P and Q are polynomials, there is no
way to let x^2 + y^2 = r^2 (a constant, independent of u) except the trivial
point solution x = constant, y = constant.
So how do they represent circles? Possibility 1: they represent them
approximately with splines. Problem: suppose you want to model another
cylinder rotating in that hole. It's quite conceivable that the clearances
involved might be 10^-3 or less of the diameter. That implies an awful lot of
spline-knots. Possibility 2: Circles (and by extension, cylinders etc.) exist
as exact entities in their system. Problem: how do you join that to a spline
solid, for instance when you put a hole in something?
If you have any ideas about this, I'd be very interested in hearing them.
Many thanks for any light you might shed. Oh yes, please reply to me directly.
John Aspinall.
------------------------------
Date: Wed 27 Feb 85 14:36:52-CST
From: CMP.BARC@UTEXAS-20.ARPA
Subject: General Research Corp. Query
General Research has two locations:
P.O. Box 6770 7655 Old Springhouse Road
Santa Barbara, CA 93160-6770 McLean, VA 22102
(805) 964-7724 (703) 893-5915
TWX-TELEX 910-334-1193
Dallas Webster
------------------------------
Date: Thu 28 Feb 85 15:22:16-PST
From: David Roode <ROODE@SRI-NIC.ARPA>
Subject: Webster service
In Volume 3 issues 17 and 23 of this digest, mention was made
of an experimental service being developed here at SRI-NIC.
The articles gave the false impression that this was a generally
available, production service. This is in fact not the case.
We have to iron out several issues before the service can be made
public, and an announcement will be made when and if the
online dictionary is available.
[My apologies for picking up the Rutgers bboard announcement and
help file for the WEBSTER program without checking into the
program's availability. It seems there may be copyright or
related issues in addition to software-related ones. -- KIL]
------------------------------
Date: Wed, 27 Feb 85 13:16:03 EST
From: Pete Bradford (CSD UK) <bradford@AMSAA.ARPA>
Subject: Re: Wally
As a `Brit', now resident in the US for some 3 years,
and an avid fan of the `Grauniad' newspaper, let me try to
answer Gordon Joly's question. To the best of my knowledge
the word `wally' does NOT exist in the United States. It app-
ears in Websters Dictionary as a Scottish word meaning `fine,
splendid, or sturdy', a long way from its modern day British
use! There is also a word `wallydraigle', or `wallydrag'(!),
also Scottish, which means `a feeble, undergrown or slovenly
creature', perhaps a lot closer to the `wally' we all know --
at least, ONE of the meanings we know!
The nearest equivalents to `wally' over here, in the
sense of a stupid or foolish person, are `jerk',or `nerd'. As
for the other meanings, well, the one meaning that which you
would buy in a Fish and Chip shop is a `dill pickle', or just
plain `pickle'. The other meaning, ie an organ which I, along
with approximately half the human race, possess, is covered
by as many terms in the US as it is in the UK - but NOT wally.
Hope this helps!
PJB
------------------------------
Date: Tue, 26 Feb 85 13:43:58 est
From: "Thomas E. Schutz" <tes%bostonu.csnet@csnet-relay.arpa>
Subject: Y'alls
Some people in Richmond, Virginia have evolved their own forms
for the singular and plural "you" :
Singular - "y'all"
Plural - "y'alls"
I discovered this while visiting Richmond over summer vacation.
A native of the town addressed me as "y'all"; perplexed, I turned
around but did not see anyone else standing there. Later on
I overheard the word "y'alls," and then everything fell into place.
- Tom Schutz
------------------------------
Date: Tue, 26 Feb 85 16:04 EST
From: White@RADC-MULTICS.ARPA
Subject: R&D Opportunity
An announcement will soon appear (estimated 7 - 14 March 85) in the
Commerce Business Daily (CBD) advertizing a research and development
(R&D) program to be undertaken by the Rome Air Development Center
(RADC). This advertisement will be in the form of a Program Research
and Development Announcement (PRDA) which is a notice which provides
information about RADC's interest in R&D in a specific area where the
method for achieving the goal is not evident. This particular PRDA
will contain many different areas of interest. Some will be of
interest to the AI community such as the application of AI technology
in software development environments. The notice you are now reading
is not a solicitation and is merely a "heads up" to encourage
participation. Responses and/or communications are not to be
directed to the address appearing on this notice and should be made
in accordance with directions contained in the CBD announcement.
------------------------------
Date: Mon, 25 Feb 85 05:50:10 cst
From: Laurence Leff <leff%smu.csnet@csnet-relay.arpa>
Subject: Medicomp AI/DB System
Electronics Week February 18, 1985 Page 59
Medicomp of Virginia Inc. is now selling a combination medical records
/artificial intelligence system. The system runs on an IBM Series/1
minicomputer with a dabase covering 3000 diseases, 1000 symptoms and 1000
medical tests. The system had its first sale to Cleveland Clinic. They
purchased it when it solved a baffling case of infectious endocarditis.
It costs $55,000.
They have been maintaining a free telephone hookup to the system for
demonstration purposes.
------------------------------
Date: 25 February 1985 1351-EST
From: Jeff Shrager@CMU-CS-A
Subject: Seminar - Direct Manipulation Interfaces (CMU)
[Forwarded from the CMU bboard by Laws@SRI-AI.]
Artificial Intelligence Lecture Series
COGNITIVE ENGINEERING of DIRECT MANIPULATION INTERFACES
Don Norman
Institute for Cognitive Science
University of California, San Diego
La Jolla CA 92093
2:00 PM, Friday, March 1
Westinghouse R & D Center Auditorium - Bldg 401
Abstract
To program by Direct Manipulation is to program by connecting together
data streams, drawing the interconnections on the screen and letting
the resulting diagram be the program. Direct Manipulation is made
possible by the combination of object-oriented programming languages
(as available on Lisp machines), high-resolution bit-mapped graphics,
and pointing devices. It promises superior performance for that class
of problems that can be characterized in this manner. It allows
concentration upon the semantics of the task rather than the structure
of the programming language. At least, so goes the rhetoric.
Examination of the nature of Direct Manipulation leads naturally to an
examination of the major issues in interface design. It turns out that
one of the major virtues of Direct Manipulation is the feeling of
control that it produces, the feeling of directly manipulating the
computational objects one cares about. This feeling depends as much
upon the skill of the user as the nature of the interface. We meet the
Gulfs of Execution and of Evaluation, gulfs that Direct Manipulation
interfaces promise to bridge more readily than conventional interfaces.
Will Direct manipulation Interfaces live up to their promise? Yes and
no.
Directions
Head east on the parkway, I-376, to the Churchill exit (13). As you
come off, keep bearing right for a few hundred yards to the first
traffic light. Proceed straight through it, across Beulah Road, and
up the R&D entrance road where the guard will point out visitors
parking and reception. You don't need to call in advance; there will
be a few technicians to escort visitors to the auditorium.
In addition to the Westinghouse talk, Don Norman will be speaking in the
Psychology Colloquium Series this Thursday, Feb 28, at 4:00 in the
Adamson Wing (BH). The talk is entitled: "From slips and mistakes to
a theory of action".
------------------------------
Date: Tue 26 Feb 85 09:39:50-PST
From: Paula Edmisten <Edmisten@SUMEX-AIM.ARPA>
Subject: Seminar - Counterfactuals (SU)
[Forwarded from the Stanford SIGLUNCH distribution by Laws@SRI-AI.]
DATE: Friday, March 1, 1985
LOCATION: Chemistry Gazebo, between Physical and Organic Chemistry
TIME: 12:05
SPEAKER: Matthew Ginsberg, Computer Science Department and
Heuristic Programming Project, Stanford University
ABSTRACT: COUNTERFACTUALS
Counterfactuals are a form of commonsense non-montonic inference that
has been of long-term interest to philosophers. In this paper, we
begin by describing some of the impact counterfactuals can be expected
to have in artificial intelligence, and by reviewing briefly some of
the philosophical conclusions which have been drawn about them.
Philosophers have noted that the content of any particular
counterfactual is in part context-dependent; we present a formal
description of counterfactuals that allows us to encode this
context-dependent information clearly in the choice of a sublanguage
of the logical language in which we are working. Having made this
choice, we show that our description of counterfactuals is formally
identical to the accepted "possible worlds" interpretation due to
David Lewis. Finally, we examine the application of our ideas in the
domain of automated diagnosis of hardware faults.
Paula
------------------------------
End of AIList Digest
********************