Copy Link
Add to Bookmark
Report
AIList Digest Volume 2 Issue 046
AIList Digest Friday, 13 Apr 1984 Volume 2 : Issue 46
Today's Topics:
Education - LISP Advice Sought,
AI Jobs - Noncompetition Clauses,
Natural Language - Metaphor,
Humor - Smalltalk-52 Seminar
----------------------------------------------------------------------
Date: Wednesday, 11 April 1984 23:08:30 EST
From: Kai-Fu.Lee@cmu-cs-g.arpa
Subject: Comments solicited
I will be teaching a LISP/AI course to a group of Pennsylvania high
school seniors who are talented in science. (part of the Penn.
Governor's School for the Sciences) I am interested in hearing from
people with similar experience or ideas for subject/assingments/
projects/text book.
The students will likely be divided into two groups, one with and one
without experience in programming. The course consists of 17 1-hour
lectures to each group. I am planning to divide the course into 3 parts :
(1) Basic Concepts of C.S. [2 lectures] (2) LISP Programming [6-7
lectures] (3) A.I. [8-9 lectures]. Since the schedule is rather tight,
it is unlikely that I could cover anything in much detail. There will be
two or three programming assignments. In addition, students may choose to
do a project in computer science.
Thanks,
Kai-Fu Lee
KFL@CMU-CS-G.ARPA
------------------------------
Date: Thu 12 Apr 84 00:36:19-PST
From: Mabry Tyson <Tyson@SRI-AI.ARPA>
Subject: Re: Noncompetition Clauses
Ken, I think you missed one problem with the noncompetition agreement.
I guess you looked at it like I did that your salary would be in the
same range if you could change jobs so it doesn't hurt too much if you
just continued on at the old salary but took the time off.
Suppose you started to work for company A at salary X. The next year (or so)
you get a much better offer (say as a manager) from company B at salary 2*X.
Now you are prevented from taking that job (assuming B is in competition with
A).
Another way of looking at that is to suppose that you do something good
in your first few years after school but that your company doesn't
want to give you the raise in salary that is commensurate with your proven
abilities. Now you can't just say that company B will pay you twice your
current salary. They'd just laugh at you and say you couldn't go. It might
be worth 3 years of your old salary to keep you from company B even if you
don't do any work for them.
I see the clause as cutting down on wage wars between companies. It also
cuts down on the mobility available to employees of that company. Finally
it probably prevents an employee from starting his own company in that
field.
I also feel that the restriction may have a negative effect on the company
requiring it. Would you go to a black hole from which no one ever could
get away? I suppose companies requiring that clause are just going to have
to settle for employees that can't find a better offer. Would you want to
work with second class people?
------------------------------
Date: 12 Apr 84 08:56:18 PST (Thu)
From: Carl Kaun <ckaun@aids-unix>
Subject: Non-competition clauses and other employment agreements
I think that anyone concerned about employment agreements would do well
to contact a lawyer. The different states consider various clauses in
employment agreements enforceable to differing degrees. I seem to remember
reading an article about six months ago that said (broadly) that California,
for example, considered clauses restricting people from continuing their
professional careers to be not generally in the public interest, and such
clauses should be carefully considered as to enforceability for that reason.
This same article (which I am trying to dig up) said something about the state
of California holding that clauses assigning all rights to all ideas, patents,
etc. arising during employment, are enforceable only to the extent that such
ideas, etc. resulted from the employment situation. Again, one should
really contact a lawyer to get a clear opinion in any given situation. There
is enough variability in this area to make any general comments suspect.
My personal experience has been that employment agreements and the
employer's approach to them are remarkably uniform in industry. When I
have questioned companies about these agreements, I am told that they
adopt this fairly standard agreement and hold to it firmly on the advice
of their corporate counsel. The general idea (from the company point of
view) seems to be to claim all you can now, and work out what you can really
enforce if the situation comes to that. They take this approach
not because of some malignant motive, but because it has proved the most
prudent course for them to take.
------------------------------
Date: Thu, 12 Apr 84 17:51:28 EST
From: Mark S. Day <mday@BBN-UNIX>
Subject: Re: Employment Agreement
Keeping someone entirely out of a field like AI for three years is an illegal
restraint of trade, I would suggest. Employment agreements must be
reasonable with respect to time and area limitations to be enforceable, and
I doubt that 3 years is a reasonable time constraint, especially given that it
seems to be sufficiently long to get completely out of touch with the field.
The fact that the company offers to pay you for those three years is
irrelevant.
--Mark
------------------------------
Date: Thu, 12 Apr 84 07:06:04 cst
From: Peter Chen <chen%lsu.csnet@csnet-relay.arpa>
Subject: Noncompetition Clauses
I think that there are quite a few companies putting on restrictions on
post-employment activities, although most of these companies are usually not as
restrictive as the company mentioned by Scott Fahlman. I think it
is fair for an employer to ask its employees to avoid future
involvement in direct competition with the company
within a short period of time (say, one year instead
of three years) and in a more narrow subject area (i.e., in the area/topics
the individual is working on, rather than a broad definition of the AI field or
the whole computer field).
If I remembered correctly, when I worked for a large computer manufacturer ten
years ago, I was required to sign an agreement that whatever
ideas or products I might develop in my spare time would belong to the company
even though the ideas/products were not related to computers. Do you think it
is fair? Do you think your computer employer has the right of the novel you
write during weekends? I think this case is much more unfair than asking
the employee not to compete with the company after he/she leaves the company
for more than a year.
As far as I understand, all these agreements/contracts are legally binding if
the contracts are signed under free will. Therefore, they can be enforced if
the companies choose to do so. However, most of time the companies just use
them as a possible protection for their interest.
Peter Chen
Computer Science Dept., LSU
<chen%lsu@csnet-relay>
------------------------------
Date: Wed, 11 Apr 84 17:45:10 PST
From: Charlie Crummer <crummer@AEROSPACE>
Subject: Metaphor
The thing about a metaphor is that it contains little explicit information.
It acts as a trigger in such a way that the hearer creates meaning for it.
Different hearers create different meanings. For example, one hearer,
drawing from his background as an environmentalist might take "Man is a Wolf"
to mean that man has a wild, misunderstood soul while another hearer, drawing
from his background as a mountain man who has had to compete with wolves might
take the metaphor to mean that he himself is a savage beast that will kill,
if necessary, to live.
It becomes pretty far-fetched to make up a model of "metaphor" that says
that this information is contained in the statement of the metaphor.
--Charlie
------------------------------
Date: 11 Apr 84 2255 EST (Wednesday)
From: Steven.Minton@CMU-CS-A.ARPA
Subject: Metaphor comprehension pointers
The following references might prove helpful if you're interested in
AI and metaphor comprehension:
Carbonell, J.G. and Minton, S. "Metaphor and Common-Sense Reasoning"
CMU tech report CMU-CS-83-110, March 83
Carbonell, J.G. "Metaphor: An Inescapable Phenomenon in Natural Language
Processing", in Strategies for Natural Language Processing, W. Lehnert
and M. Ringle (eds.), Erlbaum 1982
Carbonell, J.G. "Invariance Heirarchies in Metaphor Interpretation"
Proceedings of the 3rd Meeting of the Cognitive Science Society, 1981
There's a large body of literature on analogical reasoning and other
aspects of metaphor comprehension. Much of the relevant research
has been done within psychology and linguistics. I'd suggest looking at
these for an overview:
Ortony, A. (Ed.) "Metaphor and Thought" Cambridge Univ. Press 1979
Lakoff, G. and Johnson, M. "Metaphors We Live By" Chicago Univ. Press 1980
Gentner D. "Structure-Mapping: A Theoretical Framework for Analogy" in
Cognitive Science, Vol. 7, No.2 1983
Winston P. "Learning by Creating and Justifying Transfer Frames" in
Arificial Intelligence, Vol. 10, No. 2, 1978
I don't know of any natural language system which can handle a wide range
of novel metaphors, and I don't expect to see one soon.
Any such system would have to contain an enormous amount of
knowledge. Unlike most present-day NL systems, a robust metaphor comprehension
system would have to be able to understand many different domains.
In spite of this difficulty, metaphor comprehension remains a fertile
area for AI research. I've spent some time examining how people
understand sentences like: "The US/Russian arms negations are a high
stakes poker game". When you get right down to it, its amazing that
people can figure out exactly what the mapping between "arms negotiatations"
and "poker games" is. What's most amazing is that using and understanding
metaphors APPEARS to take so little effort. (In fact, they are often the
easiest way to to rapidly communicate complex technical information. The next
time you are at a talk, try counting the analogies and metaphors used.)
-- Steve Minton, CMU
------------------------------
Date: Thu, 12 Apr 1984 11:31:13 EST
From: Danger, Will Robinson, Danger! <AXLER%upenn-1100.csnet@csnet-relay.arpa>
Subject: Metaphoric Comparisons
Mike:
Anthropologists and folklorists have been dealing with metaphor (and
related tropes) for a long time, in terms of their use in such common forms of
speech as proverbs and riddles, both of which depend almost totally on the
use of metaphoric and metonymic comparison. One thing that's critical is the
recognition that use of metaphor is extremely context-dependent; i.e., you
cannot apply Chomskian assumptions that competence is important, because the
problem occurs in performance, which Chomsky relegates to a side issue.
I'd suggest the following references for a start:
1. Sapir and Crocker, eds., "The Social Use of Metaphor" -- an excellent
anthology, about eight years old, covering a great deal of ground.
2. The special issue of the Journal of American Folklore from the early or
mid-seventies on Riddles and Riddling.
3. Dell Hymes, "Foundations of Sociolinguistics". (A really critical book
which set the stage for many anthropologists, linguists, etc. to shift over
from competence to performance; its biggest flaw is Hymes' insistence that
communication doesn't exist without intention on the part of at least one of
the performer(s), the receiver(s), and the audience.)
4. The journal "Proverbium", which was, for its 25-year life, THE place to
look for research on proverbs and related stuff. Especially good are articles
by Nigel Barley, Alan Dundes, and Barbara Kirschenblatt-Gimblett, whose "The
Proverb in Context" is a real key article.
5. Kirschenblatt-Gimblett and Sutton-Smith, eds., "Speech Play". A very good
anthology about uses of all sorts of special speech techniques, including
metaphorical comparisons, in various cultures.
Those are the ones I can remember off the top of my head. There are lots
more stored in my bibliography hard-copy file at home, and you can drop me
a net-note if you need 'em...
--Dave Axler
------------------------------
Date: 12 Apr 1984 09:49:50-EST
From: walter at mit-htvax
Subject: Smalltalk-52
[Forwarded from the MIT bboard by Laws@SRI-AI.]
ANNALS OF COMPUTER SCIENCE SEMINAR
DATE: Friday, April 13th, 1984
TIME: Refreshments 12:00 noon
PLACE: MIT AI Lab 8th Floor Playroom
SMALLTALK-52 and the Wheeler Send
ABSTRACT
Recently discovered paper tapes reveal that J.M. Wheeler
designed the first version of Smalltalk in 1952,
intending it to run on the University of Cambridge's
EDSAC Computer. The initial implementation, however,
required the machine's entire 512-word memory and was deemed
infeasible. Wheeler, who is credited with the invention
of bootstrap code, subroutine calls, assemblers, linkers,
loaders, and all-night hacking, can now be properly
credited with inventing message passing, object oriented
programming, window systems, and impractical languages.
This fascinating historical discussion and the accompanying
Graduate Student Lunch will be hosted by Steve Berlin.
Next Week:
Lady Lovelace's Public-Key Encryption Algorithm.
------------------------------
End of AIList Digest
********************
