perm filename F84.OUT[LET,JMC] blob
sn#781388 filedate 1984-12-31 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00002 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002
C00003 ENDMK
C⊗;
∂01-Oct-84 1007 JMC
To: RA
Include in your message that they should get their own tickets.
∂01-Oct-84 1026 JMC
To: oliger@SU-NAVAJO.ARPA
We are going Friday morning from San Jose and returning Friday evening.
I don't think any particular person going is needed to persuade DARPA.
A reason to go would be if you need more information in order to
establish your own interest in the machine and this seems to be the
most convenient way to get it.
∂01-Oct-84 1408 JMC
To: RA
hep.dis[e84,jmc]
I still prefer 5:43. Others may have different choices.
∂01-Oct-84 1410 JMC
To: GLB
I'd like to see you about EKL in CS206.
∂02-Oct-84 0155 JMC Kurzweil
To: Spencer.PA@XEROX.ARPA
CC: Diebert.PA@XEROX.ARPA, Allen.PA@XEROX.ARPA
Thanks for the invitation to see the Kurzweil. Any time Monday or
Wednesday (Oct 8 or 10) would be good for me. Otherwise Tuesday
or Thursday morning or Friday afternoon.
John
∂02-Oct-84 1109 JMC
To: steel@TL-20A.ARPA
I am indeed planning to take part as indicated.
∂02-Oct-84 1449 JMC kurzweil
To: allen.pa@XEROX.ARPA
I will be at PARC at 2pm Monday. Thanks.
∂02-Oct-84 1620 JMC
To: block@SU-CSLI.ARPA
I would be interested in the lectures on GB, GPSG and LFG assuming there
is a reasonably good chance that someone who doesn't even know what phrases
these acronyms stand for can understand them.
∂02-Oct-84 1805 JMC speech
To: RA
Thanks for transcribing it. Please rename it aaai.spe[e84,jmc] and
make an entry for it on the Summer 1984 page of files[let,jmc].
∂02-Oct-84 2140 JMC update on hep
To: kahn@USC-ISI.ARPA
Six of us will visit Denelcor Friday. We plan a telephone conference
with Encore (Gordon Bell). It is possible that I may hire Lester
Earnest to come back from the business world and manage the project -
provided it turns out to need that kind of management.
∂02-Oct-84 2152 JMC Abadi
To: ZM
CC: MA
I called Reiter and he will send Martin a net message inviting him and
telling him where to send money, etc.
∂03-Oct-84 1212 JMC Kowalik workshop
To: aaai-office@SUMEX-AIM.ARPA
Claudia,
I have decided to support the Kowalik workshop. He should keep us,
especially you, informed as the plans proceed. Congratulations on
getting out the papers for the Non-monotonic Reasoning Workshop so
promptly. I think that will be a real help in making the workshop
a success. I would like another copy if you can spare one. Send
bill too.
-- John
∂03-Oct-84 1544 JMC
To: bosack@SU-SCORE.ARPA
Thanks for lending Yoram the modem.
∂03-Oct-84 1552 JMC
To: RA
I prefer them to just come by.
∂03-Oct-84 2209 JMC
To: ARK
Not quite yet, please.
∂03-Oct-84 2212 JMC
To: barwise@SU-CSLI.ARPA
I would be glad to talk on non-monotonic reasoning - either a technical
talk emphasizing circumscription or an informal talk, pointing out
non-monotonic aspects of common sense reasoning and of linguistic
conventions.
∂03-Oct-84 2305 JMC
To: RA
When you make plane reservations for me please enter them in CAL.
∂04-Oct-84 0151 JMC
To: ARK
Say again what dates Scherlis will be here.
∂04-Oct-84 1136 JMC
To: withgott@SU-CSLI.ARPA
I will be away Oct 25. About two weeks later would be possible or
thereafter.
∂04-Oct-84 1154 JMC
To: withgott@SU-CSLI.ARPA
Nov. 15 is ok provided we can avoid 1:15 to 2:30 when I teach.
∂04-Oct-84 1436 JMC Karp
To: golub@SU-SCORE.ARPA
I vote no, not from any doubts about Karp's qualifications, but
because I think we are already oversupplied in his areas of interest.
My opposition is mitigated by the fact that Guibas turned us down
and by the fact that he will be half in OR.
∂04-Oct-84 1631 JMC Karp
To: golub@SU-SCORE.ARPA
This is to vote against the Karp appointment. I have no hesitations about
his qualifications, but we have too many faculty in the areas of
his strength. The fact that the appointment is half in OR
reduces my objections a little.
∂05-Oct-84 2119 JMC
To: RTC
Let him ask Hedrick, I haven't.
∂05-Oct-84 2119 JMC
To: RTC
Wow.
∂05-Oct-84 2120 JMC
To: JDM@SU-AI.ARPA
∂05-Oct-84 1324 JDM Key to offices containing lisp machines
John, Is it OK with you if I get a key to the offices containing the
lisp machines?
Jock
OK with me. - jmc
∂05-Oct-84 2122 JMC
To: RA
∂05-Oct-84 1542 RA Did you reorder the textbook for CS206?
I think Ross said the bookstore was re-ordering Common Lisp by Steele.
If I need to formally ask them to do so, please do it for me.
∂05-Oct-84 2132 JMC Subject: Zschau and Carnoy debate: Nuclear Weapons and National Security (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Seems brave of Zschau.
∂06-Oct-84 1111 JMC
To: JJW
I have no objection to your telling the students about the HEP, although
there is competition for the machine to be used in the project. The HEP
wouldn't be a departmental machine in the sense that Department members
could automatically get accounts. However, we would certainly encourage
any usage that would advance the goals of the project. We plan to have
Burton Smith come again. The HEP material should be returned to Dave
Cheriton.
∂06-Oct-84 1459 JMC
To: genesereth@SUMEX-AIM.ARPA
10-08 Monday, 2p, PARC, Kurzweil demo
∂06-Oct-84 1957 JMC Editor base operating system
To: ME
I am now making up a proposal and budget for the IBM sponsored editor-based
operating system project. Are you interested in working on it part time?
∂06-Oct-84 1958 JMC
To: ME
Or have you more questions?
∂06-Oct-84 2134 JMC re: Zschau vs. Carnoy on nuclear weapons (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Indeed it has been gerrymandered - by the Democratic controlled
state legislature. The object of the gerrymander is to concentrate
the Republicans into a relatively small number of districts which
they totally dominate and let the Democrats win 55 percent majorities
in the others. The Republicans have an initiative on the ballot to
require compact districts. Vote for it.
∂06-Oct-84 2251 JMC
To: SMC
vetap.nar
∂06-Oct-84 2251 JMC vetap.nar
To: SMC
It seems ok. I fixed one misprint. I wonder if it wouldn't be advantageous
to say something like, "My present intention is to specialize in
equine orthopedics, but I expect to become acquainted with many other
possibilities in vet school". Where does equine orthopedics fit in
the vet schools' picture of what the priorities are?
∂06-Oct-84 2317 JMC
To: SMC
Please mail that message. It got lost.
∂07-Oct-84 0941 JMC Gordon Bell
To: "@HEP.DIS[E84,JMC]"@SU-AI.ARPA
He is will tell us about Encore's product on Tuesday. It was originally
scheduled for 10am, but we are trying to move it to 9am to give a little
more time.
∂07-Oct-84 1433 JMC
To: SMC
Martha called.
∂07-Oct-84 1541 JMC Gordon Bell
To: feigenbaum@SUMEX-AIM.ARPA
We'll meet in or near my office. My secretary, Rutie Adler, or a sign
will know.
∂07-Oct-84 1621 JMC
To: RA
decter.1 in tex
∂07-Oct-84 1623 JMC
To: RA
Please rename my AAAI speech AAAI.SPE[F84,JMC] and enter it in FILES[LET.
∂07-Oct-84 1725 JMC
To: golub@SU-SCORE.ARPA
Omit the second word. Encore is will propose a multi-processor system
as a competitor to HEP for multi-processing LISP. The meeting will start
in my office and move to a seminar room.
∂07-Oct-84 1857 JMC needed example
To: grosof@SUMEX-AIM.ARPA, LEP@SU-AI.ARPA, VAL@SU-AI.ARPA,
nilsson@SRI-AI.ARPA
James Goodwin, in his paper for the non-monotonic conference, remarks
that one sometimes jumps to conclusions that are withdrawn on further
thought without any more facts from the outside. He uses this to
justify his flavor of non-monotonic reasoning (which seems quite
interesting). However, it seems to me that we can do the same with
circumscription, e.g. by using propositional circumscription to draw
conclusions that can be in contradiction to conclusions drawn when
substitution is used. I conjecture that more than propositional
logic will be involved in many applications. However, I lack examples,
and I don't recall that any are given in Goodwin's paper.
So, this is a request for examples of common sense reasoning where one
jumps to a conclusion that is withdrawn after further thought.
∂07-Oct-84 1907 JMC preliminary remarks on your non-monotonic chapter
To: nilsson@SRI-AI.ARPA, genesereth@SUMEX-AIM.ARPA
These are in case I don't get around to the more detailed remarks
I intend to send you.
It seems to me that the part about circumscription follows the
history too closely for a textbook. I would suggest using only formula
circumcription and using it for the examples that are doable with
predicate circumscription. Bring in the question of what predicates are
taken as variable at the beginning. This will shorten the chapter and
reduce the probability of confusion.
I have some new results about reducing prioritized circumscription
to ordinary circumscription, but I don't yet know how general they are.
Incidentally, even though my 1980 paper doesn't indicate it and
the examples I gave don't require it, I was aware that varying more than
the predicate being circumscribed would sometimes be required. I was
surprised when I discovered from Reiter's and Etherington's work that I
had neglected to mention it and give examples.
∂07-Oct-84 2006 JMC
To: golub@SU-SCORE.ARPA
Gordon called on Friday while I was in Denver, said he was coming to
Palo Alto Monday night and could talk Tuesday morning about his
company's proposal for a parallel computing. My secretary suggested
10am knowing my dislike for early hours, but my message, the copy
of which aimed at you must have been garbled, said that I was
counter-proposing 9am in order to have a bit more time. I haven't
yet reached Gordon to see if 9 is feasible.
∂07-Oct-84 2025 JMC
To: cheriton@SU-PESCADERO.ARPA
The body will be here. We'll meet at my office and go to a seminar
room. You are right about preparation. If we had split up earlier
as they proposed, there would have been a more detailed discussion.
Gabriel, Bosack and Weening had a more detailed discussion later.
∂07-Oct-84 2032 JMC vet application
To: SMC
Since you give taking care of other people's horses such high priority, you
might as well get credit for it on your vet school application. Therefore,
I suggest some letters from people who will praise you for that.
∂07-Oct-84 2042 JMC
To: SMC
I didn't mean my previous message to be critical.
∂07-Oct-84 2044 JMC going directly to Turing
To: MS
There is no advantage in going to Turing through SAIL. You can dial its
numbers, which you can get from them, directly. However, the SAIL LOCK
key on the terminal shouldn't be depressed and you should find out from
Eric Ostrom what to tell Tops-20 about your terminal type. Going through
SAIL does have the advantage that you can use the SAIL line editor.
∂07-Oct-84 2237 JMC
To: genesereth@SUMEX-AIM.ARPA
That would be Bill Spencer. Shall we go together?
∂07-Oct-84 2239 JMC terminal
To: MS
I forgot about the modems. When we started giving out home terminals,
the 1200-150 modems were much cheaper. Incidentally, if you don't use
SAIL much, I would like to swap the terminal for a more conventional
terminal. It is difficult and expensive to get more of the SAIL
terminals.
SAIL is still usable, though not so conveniently, from a conventional
terminal.
∂07-Oct-84 2324 JMC
To: MS
OK about DM then. I seem to win about one game in twelve. However,
I usually give up if I don't reduce the down cards to 30 before having
to deal.
∂08-Oct-84 0113 JMC qlambda
To: fsbrn@BRL-VOC.ARPA
CC: RPG@SU-AI.ARPA
I just saw your inquiry in the AI-list digest.
QLAMBDA is described in a paper by Richard Gabriel and myself given
at the LISP conference in Austin this August. The Proceedings are
published by ACM. Rumor as to its implementation is far in advance
of fact. We are currently considering (and DARPA is considering)
acquiring a HEP and implementing it thereon. Dick, the inquiry
is in AI.TXT[2,2]/17P.
∂08-Oct-84 1059 JMC
To: RA
Ask for 75.
∂08-Oct-84 1120 JMC ti
To: RPG
If we leave at 4pm, we'll be prompt. This might help get their attention.
On the other hand, they may be busy greeting people, and in this case it
would be better to go later. What do you think?
∂08-Oct-84 1128 JMC kurzweil
To: spencer.pa@XEROX.ARPA
Bill,
OK, I'll expect a message from him. I'll be away from Tuesday of next week
through the following week.
John
∂08-Oct-84 1614 JMC
To: CLT
I'm going to SF to kick TI tires. Tire kicking ends at 8. See you later.
∂08-Oct-84 2347 JMC renaming
To: RA
I'm on f84,jmc now, because I make a new directory each quarter. Therefore,
aaai.spe should have been put in that directory as I in fact requested. Also
I requested that the file be entered in FILES[LET,JMC]. This means that it
should be entered alphabetically in the F84 page, which is the last page.
I have done these things, but in the future ...
Also, from now on please put a subject in your messages to me, and I'll do
the same in my messages to you. This makes it easier to find the message
later, since it is then mentioned in the directory page of the message
file.
∂09-Oct-84 1121 JMC
To: RA
Hurd lunch tomorrow ok.
∂09-Oct-84 1511 JMC enclosure for Decter letter
To: RA
opinio[f83,jmc] should be printed, and opinio[w84,jmc] should be
PUBbed and printed. Also make a new ccpy of the letter itself, because
I have made further changes.
∂09-Oct-84 1518 JMC my class
To: JJW
The days will be Thursday October 18 and T and Th Oct 23 and 25.
Yes, let's get together later this week.
∂09-Oct-84 1656 JMC
To: JJW
Denelcor Burton Smith (303) 337 7900. Bob Lackher (206) 775 1322.
1 hit on key "burton".
Exit
↑C
.
∂10-Oct-84 0011 JMC Subject: Ken Thompson lecture on Belle (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - They could move the lecture for the benefit of the foreign
nationals, but the poor Unix groupies seem to be absolutely
unwelcome.
∂10-Oct-84 1010 JMC
To: berg@SU-SCORE.ARPA
Have them make it 40 then.
∂10-Oct-84 1126 JMC
To: RA
Rutie,
Please send him a computer message to find his U.S. mail address
and then mail him a copy of the paper.
∂10-Oct-84 1044 SARASWAT@CMU-CS-C.ARPA Circumscription paper
Received: from CMU-CS-C.ARPA by SU-AI.ARPA with TCP; 10 Oct 84 10:44:42 PDT
Received: ID <SARASWAT@CMU-CS-C.ARPA>; Wed 10 Oct 84 13:41:47-EDT
Date: Wed 10 Oct 84 13:41:45-EDT
From: Vijay.Saraswat@CMU-CS-C.ARPA
Subject: Circumscription paper
To: jmc@SU-AI.ARPA
I wonder if it would be possible for you to send me a copy of your
paper:
"Applications of circumscription to formalizing common sense knowledge"
which you say has been submitted to Arificial Intelligence? (I am
quoting from the abstract of the talk you are to give at Maryland later
this month.)
If it is possible to retrieve it from SU-ai by an anonymous login, that would be the best (i.e. fastest) way of getting it for me.
Thank you, in advance.
Vijay A. Saraswat
-------
∂10-Oct-84 1410 JMC
To: HANRAHAN@SU-SCORE.ARPA
Thanks for the message about the phone call.
∂10-Oct-84 1628 JMC
To: RPG
CC: JJW
How about shortest?
∂10-Oct-84 2311 JMC CIS show and tell
To: lundstrom@SU-SIERRA.ARPA
CC: meindl@SU-SIERRA.ARPA
I prefer the 14th, but I will be talking at the Law School (on the same
subject) at 4:15. On the 15th I teach from 1:15 to 2:30, but I usually
reserve the morning for preparation, and I'm often tired afterwards.
∂10-Oct-84 2312 JMC Hep, etc.
To: JJW
Friday after 2pm would be ok.
∂11-Oct-84 1440 JMC
To: RA
That's Bernie Galler.
∂11-Oct-84 1816 JMC
To: CLT
∂11-Oct-84 1748 JJW HEP Lisp
This afternoon I visited Lucid and discussed HEP Lisp with them. We went
through a lot of issues, using the longest-list example. I think I know
enough now to turn it into reasonable HEP assembly code, that will test
some of the things that Lisp programs could be expected to do. It's going
to take at least another day or two for me to write the code.
∂11-Oct-84 1817 JMC
To: CLT
∂10-Oct-84 1023 JJW Test code for the HEP
To: RPG, JMC
Here are some thoughts on test code for the HEP. I've written a simple
length function, without procedure calls, and have a suggestion for a
more complex test.
1. Length of a list
This makes some assumptions about the format of assembler input,
since we don't yet have a reference manual for the assembler.
Variable registers Use
U pointer to current cons cell
X result of null test
LEN length of list
Constant registers Value
NIL representation of NIL
CAR 0 (offset of CAR given pointer) [not really needed]
CDR 1 (offset of CDR given pointer)
LENGTH .. function entry ..
CLR LEN ;length ← 0
B LENGTH2 ;unconditional branch
LENGTH1 INC LEN ;length ← length + 1
LODX U,CDR,U ;U ← cdr U
LENGTH2 TEQ X,U,NIL ;X ← if U=NIL then 1 else 0
BEQ X,LENGTH1 ;loop back unless U=NIL
.. function return ..
We could test it by initializing a region of data memory to contain a
list, and then executing code with a pointer to the beginning of it.
2. Finding the longest list
We are given a list of lists, e.g. ((A B C) (D C) () (D A B C)). The goal
is to return the longest one; in this case (D A B C).
One way to do this in parallel is to CDR down the first list and start a
process for each of the sublists that determines its length. It then
stores the length somewhere (where?), and we finish by CDRing down this
list, keeping track of the largest number so far, and simultaneously
CDRing down the original list again, so that we end up pointing to the
longest list.
a. How would you write this in QLAMBDA?
b. We could test this on the HEP without calling CONS, if we provide
a list for the processes to store the lengths in before creating
them. I'll work on the code for this later today unless you decide
to do it first.
The test should include lists that share storage, such as the one
above, to see how that affects performance.
Various ideas for using the full/empty bits come to mind for this
problem.
If we change it to finding the shortest list, then we can keep a
global variable SHORTEST-LENGTH-SO-FAR, and processes can die early
if their length is longer than it.
∂11-Oct-84 1817 JMC
To: CLT
∂10-Oct-84 1540 RPG
To: JJW
CC: JMC
Like this:
(m-defun longest (l)
(let ((longest ())
(length-of-longest 0))
(labels ((monitor
(qlambda t (list len)
(cond ((< length-of-longest len)
(setq length-of-longest len)
(setq longest list)))))
(q-length
(lambda (list)
(do ((l list (cdr l))
(count 0 (1+ count)))
((null l)
(wait (monitor list count)))))))
(qcatch 'return
(do ((l l (cdr l)))
((null l) ())
(spawn (q-length (car l)))))
(list length-of-longest longest))))
(longest '((1 2 3)(1 2)(1 2 3 4 5 6 7)(1 2 3 4)(2 3 4 5 6 7) ()))
∂11-Oct-84 1817 JMC
To: CLT
∂11-Oct-84 0911 RPG
To: JJW, JMC
(m-defun shortest (l)
(let ((shortest ())
(length-of-shortest ()))
(labels ((monitor
(qlambda t (list len)
(cond ((or (not (numberp length-of-shortest))
(> length-of-shortest len))
(setq length-of-shortest len)
(setq shortest list)))))
(q-length
(lambda (list)
(do ((l list (cdr l))
(abort () abort)
(count 0 (1+ count)))
((or (null l)
(and (numberp length-of-shortest)
(< length-of-shortest count)
(setq abort t)))
(cond ((not abort)
(wait (monitor list count)))))))))
(qcatch 'return
(do ((l l (cdr l)))
((null l) ())
(spawn (q-length (car l)))))
(list length-of-shortest shortest))))
∂11-Oct-84 1818 JMC
To: CLT
∂11-Oct-84 1746 JJW Multiprocessor comparison
To: JMC@SU-AI.ARPA, RPG@SU-AI.ARPA, cheriton@SU-PESCADERO.ARPA,
Bosack@SU-SCORE.ARPA
I've prepared a file containing some comparative statistics for the
Denelcor HEP, TI Explorer, Encore, Elxsi 6400 and Synapse N+1 systems.
It is on SPECS[1,JJW] at SAIL. Some information is missing; please
send me a message if you can provide any corrections or additions.
Joe
∂11-Oct-84 1819 JMC
To: JJW, RPG
Please include CLT in messages on coding and on evaluations.
∂11-Oct-84 1923 JMC Farmwald comments
To: "@HEP.DIS[E84,JMC]"@SU-AI.ARPA
Mike Farmwald of the S-1 Project at LLNL
will discuss the HEP and other parallel computers
at 3:30pm tomorrow. We'll start from my office.
∂12-Oct-84 1406 JMC your move after all
To: manuck@SU-SCORE.ARPA
It just occurred to me that before I can go to the Executive Council, we need
a budget for the project. You should negotiate with Claudia about your part
and, with her help, and that of Mike, make a draft budget that I will look at.
∂12-Oct-84 1413 JMC electronic library
To: aaai@SUMEX-AIM.ARPA
CC: genesereth@SUMEX-AIM.ARPA
Richard has said he wants to do it. I have asked him to call you about
negotiating his pay and to prepare a budget for the project as a whole
that can be submitted in outline to Woody and to the Executive Council.
I think I already told you that Woody said he was for the idea.
You will also have to negotiate with Len Bosack about using CSD's file
computer. I will see about getting approval in principle for its use.
∂12-Oct-84 1450 JMC New Scientist Interview
To: aaai@SUMEX-AIM.ARPA
I'll be glad to talk to him if it turns out to be convenient. I am curious
to sample the relation between what one says to New Scientist and what
appears in print.
∂12-Oct-84 1510 JMC withdrawing from orals
To: feferman@SU-CSLI.ARPA
CC: CLT@SU-AI.ARPA
With Manna agreeing to be a reader and to be on the orals committee, I believe
I can withdraw from the orals committee and therefore will attend only
the public part of the examination. I think I should get another C.S.
person if I am also to withdraw as a reader. Since reading is done
without interaction among the readers, I feel withdrawing here as less
urgent. What is your opinion?
∂12-Oct-84 1734 JMC
To: CLT
It seems like an excessively permanent way of publishing a movie review.
∂12-Oct-84 1839 JMC
To: bmoore@SRI-AI.ARPA
I missed the last meeting by forgetting, and I'll miss the next one by
being out of town. Nevertheless, I'd like a copy of the Harmon paper.
∂13-Oct-84 1350 JMC
To: DEK
I think it's WW II or earlier.
∂13-Oct-84 1356 JMC
To: FY
solitr sometimes says "bad checkout", congratulations. Why?
∂13-Oct-84 1749 JMC wrong title in list of reports
To: berg@SU-SCORE.ARPA
CC: JK@SU-AI.ARPA, JJW@SU-AI.ARPA
Not only were the two lines of the title
run together in the title page of the EKL
manual, but they were also run together in the name of the manual
on the list of reports. Presumably that latter error occurred after
the former was brought to your attention. Please try to be more
careful. In case the matter still isn't clear the correct title is:
EKL - An interactive proof checker
User's Reference Manual
∂14-Oct-84 0044 JMC solitr bug
To: FY
I sometimes type - instead of *, because of not reaching far enough.
The combination - <n> always ends the game - at least on a Datamedia.
∂14-Oct-84 2140 JMC Gabriel and McCarthy
To: RA
Please mail a copy of this paper to
Professor Eichi Goto
Computer Science Department
Tokyo University
Tokyo
JAPAN
Please include a note to the effect that I thought that he might find
it relevant to his circular pipleline machine project.
∂15-Oct-84 0110 JMC paper from library
To: RA
Please make a copy for me this morning of a paper from the Math library.
It is by George Polya and it appeared in volume 53 (1956) of the
American Mathematical Monthly, pp. 689-697.
∂15-Oct-84 0951 JMC clt reading committee
To: SF@SU-CSLI.ARPA
OK, I'll withdraw completely from the official part. Manna has already
agreed to be an official reader. Also Ullman has agreed to be a member of
the orals committee. I think I would like to try to get one more member
of CS to take part, since I would like Carolyn to be better known in the
Department.
∂15-Oct-84 1219 JMC
To: RA
Please pick up levien.xgp from the xgp and make an envelope.
∂15-Oct-84 1611 JMC
To: RA
It is biojmc[1,jmc]; if there is a .xgp copy later than the main file,
print it. Otherwise, it should be pubbed first. There is also a shorter
biography called newbio[1,jmc].
∂15-Oct-84 1744 JMC biography
To: SMC
Remember to tex my biography biojmc[1,jmc]. It should be called
biojmc.tex[1,jmc] and should be entered into files[1,jmc].
∂15-Oct-84 1758 JMC
To: RA
galler.1
∂15-Oct-84 1901 JMC
To: FY
You don't mean "brandished as a cheater".
∂15-Oct-84 2317 JMC papers on circumscription
To: chou@UCLA-LOCUS.ARPA, RA@SU-AI.ARPA
My secretary will mail you my latest paper on circumscription. Many papers
are included in the preprints of a AAAI conference on non-monotonic reasoning.
These papers may be available from the AAAI office, 445 Burgess St.
Menlo Park, CA. (415)328-3123.
∂15-Oct-84 2318 JMC junk mail
To: ARK
Arthur, I don't appreciate receiving all this junk mail about TGIF. How
about using the REPLY feature more sparingly.
∂15-Oct-84 2322 JMC junk mail
To: jf@SU-SCORE.ARPA
I do not appreciate receiving all this junk mail about TGIF.
Those of us faculty who don't know about TGIF by now are very likely immune.
How about using the REPLY feature more sparingly.
∂16-Oct-84 1054 JMC
To: RA
Please write a note for me explaining that I will be in France.
∂16-Oct-84 1044 RA III 10/25 meeting
It's ok for you not to attend the meeting but please send a note to Alex
Jacobson explaining why.
∂16-Oct-84 1055 JMC
To: RA
That's Earnest.
∂16-Oct-84 1134 JMC
To: karp@SU-SCORE.ARPA
Please stop this endless drivel about tgif.
∂16-Oct-84 1651 JMC my miscount
To: ullman@SU-AIMVAX.ARPA
CC: CLT@SU-AI.ARPA
It turns out that there are enough readers on Carolyn's committee
but she still needs you as the non-reader oral examiner.
∂16-Oct-84 1848 JMC
To: CLT
Zella said she won't be here next week.
∂29-Oct-84 2317 JMC an early bug (maybe)
To: shore@NRL-CSS.ARPA
I don't know whether you want to call this a bug, but it is said
that in the first payroll program, which was for the Univac at the
GE plant in Louisville, Kentucky, programmers took advantage of the
fact that the social security tax was two percent of wages.
Everywhere this percentage had to be calculated, they used shifts
and additions instead of multiplications to make the program faster.
When the tax rate changed to 2.2 percent ... . You might try Herb
Grosch in order to determine whether this really happened or is
hypothetical. I'm suspicious of the Fortran DO I = 1.5, because
I and others thought of it as a recreational problem, and I wonder
whether someone might have improved the story.
∂30-Oct-84 0000 JMC
To: hobbs@SRI-AI.ARPA
I would have discussed but just returned.
∂30-Oct-84 0009 JMC
To: restivo@SU-SCORE.ARPA
MAIL BOYER@UTEXAS-20
∂30-Oct-84 0015 JMC
To: RTC
Yes on more disk.
∂30-Oct-84 0017 JMC
To: RA
Yes on Caltech alumni gift.
∂30-Oct-84 0341 JMC Friday won't do.
To: rjf@MIT-MC.ARPA
It turns out that Friday won't do for getting together concerning
my questions about Macsyma, etc. I'll phone you next week to arrange
to visit you.
∂30-Oct-84 1027 JMC
To: aaai-office@SUMEX-AIM.ARPA
It sounds to me like you have done a sufficiently careful job.
∂30-Oct-84 1139 JMC
To: RA
It is inconvenient to have you gone the day I return, so I hope it's important.
∂30-Oct-84 1139 JMC contract and security
To: LLW@SU-AI.ARPA
I have been away until today. My secretary is out, but I have looked through
my mail, and I don't seem to have heard from either. Are you going to the
next Pournelle show? If so the usual would suit me.
∂30-Oct-84 1302 JMC
To: llw@S1-A.ARPA
My secretary confirms that there has been no word of either.
∂30-Oct-84 1552 JMC
To: ullman@SU-AIMVAX.ARPA
I got the Burroughts questionnaire.
∂30-Oct-84 1555 JMC Prolog in Lisp
To: CS.DENNEY@UTEXAS-20.ARPA
There is a Zetalisp implementation by Martin Nilsson of the Computer
Science Department of Uppsala University in Uppsala, Sweden.
∂30-Oct-84 1606 JMC
To: tracz@SU-SIERRA.ARPA
jmc - Try Hsi Nan on University and Emerson.
∂30-Oct-84 1620 JMC re: A quote a day keeps Reagan away. (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - I favor Reagan, but I don't think that the anti-Reagan messages
on BBOARD are improper in any way. At this point we have 568 lines
of flamage on the SAIL BBOARD, which means it isn't close to big league
flamage, and it's a week before the election. I will comment only on
inflation. I don't understand fully why it went down, but when one doesn't
understand it is rational to vote for the politician under whose auspices
a result one likes occurred - barring a convincing reason that he deserves
no credit.
∂31-Oct-84 0327 JMC
To: cheriton@SU-PESCADERO.ARPA
Either time is ok for me.
∂31-Oct-84 2350 JMC
To: RA
I don't think he does. Why do you ask?
∂01-Nov-84 1630 JMC
To: davies@SUMEX-AIM.ARPA
I don't know of work in this area.
∂01-Nov-84 1632 JMC
To: YM
Let it indeed be Dec 3 at 2pm.
∂01-Nov-84 1700 JMC re: A quote a day keeps Reagan away. (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - However, as didn't occur in many other countries, there was also
an expansion of production, and the recession of the early days of the
Reagan Administration ended.
∂01-Nov-84 1704 JMC re: Pay phones: the new enemy (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Imagine yourself to be a creative young lawyer and figure out how
the phone company was negligent.
∂01-Nov-84 2145 JMC
To: barwise@SU-CSLI.ARPA
The draft on military funding did not appear.
∂01-Nov-84 2210 JMC
To: cc.clive@UTEXAS-20.ARPA
I was away last week, and I have been in meetings all this week. I'll
get around to deciding what to do about your invitation tomorrow.
∂02-Nov-84 1403 JMC
To: RA
I don't intend to attend the Hackers' Conference.
∂02-Nov-84 1458 JMC refereeing
To: stan@SRI-AI.ARPA
Experience has shown that I do it badly and tardily, and therefore I must
decline to referee papers for IJCAI-85. Sorry. Congratulations on being
Director of the AI Center at SRI.
∂02-Nov-84 1703 JMC page 11
To: kdf@MIT-MC.ARPA
The answer to the last question is incorrect, and the correct answer probably
cannot be given by a qualitative theory. What will happen if the system is
pumped depends on how it is pumped. If it is a linear system and is pumped
sinusoidally, then there will be a stable oscillation unless the pumping
is at the resonant frequency in which case the spring will eventually break.
If the system is non-linear or if the force is not sinusoidal, then it is
more complicated to determine what will happen, but I believe that to get
a growing oscillation is an exceptional case.
∂02-Nov-84 2202 JMC
To: tracz@SU-SIERRA.ARPA
Peninsula - Chinese
Northern
*Ambassador 415 364-6400
826 5th Avenue, Redwood City
Hours: 11:30am-9:30pm Mon-Sat, 11:30am-9pm Sun. MC, VS.
Very good; in the low rent district. Excellent hot & sour soup,
Mongolian beef, and Szechwan prawns; good mui shui pork and
potstickers; so-so lemon chicken. At $12 pp., a best buy!
[LES-80:6/84]. Agreed. I've never been disappointed [KIP-80].
Excellent, but tends towards the Americanized (e.g., lemon chicken is
delicious but very sweet) [NCR-82].
Chef Chu's Mandarin Kitchen 415 948-2696
1067 North San Antonio Road, Los Altos
Hours: noon-9pm daily, till 10pm Fri-Sat. MC, VS.
Much better can be had elsewhere [SMC-80]. Good food but terrible
service. Have a pizza delivered while waiting for your reservation
[WDG-81, LES-70:5/84]. Excellent food. The usually long waits can be
avoided if you arrive early. Reservations accepted for parties of 6
or larger. They published a cookbook, and they offer cooking classes
through Foothill College [ARK-83].
China First 415 326-3900
675 El Camino Real, Palo Alto (near Holiday Inn)
Hours: 11:30am-2:30pm, 5pm-9:15pm. VS, MC.
Consistently very good, Chen pea chicken and General's Chicken are
tasty, and they will take reservations (they give you a "number" over
the phone) [PAT-78:80]. OK, not outstanding. Hot & sour soup often
bland. Mu Shu pancakes too thick. General's chicken worth ordering
[KJK-78:80]. Good luncheon special [SJF-80]. Quite variable in
quality [MOG-80]. OD'd once on MSG. Generally delicious [LES-79:82].
Quite good [WDG-81]. Very good General's Chicken. Good luncheon
specials [ARK-6/84].
China Lion 415 493-9989
3345 El Camino Real, Palo Alto
Hours: 11:30am-9:30pm, Fri-Sat till 10:30pm. MC, AX, VS.
Not bad hot and sour soup and mui shui pork the one time I was there
[PAT-80]. OK, but nothing to write home about [ARK-84].
Colonel Lee's Mongolean Bar-B-Q 415 968-0381
304 Castro, Mountain View
Hours: 11:30am-9:30pm, Sun 4pm-10pm, closed Mon. MC, VS.
An interesting variation of Chinese cuisine; quite good [LES-78:81].
Dinner now $6.95 for all you can eat. Very good. I haven't had much
trouble with lines -- you can eat in less than an hour on a weekday
[PTZ-80].
Feng Yuan 415 494-7391
3950 Middlefield Road, Palo Alto
Hours: 7:30am-9pm Mon-Fri, 11am-9pm Sat-Sun. No credit cards or
checks.
This is not a new place, but it used to be more of a coffee shop. At
dinner time it becomes a very credible Chinese restauraunt. The
twice cooked pork is the best I've ever had. The only disappointment
so far has been the potstickers. Prices are quite reasonable
[TWE-80]. I think the raves for this place were just a fad. It is,
and always has been, a greasy chopstick lunch place [JQJ-82].
Grand China 415 964-6464
5100 El Camino Real, Los Altos
Hours: 11:30pm-10pm, till 10:30pm Fri-Sat, closed Tue; dim sum
served 11am-2pm Sat-Sun only. MC, VS, chacks accepted.
Szechuan and Hunan dishes best; especially lamb dishes. One of my
favorites [SM-2/83]. Good hot-and-sour soup. Mediocre potstickers.
Good service [MOG-7/84].
House of China 415 961-7347
102 Castro Street, Mountain View
Hours: 11:30am-2:30pm 5pm-9pm daily.
Newly open, friendly family atmosphere, good honest Mandarin and
Szechuan dishes. I like to know I can get in to a restaurant any
time, so I go here instead of Chef Chu's. Serve beer and wine. $4-8
pp. [EBK-3/84].
House of Yee 415 961-9694
160 Castro Street, Mountain View
Hours: 11:30am-9:20pm daily; dim sum served 11am-2pm Sat-Sun only.
MC, VS.
We went for dim sum and had a decent meal for about $4 pp. We were
the only caucasians there. The decor is quite modern, service pretty
good [MOG-82].
Hsi-Nan the Little 415 326-6065
203 University Avenue, Palo Alto
Hours: noon-2:30pm 6pm-10pm Mon-Fri, 6pm-10pm Sat, closed Sun.
Checks OK, no cards.
Louie has done it again. Absolutely the best Chinese food in the
universe, although the prices seem to be skyrocketing. I actually
prefer the Szechuan dishes now that Mr. Fong is gone. They seem to be
less variable in quality. Avoid Twice-Cooked-Pork and Moo Shoo Pork.
The first is grossly oversweetened and the second has little more
than lettuce [SMC-82]. Still the world's best potstickers. But don't
order anything special without finding out the price in advance
[BH-82]. Everytime I go there with Ralph Gorin or John McCarthy, I
get flamed out from the food. I've never had a spicier Spicy Beef
Wonton Soup. When it's good, it is very, very good; when it's bad,
it's hotter than hell or too greasy [ARK-83].
Hunan Garden 415 948-1270
2540 California Street, Mountain View
Hours: 11:30am-9pm Sun-Thu, 11:30am-10pm Fri-Sat. AX, MC, VS.
New restaurant opened in the Old Mill. Decent Hunan cuisine although
not nearly hot enough. If they increased their hotness and put a
little more meat in their dishes, they might be simliar to Hunan in
San Francisco. The onion-cakes are a bit greasy. A promising
bridesmaid at the moment [SMC-82]. The bridesmaid has a moustache;
bad service and bland food [SM-2/83].
King Chuan 415 323-6550
75 Town and Country Village, Palo Alto
Hours: 11am-9:30pm Mon-Thu, 11am-10pm Fri-Sat, closed Sun. VS, MC.
Food okay, but unimpressive [DON-80]. Seems to vary. Sometimes good,
sometimes lousy. Hot & sour soup neither hot nor sour [KJK-80]. Food
has improved. Szechwan items particularly well done [DLO-82].
Hwei Fu Shing 415 493-0643
3691 El Camino Real, Palo Alto
Hours: 11:30am-2pm, 5pm-10pm. VS, MC, checks accepted.
Buffet lunch downhill; all cheap stuff [SM-80].
O'Mei Restraunt 415 323-2700
1438 El Camino Real, Menlo Park
Hours: 11am-2:30pm daily, 5pm-9:30pm Mon-Fri, till 10pm Fri-Sat, til
9pm Sun. MC, VS.
Unusual Diem Sum. Scallion cakes and crispy oil sticks interesting
[SM-80]. Nice change of pace in Szechuan hot & spicy. Milder than
the Little Restaurant, but decor and service are very good and the
dishes are reasonably warm with some slight hint of MSG. Pretty good
Diem Sum on weekends [SMC-80]. Consistently good [MOG-80].
Ruby King 415 941-5656
4320 El Camino Real, Los Altos
Hours: 11:30am-2pm Tue-Sun, 5pm-10pm Tue-Sat, closed Mon. AX, MC,
VS.
Fabulous atmosphere and menu. Service was atrocious the time that I
went. The waiter served our food to someone else and we ended up
leaving after the soup course since it had already taken an hour
[SMC-80]. Food is good, well-prepared, but not inspiring; even with
large portions, the prices are about $1 high, per dish. Service
seems slow [MOG-7/84].
Shanghai 415 853-9606, 324-8749
2111 El Camino Real, Palo Alto
Hours: 5pm-9:30pm, closed Mon. VS, MC.
New management and absolutely delicious, but needed coaching to get
that way. They put a dish on the menu especially for us. Dishes to
concentrate on are: Mu Shu Pork, Hunan Style Lamb, Twice Cooked Pork,
Prawns with Special Sauce, chicken dishes. Be sure to ask for the
hot dishes "squeeze-hot." They take the peppers and squeeze the
juices over! Invented for us since we were always clamoring for
hotter food. Avoid losing potstickers and soups [SMC-80]. Poor
potstickers, hot-and-sour soup ok [MOG-7/84].
Six Happiness 415 326-6565
220 University Avenue, Palo Alto
Hours: 11:30am-9pm daily. VS, MC.
Usually good and moderately priced [PAT-80]. Sometimes good, but not
consistently [KJK-79:80]. Also try the Dim Sum [WLS-80]. Good for
lunch [WDG-81]. I'm not impressed [ARK-83].
Su Hong 415 322-4631
1037 El Camino Real, Menlo Park
Hours: 4:30pm-9:30pm. VS, MC. Reservations accepted.
Reasonable twice cooked pork and potstickers. Reasonable prices and
atmosphere. Pretty good lunches [SMC-80]. New chef. Try rolling
lettuce chicken and potstickers (as good as Louie's) [SM-80].
Recently remodelled. Rolling lettuce chicken is good. Potstickers
are tasty and a good deal since there are 8 of them per order. Good
food and reasonable service [ARK-84].
Southern
The Golden Flower Restaurant 415 941-7613
2650 El Camino Real West, Los Altos
Hours: 11am-9:30pm Tue-Sun, closed Mon. MC, VS.
Fine Cantonese cuisine at a low price. Authentic Cantonese is a treat
that is not easy to find. Some special dishes: sizzling chicken in
clay pot, assorted sea food on fried egg, shrimp in black bean sauce,
A1 Cafe chow mein, chow fun, clams with black bean sauce [TOB-82].
Looks like the kind of restaurant you'd never go in, but this place
has great Cantonese food. Try the spare ribs, the stuffed bean-
cakes, the roast baby quails. I didn't like sand dabs. The owner is
obviously enthusiastic about his food; ask him what to try.
Definitely worth going back to [MOG-3/84]. Excellent till recently,
then declined substantially under new management [LES-6/84].
Furama 415 941-6328
Old Mill Shopping Center, Mountain View
Appears to be a Cantonese place that has recognized that it takes
Szechuan dishes to "make it" around here. Food is actually pretty
good, although expensive ($7/dish average) [TLR-11/83].
House of Lee 415 941-4373
633 California Street, Mountain View
Hours: 11:30am-9pm, Fri-Sat till 10pm. VS, MC.
Superb chicken salad. Quality varies a bit but their overall record
is very good. At $8 pp., very cost effective [LES-72:8/84].
Imperial Gardens 415 968-1694
2116 El Camino Real, Mountain View
Hours: 11:30am-2pm, 5pm-9:30pm daily. AX, MC, VS.
Elegant decor and a reasonable although somewhat expensive Mandarin
and Cantonese menu. Best bet is Dim Sum on weekends -- at $1.35 per
plate it isn't as cheap as Chinatown in the city, but they brought a
tasty assortment of dishes around. Weekend Dim Sum seems to be a new
experiment for this restaurant. Hard to believe, but somehow I miss
the noise and hectic atmosphere that I've come to associate with most
other places [SM-4/83].
Ming's 415 856-7700
1700 Embarcadero Road, Palo Alto
Hours: Mon-Fri 11:30am-2pm, Mon-Sat 5pm-10pm, Sun 4:30pm-9:30pm.
All major cards.
Excellent and expensive. Slick decor, efficient service. Family-
style lunches only for large parties with advance reservations
[LES-66:6/84].
Pearl River Seafood Restaurant 415 967-1689
246 Castro Street, Mountain View
Hours: 11am-11pm, till midnight Fri-Sat, closed Tue. MC, VS.
I've only eaten there once so far but I plan to go back often. The
deep fried oysters ($4.75) were a excellent appetizer: six huge
oysters, hot and juicy, each enveloped in a golden, light crusty
batter. Steamed prawns in garlic sauce ($7.00) came to the table
wearing their shells and a unthickened sauce redolent with garlic and
sporting a goodly amount of black beans. The Dragon and Phoenix bean
cake ($4.95) was the best I've every had. This dish has triangles of
bean cake, mushrooms and pieces of ham with an egg white thickened
sauce that has a seafood flavor: a must for bean cake lovers
[MZA-12/83].
Seven Seas 415 328-6885
443 Emerson, Palo Alto
Hours: 11am-2:30pm, 3:30-9pm. No cards.
The only Chinese restaurant with "Italian Spaghetti" and "Catsup a la
Dog" (sic). Christmas decorations at inappropriate times. Sadly,
they've taken the "Smile--God Loves You" sign off the milk machine.
Most evenings there's at least one crazy person at the counter
[DES-76]. Definitely the best deal in town. Try grandfather's
lemonade [PB-80]. Be sure to ask for family-style to get more food,
less rice [AAM-80]. OK for fast cheap food [KJK-80]. Fairly
inexpensive and quick. Coffee shop atmosphere. "Bus station chinese
food." Food ok for the price, and not limited to chinese [ARK-84].
Yuen Yung 415 325-7807
639 Santa Cruz Avenue, Menlo Park
Hours: 11:30am-2:30pm, 4pm-9:30pm. MC, VS.
House special fried rice is very good. The crab with ginger and
scallions is also very tasty (price varies but was only $7.50 when I
was there). Everything excellent. Good takeout, too [DON-80]. Pretty
reasonable, although they use some MSG. Avoid the sweet & sour
dishes [SMC-80]. Good but expensive [WDG-81]. Recently remodelled.
Food is rather good, but service is variable. The ventilation is
good enough to cool the air but not good enough to remove the smoke.
The dim sum is good, but you have to order it from the menu
[ARK-8/84].
∂02-Nov-84 2213 JMC
To: tracz@SU-SIERRA.ARPA
It's not a closed circle, but I don't know if anyone is presently engaged
in making a new edition. Printed copies of the whole guide can be obtained
from Rutie Adler in 358 Margaret Jacks Hall, 497-2800.
∂03-Nov-84 1223 JMC
To: ROB@SU-AI.ARPA
If you are physically close by, it would be best to phone some afternoon
and come. Otherwise phone, and we'll make an appointment.
∂03-Nov-84 1225 JMC
To: ramin@SU-STAR.ARPA
It's at the Business School.
∂03-Nov-84 1331 JMC
To: SMC
Martha will pass by around 7pm on her way to Watsonville.
∂03-Nov-84 1347 JMC
To: SMC
n028 0952 03 Nov 84
BC-VECSEY-COLUMN 2takes
SPORTS OF THE TIMES
New Method Saves Saratoga Six
By GEORGE VECSEY
c. 1984 N.Y. Times News Service
NEW YORK - The bone snapped like a tree limb. A few lengths behind,
D. Wayne Lukas, the trainer, heard the cracking sound above the soft
thudding of hooves in the morning workout. Then he saw his prize
2-year-old, Saratoga Six, pull up lame.
It was one of those injuries that owners and trainers fear, the
sudden, inexplicable break that makes racing a gamble all the way
around, not just for the long-suffering bettors in the stands. It was
a disruption of the suspensory apparatus that holds the ankle at a
precarious 45-degree angle to the hoof.
It was the same injury that sent Ruffian staggering in her tragic
match race with Foolish Pleasure at Belmont in 1975, that sent her to
a futile operation and her burial at the Belmont infield a day later.
It was an injury that killed 90 percent of the horses who suffered
it, until recently.
But Saratoga Six is alive today, 13 days after suffering the injury,
and there is every chance he will endure - never to race again, but
to distribute his Alydar genes to new generations. He has survived
because of an innovative steel-plate procedure by Dr. Larry Bramlage,
an assistant professor of equine surgery at the College of Veterinary
Medicine at Ohio State University.
The saving of Saratoga Six is one of the most hopeful moments of
this season, in which Devil's Bag never won a Triple Crown race, and
Swale died suddenly a few days after winning the Belmont, and John
Henry, the 9-year-old people's champion, may have to drop out of the
Breeders' Cup next Saturday at Hollywood Park because of an injury.
The seven-stakes Breeders' Cup grand finale is certainly the most
exciting racing event of the fall season, but new surgical techniques
like Bramlage's may do more to ''improve the breed'' than any one-day
extravaganza. However, as long as racing spends millions on purses
and pennies on research in genetics and medicine, the fragile basis
of the industry will always be threatened.
Racing needs more success stories like that of Saratoga Six, which
began with the crack of bone - ''like a rifle shot,'' said Lukas. One
of six owners of the colt, Lukas was riding an exercise pony during
the workout at Santa Anita on Oct. 22.
''They told me to clear the track so the other horses could keep
going,'' Lukas recalled the other day. ''I said: 'Not for this colt,
you don't.' And we kept him at the rail until we calmed him down.''
Lukas said the colt was insured for $15 million, but added: ''As
good as it might sound to the layman, we did not want the $15
million. The colt was worth more to us alive. The insurance company
wanted us to have the operation and they wanted to pick the team,
with our approval, of course.''
The surgical team included Bramlage, 33 years old and a native of
Kansas, who had studied with the Association for the Study of
Internal Fixation, a medical group involved in using plates and
screws on human bone. He had worked in Davos, Switzerland, and Ulm,
West Germany, with ''a good friend, Caius Burri, a medical doctor,
who helped me a lot.''
The veterinarian assembled a seven-member team on Oct. 24,
encouraged that Saratoga Six was in a good mood. Contrary to the
image of a thoroughbred kicking in his stall because of an injury -
which happened with Ruffian - Bramlage says most seem to calm down
when medical attention is being prepared ''because they are athletes
and they know something is wrong.''
Lukas led the colt into the operating room but left when the
anesthesia was given because ''I wanted to remember him as he was. I
didn't want to see them cutting him open and drilling
one-and-a-half-inch screws into him.''
As the two-and-a-half-hour operation began, the team made an
incision of roughly 14 inches in front of the bone. A bending press
was used to fashion a metal plate 12 inches long that would fit the
colt's sesamoid. The bone was scraped to make the plate fit. Then two
12-inch wires were placed along the surface of the bone to replace
the ligaments that had snapped during the accident.
Next, the team attached the plate with 20 screws, after drilling
threaded holes with a pneumatic instrument. The screws were also put
in place with a pneumatic instrument because ''that bone is extremely
hard, like ivory.'' After that, the incision was closed. And finally,
a cast was placed on the leg to protect the colt for the next 30-40
days.
''With a human being, you can say, 'Please stay in bed for six weeks
and then we'll give you crutches for six weeks,''' Bramlage says.
''With a horse, you only have 45 minutes and they've got to be
standing again.''
When Saratoga Six was revived, Lukas walked him to a recovery area.
The colt seemed calm, a good sign to the doctor, who says: ''It's
just like a patient. If you go into an operation scared, the first
thing you'll remember when you wake up is that you're scared. But
this colt was very well-behaved.''
Bramlage says he has done ''around 47 of these operations in the
last four or five years. We've had a 50 percent success rate, and
it's actually around 75 percent for horses with no damage to the
other foot and no damage to the blood supply.''
The doctor notes that his procedure costs at least $2,000, and that
many pet horses and work horses ''can be replaced for that amount of
money. You can't lose sight that with veterinary procedures there is
always an economic decision to be made with horses.'' And he adds
that the procedure, which is available to other veterinarians, is
aimed mostly ''for horses with breeding value.''
Lukas, who is busy preparing five horses for the Breeders' Cup
series, lost Landaluce, a promising 2-year-old, to a mysterious virus
before the 1983 season, and has fallen short of his expectations with
such horses as Codex, Marfa and Althea. He is not shy about talking
up the breeding potential of Saratoga Six, who had won all four
career starts, and he says, ''I just count my blessings that the colt
is alive.''
nyt-11-03-84 1255est
***************
∂03-Nov-84 1538 JMC
To: HST
Professor Thomas Kurtz
Kiewit Computation Center
Dartmouth College
Hanover, New Hampshire 03755
home (603) 448-1780
∂04-Nov-84 1725 JMC
To: HST
Sure, mail them.
∂04-Nov-84 1830 JMC
To: kdf@MIT-MC.ARPA
I don't even agree with your point 1, although I'm not in the office
where your article is available. The difficulty is that exerting a
force doesn't add energy if the velocity is in the opposite direction
to the force. If you apply a sinusoidal force to a frictionless system,
you succeed in adding energy until the amplitude is such that the
forces opposite to the direction of the applied force become sufficiently
important. The mechanical situation is just the same as the electrical
which you have been presumably obliged to study in the EE department.
Recall how impedance depends in frequency when there is capacitance
(analogous to springs), inductance (analogous to mass) and resistance
(analogous to friction). If there is no resistance, the impedance
becomes zero at the resonant frequency. At other frequencies, the
impedance is not zero and the amplitude of oscillation is finite.
The relevant fact for qualititative reasoning is that simple physical
intuition doesn't tell the truth about what will happen unless adjusted
to conform with the mathematics.
∂04-Nov-84 1834 JMC
To: SMC
Christian Newman called and will call again
∂05-Nov-84 0920 JMC ignorance, etc.
To: llw@S1-A.ARPA
CC: pmf@S1-A.ARPA
I can't come Tuesday or Thursday pm because of classes from 1:15 to 2:30.
Other days would be fine. However, because of having not received any
contract renewal, my badge has expired, so a special arrangement would
be necessary.
∂05-Nov-84 0924 JMC
To: llw@S1-A.ARPA, pmf@S1-A.ARPA
For example, this Wed or Fri or following Mon.
∂05-Nov-84 1145 JMC
To: stan@SRI-AI.ARPA
It turned out that Bell Labs might be willing to support her work
on this, and an application has been made. I forget when there is
supposed to be a reply. I take it that you might be interested,
and I'll mention it to her.
∂05-Nov-84 1243 JMC McCarthy's topic
To: Bernard←Galler%UMich-MTS.Mailnet@MIT-MULTICS.ARPA,
wilensky%ucbdali@UCB-VAX.ARPA
New generation computing, knowledge information processing and artificial
intelligence
The New Generation Computing system objective of moving to "knowledge
information processing systems" as proposed by Kazuhiro
Fuchi of ICOT has met quite general approval and motivates projects
all over this world. This lecture concerns
the problems of artificial intelligence that must be solved in order
to achieve the goal. The present rate of progress in achieving
computerization of common sense knowledge and common sense reasoning
isn't adequate to promise powerful knowledge information processing
systems by the 1990s. The lecture will identify some problems and
propose solutions for some and lines of research for others.
∂06-Nov-84 0959 JMC
To: RA
What are these workshops?
∂06-Nov-84 1756 JMC
To: RA
∂06-Nov-84 1631 RA Meeting at Xerox Wed., Nov. 7, @ 10:00
Doreen from Bill Spencer's office called to let you know that the meeting
which was scheduled for tomorrow must be rescheduled because the fellow
who was supposed to give the demonstration is going to be out. Two possibilities:
Mon. Nov 19, or Wed. Nov. 21. Please let Bill Spencer know what will be
convenient for you, or let me know and I will call his office.
Rutie:
Please check whether Monday the 19th is convenient for Mike Genesereth.
If so make the appointment and put it in my calendar file. Otherwise,
find out from III (Doris Hyde) whether the III board meeting has been
moved and if so to when. Actually please find that out anyway. Their
number is 212 390-8611.
∂06-Nov-84 1920 JMC Kurzweil postponed
To: genesereth@SUMEX-AIM.ARPA
I don't know if you got a message confirming our visit tomorrow but I
didn't. I did get a message postponing and proposing other dates.
Rutie will be in touch with you about the other times.
∂07-Nov-84 1241 JMC lisp history
To: HST
It occurs to me that you may not know about the file LSPARC.RPG[UP,DOC] here
at SAIL that includes Maclisp documentation going back to 1969. In
her thesis Carolyn is using it as a historical reference for CATCH and
THROW.
∂07-Nov-84 1329 JMC
To: jmc@S1-A.ARPA, llw@S1-A.ARPA, gap@S1-A.ARPA, pjb@S1-A.ARPA,
pmf@S1-A.ARPA
21st is fine with me for Mark IIB design review.
∂07-Nov-84 1337 JMC reprinting old lisp stuff
To: HST
I have no objection to your reprinting that material given that you
and your publisher think that the historical interest justifies reprinting
such scientifically obsolete material. I suppose I would like to see it
before it's printed just in case there are misprints or implicit references
to other documents that might lead to substantial misinterpretations.
∂07-Nov-84 2303 JMC
To: llw@S1-A.ARPA
The consultant agreement arrived a couple days ago and has been mailed.
I don't know the agenda of Jerry's meeting. I presume we'll receive
an advance agenda, but failing that we can MAIL pournelle@MIT-MC with
an inquiry.
∂08-Nov-84 0034 JMC
To: SG
We'll see you then.
∂08-Nov-84 0102 JMC re: Flame: What the Democratic Party did wrong (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - I am curious to know what part of the Constitution Jeff Mogul is
referring to and what conclusions he draws from it regarding the issues
of the election.
As to the opinion that voters agreed with Mondale about issues but voted
for Reagan, this refers to the "issues" defined by the pollsters. For
example, is inflation an issue? Now that it is much reduced, it isn't
perceived as an issue. Nevertheless, a voter who credited Reagan with
reducing inflation, might vote for him for that reason even though when
asked he would say that it wasn't an issue.
I agree with William Safire's column that says that if the Democrats
think they can win in the future with their present form of coalition
politics, they are mistaken. Sixty percent of the delegates to the
Democratic convention came from NEA, AFL-CIO and NOW. These groups
controlled the platform, but their activists scarcely represent even
the opinions of the people in whose interests they purport to speak.
Reagan had a majority among women and among blue collar workers; I
didn't see anything about how well he did among teachers. The Demmocrats'
problem is that members of these groups don't only think about
issues their self-styled leaders propose as dividing them from the
rest of the population.
I'm also puzzled by Jeff's contrast between "majority rule" and "rule
of law". The majority creates the law including those laws that
protect minorities from majority opinion. The two rules operate
mainly at different times.
Finally, is Jeff ashamed of being a white male, and, if Protestant,
is he ashamed of that?
∂08-Nov-84 1157 JMC
To: nilsson@SRI-AI.ARPA
Nils, do you know any work on formalizing prevention or impossibility?
Prevention
In the trash can problem we want to prevent the dogs from
overturning the trash cans, i.e. we want to make it impossible for
them to do it. It seems to be harder to express our common sense
knowledge about when something is impossible than knowledge about
what actions will achieve a goal. Anyway I don't know any work on
it.
∂08-Nov-84 1407 JMC incompleteness
To: minker@MARYLAND.ARPA
Is the new incompleteness result simply Davis's modified for
formula circumscription? Even if that's it, I'd like more
information about it - including who did it.
∂08-Nov-84 1409 JMC Wu
To: pratt@SU-NAVAJO.ARPA
I must have agreed to something suggested by Victor Kuo (Kuo@sail) (Guo Weide).
Anyway I'll let you know when I find when Wu will appear.
∂08-Nov-84 1446 JMC
To: bscott@SU-SCORE.ARPA
I'd like to find out what Chris's intentions are, but he's on a trip.
Would it would be convenient to do nothing till the week after next; his
secretary at Silma has promised to have him call me then.
∂08-Nov-84 1450 JMC
To: RA
Thursday morning any time between 10:30 and noon would be convenient to see Hogan.
∂08-Nov-84 1748 JMC
To: RA
Rutie:
Please tell Dutton I have a conflict with that date.
∂08-Nov-84 1801 Mailer failed mail returned
To: JMC
In processing the following command:
MAIL library
The following message was aborted because of a command error,
namely, nonexistent recipient(s):
library
------- Begin undelivered message: -------
∂08-Nov-84 1801 JMC
I will loan both for photocopying.
------- End undelivered message -------
∂08-Nov-84 1801 JMC
To: library@SU-SCORE.ARPA
I will loan both Lisp conference proceedings for photocopying.
∂09-Nov-84 0736 JMC
To: CLT
READ CONFER[2,2]/16P. Perhaps you should go.
∂09-Nov-84 1210 JMC
To: allen.pa@XEROX.ARPA
2:30 is fine with me - John
∂09-Nov-84 1210 JMC Confirmation of Kurzweil Demo Date
To: genesereth@SUMEX-AIM.ARPA
∂09-Nov-84 1159 Allen.PA@Xerox.ARPA Confirmation of Kurzweil Demo Date
Received: from XEROX.ARPA by SU-AI.ARPA with TCP; 9 Nov 84 11:59:38 PST
Received: from Chardonnay.ms by ArpaGateway.ms ; 09 NOV 84 11:59:25 PST
Date: 9 Nov 84 11:39:14 PST (Friday)
From: Allen.PA@XEROX.ARPA
Subject: Confirmation of Kurzweil Demo Date
To: JMC@SU-AI.ARPA
cc: Diebert.PA@XEROX.ARPA, Ritchie.PA@XEROX.ARPA, Spencer.PA@XEROX.ARPA,
Allen.PA@XEROX.ARPA
I believe that Monday, 19 November, works with everyone's schedule for
this much delayed demonstration. Can we make this for 2:30 p.m. as
Spencer had made a lunch commitment I wasn't aware of when I said 1:30.
Thanks,
Dorene
2:30 is fine with me - John
∂09-Nov-84 1250 JMC my talk
To: meseguer@SRI-AI.ARPA
Sorry about delay in title and abstract. I have to confess that I somehow
got this commitment mixed up with talking to CIS on the previous day.
There are several things I could talk about, and it really should depend
to some extent on who goes to those seminars. I'm tempted, imagining that
linguists are in the audience, to use the following title and abstract
Natural language from the Standpoint of Artificial Intelligence
Abstract: An intelligent individual, human or computer program, must
act on the basis of what it believes in advance modified by what it
observes and what it learns from linguistic communication. Thinking
about how the achievement of goals is helped by communication leads
to a somewhat different point of view from one derived mainly from
study of the corpus of spoken and written language. Namely,
1. Communication should be regarded as a modifier of state of mind.
2. The most basic form of communication is the single word sentence
uttered under conditions in which the speaker and hearer share enough
knowledge so that the single word suffices. The complete sentence
develops under conditions in which the speaker and the hearers share
less context.
3. Many of the characteristics of language are determined by so far
unrecognized requirements of the communication situation. They will
apply to machines as well as people.
4. An effort to make a Common Business Communication Languages for
for commercial communication among machines belonging to different
organizations exhibits interesting problems of the semantics of
language.
**** end of abstract
This talk would be somewhat off the cuff. If linguists aren't likely
to come, I could talk about "Modifying programs without reading them"
or about "Non-monotonic reasoning" or "Formalizing common sense knowledge".
Please phone me if you think something other than the natural language
talk would be better.
2:30 is the right time. If you decide to go ahead with this title and
abstract, let me and CSLI know.
∂09-Nov-84 1314 JMC disagreement
To: SJG
The first thing I disagree with is the title. It suggests that someone
claims that probability is never of use. Why not a more agressive title
like "Is there any useful non-monotonic reasoning without probability?"
or a neutral title like "The place of probability in nmr".
∂09-Nov-84 1430 JMC
To: meseguer@SRI-AI.ARPA
everything is ok as arranged
∂09-Nov-84 1436 JMC Lisp conference proceedings
To: library@SU-SCORE.ARPA
The 1980 and 1982 Proceedings are on the file cabinet in my office next
to the door. I assume you have the 1984 Proceedings.
∂09-Nov-84 1630 Mailer failed mail returned
To: JMC@SU-AI.ARPA
In processing the following command:
MAIL/su
The following message was unsent because of a command error:
------- Begin undelivered message: -------
∂09-Nov-84 1630 JMC >
To: CC.Clive@UTEXAS-20.ARPA
declination
I have to decline. I'm being invited to too many reminiscent events of
one kind or another. I figure that if I work instead I may be invited
to the 20th anniversary of some work done in the 1980s.
------- End undelivered message -------
∂09-Nov-84 1631 Mailer failed mail returned
To: JMC@SU-AI.ARPA
In processing the following command:
MAIL/su
The following message was unsent because of a command error:
------- Begin undelivered message: -------
∂09-Nov-84 1631 JMC >
To: CC.Clive@UTEXAS-20.ARPA
declination
I have to decline. I'm being invited to too many reminiscent events of
one kind or another. I figure that if I work instead I may be invited
to the 20th anniversary of some work done in the 1980s.
------- End undelivered message -------
∂09-Nov-84 1632 JMC declination
To: CC.Clive@UTEXAS-20.ARPA
I have to decline. I'm being invited to too many reminiscent events of
one kind or another. I figure that if I work instead I may be invited
to the 20th anniversary of some work done in the 1980s.
∂09-Nov-84 1706 JMC
To: trudy@SU-CSLI.ARPA
Accept for Dec 4 A.M. only.
∂09-Nov-84 1721 JMC NSF and education
To: faculty@SU-SCORE.ARPA, bscott@SU-SCORE.ARPA
NSF's position is ultimately determined by Congress. From time to time,
Congress has appropriated money for various educational programs to be
run by NSF. NSF will cheerfully provide any number of brochures on the
programs that exist and what Congressional initiatives they correspond
to. No doubt the relevant officials, e.g. Kent Curtis, will respond
to telephoned inquiries. Speculation is pointless.
∂09-Nov-84 1924 JMC re: Flame: What the Democratic Party did wrong (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - A free market reaches an optimal result in a definite technical
sense. As I remember it from an economics class of about 40 years ago,
it's something like this. Forget about money and imagine that goods
were directly exchanged. An exchange of goods is optimal if there are
no additional deals that all the participants in the proposed deal
would prefer. If you compute this kind of optimum it is a constrained
optimization problem and its solution involves the use of Lagrange
multipliers. These Lagrange multipliers can be identified as prices.
The theorem of mathematical economics is that (under addtional assumptions
that I forget) a free market results in an exchange that is optimal in
this sense, i.e. money and market prices play the role of setting up the right
complicated exchanges. The assumptions involve (among other things)
that the number of traders, producers and consumers of each commodity
is large.
The theorem doesn't assert that everyone will like the result. Someone
who can enforce a monopoly, e.g. by a domestic content law, may like
the result better than the result of freer trade. However, this is
at the expense of someone who would like to buy foreign goods at
a free market price rather than at the price determined by an artificial
scarcity. While the social utility of small deviations from market
prices may be argued, a practice, common when socialists have power,
of setting prices arbitrarily often has results that everyone regards
as bizarre. For example, it was part of Soviet doctrine to subsidize
the price of bread and to keep down the payments peasants received
for their wheat. The result was that the police had to enforce laws
against peasants buying bread to feed to their cows. (The peasants
weren't allowed to keep their own wheat for their cows). My opinion
is that equal stupidities would be likely if certain Stanford radical
faculty and students achieved enough power.
I don't think the economic theorem says it's not optimal to tax me to feed
the unfortunate or other groups you favor.
On another matter, while economic theory doesn't decide the matter,
Government intervention in energy has often turned out unfortunately.
For example, the synthetic fuel initiative seems to have been premature,
judging from the fact that OPEC can now sell less than half the oil
they were selling at the height of their power. Certainly economic
theory by itself doesn't say that new energy technologies will actually
be available. Moreover, even if they are available, there may be
temporary suffering and very high prices if a prompt start isn't made.
A few years ago Scientists and Engineers for Secure Energy (at my suggestion)
held a conference on what could be done in a crash program if there
were a total cutoff of Middle East oil. The assumption was that the
suffering would be sufficient to remove bureaucratic obstacles.
One of the speakers was Robert Nathan who had been head of the War
Production Board during World War II and who told about how various
emergency programs were carried out and how long they took. It turned
out that there was a lot that various people knew could be done.
Natural gas could substitute rapidly, nuclear power plants can
be operated at 20 percent above rated power for substantial periods,
and the expert said that new ones could be built in two and half
years even without World War II three shift operations.
My opinion, after study of the technlogical possibilities,
is that energy is assured for a few billion (yes billion) years
by technologies based on present science. Of course, some may
posture that they would prefer the demise of civilization to the
use of nuclear technology, but I think this posture would collapse
in face of the prospect of even a halving of the American standard
of living, e.g. going back to the standard of living of 1940 when
I was a child.
A hundred years ago, it not enough science was known to say that
civilization could survive far into the future, and Jevons's pessimism
about what would happen to English civilization when the coal ran
out was far more plausible even though he did underestimate the
amount of coal England had.
I can't do anything about people who will criticize me for having
an ideology of blind optimism, but I am willing to argue about
details for a certain sample of proposed disaster scenarios.
Returning finally to politics, I think Reagan's optimism is justified
by the state of technology, whether or not he knows any details.
I am also much encouraged that we will adopt necessary technology
when the pain of doing otherwise is high enough. Fortunately, there
isn't a world government, so stupidity in one country, be it the U. S. A.,
is eventually made clear by the success of the countries that
avoid that particular stupidity. For example, the French now
in 1984 are getting 55 percent of their electricity from nuclear
plants each of which comes in at a lower cost than the preceding ones.
I believe that they will be rewarded for their enterprise in the
next twenty years and as we are being punished for our stupidity.
We will learn from their experience. Of course, if we hadn't
been stupid about this and other matters our per capita income
might well be twice what it is and the number of poor a lot lower.
∂09-Nov-84 1944 JMC re: To Calif. Democrats, District 12 (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Christopher Stuart's view is one reason why the present system
of alliances the U. S. is involved in (and regrettably leads) may
not be able to persist. At one time Australia its alliance with us
seriously and helped South Vietnam defend such freedom as it had.
Now the view that these alliances are merely a matter of U.S. "hysteria"
is quite widespread, and this leads to competition among our "allies"
as to who can do the least. At a certain point the alliance will
be unsustainable, and each country will have to defend itself as
best it can. Many Americans, both liberals and conservatives, think
that NATO has reached this point, and that the Europeans would do
more for their own defense if the 300,000 American troops were withdrawn
from Europe. Of course, this might have consequences many would
consider undesirable, e.g.
(1) West German nuclear weapons. or
(2) Finlandization. For example, Finland has a policy of arresting
and giving back anyone who escapes from the Soviet Union. Not all
Finnish farmers act in accordance with this policy, however. or
(3) Some country losing its freedom, e.g. Yugoslavia or
Greece.
∂09-Nov-84 1945 JMC
To: stuart@SRI-AI.ARPA
jmc - Christopher Stuart's view is one reason why the present system
of alliances the U. S. is involved in (and regrettably leads) may
not be able to persist. At one time Australia its alliance with us
seriously and helped South Vietnam defend such freedom as it had.
Now the view that these alliances are merely a matter of U.S. "hysteria"
is quite widespread, and this leads to competition among our "allies"
as to who can do the least. At a certain point the alliance will
be unsustainable, and each country will have to defend itself as
best it can. Many Americans, both liberals and conservatives, think
that NATO has reached this point, and that the Europeans would do
more for their own defense if the 300,000 American troops were withdrawn
from Europe. Of course, this might have consequences many would
consider undesirable, e.g.
(1) West German nuclear weapons. or
(2) Finlandization. For example, Finland has a policy of arresting
and giving back anyone who escapes from the Soviet Union. Not all
Finnish farmers act in accordance with this policy, however. or
(3) Some country losing its freedom, e.g. Yugoslavia or
Greece.
[JMC - Forwarded to OTHER-SU-BBOARDS from line 92]
∂10-Nov-84 0006 JMC
To: SMC
I fixed your einit.cmd so as not to create a subjob.
∂10-Nov-84 0019 JMC
To: RPG
When you come back please phone.
∂10-Nov-84 1250 JMC proposal to spend Stauffer money on SUNs
To: csdfacilities@SU-PESCADERO.ARPA
Unfortunately, I can't vote for the SUN proposal without more discussion.
In my present opinion McSun is a far more cost-effective use of the
Stauffer money. The McSuns are not a replacement for Datadisk which
isn't broken. Instead they provide a cost-effective way, $1200 per
channel instead of $10K apiece, to access all the Department's computers,
because they will be on the Ethernet.
The Suns we have are still being used by many people as terminals to
VAXen.
∂10-Nov-84 1330 JMC re: Flame: What the Democratic Party did wrong (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Indeed it behooves people to plan. The more plans the better.
However, an official plan is another matter. Sometimes it's necessary,
but people with power or who aspire to power often do harm with
official plans. In our own field, I think the DARPA speech recognition
project run by the very best possible committee was ineffective compared
to an announced willingness by DARPA to consider unsolicited proposals.
Almost all effort in the country was chanelled into essentially one
approach, and that approach didn't turn out to be successful enough
to have warranted the suppression of other approaches. In energy
an official plan is likely to be dominated by political arguments
about who are the good guys and bad guys (do you like corporations)
struggles for monopoly, struggles for the preservation of ill-considered
tax advantages and by lawyers and publicists' science (Jeremy Rifkind,
Denis Hayes and Amory Lovins). Look at the windmills in Altamont Pass.
They don't work most of the time, and their racket makes it impossible
even to graze cows on the same land, but the tax credits are such that
the rich dentists, etc., who have invested in them, can't be sure they
aren't winning anyway.
∂10-Nov-84 1831 JMC (on TTY161 1831)
To: YM
What is date and time of your oral?
∂10-Nov-84 1857 JMC reply to message
To: YM
[In reply to message rcvd 10-Nov-84 18:49-PT.]
Thanks. Turns out I already had it in my calendar.
∂11-Nov-84 0148 JMC
To: SMC
What file lists the TEX files you have done for me?
∂11-Nov-84 1317 JMC
To: boyer@UTEXAS-20.ARPA
At what number may I phone you?
∂12-Nov-84 1445 JMC
To: SMC
Sarah, this appointment is apparently for you, because it has to do
with a '71 International truck.
∂12-Nov-84 1145 SMC phone message
I found a message that Zella took for you. You are asked to call someone
(no name given) at 247-5405 to confirm an appointment fo 12:30 on Tuesday.
∂12-Nov-84 1605 JMC
To: RPG
11-26 Monday 10am, MAD Computer with RPG
*Nafeh, Dr. John (56) 943 1711.
> MAD Computer, 2950 Zanker Road, San Jose, CA 95134
They'll call you with directions, and we should go separately.
∂12-Nov-84 2113 JMC McSun vs. workstation
To: csdfacilities@SU-PESCADERO.ARPA
I am inclined to agree that our positions are unlikely to change, since
they compare a larger number of McSuns to a smaller number of more
powerful workstations. Therefore, I propose that we split the money,
spend the workstation share now, and reserve the other half for some
reasonable length of time for McSuns. If the McSun gets unreasonably
delayed, then we should compare getting more workstations with
distributing ordinary terminals to grad student desks. While I
like Lisp machines, I can't really compare them with Suns for
the kind of thing most people do here. To what extent is the
connection of the original Sun development with Stanford
important, e.g. are many students feeling deprived of getting
the benefit of work they have done in the past.
∂12-Nov-84 2138 JMC Kahn
To: RPG
Sometime soon I would like to forward a comparison of machines looked
at to Bob Kahn. The idea is to let him know that we are being reasonably
diligent and to ask him if there are additional considerations from his
point of view that should influence the decision. I'm rather inclined
to forward your comparison together with some remarks of my own after
you have probed the one additional company.
∂12-Nov-84 2337 JMC
To: csdfacilities@SU-PESCADERO.ARPA
I agree with Cheriton proposal 1984 Nov. 12 23:14.
∂12-Nov-84 2339 JMC
To: cheriton@SU-PESCADERO.ARPA
When you mail to jmc and csdfacilities, I get two copies.
∂13-Nov-84 1807 JMC re: What are McSuns?
To: JDLH@SU-SCORE.ARPA
The McSuns are being designed and built under Len's direction, although
the original proposal is mine. They are bit map displays of relatively
high resolution that will sit on the Ethernet and be accessible by any
computer on it. A 68000 puts up characters and pictures on multiple
screens for multiple users. Their virtue is cost. McSuns are projected
to cost $1200 apiece compared to $10,000 for a workstation. In so far
as the workstations are used as terminals, the McSuns will be better.
The McSun goal is to put one on every desk in the Department.
[In reply to message sent Tue 13 Nov 84 08:13:24-PST.]
∂13-Nov-84 2110 JMC
To: CLT@SU-AI.ARPA, LES@SU-AI.ARPA
∂12-Nov-84 1522 RPG Multiprocessors
To: JMC@SU-AI.ARPA, JJW@SU-AI.ARPA, cheriton@SU-PESCADERO.ARPA
CC: rpg-q@SU-AI.ARPA
The following companies are `in the running' as far as I'm concerned:
Denelcor
Synpase
Encore
Symbolics
BBN
Sequent
Let me discuss the pros and cons of each:
Denelcor:
We all looked at the HEP and have a pretty good idea of what it can do.
Pros:
1. Friendly and co-operative staff: they believe in what we're doing;
2. cute architecture; and
3. they are giving DARPA a `good deal.'
Cons:
1. The hardware implementation of the architecture make it very
difficult to do a cheap, fast Lisp, and
2. Lucid is less interested in doing it than some of the others,
because it is probably not a `typical' multiprocessor, such as
we might see in the next 10 years in common use.
Encore:
I talked at length to Gordon Bell, Ike Nasso, Steve Amarant, and Henry Weigart
from Encore in Massachusetts. I think that of the bunch-of-micros-on-a-bus
machines, Encore's has the highest potential in terms of being relatively high
performance and expandable in the near term to more processors than either of
Synpase or Sequent. This is because they have a 100 megabyte per second bus,
with separate address, data, interrupt, and parity paths.
There are 4 different board types in the machine: processor, memory, IO, and
Ethernet. I saw a physical example of each board on the workbench, with
logic analyzers hooked up to each. That is, they have gotten board back and
are now debugging them.
They use NS32032's, and appear to have them; they are running at 6 Mhz instead
of the nominal 10 mhz. The memories and bus support 32-bit addresses, and the
memories are 4-way interleaved
Each processor board has 2 cpu's and a shared cache between them. The
cache is 32kbytes. The memories support test-and-set. In the future each
board will have 4 processors, and National is going to a 2.5x faster chip
in a year or so. They have a software group that it porting Berkeley 4.2
to it. They have several functional emulators (made up of NS32016 boards)
on which they are doing software development.
They claim that the first 2-processor (1 cpu board) machine will be put together
this week, and having seen the state of the hardware as of last week, I think it
might be possible. In addition, they claim that they could get us a prototype
machine mext April-June.
In order to do a Lisp on that machine, Lucid would have to port our Sun Lisp to
the 32032. I think this could be done using a uni-processor between now and next
summer. With the uncertainties of the DARPA contract, the Encore time schedule
might not be a problem.
Pros:
1. 32032 is a nice machine;
2. fast bus makes for future expansion;
3. Bell has a strong interest in the project and has been
associated with multiprocessors for many years;
4. demand paging;
5. Unix; and
6. Lucid would want to do it.
Cons:
1. the Encore schedule could slip; and
2. Encore may collapse, though I don't see this right away.
Synapse:
I agree with everything that Cheriton said, except for one. That is, I
think that Synapse is a little closer to the brink than he thinks. Synapse
shipped machines too early a while back, and the investors have slapped their
wrists. They may be on a prove-yourself-or-die cycle.
I want to emphasize that Synapse will have *no* interest in helping us
at all unless we convince them that there is a market they could attack
with a parallel Lisp. Otherwise, I think, they will proceed in the database
area until their investors are happy.
Pros:
1. 68000 Lisp is a simple port of Lucid's Sun Lisp;
2. basically a good architecture;
3. they will accept a purchase order today;
4. reliable hardware;
5. nearby; and
6. Lucid would want to do it.
Cons:
1. They will not help us in any way unless we change their
marketting strategy to include AI;
2. operating system is not something Lucid wants to do; and
3. perhaps they're not long for this world also.
BBN:
BBN appears to be about to get strategic computing money to put a Common Lisp
on the Butterfly along with a multiprocessing version of it. Poor lads, they
thought I (rpg) was going to do the Common Lisp for them for free. I talked to
a panic-stricken Don Allen last week, and he strongly suggests we use the
Butterfly. He and Randy Rettberg, a manager, are willing to fly out to Stanford
for a show-and-tell. They hope that we select the Butterfly and get DARPA to pay
for Lucid to put Common Lisp and the low-level primitives for Qlambda on the
Butterfly. They would then take the Lisp and the primitives and do something
different from Qlambda for DARPA.
It sounds like there are a few things wrong with their suggestion: 1), they are
suggesting that DARPA pay twice for some of the work, namely the multiprocessing
primitives, and 2), I think DARPA is more nearly sold on Qlambda than it would
be on a parallel Lisp to be specified later.
Nevertheless, the Butterfly is a 68000-based machine and it exists.
Pros:
1. 68000-based machine;
2. Sun Lisp easy to port;
3. butterfly switch is not so bad;
4. machine exists;
5. BBN *wants* to co-operate; and
6. Lucid is not against doing this.
Cons:
1. Unknown operating system; and
2. the switching network may need to be micro-coded to
handle a synchronization primitive.
Let me elaborate on this last point. Each cpu board has a micro-codable
memory-watcher. This memory-watcher implements the butterfly network, and there
is more control store in which to do other things. For example, references to
certain memory locations can be interpreted as other sorts of instructions
by this watcher. I'm not sure how much programming of it must be done.
Symbolics:
I talked to them about multi-processors, and I was surprised when they said that
they had proposed a multi-processor to Arvind, which would be a memory-bus or
multi-ported memory scheme. Another possibility would be a gigabyte serial
port amongst some 3600's. Symbolics volunteered to send out someone to Stanford
to chat about it.
Pros:
1. The 3600 already has a Lisp.
Cons:
1. I'm not at all sure whether they could deliver this in a
reasonable length of time.
Sequent:
They are left to look at. They have delivered, and use 32032's. Gordon Bell
says that except for a slower bus, they are similar to the Encore machine.
Tomorrow I am starting my sequence of inquiries which ought to land us some
information soon.
-rpg-
∂14-Nov-84 0119 JMC re: What are McSuns?
To: JDLH@SU-SCORE.ARPA
[In reply to message sent Wed 14 Nov 84 01:16:39-PST.]
Yes. Talk to Len about the present status. However, I am still
involved in proposing goals of the project.
∂14-Nov-84 0832 JMC Principals' meeting
To: ingrid@SU-CSLI.ARPA
Any of the proposed times on November 27 are feasible for me. However,
I dislike 8am, and 5pm might be a squeeze.
∂14-Nov-84 1124 JMC re: Couple of things
To: BSCOTT@SU-SCORE.ARPA
It turns out it wasn't a phone call. Chien was here and told Rutie. He
didn't mention a proposed start date. The start date in the proposal was
probably July 1, but then the proposal was quite late.
I think we should start on the consulting professor appointment, but
Gene must act. I believe Feigenbaum said he would initiate it, but
if he didn't say so or can't, I'll do it.
[In reply to message sent Wed 14 Nov 84 11:18:17-PST.]
∂14-Nov-84 1432 JMC re: Meeting 11/20/84
To: FISHER@SU-SCORE.ARPA
[In reply to message sent Wed 14 Nov 84 11:51:21-PST.]
10am on Nov. 20 is ok for me and in my calendar.
∂14-Nov-84 1831 JMC re: IP/TCP Update for Symbolics 3600's
To: Rindfleisch@SUMEX-AIM.ARPA
[In reply to message sent Wed 14 Nov 84 16:17:46-PST.]
I agree to pay my share, i.e. $4,050, and will try to charge it to DARPA. Please
ask my secretary, Rutie Adler, RA@SAIL, 7-2800 about charge number.
∂15-Nov-84 0933 JMC
To: jody@SU-CSLI.ARPA
∂25-Sep-84 0008 JMC Zingheim
To: ostrom@SU-CSLI.ARPA
Better ask Les Earnest at Imagen or Ted Panofsky at Machine Intelligence
Corp.
∂15-Nov-84 1137 JMC
To: RA
Please put more copies of Queue based multi-processing Lisp in my file cabinet.
∂15-Nov-84 1222 JMC
To: RPG
∂15-Nov-84 0934 RPG Russian Words
What do the following mean?:
notniye (spelled ``HOTH61E'')
fondahk (spelled ``FOHDAX'')
They are part of the title on a catalog of piano pieces, perhaps a catalog
of recordings.
-rpg-
notniye is an adjective referring to musical notes.
My Russian English dictionary gives examples
notniye pismo meaning musical notation
notniye bumaga - music paper
notnaya linyeka - musical line.
fondakh is the locative plural of the noun fond meaning fund or supply.
Therefore the word might mean "at the supply". It would help to have more
context.
∂15-Nov-84 1652 JMC reply to message
To: RPG
[In reply to message rcvd 15-Nov-84 15:03-PT.]
The title translates to something like
Printed Scores in the collections in the Lenin Library catalog.
and the second item to something like
The piano student-pedagogical repertory. This suggests it includes things
that students learning to be piano teachers study.
However, the Stanford library system includes people who know Russian
far better than I do.
∂15-Nov-84 1655 JMC re: Your Salary
To: BSCOTT@SU-SCORE.ARPA
[In reply to message sent Thu 15 Nov 84 15:02:51-PST.]
Thanks for catching it. I hadn't noticed yet, because I've been maintaining
a fairly large balance recently.
∂15-Nov-84 1656 JMC reply to message
To: GLB
[In reply to message rcvd 15-Nov-84 16:01-PT.]
Yes, those weights will be ok.
∂15-Nov-84 1751 JMC re: Terminal use (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - The solution is to have students in all Departmental offices for those
with offices and to lend home terminals to those who don't have them. This
way the requirement for public terminals will be minimized.
∂16-Nov-84 1435 JMC EKL to Digital
To: RA
CC: RTC, JJW, JK
[In reply to message rcvd 16-Nov-84 14:30-PT.]
The answer is yes, and Ross will help him do it, perhaps with additional
help from Joe Weening.
∂16-Nov-84 1444 JMC
To: JK, RTC, JJW
Sorry, I didn't use the reply macro correctly.
∂16-Nov-84 1430 RA
EKL program
Ron Burbak, who is a TV student in Colorado, wanted to know whether he can
get EKL program that you mentioned in your class @ Digital Corp. in
Colorado Springs. Please let me know, and I'll relay the inf. to him.
me know and I'll relay the information to him.
∂16-Nov-84 1822 JMC
To: RA
Pls. pub etc neel.1[let,jmc].
∂17-Nov-84 1608 JMC non-monotonic bibliography
To: RA
∂17-Nov-84 1252 GCOLE@SU-SCORE.ARPA Bibliography on Non-Monotonic Reasoning
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 17 Nov 84 12:52:43 PST
Date: Sat 17 Nov 84 12:52:24-PST
From: George Cole <GCOLE@SU-SCORE.ARPA>
Subject: Bibliography on Non-Monotonic Reasoning
To: JMC@SU-AI.ARPA
Dear Prof. McCarthy,
Last Wednesday, at the Law and Computers Seminar, you stated that you
would be glad to send out a bibliography on non-monotonic reasoning to any
attendee who was interested. I would be grateful if you would forward that
bibliography to me (GCole@Score).
George Cole
-------
Rutie,
Please do the following. Update the shelf list of the non-monotonic
reasoning library. Get the bibliography from non-monotonic conference,
compare the two and make a combined list. Mark in some way the items we
don't have. Send a copy to Cole and put others in the non-monotonic file.
If you want an electronic copy of the Perlis bibliography, a request to
PERLIS@MARYLAND will probably succeed.
∂18-Nov-84 1244 JMC re: presidential address
To: ENGELMORE@SUMEX-AIM.ARPA
[In reply to message sent Sun 18 Nov 84 11:08:07-PST.]
Well this makes two. I already had my secretary make a transcription.
My plan is to rewrite the article from scratch and I have
started. At the rate I'm going it will take a few months.
If that goes too slowly I'll edit the transcription.
∂18-Nov-84 1320 JMC Science Digest young scientists
To: golub@SU-SCORE.ARPA
Robert Bolles
Randall Davis
Reobert Ewald, Cray
Ira Goldstein, HP
Richard Greenblatt, LMI
Daniel Hillis, Thinking Machines inc
Anthony Lauck, D.E.C.
Douglas Lenat
Thomas McWilliams
Brian Reid
Bruce Shimano, Adept Technology
Harry Tennant, MCC
The nominator most likely to have designated the Stanford people is
Bill Miller. If you don't have it already, I'll bring in the magazine.
∂18-Nov-84 2207 JMC (on TTY20 2207)
To: CLT
How long is Samson and Delilah likely to be?
∂19-Nov-84 0751 JMC
To: VAL
I can get you a desk whenever you like.
∂19-Nov-84 0826 JMC
To: restivo@SU-SCORE.ARPA
misprints
Chuck, there are obvious misprints in your axioms from C. Morgan. For
example, the parentheses don't match. I think 1 and 3 should be as
follows:
1.] P(i(x,i(y,x)))
3.] P(i(n(y),i(y,x)))
∂19-Nov-84 1022 JMC neel or something
To: RA
I see that the letter, called seel.1, has two versions of his name. I
believe I put his letter in the box to be filed. Please check which
is right and fix it.
∂19-Nov-84 1119 JMC Wu not coming.
To: VRP@SU-AI.ARPA, ZM@SU-AI.ARPA, KUO@SU-AI.ARPA
I received a letter this morning saying that he couldn't come to
Stanford this trip, although he hopes to make it in May or June.
He sends regards to each of you. The return address is
Wu Yunzeng
33 Fanshaw Avenue
Yonkers, N.Y. 10705,
but there is no indication of how long he will be there.
∂19-Nov-84 1323 JMC
To: RA
reinfr.1[let,jmc]. Note promised enclosures.
∂19-Nov-84 1652 JMC re: Proposed Workshop
To: cheeseman@SRI-AI.ARPA
CC: aaai@SUMEX-AIM.ARPA
[In reply to message sent Mon 19 Nov 84 16:45:17-PST.]
Peter, I'm no longer President of AAAI; Woody Bledsoe is. Also note
the spelling of my name. However, as it happens, I'm still in charge
of support for workshops, so your message went to the right place.
The idea seems appropriate, but I would like to know who, besides
yourselves, is on the organizing or program committees and some idea
of whom you plan to invite.
John
∂19-Nov-84 2225 JMC
To: CLT
Please phone if you return before 1am.
∂20-Nov-84 1038 JMC
To: RWW
Yes, I would like to be on the list. Good luck with it. Work hard.
∂20-Nov-84 1217 JMC FOL interest mailing list
To: YK, RWW
Please enter me as jmc-lists@su-ai on that list.
∂20-Nov-84 1259 JMC Industry lecturers
To: faculty@SU-SCORE.ARPA, su-bboards@SU-AI.ARPA, LEP@SU-AI.ARPA
It is again time to solicit applications for people in industry to teach a
course in the Computer Science Department. There will be three lecturers
for 1985-86 and each will teach a course for one quarter. An application
should include a curriculum vitae and a course description, suitable for
inclusion in the 1985-86 Stanford catalog. The lecturers this year are
Clarence Ellis of Xerox, Joseph Halpern of IBM and Richard Waldinger of
SRI. Their courses are described in the current catalog. Applications
are due January 15 in order that lecturers can be selected in time for the
courses to be in the catalog. The course should be in a scientific
specialty of the lecturer, and areas of computer science in which the
regular CSD offerings can usefully be supplemented are especially welcome.
The chairman of the Industry Lecturer Committee is John McCarthy
and suggestions and applications should go to him.
∂20-Nov-84 1444 JMC re: jmc-lists
To: RWW
[In reply to message rcvd 20-Nov-84 13:05-PT.]
Marty made it for me. I think it has to be done by him.
∂20-Nov-84 1647 JMC re: Next Tuesday's meeting
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Tue 20 Nov 84 16:34:19-PST.]
What was the scope of the "original conception"?
∂20-Nov-84 1951 JMC re: Original conception
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Tue 20 Nov 84 17:16:16-PST.]
Oh, I imagined the original conception had something to do with the
areas being covered rather than with an amount of money.
∂20-Nov-84 2027 JMC
To: CLT
Coffee later?
∂21-Nov-84 0024 JMC re: Common Lisp
To: RTC
[In reply to message rcvd 20-Nov-84 22:13-PT.]
Me and Joe Weening anyway. Joe knows about the new EKL Manual. Please
get it reproduced as quickly as possible.
∂21-Nov-84 0034 JMC re: Social Impacts of Computing: Graduate Study at UC-Irvine (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - My impression is that this program will do some harm, though perhaps
not much. The following approximate quotation from an
article by Kling on electronic funds transfer exemplifies what bothers
me about this program. "Shall we allow this technology to be introduced
before we full understand its consequences?"
No one ever understands fully the consequences of any technology, and
the Kling article amounted to mere obstructionism. I suspect that this
is what this program will come to.
∂21-Nov-84 1016 JMC re: Industrial Lecturers
To: ullman@SU-AIMVAX.ARPA
[In reply to message sent Wed, 21 Nov 84 08:39:25 pst.]
We could use Nelson to teach a regular course, but so far the industrial
lecturers have taught courses of their own devising. If there is some
strong reason to use the industrial lecturer program to get regular
courses taught, there is no bureaucratic obstacle, but
it is also possible to have an outsider teach a regular course
without using the industrial lecturer program. The point of the
program is that we get the descriptions of the one shot courses
into the catalog. Where does Nelson work now?
∂21-Nov-84 2309 JMC reply to message
To: RA
[In reply to message rcvd 21-Nov-84 14:49-PT.]
Yes, the circumscription paper.
∂21-Nov-84 2314 JMC re: Symbolics 3600 Access
To: Rindfleisch@SUMEX-AIM.ARPA
[In reply to message sent Wed 21 Nov 84 17:00:55-PST.]
There is plenty of time available now that I have lost
Dick. They are already using the machines on an informal
basis. I am agreeable to their continuing to do so in
exchange for HPP paying maintenance costs during the period
of HPP use. I would just as soon leave them in the offices
they now occupy even though it is somewhat of a nuisance for
the occupants.
∂21-Nov-84 2326 JMC re: Hazeltine 1510 manual
To: LYNCH@SU-SIERRA.ARPA
[In reply to message sent Wed 21 Nov 84 15:05:59-PST.]
LOTS used them for a while, so try the LOTS office.
∂21-Nov-84 2330 JMC re: Seeing Ethiopia on a Marxist Diet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - I think David has a great idea, and if Canada invades Ethiopia
I think the U.S. should veto any Security Council resolution condemning
it.
∂23-Nov-84 0012 JMC re: Seeing Ethiopia on a Marxist Diet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Some comments.
1. To Joan Feigenbaum. There is an enormous difference between what
Americans are taught about the Soviet Union and what Soviets are taught
about America. What Americans are taught about the Soviet Union is
on the average somewhat more favorable than reality, while what the
Soviets are taught contains many deliberate lies. On the other hand
there is much more interest about America in the Soviet Union than
vice versa, so the number of well informed Soviets about America may
be larger than the converse. Apparently the most recent pilgrim to
the Soviet Union is Norman Mailer, but the revelations he reports
would not have surprised him had he bothered to read any of the
recent books on the Soviet Union by newspaper correspondents. In
order of merit these are the books by Robert Kaiser, Hedrick Smith
and David Shipler, but they're all good.
The ban on organized prayer in the schools is quite recent -
1960s. When I went to school prayers were common, but atheists
like me were tolerated.
2. Back in the days when the civilized countries were self-confident
about their civilization, it sometimes occurred that populations
were rescued from tyrants like Mengistu, but then these tyrants weren't
supplied with modern weapons and training by a superpower.
3. Apartheid per se isn't responsible for the relative prosperity
of South African blacks. However, maintaining the economic system
of capitalism and avoiding the corruptions of tribal dictatorship
has played a big role in it. The Zimbabwe blacks are the most
recent to pay a big economic price for the political power they
have won. The South African whites won't give up their black
servants as long as having servants is safe. Then there will probably
be a partition, i.e. apartheid, but it can't involve the whites
having 87 percent of the land and the blacks having the 13 percent that
is allocated to their "homelands" by the whites. The whites
will then have to do their own housework and mining.
∂23-Nov-84 1731 JMC
To: CLT
I had a gum abcess treated and expect to be somewhat under the weather.
∂23-Nov-84 1824 JMC tooth
To: RA
Please check with Stanford benefits on whether I have dental coverage, and
if so, what info I need to give the dentist about it.
∂24-Nov-84 2345 JMC re: Seeing Ethiopia on a Marxist Diet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Although I haven't audited courses on development economics
and have read only one book on the subject, I have read quite a few
articles. Even if I had read only the newspapers I would still be
inclined to comment, and I don't think David Cheriton should be censured
for stating his opinion either. If JMM wants to start a refereed
BBOARD on third world development it would be interesting, but I
fear it wouldn't have an interesting number of contributors. Moreover,
JMM's own opinions about famine coupled with supporting arguments would be more
interesting than his opinion that others expressing theirs is "almost
criminal".
The articles and book reviews I have read are sufficient to report that
David's opinions are in accordance with some opinions by professionals
in the field. Others differ. In particular, P. T. Bauer, head of the
London School of Economics has expressed opinions in which David
might take comfort in several of his books, which I list at the end of
this flame together with their locations in the Stanford Libraries.
I believe "Dissent on Development" is his best known work.
It argues that development aid has harmed the third world by creating
dependence.
Two of the three conditions to which JMM ascribes South African economic
success limit each other now that banks (including the World Bank and
the International Monetary Fund) are beginning to insist as a condition
for new loans that the economy be put in a state that makes repaying them
somewhat more likely. Namely,
" 2. Relative independence from colonial interference since about 1910
or so to develop own industrial base.
3. Relatively free access to western capital,credit markets etc.
.
.
.
If the superpowers were to stick to their own affairs, it might help some!"
I suspect that the basic disagreement between me and JMM, as between P. T.
Bauer and the conventional development economists, is the question of
whether the social values and the political and economic system of the
advanced capitalist countries are essential to prosperity. Many people in
former colonial countries regard these values and political and economic
system as mere relics of colonialism. This view is supported enthusiastically
by the Soviet Union and Marxists in all countries. Unfortunately, it isn't
easy to learn better from experience, because it is always possible to offer
the explanation that reductions in production after independence are
consequences of further exploitation of some sort.
My own opinion is that the best favor the advanced countries
can do the underdeveloped world is to advance the values of political
democracy, free trade and capitalism. In this respect the Reagan
Administration took the right position at a recent conference on
development. Moreover, we must reject the theory that the poverty
of the third world is a consequence of continued "exploitation" and
cite the abundant evidence to the contrary, i.e. that the prosperity
of a third world country is positively correlated with the extent
of its trade with the third world and with the extent of the activities
in the country of multi-national corporations.
To put the probable disagreement most sharply. I believe that India has
harmed its prosperity by its socialist schemes, by its restrictions
on foreign trade and investment and by its valuing government
officials more highly than businessment. Perhaps its continued
communal strife is also related to rejection of Western values.
(I also reject blaming the communal strife on the British).
I'll spare you this time a detailed comparison of the third world
activities of the U.S. and the Soviet Union, but I reject JMM's
equation of them.
Finally, perhaps I should spell out my previous wisecrack about Canada
as referring to the fact that David is Canadian. More seriously, it
is politically impossible for the U.S. to serve as world policeman.
There would have to be agreement and collaboration among the democratic
countries for international police actions to make sense. There was
such collaboration in Korea, and I thinks this substantially accounts
for the fact that the Communist invasion of South Korea was defeated
while that of South Vietnam was successful.
Here are the references to Bauer:
3) AUTHOR: Bauer, P T
3.1) PIONEERS IN DEVELOPMENT (New York : 1984.)
LOCATION: HD74.P56 1984: Green Stacks;
3.2) Bauer, P. T. REALITY AND RHETORIC (Cambridge, Mass. : 1984.)
LOCATION: HD82.B333 1984: Jackson Business (bl 11);
LOCATION: HD82.B333 1984: Green Stacks;
3.4) Bauer, P. T. EQUALITY, THE THIRD WORLD, AND ECONOMIC DELUSION (Cambridge,
Mass. : 1981.)
LOCATION: HB171.B323: Hoover;
LOCATION: HB171.B323: Meyer;
LOCATION: HB171.B323: Green Stacks; Food Research;
LOCATION: HB171.B323: Jackson Business (bl 11);
3.5) THE FIRST WORLD & THE THIRD WORLD : 1st ed. (Rochester, N.Y. : 1978.)
LOCATION: HF1411.F527: Meyer;
3.7) Bauer, P. T. WEST AFRICAN TRADE; (1963.)
LOCATION: HF3896.B3: Meyer;
3.8) Bauer, P. T. THE ECONOMICS OF UNDER-DEVELOPED COUNTRIES, (1957.)
LOCATION: HD82.B332: Meyer;
3.9) Bauer, P. T. ECONOMIC ANALYSIS AND POLICY IN UNDERDEVELOPED COUNTRIES,
(1961.)
LOCATION: HD82.B33: Meyer;
3.10) Bauer, P. T. DISSENT ON DEVELOPMENT; (Cambridge, Mass., 1972.)
LOCATION: HD82.B328: Meyer;
3.11) Institute of Economic Affairs, London. TWO VIEWS ON AID TO DEVELOPING
COUNTRIES, (1968.)
LOCATION: HC60.I47: Meyer;
3.12) Bauer, P T OVERCOMING WORLD POVERTY. (Boston, Mass., 1981.)
AUTHOR: Bauer, P. T. (P`eter Tam`as)
TITLE: Dissent on development; studies and debates in development economics
[by] P. T. Bauer.
IMPRINT: Cambridge, Mass., Harvard University Press, 1972.
550 p. 23 cm.
LOCATION: HD82.B328: Meyer;
TOPICS: 1. Economic development.
2. Underdeveloped areas.
∂25-Nov-84 0059 JMC re: JMM's and your remarks on attaining the "Correct" view of Ethiopia
To: FOGELSONG@SU-SCORE.ARPA
[In reply to message sent Sun 25 Nov 84 00:24:18-PST.]
∂25-Nov-84 0106 JMC re: JMM's and your remarks on attaining the "Correct" view of Ethiopia
To: FOGELSONG@SU-SCORE.ARPA
[In reply to message sent Sun 25 Nov 84 00:24:18-PST.]
Sorry. I must have misused the REPLY macro. I'll try again.
In brief, I think that as human beings we find ourselves in
a rather hopeless situation when trying to find out the "true nature
of the world."
I suppose you would qualify this remark if it referred to
a common sense question such as whether the computers in MJH are
currently operating or a scientific question of whether heavy and
light objects fall at approximately the same speed. At one time
biology and medicine were as confused as politics and political
economy are today. While many controversies remain, there is
agreement among the learned about many questions and even wider
agreement about the nature of the remaining controversies. My
view is that politics and sociology are even more complicated
than the natural sciences and will take longer to understand.
There many important question on which there is no consensus
even among intelligent and experienced professionals.
Moreover, the objective investigation of these questions
is hindered by their involvement in struggles for power among
groups in all societies. This means that even where there is
good objective evidence about some point, it may still not be
accepted by people on the other side of a power struggle.
Nevertheless, the consensus about a few matters is quite
strong among people free to read a variety of opinions and form
their own. These include
(1) the relative attractiveness of strongly different
societies among people free to migrate and experience them.
(2) the preference for personal freedom.
(3) that war is a disaster. This isn't quite a truism;
before World War II, there were many writers and politicians
who praised war for its effect on character.
∂25-Nov-84 1442 JMC re: Discussion on bboard
To: JMM
[In reply to message rcvd 25-Nov-84 14:19-PT.]
Sure. Sorry I just missed you, because this afternoon would have been
a good time. This evening or tomorrow evening is also possible, and
Wednesday evening might be. Otherwise, I'm tied up this week including
next weekend.
∂25-Nov-84 1515 JMC capitalist hearts
To: su-bboards@SU-AI.ARPA
.
.
.
Humana Inc. owns 89 hospitals in the United States and abroad. The
company, based in Louisville, hopes the artificial heart program and
its recently begun heart transplant program will increase its
visibility.
''If we're successful, we'll attract patients from all over the
world,'' said David Jones, chairman of the profit-making company, in
a videotaped interview shown to reporters Sunday.
''There are good business reasons for participating in this
program,'' he said.
*****
Here's a classic case for us capitalism fans. The University of
Utah chickened out and a profit-making firm took over. I hope
Schroeder survives and that Humana makes large profits from its
enterprise. What do you say, socialists?
∂25-Nov-84 2139 JMC re: Seeing Ethiopia on a Marxist Diet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - To be specific, India has got a far smaller share of the world
manufacturing market in electronics, consumer products generally,
and other manufactured products than its level of manpower, capital
and scientific and engineering skills would warrant. The countries
that have been less protectionist and regulatory have done far better -
not merely in proportion to population but absolutely.
For example, one of the products whose import restricition Sankar
praised is computer terminals. He seems to believe that the
way to build industry is to restrict imports. When I visited Taiwan
in 1978 and visited a certain company I suggested that they build
computer terminals on the grounds that the technology was similar
to that of the black and white TVs they already built and that
terminals were a commodity almost without brand loyalty. When I
visited that company again in 1983, the first thing anyone said
was, "Did you know that Taiwan now makes 1/3 of the computer
terminals in the world?" They didn't get there by bureaucrats
cleverly deciding what imports to restrict.
I believe that India lost by throwing out IBM, although certain
local manufacturers may have gained something. I also believe
the U.S. loses by restricting Japanese cars. As for Coca-Cola,
it's hard to tell what benefits to ascribe to soft drinks, although
people certainly like them, especially in hot climates. The Indians
An Indian company (whom did it pay off for the privilege) produced
a drink called Campa-Cola with such a good imitation of the Coca-cola
logo that I bought a bottle under a mistaken impression. I didn't
like it at all, and I suspect it wouldn't survive free competition
with Coca-Cola.
India is sovereign and can make any restrictions it wishes. However,
I suspect that the nationalist arrogance of Indian intellectuals has
resulted in the death of a large number of Indian children whose
parents would otherwise have jobs.
I knew someone would ask me to define "western values", and I put off
writing the required micro-essay. Fortunately, JMM has done a good
job of it earlier on this page. My only additional remark concerns
the doctrine of "white man's burden".
No-one holds this doctrine - that the whites should rule
the East for the benefit of its inhabitants - in the forms
so poetically expressed by Kipling.
However, there is a modern version of the "white man's burden" doctrine
that is very harmful and is often expressed at conferences held
here in the U.S. Someone will say, "What should we Americans
do about the population problem of India or Bangladesh?" Elaborated,
the doctrine asserts that the Western countries, especially the
U.S., is responsible for curing the ills of the third world. Not
only is the doctrine prevalent among U.S. academics, leftists and
government officials, but pervades the speaches of "third world
spokesmen" at the U.N. and other international organizations.
It is used by politicians from these countries to blame their
lack of success and the consequences of their blunders on the
West. One will find a large negative correlation between the
amount accusation that the U.S. or the multi-national corporations
are responsible for a country's problems and its ability to
make progress.
There are more points I might answer, but I won't.
∂26-Nov-84 1404 JMC re: Veronica Dahl's talk
To: RA
[In reply to message rcvd 26-Nov-84 10:50-PT.]
Veronica Dahl's talk concerned "The logic of natural language" and
we should pay an honorarium of $200 out of DARPA money. It was also
university related if there should be a problem about paying with
DARPA money.
∂26-Nov-84 1406 JMC
To: RA
Make bill to MAD computer for 2 and a half hrs - today.
∂26-Nov-84 1408 JMC re: John Nafeh
To: RA
[In reply to message rcvd 26-Nov-84 13:50-PT.]
Schedule Nafeh 1pm next Friday.
∂26-Nov-84 1410 JMC re: nonmonotonic biliography
To: RA
[In reply to message rcvd 26-Nov-84 10:44-PT.]
There is a copy of the conference proceeding in the back of the non-monotonic
library. The bibliography (by Perlis) is the last paper. Yes, please
add the listed papers. I forgot what I wanted from Perlis, but you can
ask him to computer MAIL a copy of the bibliography.
∂26-Nov-84 1529 JMC desk
To: VAL
CC: RPG, RA
Dick Gabriel's desk in room 360 is now available for your use.
He will move his things shortly but will sometimes around in the
evenings finishing his LISP timing report and use that or some other
terminal.
∂26-Nov-84 1622 JMC
To: RG@SU-AI.ARPA
I presume this means your Stanford office.
∂26-Nov-84 2110 JMC re: AI info for annual report
To: HULLMAN@SU-SCORE.ARPA
[In reply to message sent Mon 26 Nov 84 19:47:57-PST.]
I don't work on expert systems, and I don't write surveys about anything,
so I guess not.
∂26-Nov-84 2321 JMC re: Seeing Ethiopia on a Marxist Diet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - This is all a country is justified in doing - at least the first
fifty times it happens.
TOKYO (AP) - The Foreign Ministry called in a Soviet Embassy
official to protest Friday's violation of Japanese air space by
Soviet aircraft, the ministry said.
Takehiro Togo, counselor of the ministry's European and Oceanic
Affairs Bureau, conveyed Japan's protest Monday to Lyudvig A.
Chizhov, minister-counselor at the Soviet Embassy, the ministry said.
Defense Agency reports said two Soviet Tu-95 bombers violated
Japanese air space off Japan's southernmost main island of Kyushu on
Friday, and 34 Japanese warplanes scrambled against the intrusion.
The intrusion was the second in 11 days. A Tu-16 bomber, part of a
nine-bomber convoy, entered Japanese territory in the same area Nov.
12, and the ministry protested the Soviet violation on Nov. 15.
∂27-Nov-84 0905 JMC re: books
To: BERG@SU-SCORE.ARPA
[In reply to message sent Tue 27 Nov 84 08:21:29-PST.]
I plan to use a collection of papers that will have to be reproduced.
It isn't quite ready.
∂27-Nov-84 1312 JMC
To: SMC
Push 73# on the phone to cancel call-forwarding.
∂28-Nov-84 0041 JMC re: Resume Joke in "Miss Peach" (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Many here don't. Even more don't know how to start writing
on the blackboard at the extreme upper left.
∂28-Nov-84 0045 JMC re: IBM in India (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Perhaps some Indian can correct me, but as I recall it, IBM did
indeed sell then outmoded models in India. The reason was that they
weren't allowed to get dollars for more modern computers, so they
sold the old ones for unconvertible rupees.
∂29-Nov-84 1051 JMC Japanese Lisp machines
To: feigenbaum@SUMEX-AIM.ARPA
Thanks for the copy of the report. Please ask your group to keep me
on the list for anything similar that comes out. I have as a visitor
for the year Shigeki Goto who was one of the main designers of the ELIS
at NTT and also TAO. He would be glad to give a talk on this subject.
Was their any discussion of the Japanese machines becoming commercially
available. I was glad to hear that FLATS is now debugged and in use,
because I have a high regard for Eichii Goto.
∂29-Nov-84 1110 JMC teaching a course next year
To: rwg@MIT-MC.ARPA
Would you be willing to teach a course in symbolic computation in
mathematics at Stanford next year? This would be part of our industry
lecturer program. These courses are scheduled in January of the
preceding year so that the course description can be put in the
catalog. You can make the content what you please and assume what
prerequisites you please. I would need a course description suitable
for inclusion in the catalog some time in January as well as your agreement
to teach the course. The pay is min($3K,1/16 annual salary); at least
that is what it was this year. I believe we could round up the resources
to give the students access to MACSYMA, although we might have to limit
the size of the class.
What is your preliminary reaction?
∂29-Nov-84 1117 JMC teaching a course next year
To: rwg@MIT-MC.ARPA
I forgot to say that the course would be for one quarter. If you agree
quickly you can have your choice of quarters. However, if you don't
care, I'd prefer to keep the flexibility for our other two industry
lecturers. The three this year are Clarence Ellis from PARC, Joe
Halpern from IBM and Richard Waldinger from SRI.
∂29-Nov-84 1522 JMC Japanese lisp machine speeds
To: RPG
The Fujitsu alpha is said to be 3 times faster than 3600 and 5 times
faster than 2060 Maclisp. Do you believe that, and does it correspond
to a plausible ratio of speeds between 3600 and 2060?
∂29-Nov-84 1810 JMC special seminar
To: faculty@SU-SCORE.ARPA
Monica M. Strauss from MIT Laboratory for Computer Science will give
a special seminar on "Computer Science Research Environments in the
US and Japan", on Friday November 30, at 3:00 pm, room 301.
∂29-Nov-84 1814 JMC change of mind
To: MS
CC: RA
I have changed my mind and will contribute $150 as an honorarium to your
proposed visitor from Vancouver. Tell Rutie the details and ask her to
check on whether something extra has to be done to pay an honorarium to
a Canadian. It will come from the DARPA contract unless there is a
foreigner problem, in which case my unrestricted.
∂30-Nov-84 0014 JMC
To: llw@S1-A.ARPA
I probably won't go, but if I do I'll ride back with you.
∂30-Nov-84 0629 JMC
To: RA
We will pay $100 honorarium to Monica Strauss for seminar.
∂30-Nov-84 1109 JMC Goad
To: bscott@SU-SCORE.ARPA
I was about to call him when it occurred to me to ask you whether his
status would affect NSF support of his participation in the joint
study with the Japanese. Both he and Jussi Ketonen are included in
the plan, and Sato talked to both, and they both want to participate.
Perhaps it would be simplest to extend Chris's leave for another year.
What do you think?
∂30-Nov-84 1115 JMC
To: feigenbaum@SUMEX-AIM.ARPA
Monica M. Strauss from MIT Laboratory for Computer Science will give
a special seminar on "Computer Science Research Environments in the
US and Japan", on Friday November 30, at 3:00 pm, room 301.
∂30-Nov-84 1409 JMC
To: bscott@SU-SCORE.ARPA
OK, terminate Chris.
∂30-Nov-84 1531 JMC re: Program Summary
To: OHLANDER@USC-ISI.ARPA
[In reply to message sent 26 Nov 1984 14:03-EST.]
Sorry about the delay in replying.
1. I'll get to the summary of research shortly - this weekend I hope.
2. I am expecting Dick to tell me what he wants to do and expect to
agree to it.
3. The problem didn't concern parallel Lisp. We (including Dick) are
pursuing that in discussions with parallel processor companies and
with Bob Kahn. It concerned budget but seems to be resolved for now.
∂30-Nov-84 2001 JMC Workshop on probability and uncertainty in AI
To: cheeseman@SRI-AI.ARPA, aaai-office@SUMEX-AIM.ARPA,
bledsoe@UTEXAS-20.ARPA
This is sufficient information, and I agree that AAAI should support
the workshop. I probably won't participate but might if I can
formulate a position on where probabilistic reasoning fits into
a more general picture of non-monotonic reasoning.
∂30-Nov-84 2005 JMC Matt Ginsberg
To: cheeseman@SRI-AI.ARPA
He is quite energetic and a promoter of probabilistic as opposed to
other methods of non-monotonic reasoning. SJG@SAIL.
∂01-Dec-84 0025 JMC technical reports
To: berg@SU-SCORE.ARPA
∂30-Nov-84 2348 KROVETZ@SUMEX-AIM.ARPA technical reports
Received: from SUMEX-AIM.ARPA by SU-AI.ARPA with TCP; 30 Nov 84 23:48:42 PST
Date: Fri 30 Nov 84 23:49:03-PST
From: Bob Krovetz <KROVETZ@SUMEX-AIM.ARPA>
Subject: technical reports
To: jmc@SU-AI.ARPA
Professor McCarthy,
Some time ago I made some inquiries about who could be contacted
for ordering tech reports via the net, and which sites maintained
online bibliographies. Diana Hall told me that a bibliography of
AI lab reports existed under [BIB,DOC]AIMLST. At the time I was
able to ftp this file, but since then the directory appears to
have been protected. On my primary machine (NLM-MCS, the National
Library of Medicine) I can get a listing of the directory, but I'm
unable to transfer any files. On SUMEX I can't even list the directory!
Is there anything you can do about this?
Thank you,
Bob Krovetz
-------
∂01-Dec-84 1120 JMC re: IBM
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Sat 1 Dec 84 10:19:02-PST.]
The Winograd I know at Yorktown is Shmuel Winograd, sometimes called
Sam Winograd. He is the head of their Mathematics Department and
Ralph Gomory (not Gomery) relies on his opinion. Gomory's position
at IBM is probably stronger than Pake's at Xerox. I'll check my
IBM phonebook to see if there is a plausible Joe Winograd.
∂01-Dec-84 1122 JMC
To: nilsson@SU-SCORE.ARPA
There is also Richard Fateman at Berkeley.
∂01-Dec-84 1745 JMC re: IBM
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Sat 1 Dec 84 17:13:01-PST.]
I don't think I know him well enough to be helpful; e.g. I don't recall
ever having a private conversation. I also don't know CSLI well enough
to be an advocate. I think the IBMers you have talked to will probably
do better. Someone I do know well enough and who is influential is
John Cocke, who comes out here fairly often. I could introduce you, but
you would have to do your own persuading.
∂03-Dec-84 1143 JMC
To: golub@SU-SCORE.ARPA
What about Jeff Ullman or some other prolific textbook writer?
∂03-Dec-84 1145 JMC
To: golub@SU-SCORE.ARPA
Don Knuth's Art of Computer Programming is also a candidate for education honors.
∂03-Dec-84 1208 JMC Dyson review
To: RA
Please send cc of my Dyson review to:
Professor Freeman Dyson
Institute for Advanced Study
Princeton, New Jersey
Professor Sidney Drell, SLAC
Professor Sidney Hook, Hoover
Miss Midge Decter, Committee for a Free World (see phon)
∂03-Dec-84 1346 JMC
To: RA
∂03-Dec-84 1245 RA mileage
Any idea how many mile round trip Livermore-PA?
I've been claiming the standard amount, and I guess it's 100 miles,
but I guess you better ask Dick Gabriel if you can't find it somewhere.
∂03-Dec-84 1349 JMC reply to message
To: GOLUB@SU-SCORE.ARPA
[In reply to message sent Mon 3 Dec 84 13:32:11-PST.]
No, I have neither the time nor the eloquence.
∂03-Dec-84 1412 JMC reply to message
To: RA
[In reply to message rcvd 03-Dec-84 13:56-PT.]
Make it page 3 of the existing file chudno[f84,jmc].
∂03-Dec-84 1723 JMC
To: ZM@SU-AI.ARPA, YM@SU-AI.ARPA, waldinger@SRI-AI.ARPA
malach[f84,jmc] Concerning Yonatan Malachi's thesis
On the basis of the oral examination December 3, I would like
to see the following questions treated in thesis.
1. There should be a better example to illustrate those cases in
which writing in TABLOG is more convenient than writing in LISP.
I think such examples can be found that go beyond the fact that
multiple output functions don't exist in basic pure LISP. Common
Lisp and even the Maclisp within which TABLOG is written do allow
multipile output functions although their notation for them is
rather barbarous. Here is quicksort in the notation that I
currently use in my classes.
quicksort u ← if n u ∨ n d u then u
else [λv1 v2.quicksort v1 * [a u . quicksort v2][filter[d u,a u]]
where * is an infix standing for append.
filter[u,x] ← if n u then values[nil,nil]
else [λv1 v2.if x ≤ a u then values[a u . v1,v2] else values[v1,a u . v2]]
[filter[d u,x]].
2. The initial propaganda for TABLOG makes much of the fact that TABLOG
programs have both a logical and a procedural interpretation. Yet little
seems to be known about the relation between the two, although it is
clear that they correspond in the simple examples given. The thesis
should give some sufficient conditions for the correspondence.
Of course, this requires settling on some definite interpreting algorithm
or discussing the correspondence for several.
In the oral we discussed the pure Lisp subset of TABLOG. There
the results of Vuillemin and Cadiou showing that the call-by-name
interpreter gives the least fixed point of the functional equation and
the call-by-value interpreter gives the least fixed point of a "strictification"
of the functional equation.
In EKL today we introduce Lisp functions as definitions (not
yet in the most general case). This means we must prove the existence
of a least function satisfying the functional equation, and there are
axioms permitting this for a wide class of recursion schemes - soon
to be extended to arbitrary Lisp functions. This means that EKL takes
the responibility for the fact that an object satisfying the functional
equation exists. Can the like be done for TABLOG?
3. There should be a TABLOG interpreter written in TABLOG in the thesis
for the thesis readers' inspection. This requires a Lisp-like internal
notation for TABLOG programs, but perhaps the notation used by the
interpreter written in Maclisp will do.
4. TABLOG achieves its extended language over Lisp and probably over
Prolog at the cost of giving up a well-defined correspondence between
the function computed by the program and the equations defining the
program interpreted logically. The thesis should discuss whether some
such relaxation is required by the task of extending Lisp and Prolog
or whether this is just a mistake in the design of TABLOG.
5. Whether the other attempts to combine Lisp and Prolog have the same
deficiencies should be discussed.
∂03-Dec-84 1853 JMC re: Dinner?
To: SMC
[In reply to message rcvd 03-Dec-84 18:51-PT.]
Not tonight.
∂03-Dec-84 2122 JMC
To: CLT
12-14 4pm, Fateman there, 519 Evans
∂03-Dec-84 2127 JMC Fateman conference
To: DAC
I didn't realize that his conference is that same summer - in July in Waterloo.
I suggested to him that some papers would appropriately be given in both
conferences, since the audiences will be quite different.
Incidentally, he was asked about your conference by someone
who thought he was on its program committee, and he hadn't agreed to it,
considering that he was General Chairman of a competing conference.
I will visit him at the end of next week on Friday, so I can carry
any messages.
∂03-Dec-84 2138 JMC
To: DAC
Wrong file
∂03-Dec-84 2138 JMC wrong file
To: DAC
I mentioned the wrong file. The one I mentioned contained just notes
for some earlier conversations. The actual comments on your proposal
are contained in the file CHUDNO.3[LET,JMC] which was a draft of a
letter to you. The new one also contains a slight revision of your
draft.
∂03-Dec-84 2147 JMC
To: DAC
The file is now chudno.3[f84,jmc].
∂03-Dec-84 2322 JMC committee decision
To: csdfacilities@SU-PESCADERO.ARPA
Please review the entire proposal including what is proposed to
be spent on other things. Also I would like to take an informal
poll in the Department lunch tomorrow to see what pinches the
most in the way of facilities. I'm sorry not to have the whole
situation more firmly in mind.
∂04-Dec-84 2136 JMC re: faculty liaison
To: TAJNAI@SU-SCORE.ARPA
[In reply to message sent Tue 4 Dec 84 14:43:34-PST.]
What city in Idaho?
∂05-Dec-84 0014 JMC [Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>: [Andy Freeman <FREEMAN@SUMEX-AIM.ARPA>: [Glenn Adams <glenn@ll-xn.ARPA>: Re:
To: les-here
∂22-Oct-84 0919 FEIGENBAUM@SUMEX-AIM.ARPA [Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>: [Andy Freeman <FREEMAN@SUMEX-AIM.ARPA>: [Glenn Adams <glenn@ll-xn.ARPA>: Re:
Received: from SUMEX-AIM.ARPA by SU-AI.ARPA with TCP; 22 Oct 84 09:19:02 PDT
Date: Mon 22 Oct 84 09:22:19-PDT
From: Edward Feigenbaum <FEIGENBAUM@SUMEX-AIM.ARPA>
Subject: [Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>: [Andy Freeman <FREEMAN@SUMEX-AIM.ARPA>: [Glenn Adams <glenn@ll-xn.ARPA>: Re: Tightly-coupled multiprocessor UNIX]]]
To: jmc@SU-AI.ARPA
Mail-From: DELAGI created at 20-Oct-84 07:36:24
Date: Sat 20 Oct 84 07:36:24-PDT
From: Bruce Delagi <DELAGI@SUMEX-AIM.ARPA>
Subject: [Andy Freeman <FREEMAN@SUMEX-AIM.ARPA>: [Glenn Adams <glenn@ll-xn.ARPA>: Re: Tightly-coupled multiprocessor UNIX]]
To: nii@SUMEX-AIM.ARPA, brown@SUMEX-AIM.ARPA, feigenbaum@SUMEX-AIM.ARPA,
davies@SUMEX-AIM.ARPA, andy@SU-SCORE.ARPA, rpg@SU-AI.ARPA
Based on my work experience with Dave Rogers and Dave Schanin (the guy I
imagine who is responsible for the ideas in Encore's machine) the Sequent
machine seems worth knowing more about. (I mean no derogation of Schanin:
it's just that Rogers -- at least when I knew them both -- was the deeper
technically, ranged over a broader set of design issues, and had better
human interaction skills. If these qualities showed up in the design of
the Sequent, it's a machine we should understand better)......./bruce
---------------
Mail-From: FREEMAN created at 17-Oct-84 14:57:51
Date: Wed 17 Oct 84 14:57:50-PDT
From: Andy Freeman <FREEMAN@SUMEX-AIM.ARPA>
Subject: [Glenn Adams <glenn@ll-xn.ARPA>: Re: Tightly-coupled multiprocessor UNIX]
To: delagi@SUMEX-AIM.ARPA
Properly programmed, this may be a decent multi-processor simulator.
-andy
---------------
Return-Path: <unix-wizards-request@BRL-TGR.ARPA>
Received: from BRL-TGR by SUMEX-AIM.ARPA with TCP; Wed 17 Oct 84 14:49:57-PDT
Received: from ll-xn.arpa by BRL-TGR.ARPA id a006584; 17 Oct 84 16:45 EDT
Received: by ll-xn.ARPA (4.12/4.7)
id AA24683; Wed, 17 Oct 84 16:45:53 edt
Date: Wed, 17 Oct 84 16:45:53 edt
From: Glenn Adams <glenn@ll-xn.ARPA>
Message-Id: <8410172045.AA24683@ll-xn.ARPA>
To: david@COLUMBIA-20.ARPA
Subject: Re: Tightly-coupled multiprocessor UNIX
Cc: ted@COLUMBIA-20.ARPA, unix-wizards@BRL-TGR.ARPA
Dave,
I would call your attention to the recently released Balance 8000
system from Sequent Computer Systems, Inc., Portland, OR. They have taken
the 4.2BSD system and modified it extensively to operate in a tightly
coupled environment using a processor pool architecture. Their system uses
the 32XXX series of CPUs from National Semiconductor with the FPU and MMU
from that same series. In addition, they add an 8-KByte, 2-way, set
associative cache and a high speed 8 Kbyte local memory used for read-only
kernel data. Two autonomous CPUs are packaged per board with up to twelve
supported in the system. A custom VLSI chip, the System Link and Interrupt
Control (SLIC), handles interrupt handling contention and implements mutual
exclusion semaphores.
A distributed arbitration, 40 Mbyte/sec, synchronous bus (an enhanced VAX SBI?),
handles communication between processors, memory, a multibus adapter, and a
SCSI Bus Adaptor/Ethernet Module. A separate synchronous serial bus ties each
SLIC together, each unit on the main bus possessing a SLIC.
Software-wise, you get a 4.2BSD system with its Virtual Memory system completely
replaced with one of Sequent's design. A new model was employed in its design
that results in processes paging against themselves, rather that against the
system at large, etc.
Their performance figures show near linear performance up to the full complement
of twelve processors over a variety of benchmarks. They are currently shipping
with 10Mhz 32016's (half a VAX750), but soon hope to be using the 32032 as soon
as they can get reliable, i.e., full temperature range, chips from National.
With the latter, the system is fully worth a dozen VAX750s.
If some of the above sounds VAXEN like, or an improvement thereof, it is
probably due to their V/P of engineering, Dave Rogers, who was on the VAX780
architecture team at DEC.
Personally, I am going to watch this company carefully. I think they have a
handle on a good architecture and have did all the leg work to produce a
working system. If there is anything lacking, it may be the limitations (!?!)
of the 4.2BSD network system in producing a viable networked file-system.
However, given that multiplexor files are still warm in the grave, I think
that a lot can be done with the generic network architecture in the 4.2BSD
implementation. This reflects little on Sequent's system and I urged Mr. Rogers
to continue in the same vein and add a good networked file-system on to the
Balance 8000.
For further information, see Electronics Design, September 6, 1984, pp. 153-168.
Glenn A. Adams
glenn@LL-XN.ARPA
MIT - Lincoln Laboratory
-------
-------
-------
∂05-Dec-84 0015 JMC
To: LES@SU-AI.ARPA
∂29-Oct-84 1348 RPG Subcontracts
There was some discussion among the ARPA folks and me about Stanford
subcontracting the Common Lisp work at Stanford to Lucid. This would
involve the blue pages work and the yellow pages work. I think we're in
a pretty good position to take that on. How shall we proceed?
We have been looking at the various pieces of multiprocessor hardware, and
we've pretty-much decided that the way to do the HEP is with a byte-code
compiler and interpreter. That is, we will compile Lisp down to some
abstract machine code, and we will hand-code an interpreter for that
language in HEP assembly language (or PASCAL). The interpreter will
run in the program memory. Unless we do this, it will take a very long
time to get a Lisp running.
We are pursuing the other manufacturers, such as Synapse, Sierra,
and Sequent.
-rpg-
∂05-Dec-84 0016 JMC
To: les-here
∂12-Nov-84 1522 RPG Multiprocessors
To: JMC@SU-AI.ARPA, JJW@SU-AI.ARPA, cheriton@SU-PESCADERO.ARPA
CC: rpg-q@SU-AI.ARPA
The following companies are `in the running' as far as I'm concerned:
Denelcor
Synpase
Encore
Symbolics
BBN
Sequent
Let me discuss the pros and cons of each:
Denelcor:
We all looked at the HEP and have a pretty good idea of what it can do.
Pros:
1. Friendly and co-operative staff: they believe in what we're doing;
2. cute architecture; and
3. they are giving DARPA a `good deal.'
Cons:
1. The hardware implementation of the architecture make it very
difficult to do a cheap, fast Lisp, and
2. Lucid is less interested in doing it than some of the others,
because it is probably not a `typical' multiprocessor, such as
we might see in the next 10 years in common use.
Encore:
I talked at length to Gordon Bell, Ike Nasso, Steve Amarant, and Henry Weigart
from Encore in Massachusetts. I think that of the bunch-of-micros-on-a-bus
machines, Encore's has the highest potential in terms of being relatively high
performance and expandable in the near term to more processors than either of
Synpase or Sequent. This is because they have a 100 megabyte per second bus,
with separate address, data, interrupt, and parity paths.
There are 4 different board types in the machine: processor, memory, IO, and
Ethernet. I saw a physical example of each board on the workbench, with
logic analyzers hooked up to each. That is, they have gotten board back and
are now debugging them.
They use NS32032's, and appear to have them; they are running at 6 Mhz instead
of the nominal 10 mhz. The memories and bus support 32-bit addresses, and the
memories are 4-way interleaved
Each processor board has 2 cpu's and a shared cache between them. The
cache is 32kbytes. The memories support test-and-set. In the future each
board will have 4 processors, and National is going to a 2.5x faster chip
in a year or so. They have a software group that it porting Berkeley 4.2
to it. They have several functional emulators (made up of NS32016 boards)
on which they are doing software development.
They claim that the first 2-processor (1 cpu board) machine will be put together
this week, and having seen the state of the hardware as of last week, I think it
might be possible. In addition, they claim that they could get us a prototype
machine mext April-June.
In order to do a Lisp on that machine, Lucid would have to port our Sun Lisp to
the 32032. I think this could be done using a uni-processor between now and next
summer. With the uncertainties of the DARPA contract, the Encore time schedule
might not be a problem.
Pros:
1. 32032 is a nice machine;
2. fast bus makes for future expansion;
3. Bell has a strong interest in the project and has been
associated with multiprocessors for many years;
4. demand paging;
5. Unix; and
6. Lucid would want to do it.
Cons:
1. the Encore schedule could slip; and
2. Encore may collapse, though I don't see this right away.
Synapse:
I agree with everything that Cheriton said, except for one. That is, I
think that Synapse is a little closer to the brink than he thinks. Synapse
shipped machines too early a while back, and the investors have slapped their
wrists. They may be on a prove-yourself-or-die cycle.
I want to emphasize that Synapse will have *no* interest in helping us
at all unless we convince them that there is a market they could attack
with a parallel Lisp. Otherwise, I think, they will proceed in the database
area until their investors are happy.
Pros:
1. 68000 Lisp is a simple port of Lucid's Sun Lisp;
2. basically a good architecture;
3. they will accept a purchase order today;
4. reliable hardware;
5. nearby; and
6. Lucid would want to do it.
Cons:
1. They will not help us in any way unless we change their
marketting strategy to include AI;
2. operating system is not something Lucid wants to do; and
3. perhaps they're not long for this world also.
BBN:
BBN appears to be about to get strategic computing money to put a Common Lisp
on the Butterfly along with a multiprocessing version of it. Poor lads, they
thought I (rpg) was going to do the Common Lisp for them for free. I talked to
a panic-stricken Don Allen last week, and he strongly suggests we use the
Butterfly. He and Randy Rettberg, a manager, are willing to fly out to Stanford
for a show-and-tell. They hope that we select the Butterfly and get DARPA to pay
for Lucid to put Common Lisp and the low-level primitives for Qlambda on the
Butterfly. They would then take the Lisp and the primitives and do something
different from Qlambda for DARPA.
It sounds like there are a few things wrong with their suggestion: 1), they are
suggesting that DARPA pay twice for some of the work, namely the multiprocessing
primitives, and 2), I think DARPA is more nearly sold on Qlambda than it would
be on a parallel Lisp to be specified later.
Nevertheless, the Butterfly is a 68000-based machine and it exists.
Pros:
1. 68000-based machine;
2. Sun Lisp easy to port;
3. butterfly switch is not so bad;
4. machine exists;
5. BBN *wants* to co-operate; and
6. Lucid is not against doing this.
Cons:
1. Unknown operating system; and
2. the switching network may need to be micro-coded to
handle a synchronization primitive.
Let me elaborate on this last point. Each cpu board has a micro-codable
memory-watcher. This memory-watcher implements the butterfly network, and there
is more control store in which to do other things. For example, references to
certain memory locations can be interpreted as other sorts of instructions
by this watcher. I'm not sure how much programming of it must be done.
Symbolics:
I talked to them about multi-processors, and I was surprised when they said that
they had proposed a multi-processor to Arvind, which would be a memory-bus or
multi-ported memory scheme. Another possibility would be a gigabyte serial
port amongst some 3600's. Symbolics volunteered to send out someone to Stanford
to chat about it.
Pros:
1. The 3600 already has a Lisp.
Cons:
1. I'm not at all sure whether they could deliver this in a
reasonable length of time.
Sequent:
They are left to look at. They have delivered, and use 32032's. Gordon Bell
says that except for a slower bus, they are similar to the Encore machine.
Tomorrow I am starting my sequence of inquiries which ought to land us some
information soon.
-rpg-
∂05-Dec-84 0017 JMC
To: les-here
ebos[f84,jmc] Editor based operating system proposal
Stanford proposes to develop technology for editor based
operating systems for the IBM ROMP microprocessor. A particular
such operating system will be the concrete result of the work, but
our proposal includes technical reports and scientific publications
on the technology in general. The work will cost $115,000 per
year for personnel, and we propose a 4 year project. In addition to the
personnel cost, we will need three ROMP microprocessor systems,
each with 4 megabytes of memory (not necessarily immediately),
and ethernet interface. Unless a substantial disk for file storage
is included, there will be (not large) costs for use of Stanford University
Computer Science Department computer facilities.
The part of the project that involves experimenting with keyboards
will require the ability to order additional keyboards beyond
the number of processors. Since most of these will have to be
ordered specially, we anticipate subcontracting $15,000 worth
of keyboards unless IBM is in a position to provide them.
Goals of the project:
The main premiss of the project is that the proper environment
for the interactive user of a computer is a display editor. This provides
the ability to re-edit both previous input and new output in
order to generate new input and also to file the output as desired.
Various other improvements in editors and operating systems are planned.
Thus the goals of the project include the following.
1. The ROMP executive and interactive programs will be
normally operated out of a new advanced display editor. Programs
for which this is not suitable will be able to control the interaction
themselves.
2. We have studied the CPR operating system and find it
suitable as a base for building our editor and interactive system.
Thus our programs will use its system calls. We don't presently
anticipate having to change it in important ways, but very likely
something will turn up.
3. The editor and our other programs will be written in
Common Lisp. The full facilities of Common Lisp will be available
for macros (execs in certain IBM dialects) for controlling the editor
itself, the operating system and user programs. Our experience
at Stanford has shown us that if the editor provides good enough
interactive facilities, then many user programs can rely on them
thus allowing simpler programming of interactive programs.
4. The editor and the operating system will be kept fully
programmable. Thus anything a person can do interactively, he will
be able to write programs to do. This means that all output seeable
by a user must also be readable by programs. Interactive programs
that use the display in a non-standard way will be able to violate
this condition.
System status information will be maintained in editable
files.
5. The editor and file system will accomodate arbitrary
character sets. This advance is easy to make in the editor itself
and in bit map (IBMese APA) displays and modern printers, but keyboards
present a problem for which we intend to provide and explore a variety
of solutions.
6. Our planned solution to the arbitrary character set problem
will allow both for standard keyboards and for keyboards adapted
to special tasks, e.g. the use of mathematics, APL or foreign
languages. Thus a special character can appear on the screen
either because the user has pressed a key on a special keyboard
or because he has pressed a suitable sequence of keys on an
ordinary keyboard. The same sequence of bytes will inhabit the
file in either case.
The pace of the project will depend on how soon we can
interest students in the project after it gets started. We should
have something working within a year of the time we get people,
computers, budget and Common Lisp.
Personnel:
John McCarthy, 15 percent
Executive officer and project manager, Lester Earnest, 25 percent
Research associate, Carolyn Talcott, 10 percent.
2 graduate research assistants, half time academic year, full time summer.
Stanford computer facilities $10K per year
This includes both computer time, disk storage and help with installation.
Research Programmer, Martin Frost 30 percent.
∂05-Dec-84 0024 JMC print and mail file
To: RA
Please print the file lewis.ns[f84,jmc] and mail it to Professor Wayne
Vucinich, History Department. Then please delete the file.
∂05-Dec-84 1622 JMC re: Final
To: DIEP@SU-SCORE.ARPA
[In reply to message sent Wed 5 Dec 84 14:23:57-PST.]
It is not possible to take a final in advance.
∂05-Dec-84 1652 JMC re: Kennedy and Drell
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Wed 5 Dec 84 16:25:25-PST.]
I have done my share of taking part in debates about what harm or good
military funding of research might do. I haven't been following the CSLI
discussions, and I would like to know what policy matters are at issue.
Howevr, if CSLI were to take a position that military funding is bad
per se, I would feel morally obliged to sever my connection with CSLI.
∂05-Dec-84 1757 JMC
To: DAC
Glad your visit was promising.
∂05-Dec-84 1805 JMC help command
To: DAC
If you type HELP followed by a topic, such as MAIL, you will get
help on that topic. Typing just HELP gives a list of topics.
∂05-Dec-84 2001 JMC re: letter for Mr. Weide Guo
To: ARIADNE@SU-SCORE.ARPA
[In reply to message sent Wed 5 Dec 84 17:26:37-PST.]
Guo Weide, aka Victor Kuo, has been receiving mail here regularly for the
last year and a half. His office is MJH 362.
∂05-Dec-84 2006 JMC re: Draft
To: BARWISE@SU-CSLI.ARPA
[In reply to message sent Tue 13 Nov 84 19:48:26-PST.]
I forgot that I had read the draft policy of November 13 and was
entirely soothed.
∂05-Dec-84 2022 JMC The L.A. Times is getting too big, and actresses' dogs are getting too small.
To: su-bboards@SU-AI.ARPA
a284 1919 05 Dec 84
AM-Dog Death,0125
Actress's Dog Killed By Flying Newspaper
LOS ANGELES (AP) - Barbara Bain's dog was killed when a delivery
person heaved a copy of the Los Angeles Times toward the actress'
front lawn and it landed on the animal, the actress's agent said
Wednesday.
The accident occurred last week, Marty Blumenthal said.
''They offered to make restitution, but how do you put a price on a
pet you've had for 14 years?'' Miss Bain told the Los Angeles Herald
Examiner.
A man contacted late Wednesday afternoon in the Times' circulation
department declined to comment on the incident, saying he didn't know
anything about it. He didn't give his name.
Blumenthal was unable to provide additional details Wednesday and
said Miss Bain was not available for comment.
AP-NY-12-05-84 2216EST
***************
∂05-Dec-84 2301 JMC Breaking the boat
To: VAL
I haven't solved it, but the problem seems somewhat analogous to those
that arise when we try to circumscribe equality. I try the axiom
∀x s'.¬ab aspect3(x,s') ⊃ ∀s.s' ≠ result(break x, s),
but I don't think it quite does what we want. Maybe we can solve it
by reifying.
∂06-Dec-84 1406 JMC re: TA for 224
To: REGES@SU-SCORE.ARPA
[In reply to message sent Thu 6 Dec 84 12:25:05-PST.]
The course is 226, and the TA will be Gian-Luigi Bellin, a graduate
student in philosophy.
∂06-Dec-84 1416 JMC re: invoice inference
To: RA
[In reply to message rcvd 06-Dec-84 14:12-PT.]
Send it ATTENTION: Alex Jacobson. Add a note asking whether it should
go to someone else.
∂06-Dec-84 1435 JMC re: TA for 224
To: GLB
∂06-Dec-84 1433 REGES@SU-SCORE.ARPA re: TA for 224
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 6 Dec 84 14:32:56 PST
Date: Thu 6 Dec 84 14:26:14-PST
From: Stuart Reges <REGES@SU-SCORE.ARPA>
Subject: re: TA for 224
To: JMC@SU-AI.ARPA
In-Reply-To: Message from "John McCarthy <JMC@SU-AI.ARPA>" of Thu 6 Dec 84 14:06:00-PST
Office: Margaret Jacks 210, 497-9798
Could you ask Gian-Luigi to come in and fill out the "TA/RA Information Sheet"
(available at the Receptionist Desk)? We need some information about address,
social security number, etc., to process the paperwork.
-------
∂06-Dec-84 1818 JMC re: CS206 exam
To: RTC
[In reply to message rcvd 06-Dec-84 18:15-PT.]
We cannot give the exam early, as I have already told another student.
We might give him one late.
∂06-Dec-84 1851 JMC
To: RTC
Grades will have to be submitted the following Monday.
∂06-Dec-84 1952 JMC re: Terminals on desks. (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - That's so, but terminals are much cheaper than space for desks.
∂07-Dec-84 0103 JMC
To: RA
Did you send Dyson review to Drell?
∂07-Dec-84 0937 JMC re: A.J. Thomas
To: FOGELSONG@SU-SCORE.ARPA
[In reply to message sent Fri 7 Dec 84 01:59:04-PST.]
I saw him at IJCAI-83, and he told me what he was doing, but it completely
slips my mind at present. I don't even remember if he is living in England
or in the U.S. You might try to check if he is a member of AAAI or AISB.
∂07-Dec-84 1746 JMC re: summary
To: withgott.pa@XEROX.ARPA
[In reply to message sent 7 Dec 84 17:02 PST.]
I assume that I am to approve the summary and Bob is to supply a summary
of his comments. It seems ok with correction of misprints.
∂07-Dec-84 1810 JMC
To: CLT
a004 2206 06 Dec 84
PM-Topic-La Scala, Bjt,0863
TODAY'S TOPIC: Problems Facing Venerable Opera House Unchanged Since
Verdi
Laserphoto NY7
By VICTOR L. SIMPSON
Associated Press Writer
MILAN, Italy (AP) - Giuseppe Verdi would feel at home amid the
highly charged controversies buffeting La Scala Opera House these
days.
The 206-year-old temple of bel canto is plagued by many of the same
problems that bedeviled the theater when the composer was producing
his masterpieces more than 100 years ago.
Another year of budget deficits has forced a relatively lean
program, the irrepressible Italian opera critics are flinging charges
of mismanagement and the temperamental performers are preparing to
face one of the world's most demanding audiences.
Except for modern-day labor problems - only an eleventh-hour
settlement with the orchestra allowed the opening night to go on this
season - the atmosphere isn't all that changed from Verdi's day, when
he railed at the critics and threatened to block the premiere of his
''Macbeth'' unless the management changed its ways.
The theater, considered an acoustical gem, is a cultural landmark in
the center of this northern financial capital. Across Piazza della
Scala is the famed arcade where elegant Milanese go to see and be
seen; down the street is the city's Gothic cathedral.
But the problems of money and planning have raised concern about the
future of La Scala as a magnet for talent and innovation, especially
in light of increased competition from U.S. opera houses to sign big
names.
Carlo Mazzonis, La Scala's artistic director, says that the money
demands of the artists, often in dollars, ''have reached unbearable
levels.'' He has called for a united stand by Europe's leading
theaters against costly contracts.
Many Italian critics agree, but also say the theater's management
has to bear responsibility for lack of long-term planning.
''In this business you have to plan four or five years in advance.
The Met is the Met but now Los Angeles and Dallas are going after the
big voices,'' said Guido Vergani, a critic for the newspaper La
Repubblica.
This year's gala opening on Dec. 7, the traditional opening date on
the feast day of Milan's patron saint, features Claudio Abbado
conducting Georges Bizet's ''Carmen,'' starring American soprano
Shirley Verrett and Spanish tenor Placido Domingo.
But it is one of the few new productions of the year, and the two
stars were signed for only five of the nine performances.
Other big names signed this season include Frederica von Stade,
Agnes Baltsa, Jose Carreras, Yoko Watanabe, Olivia Stapp, Eva Marton
and Sylvia Greenberg.
And in one of the few new works, Soviet expatriate Yuri Ljubimov
will direct the La Scala company in June in a version of Johann
Sebastian Bach's ''The Passion According to St. Matthew.''
Missing for a second straight year, however, is Luciano Pavarotti,
the popular Italian tenor, whose last appearance at La Scala drew a
chorus of catcalls from the demanding gallery when he suffered voice
problems.
The 1984-85 program is still not completed. Carlo Maria Badini, who
has run La Scala since 1977, said this is because the government
failed to disclose the exact amount of La Scala's subsidy and
management doesn't know how much money will be available in 1985.
La Scala and Italy's 12 other leading opera houses are financially
dependent on the government's slow-moving bureacracy to offset the
ever-increasing costs of opera productions.
MORE
AP-NY-12-07-84 0103EST
- - - - - -
a006 2216 06 Dec 84
PM-Topic-La Scala, Bjt-1st Add, a004,0313
MILAN, opera productions.
For this year, the state allocated $150 million, with La Scala
receiving $18 million, the largest single chunk.
The financial situation was especially bleak from the mid-1970s
until two years ago, when Badini embarked on an ''austerity plan,''
reducing the number of new productions to make use of available sets
and costumes and seeking new sources of revenue.
Commercial sponsors now have their trademarks in the program despite
opposition from ''purists''; an insurance company partially financed
opening night and a washing machine company is sponsoring the ballet
season.
Thus Badini was recently able to report that La Scala's projected
deficit will be cut to $600,000 this season, to be made up by bank
borrowing.
If it's any consolation for the current management, the theater was
already running in the red only a decade after it opened its doors in
1778. Performers and composers throughout its history have complained
about the poor reception given works that were applauded in theaters
abroad.
Early this century, La Scala went into serious decline until rescued
by the iron hand of Arturo Toscanini. ''I am La Scala,'' the
conductor often proclaimed.
''In the past there were many great directors and among the few good
ones of today there is nobody who can compete on both an
organizational and artistic level as those of the past at La Scala,''
wrote Paolo Isotta, music critic of Milan's leading Corriere della
Sera.
Despite the problems, the theater's 2,015 seats are often filled
even with constantly rising ticket prices. The best seats now go for
$87.50 while orchestra tickets for opening night cost $308.
Several major changes may be in the offing: Badini's contract comes
up for renewal next year while in 1986 Riccardo Muti, the conductor
of the Philadelphia Orchestra, is due to take over as regular
conductor from Claudio Abbado, who is moving to Vienna.
∂07-Dec-84 2158 JMC re: Terminals for grad students (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - I suggest you address your proposal to someone with the power to
act, e.g. FACULTY@SCORE or Nils Nilsson.
∂08-Dec-84 1229 JMC re: Car
To: TIBSHIRANI@SU-SCORE.ARPA
[In reply to message sent Sat 8 Dec 84 12:19:45-PST.]
She has bought a car.
∂08-Dec-84 2145 JMC inquiries
To: su-bboards@SU-AI.ARPA
The bulletin boards would be much less cluttered if there were people
willing to be identified as consultants on various topics. Perhaps
the student bureaucracy could organize this. The messages would be
sent to the consultant rather than to bboards.
∂10-Dec-84 1118 JMC
To: RA
The answer for Keith Clark is yes, and he should send me some
recent reprints.
∂10-Dec-84 1121 JMC
To: library@SU-SCORE.ARPA
The searches were in lookup mode and done from a Green Library terminal.
I'll try to recreate the phenomenon.
∂10-Dec-84 1144 JMC
To: RA
∂10-Dec-84 1139 RA
Are you going to be in today? Generally, could you let me know when
(and if) you are going to be in.
I expect to be in every day this week. Today, early afternoon.
∂10-Dec-84 1551 JMC
To: RA
TV students should take the final in class.
∂10-Dec-84 2157 JMC re: Vote: should we buy Dept. SUN workstations
To: cheriton@SU-PESCADERO.ARPA
[In reply to message sent Mon, 10 Dec 84 21:51:03 pst.]
Oh well, go ahead, i.e. "yeah".
∂11-Dec-84 1535 JMC
To: RTC
Please move all CS206 files to 206,jmc.
∂11-Dec-84 2255 JMC
To: RV
Statistics Department graduate students run an advisory service
on how to analyze data and presumably would know about such packages.
∂11-Dec-84 2300 JMC re: Funny bugs (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Anyone who has lived in East Coast apartments will tell you that
when you have mice, you often also have cockroaches. What was the effect
on the program of clicking the cockroach with the index finger? I suppose
for full programming flexibility you need three cockroaches.
∂12-Dec-84 0128 JMC energy[f84,jmc]
To: RA
Please copy that file and convert the copy to SCRIBE and print it.
You may want to pub it and print it to see what it is supposed to look like.
∂12-Dec-84 0200 JMC
To: RA
Please xgp lipset.xgp[let,jmc].
∂12-Dec-84 1351 JMC re: Your visti to MCC
To: AI.Woody@MCC.ARPA
[In reply to message sent Wed 12 Dec 84 15:06:22-CST.]
Yes, please make reservations for me for Thursday and Friday nights,
although I may not use the Friday.
∂12-Dec-84 2343 JMC re: unique names
To: VAL
[In reply to message rcvd 12-Dec-84 23:32-PT.]
It seems to me that if we are using names purely as a technical device,
we don't have to assume that anyone knows the names of every bird,
merely that the bird has a name, e.g. "bird number 34567891234567890".
∂13-Dec-84 1707 JMC re: Austin flight
To: RA
[In reply to message rcvd 13-Dec-84 16:11-PT.]
I prefer the 1:45 going out.
∂13-Dec-84 1841 JMC cs206 chore
To: RTC
As soon as we have the finals, please compare the list of people who
have taken it with our list of students. Some will take it later, but
others should be asked whether they have dropped.
∂13-Dec-84 1844 JMC changes
To: SMC
Some minor changes are required in the front pages of the files you
converted to TEX. The files should be in the same areas as the files
from which the originals came, and the mentions of the files on the
front pages should refer to the correct files. Also the file names
should be entered in FILES[LET,JMC] next to the PUB files you used.
∂14-Dec-84 0910 JMC parrallel lisps
To: RPG, CLT
∂14-Dec-84 0849 brand%ucbarpa@Berkeley parrallel lisps
Received: from UCB-VAX.ARPA by SU-AI.ARPA with TCP; 14 Dec 84 08:48:49 PST
Received: from ucbarpa.ARPA by UCB-VAX.ARPA (4.24/4.40)
id AA23303; Fri, 14 Dec 84 08:49:32 pst
Received: by ucbarpa.ARPA (4.24/4.40)
id AA22419; Fri, 14 Dec 84 08:49:47 pst
Date: Fri, 14 Dec 84 08:49:47 pst
From: brand%ucbarpa@Berkeley (Russell L. Brand)
Message-Id: <8412141649.AA22419@ucbarpa.ARPA>
To: jmc@su-ai
Subject: parrallel lisps
Richard Fateman has mentioned to me that you are interested in bringing
up a parrallel lisp for super-computers. It is soon to be the case that
Lawrence Livermore National Labs MAY have an unwanted 205 (cyber) looking
forlove care and a good home...
...would you be interested in providing it a home?
Russell L. Brand
LLL Computation Research Group
∂14-Dec-84 1159 JMC re: Austin Texas
To: RA
[In reply to message rcvd 14-Dec-84 11:34-PT.]
Please see what is available Saturday morning.
∂14-Dec-84 1438 JMC
To: les-here
\input macros[pap,rpg]
\magnification\magstep1
\paper:Outline proposal for Stanford Hep and parallel Lisp.
It is generally agreed that the main hope for large increases
in computer speed, whether for numerical work or artificial
intelligence, lies in massive parallelism. Projects are being undertaken
that will involve hundreds or even thousands of processors.
However, no-one has yet demonstrated the ability to make general purpose
use of even two processors on symbolic computation for a single problem.
We, Richard Gabriel and John McCarthy, have a variant of Lisp
called Qlambda, described in our paper (Gabriel and McCarthy 1984),
which we believe will allow Lisp to effectively use many processors.
It is based on the following ideas:
\numitem{1.}The programmer can tell when a computation will allow
parallelism and can indicate this by using appropriate constructs
in the source language. In our case these are called QLET and QLAMBDA.
\numitem{2.}When tasks that may be done in parallel are found they are
placed on a queue. Free processors take tasks from the queue.
The source and compiled programs do not require any specific number
of processes.
\numitem{3.}It is important to control the amount of multi-processing
and limit it to not much more than the number of processors available.
This is because handling the multi-processing involves computational
overhead, and if allowed to occur at a low level, this overhead could
swamp the useful computation. The control is accomplished in Qlambda
by runtime parameters to QLET and QLAMBDA that determine whether
parallelism is actually to be allowed.
We believe this simple scheme will work for effective Lisp
multi-processing in AI and symbolic mathematics applications. Our
paper includes simulations that indicate that this is true.
We propose to acquire a Hep, develop Qlambda, and test it
on applications that will provide a good test of how much
speed gain can be obtained from multi-processing using the features
of Hep which include a certain amount of data-flow architecture.
The results should help DARPA guide its more ambitious multi-processing
efforts by determining what additional architectural features are important
for multi-processing in Lisp.
Besides implementing Qlambda, we propose to implement Macsyma
and some large Macsyma applications. Implementing Macsyma and Macsyma
applications has the advantage of providing a very large and varied collection
of Lisp functions each of which is computationally expensive.
Making them run in parallel is a well-defined task that will nevertheless
fully test Qlambda and the Hep.
Besides this, HPP has proposed some major AI applications that they will
program in Qlambda and test on simulators; if Qlambda were to exist on
the Hep, they would mount the applications on it.
Acquiring and operating a Hep will be expensive in terms of
acquisition cost, space (Welch Road where HPP is located), and personnel.
Substantial personnel will be required for the Macsyma and symbolic
computation projects. Nevertheless, we believe that this project
provides and opportunity to get a preview of how to solve the problems
that will be faced by the much more expensive parallel processor
projects now being undertaken.
If there is sufficient interest a detailed proposal with costs
will be prepared.
\bye
!85k per month lease
50% trade on new machine
rental $600k first
buy 1.5 million, 50%rebate+25%discount +100k maint.
Jim Hill is president
The machine will include 32 megabytes of extended memory. It
is supposed to give good performance operating out of this memory.
A bargain on the faster HEP2 will be part of the deal.
Stanford HEP coalition
1. DARPA would buy (rent) (lease) a HEP from Denelcor.
2. DARPA would pay Stanford to contract with Lucid (Dick Gabriel, Pres.)
to put Qlambda on the HEP.
3. DARPA would pay installation and maintenance for software and
hardware and such software development as agreed. Denelcor might
provide some personnel or other help for software development
according to an agreement to be negotiated.
The object of the project is make available promptly a tool for
the development and experimental use of languages for parallel
computing. More powerful parallel computers will benefit from
being able to get experience promptly.
Stanford groups
1. The formal reasoning group under John McCarthy would manage the
project and the development of Qlambda and a suitable operating
system.
2. People would be hired to do Macsyma or something similar.
3. HPP would put an AI system on the machine.
4. Oliger
5. ERL
6. Ullman
1984 Sept 12, conversation with Burton Smith
returns oct 8 monday, communicate with Jim Hill, x201 Jennifer
!Personnel
manager
system programmer for op system
hardware hacker for connections ?
qlambda hacker
macsyma builders (2)
∂14-Dec-84 2003 JMC parrallel lisps
To: REG
∂14-Dec-84 0849 brand%ucbarpa@Berkeley parrallel lisps
Received: from UCB-VAX.ARPA by SU-AI.ARPA with TCP; 14 Dec 84 08:48:49 PST
Received: from ucbarpa.ARPA by UCB-VAX.ARPA (4.24/4.40)
id AA23303; Fri, 14 Dec 84 08:49:32 pst
Received: by ucbarpa.ARPA (4.24/4.40)
id AA22419; Fri, 14 Dec 84 08:49:47 pst
Date: Fri, 14 Dec 84 08:49:47 pst
From: brand%ucbarpa@Berkeley (Russell L. Brand)
Message-Id: <8412141649.AA22419@ucbarpa.ARPA>
To: jmc@su-ai
Subject: parrallel lisps
Richard Fateman has mentioned to me that you are interested in bringing
up a parrallel lisp for super-computers. It is soon to be the case that
Lawrence Livermore National Labs MAY have an unwanted 205 (cyber) looking
forlove care and a good home...
...would you be interested in providing it a home?
Russell L. Brand
LLL Computation Research Group
∂14-Dec-84 2006 JMC re: your chess bet
To: TOB
[In reply to message rcvd 14-Dec-84 17:01-PT.]
I never bet that a computer would be world champion. I bet in 1968
with David Levy that a computer program would beat him by 1978. He
won two games to one against, I believe, Chess 4.7.
∂15-Dec-84 1003 JMC re: paper & no.
To: RESTIVO@SU-SCORE.ARPA
[In reply to message sent Sat 15 Dec 84 10:00:09-PST.]
It was.
∂15-Dec-84 2216 JMC re: terminal
To: SMC
[In reply to message rcvd 15-Dec-84 22:11-PT.]
Some exchanges are worse than others. It may be difficult to reach
SAIL from Whit's house. Repeated attempts may help. There was
information about this in the file \gripes. You can also try
asking Martin Frost or Joe Weening.
∂15-Dec-84 2358 JMC re: Net address of Brown University (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - It has changed its name to Jim Jones University.
∂16-Dec-84 1337 JMC re: party
To: HULLMAN@SU-SCORE.ARPA
[In reply to message sent Sun 16 Dec 84 13:27:31-PST.]
Unfortunately, I will be in Austin that night. Please transmit my regrets
to Gene.
∂16-Dec-84 2219 JMC
To: SMC
∂21-Sep-84 0141 DON (on TTY24) 1200-baud lines are lousy
To: BUG-phone, "#GRIPES.TXT[2,2]"
The 1200-baud dialups (324-9030) have been terrible the last couple days.
I'm getting about a 1:1 ratio of noise to signal. Bleah! This has
happened on two different lines (9031 and 9032), on three different
terminals, and using two different modems. Is this aftermath from the
thunderstorm, or is something else wrong?
ARK - I'm now on TTY23 (324-9030). It's 9/21 03:44. There's absolutely
no noise. I haven't seen any noise the last few days either. I'm on
the 494 exchange. Must be the line between your phone and the central
office, Don.
DON - I've seen this when calling from two different phones, one in 941 and the
other in 365.
ARK - If it means anything, I am using a VA3455 modem. And it is working
clear as a bell.
DON - Well, I'm using a Bell, and it is working as poorly as a . . . Actually,
I've got a DEC DFO3. It might well be modem-modem trouble, though as I said
I've had the same trouble using someone else's modem. I'll ask him what his is.
I admit I don't know for certain that either of these modems has ever
successfully talked to Stanford at 1200 baud. By the way, I have the same
problem with SCORE (both the 322 and the 497 numbers).
DON - Here's a summary of what I've found out about the 1200-baud lines. It
looks like it might be a phone company problem, but even so it's probably
easier for somebody "in charge" at Stanford to follow up on this.
Calling from 941 exchange to SAIL (324) or to SCORE (322) using DEC DFO3 (Bell
212 protocol) loses horribly. Modem receives noise (mostly left-curly-braces)
in bursts of 1 to 20 characters every 1 to 3 seconds. Calling SCORE's 497
number fares slightly better; noise is about 1 to 3 characters every 3 to 10
seconds. Terminal type seems irrelevant (I've tried a Concept-100 and a
VT102). 300 baud works fine. The same terminals/modem/phone work fine if I
call DEC, which is a 94x number for some value of x.
Calling from 365 (Redwood City) to SAIL (324) using a UDS 212LP modem on a
DataMedia terminal has the same lossage as above. Calling SRI (859?) with
that modem/phone works fine.
Arthur claims that calling SAIL from 494 using a VA3455 works fine.
Help?
DON - Further datum: Calling Glacier (325 exchange) from 941 fails horribly.
Glacier uses Vadic on the receiving end; I don't know what SAIL and SCORE use.
Is anyone looking into this?
ARK - SAIL uses a VADIC, but Score doesn't.
DON - Well, I finally found a modem that works with minimal noise (one or two
characters every minute or so. It appears therefore that (a) the phone lines
are marginal, (b) the DEC modem is marginal, (c) the UDS 212LP is also marginal,
and (d) marginal times marginal equals terrible.
∂17-Dec-84 1003 JMC re: Brown -> Jim Jones
To: g.Ryland@SU-SCORE.ARPA
[In reply to message sent Mon 17 Dec 84 07:54:31-PST.]
I am referring to the student vote urging the university health service
to stock cyanide for use if there is a nuclear war and its similarity
with the mass suicide with cyanide by the followers of the Reverend
Jim Jones.
∂17-Dec-84 1530 JMC re: Sequent
To: RPG
[In reply to message rcvd 17-Dec-84 14:17-PT.]
How about 3:30? I have another meeting from 1:15 to 3:15.
∂17-Dec-84 2328 JMC re: invited talk
To: HST
[In reply to message rcvd 17-Dec-84 23:25-PT.]
I got the photographs. Tell me more about the meeting in Germany.
∂17-Dec-84 2354 JMC re: gwai85
To: HST
[In reply to message rcvd 17-Dec-84 23:35-PT.]
It would be interesting to meet the German AI community, and I expect
to be still working mainly on non-monotonic reasoning. However, I
can't guarantee to have a written paper for the proceedings.
∂18-Dec-84 0037 JMC re: DoE Follies
To: LLW@S1-A.ARPA
[In reply to message sent 18 Dec 84 0013 PST.]
Nothing but the letter from Melendez (or Menendez) of the consultant office
with the form to start the Q clearance process over. It is as though I
had just become a consultant.
∂18-Dec-84 0157 JMC
To: LES@SU-AI.ARPA
Give me a call when you get a chance. Imagen said you were in Alaska.
∂18-Dec-84 0157 JMC call
To: les-here
Give me a call when you get a chance. Imagen said you were in Alaska.
I forgot and sent this message to LES and it got forwarded.
∂18-Dec-84 0959 JMC testing 1 2 3 4
To: elliott%slacvm.bitnet@SU-FORSYTHE.ARPA
Elliott,
Here's how I'm told I can send you mail. The return path to
JMC@SU-AI presumably goes by bitnet to FORSYSTHE and on to SU-AI,
but I don't know the grammar.
John
∂18-Dec-84 1020 JMC grades
To: RA
I'm still grading. They'll be ready later today.
∂18-Dec-84 1048 JMC re: fateman
To: CLT
[In reply to message rcvd 18-Dec-84 10:29-PT.]
Since he got us one, we should get him a one day. I believe they can
be gotten from the receptionist or Betty knows from whom. Ask Rutie
to do it, since she should learn who has them.
∂18-Dec-84 1347 JMC Clearance delays
To: llw@S1-A.ARPA
It occurs to me that there might be an impasse of the following kind.
The PSQ asked only about organization memberships for the past ten
years, and that's what I gave them. On the other hand, the preliminary
questionnaire that I filled out when I became a summer employee several
years ago asked for memberships without time limit, and I gave them that.
All activities that they might consider relevant to a clearance are more
than ten years old, and there may be some law or regulation that inhibits
them from asking. I am entirely willing to waive such a restriction if
it exists and if they are allowed to accept such a waiver. Perhaps you
could ask the LLNL security people or the DOE people if there is such
a problem. Needless to say, I will be guided by your recommendation as
to the right strategy, but you don't know everything either.
∂18-Dec-84 1622 JMC re: Nonmonotonic Symposium / Is the proceeding available?
To: SG
[In reply to message rcvd 18-Dec-84 14:58-PT.]
The meeting at Maryland wasn't about non-monotonic logic. Perhaps you
mean the meeting at Lake Mohonk, New York. The AAAI in Menlo Park may
have some copies of the Proceedings left.
∂18-Dec-84 1628 JMC catalog pages to John Williams
To: RA
Please mail John Williams at IBM a copy of the pages in the Stanford
catalog giving the industry lecture courses for this year. The
K, etc. in the PHON entry speeds up deliver within IBM.
∂18-Dec-84 1736 Mailer failed mail returned
To: JMC
In processing the following command:
MAIL
The following message was aborted because of a command error,
namely, nonexistent recipient(s):
olender
------- Begin undelivered message: -------
∂18-Dec-84 1736 JMC re: announcements
[In reply to message sent Tue 18 Dec 84 17:11:13-PST.]
Include also Adm. Bobby Inman and Dr. Woodrow Bledsoe at MCC in Austin,
Kazuhiro Fuchi at ICOT, Ralph Gomory, John Cocke, Shmuel Winograd at
IBM (in addition to forum addressees), computer science departments at
Imperial College London (Kowalski), University of Edinburgh, University
of Aix-Marseille (Prof. Alain Colmerauer). All the above will be interested.
------- End undelivered message -------
∂18-Dec-84 1740 JMC re: annoncements
To: olender@SU-SCORE.ARPA
[In reply to message sent Tue 18 Dec 84 17:11:13-PST.]
Include also Adm. Bobby Inman and Dr. Woodrow Bledsoe at MCC in Austin,
Kazuhiro Fuchi at ICOT, Ralph Gomory, John Cocke, Shmuel Winograd at
IBM (in addition to forum addressees), computer science departments at
Imperial College London (Kowalski), University of Edinburgh, University
of Aix-Marseille (Prof. Alain Colmerauer). All the above will be interested.
∂18-Dec-84 2104 JMC mail to SLAC
To: bosack@SU-SCORE.ARPA
It worked - as soon as I surrounded the part of the address beyond Forsythe
with quotes.
∂18-Dec-84 2331 JMC re: mail to SLAC
To: BOSACK@SU-SCORE.ARPA
[In reply to message sent Tue 18 Dec 84 23:27:02-PST.]
Yes, I got a reply.
∂18-Dec-84 2335 JMC Please put the following on paper
To: RA
Note to Sidney Hook at Hoover.
If the student you spoke about wants to telephone me, I'll give him
what advice I can.
∂19-Dec-84 0143 JMC
To: HST
Yes, I think it would be interesting.
∂19-Dec-84 0149 JMC re: Speeding (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Speeding is sometimes dangerous, but self-righteousness can be much more
dangerous.
∂19-Dec-84 0222 JMC visit
To: ai.woody@MCC.ARPA
I will arrive in Austin on Muse 872C at 7:30pm on Thursday and depart
on American 139V at 9:24am on Saturday. If anyone wants to have dinner
on Thursday, I won't have eaten. I'll be glad to give a talk if there
is interest on non-monotonic reasoning. Carolyn will make a trip later;
she still has to finish her thesis although she has passed her final
orals. I could also talk briefly about QLAMBDA, although Dick Gabriel
could give a more detailed lecture.
∂19-Dec-84 0231 JMC
To: RA
I did it this time, but you should put plane reservations in CAL.
∂19-Dec-84 1047 JMC
To: SMC
Time is getting short for finishing the papers.
∂19-Dec-84 1503 JMC Kurzweil Machine
To: genesereth@SUMEX-AIM.ARPA
∂19-Dec-84 1336 Allen.PA@Xerox.ARPA Kurzweil Machine
Received: from XEROX.ARPA by SU-AI.ARPA with TCP; 19 Dec 84 13:36:21 PST
Received: from Cabernet.MS by ArpaGateway.ms ; 19 DEC 84 13:35:47 PST
Date: 19 Dec 84 13:40:32 PST (Wednesday)
From: Allen.PA@XEROX.ARPA
Subject: Kurzweil Machine
To: McCarthy@SU-AI.ARPA
cc: Allen.PA@XEROX.ARPA, Spencer.PA@XEROX.ARPA
John,
There is an outside chance Xerox could supply a Kurzweil machine to you,
the CS Department and Stanford. Could you send me a brief, one-page note
on what your (collective) interests would be and what reserch might be
forthcoming.
Happy Holidays
Bill
∂19-Dec-84 1506 JMC re: visit
To: AI.Woody@MCC.ARPA
[In reply to message sent Wed 19 Dec 84 15:52:50-CST.]
The proposed schedule is fine.
∂19-Dec-84 1508 JMC EG&G
To: CLT
∂19-Dec-84 1218 TAJNAI@SU-SCORE.ARPA EG&G
Received: from SU-SCORE.ARPA by SU-AI.ARPA with TCP; 19 Dec 84 12:18:03 PST
Date: Wed 19 Dec 84 12:16:07-PST
From: Carolyn Tajnai <TAJNAI@SU-SCORE.ARPA>
Subject: EG&G
To: jmc@SU-AI.ARPA
Shall I give EG&G to HPP? or are you interested?
-------
∂19-Dec-84 1610 JMC dithering about EG&G
To: CET@SU-AI.ARPA
My wife Carolyn Talcott is from Idaho. Perhaps she would like an occasion
to make a trip back, and that's what would induce me to accept. If you
don't hear from me by Monday, give up on me.
∂19-Dec-84 1744 JMC
To: churd@SU-SCORE.ARPA
Encore Gordon Bell, 617 237-1022, Pittsburgh, (412)361 8600
15 Walnut ST. Wellesley Hills 02181
∂19-Dec-84 2121 JMC re: Answer (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - It is a convention of natural language that when a correct statement is
made from which the usual non-monotonic inference rules lead to a conclusion
that the speaker believes to be incorrect, then the speaker should supplement
the statement by additional ones that prevent the incorrect conclusion.
In this case the statement was that Shawn was cited for going 95 with
picturesque detail about the helicopter. Curiously, failure to observe
this convention is regarded as less serious an offense than actually
uttering an untruth. One could imagine, however, that Shawn has had
legal advice since his original BBOARD message.
∂22-Dec-84 2036 JMC re: Times for another visit
To: ATP.Bledsoe@UTEXAS-20.ARPA
[In reply to message sent Sat 22 Dec 84 20:55:37-CST.]
How about Friday, Feb 15 and Saturday Feb 16?
∂22-Dec-84 2334 JMC EG&G
To: CET@SU-AI.ARPA
I have decided I don't want to be liaison to EG&G, with the proviso
that if they call me and explain our common interests, I might
change my mind. However, if Doug Lenat can no longer serve as
liaison to Texas Instruments, I might be interested in that.
∂22-Dec-84 2336 JMC
To: CET@SU-AI.ARPA
Incidentally, is MCC in Austin a forum member? I have started consulting for them.
∂23-Dec-84 1055 JMC Pournelle's story
To: shahn@SUMEX-AIM.ARPA
The story conflates two facts.
First in our first robotics proposals we did discuss a robot that
could put together a Heathkit. I thought in 1965 that there was some
possibility of doing it in two years. It seems to me that this is an
engineering goal that could be achieved by an intensive effort, even by a
few people. However, it doesn't seem that it will be achieved anytime
soon.
Second a few years later we bought a Heathkit color TV to serve
as a color display for the computer. It was perhaps the first color
display on a computer. However, we intended to put it together by hand
and did.
Jerry Pournelle, as a professional science fiction writer, is
hereby granted a literary license to continue telling the story in its
original form. He could add corroborative detail to the effect that
the unopened box with the Heathkit is still somewhere in the air conditioning
ducts of the Donald C. Power Laboratory waiting for a sufficiently
ambitious robot. He could even add that there is a prize for its
robotic assembly. Indeed perhaps Ed Fredkin could be persuaded to offer
such a prize.
∂23-Dec-84 1056 JMC Pournelle's story
To: shahn@SUMEX-AIM.ARPA
CC: pournelle@MIT-MC.ARPA
The story conflates two facts.
First in our first robotics proposals we did discuss a robot that
could put together a Heathkit. I thought in 1965 that there was some
possibility of doing it in two years. It seems to me that this is an
engineering goal that could be achieved by an intensive effort, even by a
few people. However, it doesn't seem that it will be achieved anytime
soon.
Second a few years later we bought a Heathkit color TV to serve
as a color display for the computer. It was perhaps the first color
display on a computer. However, we intended to put it together by hand
and did.
Jerry Pournelle, as a professional science fiction writer, is
hereby granted a literary license to continue telling the story in its
original form. He could add corroborative detail to the effect that
the unopened box with the Heathkit is still somewhere in the air conditioning
ducts of the Donald C. Power Laboratory waiting for a sufficiently
ambitious robot. He could even add that there is a prize for its
robotic assembly. Indeed perhaps Ed Fredkin could be persuaded to offer
such a prize.
∂24-Dec-84 1320 JMC former Gabriel large monitor
To: ME
I have move it to Art Samuel's office. He was finding the character size
of the regular monitor too small.
∂24-Dec-84 1556 JMC
To: vardi@SU-AIMVAX.ARPA
jmc - Here's how it works from SAIL. The final t of yktvmt may be replaced
by the Yorktown designation of one of their other 308x machines. The quotes
are required from SAIL, but other systems may not require it.
"loser.yktvmt.ibm"@csnet-relay
∂24-Dec-84 1556 JMC re: CSnet (from SAIL's BBOARD)
To: OTHER-SU-BBOARDS@SU-AI.ARPA
jmc - Here's how it works from SAIL. The final t of yktvmt may be replaced
by the Yorktown designation of one of their other 308x machines. The quotes
are required from SAIL, but other systems may not require it.
"loser.yktvmt.ibm"@csnet-relay
∂24-Dec-84 2128 JMC
To: CLT
%3McCarthy, John (1961)%1: "A Basis for Mathematical Theory of Computation",
in %2Proc. Western Joint Computer Conf.%1, May 1961, pp. 225-238.
Later version in Braffort, P. and D. Hirschberg (eds.) %2Computer
Programming and Formal Systems%1, North-Holland Publishing Co. (1963).
∂25-Dec-84 1200 JMC fantasy in HCN
To: cs.rich@UTEXAS-20.ARPA
Dear Elaine,
Can you supply me with the name and address of the Brown
University Alumni Magazine. I contemplate sending them the following:
"The University has been unnecessarily rigid in its outright
rejection of the student request that the Health Service stock cyanide.
We propose a compromise. While the University cannot stock cyanide for
reasons of safety and law, it should agree to stock the Kool-Aid and a
suitable number of large kettles. Students would then have only to supply
their own cyanide.
"The proposal to change the name of the University to the Reverend
Jim Jones University should be examined very cautiously. It might cause
confusion with the Bob Jones University and thereby jeopardize Brown's tax
exemption."
Of course, if an alumna were to associate herself with some
variant of such a tease, they would be more likely to print it. I
have yet to check the spelling of Kool-Aid.
I look forward to seeing you again on my next visit to MCC,
which will probably be in February.
Regards,
John McCarthy
∂26-Dec-84 1149 JMC re: Sequent
To: RPG
[In reply to message rcvd 26-Dec-84 08:58-PT.]
1985 Jan 7 2pm
∂26-Dec-84 1308 JMC Common Lisp sponsored by another organization
To: RPG
My opinion is that Common Lisp should not be sponsored by any other
organization as long as the five are willing to keep it alive and
unless there arises a political difficulty making it desirable to
find a neutral sponsor, e.g. if the five are considered unneutral
by some force that it is important to mollify. It will actually
be less trouble for you than explaining things to the sponsoring
organization. If you need another individual to lend weight, I'm
available provided it isn't too much work.
If you want to circulate this, go ahead.
∂26-Dec-84 1956 JMC reply to message
To: JK
[In reply to message rcvd 26-Dec-84 18:06-PT.]
See you then.
∂27-Dec-84 0949 JMC reply to message
To: RA
[In reply to message rcvd 27-Dec-84 09:41-PT.]
There will be a consulting bill to MCC for the Texas trip.
∂28-Dec-84 1223 JMC for inclusion in Prolog Digest
To: restivo@SU-SCORE.ARPA
CC: KUO@SU-AI.ARPA
I haven't followed the details of the discussions of the Mr. S and Mr. P
problem in the Prolog Digest, but a cursory look suggests that they miss
the original point of the problem. I found the problem on a bulletin
board and haven't been able to trace its origin, but there it was posed
as an ordinary puzzle, and it wasn't suggested that it had anything to
do with computing or AI or programming languages. I posed it as a problem
in the formalization of knowledge, i.e. making a language and a set of
axioms that include the following.
1. It can express that a person knows the value of a number or other
quantity.
2. It can express that a person knows that another person knows or
doesn't know a fact.
3. It can express the effects of a person learning something, i.e.
in a later situation he knows something he didn't know before.
4. The language should be sufficient to express the facts of Mr. S
and Mr. P and the earlier problem of the three wise men with spots
on their foreheads.
5. In the case of Mr. S and Mr. P, the transformation of the problem
into purely arithmetic terms should be a consequence of the general
axioms of knowledge and the axioms of the particular problem. For example,
it should follow from the statement that Mr. P doesn't know the numbers
that the product is not the product of two primes. It is not permissible
to take this fact about the product as the original expression
of Mr. P's statement that he doesn't know the numbers.
I wrote some axioms for knowledge and the problems after having
solved the problem by a combination of hand calculation and a small
Lisp program. Ma Xiwen of Peking University, who was visiting Stanford for
a year, revised my axioms and used Richard Weyhrauch's FOL proof checker
to transform the problem into an arithmetic problem, i.e. get rid of
the knowledge operators. I discovered later that there was a bug in
his axioms. Ma and Guo Weide (Victor Kuo) wrote a subsequent paper
about the problem using a modal logical formalization that was presented
in IJCAI-83 and was referred to in the Prolog Digest.
If anyone is interested I have some files with formalizations
of knowledge and also files with Ma's FOL axiomatization. Also
KUO@SU-AI.ARPA is the Guo Weide of the cited paper.
Perhaps I am missing some important point, but I don't see
that the Prolog formulations in the Digest solve the problem I posed.
They seem to be mere programming exercises in which the human
transforms the statements about knowledge to computational statements.
It would be more interesting if there were a way to express statements
about knowledge in Prolog and if the solution of Mr. S and Mr. P
was obtainable by giving these statements about knowledge a computational
interpretation. I am skeptical that this can be done in any honest
way, i.e. that isn't ad hoc to the particular problem.
∂28-Dec-84 1233 JMC "after I showed him the recent mail ..."
To: RPG
Why did you say "after I showed him the recent mail ..."? You didn't
show it to me. I have been skimming the Common Lisp Digest right along,
even though this is the first time I have felt impelled to comment.
I could explore AAAI's interest with Woody or the Council, but I am
quite certain that the quinquevirate will not save itself work by
getting an outside sponsor. It just means that everything will
have to be explained twice and then checked to make sure AAAI or
whoever got it right.
∂31-Dec-84 1102 JMC re: mtg
To: CLT
[In reply to message rcvd 31-Dec-84 10:47-PT.]
∂31-Dec-84 1110 JMC re: so what's the reply?!!!
To: CLT
[In reply to message rcvd 31-Dec-84 11:08-PT.]
It was yes, but evidently I goofed somewhow.
∂31-Dec-84 1434 JMC
To: CLT@SU-AI.ARPA, LES@SU-AI.ARPA, RPG@SU-AI.ARPA
See intro.pro[f84,jmc]
∂31-Dec-84 1720 JMC testr
To: RWW
testr[x,p,f,u] ← if p x then f x
else if atom x then u[]
else testr[car x,p,f,λ[[].testr[cdr x,p,f,u]].
Here u is a function of no arguments. If there is a subexpression
of x that satisfies p it returns f of that subexpression. Otherwise
it executes the function u of no arguments. Jim Slagle (sometime in 1959
or 1960) complained that this didn't work. The reason is that
cdr x was interpreted dynamically by Lisp 1. I understood the
problem, but didn't follow the discussion of how to fix it. Steve
Russell, I think, invented the FUNARG device that packaged the
current alist with the functional arguments.
∂31-Dec-84 1857 JMC
To: bscott@SU-SCORE.ARPA
It is very cold in MJH now.
∂31-Dec-84 1918 JMC complaint about heat withdrawn in confusion
To: bscott@SU-SCORE.ARPA
Jiggling the thermostat produced heat even though the thermostat
was set far above the approximately 63 degree room temperature.