JAN.OUT[LET,JMC] - www.SailDart.org

perm filename JAN.OUT[LET,JMC] blob sn#143802 filedate 1975-02-02 generic text, type T, neo UTF8

∂02-FEB-75 1452 ESS,JMC
I purged them myself.
␈ CC: REG

∂02-FEB-75 0359 ESS,JMC
1. It seems to me that it is quite easy to get proofs into the form
I want. Namely, if you always write each step onto the backup.tmp
file as it is produced, then the backup.tmp file will have what I
want. You can put a character in front of each proof step that will
cause it to be ignored when a backup.tmp file is read back in.

2. summerize consists in draining the antifreeze from a car and putting
a heavier grade of oil in. You want "summarize".

3. One should be able to SHOW part of a formula to see if one has got
the part selection right, e.g. SHOW :7#1#2.

4. If backup.tmp has all your erroneous steps so that when you fetch it
back you get errors and FOL stops. Is there a way to continue skipping
the bad step?
If so, is it in the manual.

5. Consider a command HIDE with a step range as an argument. Its effect
is to label the steps as done with. The only effect is that a simple
SHOW command shows only the unhidden steps. This makes it easier to
see what is available and intended for further use. A FULL SHOW would
show hidden steps also. - Not a very high priority yet.
␈ CC: RWW

∂01-FEB-75 2148 ESS,JMC
What are you currently doing for us?
␈ CC: jj

∂01-FEB-75 2147 ESS,JMC
I see that Schukin hasn't logged in since 13 Jan. What is he doing?
␈ CC: tob

∂01-FEB-75 2139 ESS,JMC
This microfiche reader gets very hot and glares in my eyes, so we should
definitely look at the Kodak readers.
␈ CC: paw

∂01-FEB-75 1959 ESS,JMC
What are the chances of getting UNIFY working in time to produce a proof by
Tuesday, or should I prove around it?
␈ CC: rww

∂01-FEB-75 0451 ESS,JMC
Now UNIFY doesn't seem to work properly in either FOL or CFOL, but the
error messages are different. CFOL's UNIFY always fails for lack of
UPRECHECK, but FOL succeeds in converting ¬∀N1 K.¬Q(K,N1) to ∃N1 K.Q(K,N1),
but fails in converting ¬∀N1 K.¬(Q(K,N1)∧((N1=0∧K=(M*0))∨(¬(N1=0)∧Q(K+M,PRED N1))))
to ∃N1 K.(Q(K,N1)∧((N1=0∧K=(M*0))∨(¬(N1=0)∧Q(K+M,PRED N1)))).
␈ CC: RWW

∂01-FEB-75 0413 ESS,JMC
We will do some thinking about sharpening the objectives, but I interpret
what you said as promising a more detailed reaction from you before we
try to put it in writing. I am scheduled to be on sabbatical in Japan
between March 25 and June 30, so I hope we can get it all settled by then.
Please keep both Les and me informed, however you do it, because we keep
somewhat asynchronous hours and sometimes one will be able to react
faster than the other. Otherwise, the carbons of ARPAnet messages are
just as legible as the originals.
␈ CC: licklider%ISIA

∂31-JAN-75 1250 ESS,JMC
Lick:
We are about to lay off Jim Hieronymus, because our attempts to
get support for vision in coal mining have come to nothing. Cordell tells
me that the speech committee considers that all effort must go into
short term demos in order to justify a new five year plan and that no
support for basic work in acoustic phonetic recognition is available
until the start of the new period. This seems short-sighted, but I
can understand how the speech committee got itself into this position.
I have heard that indirectly that you and others have a good opinion
of Hieronymus, but I don't know it for a fact. We could support him
for a while and add him to our proposal, but I cannot independently
evaluate how his work compares with other speech work. If you have
an opinion let me know.
␈ CC: licklider%ISIA

∂31-JAN-75 0000 ESS,JMC
Letter to zohar
␈ CC: :TASK.

∂30-JAN-75 1710 ESS,JMC AT TTY36 1710
Can you come by for a minute?
␈ CC: tw

∂30-JAN-75 0007 ESS,JMC
Got bit by that bug again in trying to prove ∀N.0+N=N.
␈ CC: RWW

∂29-JAN-75 2328 ESS,JMC
I was that very JMC - John McCarthy.
␈ CC: klh%AI

∂29-JAN-75 2327 ESS,JMC
I was that very JMC - John McCarthy.
␈ CC: klh

∂29-JAN-75 2321 ESS,JMC
INTFOL.DMP[258,JMC] is a FOL core image with 6 steps of a proof based on
INTEGE.AX[258,JMC]. An attempt to finish the proof by "∀I 6 M N;" results
in a complaint "attempt to generalize on a variable appearing free in
an assumption or axiom. Of course, M and N are free in step 1, but step
1 has been removed as an assumption by the ⊃I. So why can't I generalize.?
␈ CC: RWW

∂29-JAN-75 2229 ESS,JMC
Sorry to have missed the meeting on AI and data bases. I had a meeting
in New York early in the week, and I got confused as to the days of this
meeting and now my class at 11am tomorrow cannot be called off.
I would like to see the proceedings if any, and some of the ideas I
have been slowly pursuing on data description may be relevant.
␈ CC: FIELDS%ISIA

∂29-JAN-75 1421 ESS,JMC
My secretary remembers it as a big yellow thing, but we can't find it, so
please send another. It is my custom, however, not to answer questions
in which no choice expresses my opinion.
␈ CC: walker%SRI-AI

∂29-JAN-75 0354 ESS,JMC
Well, I have to turn it off to sleep. Hope it comes on again.
␈ CC: TAG

∂29-JAN-75 0342 ESS,JMC
The fan in the Imlac is making a truly awful noise, but I am reluctant to turn
the Imlac off since it doesn't like that. Should I do so?
␈ CC: TAG

∂29-JAN-75 0153 ESS,JMC
Thanks.
␈ CC: tag

∂29-JAN-75 0058 ESS,JMC
I want to subscribe to
Encounter
Encounter Ltd.
59 St. Martin's Lane
London WC2N4JS
I will give you a check for $15.50 and they should send it to my home.
␈ CC: paw

∂29-JAN-75 0051 ESS,JMC
Please mail a copy of sheldo.ns[let,jmc] to Margaret Young, 2505 Greer Rd.,Palo alto.
␈ CC: paw

∂29-JAN-75 0039 ESS,JMC
Please decorate bremer.le1 which is to Prof. Hans Bremermann,math.dept. UCB.
␈ CC: paw

∂29-JAN-75 0027 ESS,JMC
Please put car.ess[ess,jmc] in a nice font, poxify it, and send
a copy to Robert S. Foote, Port Authority of New York and New Jersey,
suite 5614, One World Trade Center, New York N.Y. 10048.
␈ CC: paw

∂29-JAN-75 0026 ESS,JMC
Janet Kreiling and Bob Caspe say hello.
␈ CC: bo

∂29-JAN-75 0023 ESS,JMC
Please decorate Norbye.le1[let,jmc].
␈ CC: paw

∂29-JAN-75 0007 ESS,JMC
Erik:
It was I who told you that. I don't know that there is any general
principle, but I have two instances: Zohar and a Stanford Israeli geophysicist
name Amos Nur. In Zohar's case, he was invited by Ershov without consulting
any higher authority, either because Ershov forgot that Manna is an Israeli
(I am sure I had so referred to him in a letter when I first became interested
in Manna's work), or because he didn't know that it would offer difficulty,
or because he didn't want to do the system's nasty work for it. After the
difficulty became apparent, he and others at Akademgorodok made some effort
to get Manna a visa, and spoke hopefully up to the last minute, but maybe
that was partially to avoid confronting other participants with a clear
choice.
In my opinion, one should refrain from holding international
meetings in the Soviet Union that are nominally Soviet meetings. If
the meeting is almost all Soviet, they can do as they choose, but if
the meeting is, say, half Soviet so that a considerable amount of the
communication is between non-Soviet people, then one shouldn't go unless
the meeting is formally international or there is some other guarantee
that Israeli's can come. I suppose that if the issue was moot, because
there were no invitable Israelis, this would be a borderline case, and
one's decision might depend on how much effort one felt like investing
in trying to make the Russians behave better.
␈ CC: erik%AI

∂26-JAN-75 1853 ESS,JMC
See horse.msg[1,elf]
␈ CC: elf

∂26-JAN-75 1727 ESS,JMC
That was suzuki.rec
␈ CC: paw

∂26-JAN-75 1526 ESS,JMC
susuki.rec[let,jmc] should go to Joseph Traub at CMU, Jack Schwartz and
Courant Institute, New York Univ., Mike Dertouzos at Project MAC (there
should be a P.S. in the note to Dertouzos asking him to forward the
note if he is not the right person). Suzuki will give you a resume to
include, and please send it right away without waiting for my signature.
␈ CC: paw

∂26-JAN-75 0128 ESS,JMC
Please decorate hobbs.le1.
␈ CC: paw

∂25-JAN-75 0216 ESS,JMC
I think I would like to talk in the next ai circle meeting late in feb.
␈ CC: tw

∂25-JAN-75 0024 ESS,JMC
The rest of previous message is "Friday except for wednesday evening.
The idea is for you to try to define a complete set of chessboard and
chess position predicates together with the axioms relating them. This
will help both the search for chess patterns and FOL proofs of chess
assertions. These two tasks have further parallels.
␈ CC: dew

∂25-JAN-75 0023 ESS,JMC
I have an idea for you, but I shall be away from Sunday through
␈ CC: dew

∂25-JAN-75 0002 ESS,JMC
Include me on next organized blood trip.
␈ CC: bpm

∂24-JAN-75 1439 ESS,JMC
I need to talk to you about Morales.
␈ CC: dcl

∂24-JAN-75 0002 ESS,JMC
Don't change POX so that present files don't work without Ralph's permission.
␈ CC: rem

∂23-JAN-75 1343 ESS,JMC
The first meeting of the 3850 club was held yesterday. Xerox-parc
seems to have decided that they want nothing to do with IBM and
all its works, and, besides, they have no current problem that this
solves. The other six (su-ai,sri-ai,sumex,imsss (Pat Suppes), the
academic computation center of Stanford, and presumably ARPA are
interested enough so that we have placed an order with IBM to get on
the queue, which is already 22 months long. Of course, won't be
any financial commitment until a few months of delivery so there is
plenty of time to see if the demand is effective. The biggest question
is whether there is real demand backed by money - the rental being
about $22K per month, and the biggest open technical question at the
moment is what can be done about high speed communication. We may
propose to you a large files project that will involve organizing
all unclassified ARPA reports (or even encrypted classified reports)
and whatever other useful material can be found in computer accessible
form. We are also looking into the possibility of piggybacking on
one of IBM'a other 3850 customers in this area.
␈ CC: fields%ISIA

∂23-JAN-75 1342 ESS,JMC
∂23-JAN-75 1251 network site BBN
Date: 23 JAN 1975 1552-EST
From: FIELDS at BBN-TENEX
Subject: STANFORD PROPOSAL
To: JMC at SU-AI
cc: LICKLIDER

JOHN,
I READ YOUR PROPOSAL. I AM VERY INTERESTED IN
TWO ITEMS. WOULD IT BE POSSIBLE TO GET A COPY OF
LOW'S THESIS (AIM 242) BY US MAIL? IT WOULD BE TOO
LONG TO PRINT ON MY TERMINAL. ALSO, ON PAGE 20
YOU INDICATE THAT IN JANUARY, 9175 WINOGRAD SHOULD
HAVE A COPY OF INITIAL SPECIFICATIONS FOR A KNOWLEDGE
REPRESENTATION LANGUAGE. COULD I SEE A COPY OF THAT
AS WELL? THANKS FOR BOTHIN ADVANCE.

A WORD OF FRIENDLY ADVICE. ON PAGE 41 YOU
INDICATE THE NEED FOR A SUBSTANTIAL AMOUNT OF TENEX
TIME AND ARPANET ACCESS. BOTH COST REAL DOLLARS
AND I SUGGEST THAT IN YOUR NEGOCIATIONS WITH
LICK THESE ITEMS ARE NOT OVERLOOKED. THE TENEX
TIME WILL POSE THE GREATEST PROBLEM. ALTHOUGH
NEW TENEXES ARE EXPECTED, THE DEMAND WILL EXCEED
THE SUPPLY, AND LICK WILL HAVE TO MAKE A DECISION ON
ALLOCATION SO AS NOT TO GIVE AWAY MORE THAN 100%
OF TENEX AVAILABLE. THIS WAS JUST TO HIGHLIGHT
A POTENTIAL PROBLEM TO ASSIST YOU.

BEST
CRAIG

P.S. HOW ARE THINGS IN THE 3850 CLUB?
-------

␈ CC: les

∂23-JAN-75 1339 ESS,JMC
There is a possibility that we will be able to support you, but we
will have to discuss it. I can't meet at the end of next week, because
I will be out of town almost all week, but the following week or even
the weekend are possible for me.
␈ CC: ajt

∂22-JAN-75 1440 ESS,JMC
Balint Domolki, who has been visiting us for a few months from Hungary,
would like to visit M.I.T., including the AI Lab and Project MAC especially
Jack Dennis on his way back to Hungary - for about two non consecutive days.
He has been actually useful here, because he has caught on to FOL and
our work in mathematical theory of computation very quickly. He is called
DBX here, and he will send you a message suggesting when he might come
and asking if that is ok. In AI Lab, he will want to talk to the ACTOR
people mainly.
␈ CC: phw%AI

∂22-JAN-75 1437 ESS,JMC
Do you know if the Honeywell 6000 series is the same as the GE645 and
if so what is the LISP situation on this machine? Balint Domolki,
a visitor from Hungary here, who is called DBX would like to know.
␈ CC: cdr

∂22-JAN-75 1355 ESS,JMC
call coates

␈ CC: :TASK.

∂22-JAN-75 0140 ESS,JMC
Note to Dr. David G. Hays, Twin Willows, Wanakah, New York 14075

Dear David:

I have just received an issue of the \F1American Journal
of Computational Linguistics\F0, and I would like to congratulate you
on its form. Now I really have to get a couple of microfiche readers -
one for my office and one for home. By the way, are you enthusiastic
about any particular reader?

Best Regards,
␈ CC: paw

∂22-JAN-75 0107 ESS,JMC
When can you fix my imlac? The run light won't go on.
␈ CC: tag

∂21-JAN-75 1427 ESS,JMC
Unfortunately, the man I was counting on to work on the car finding
problem finds it scientifically uncongenial. I have some other ideas,
and I'll let you know if any of them pan out.
␈ CC: carlstrom%ISIA

∂21-JAN-75 1405 ESS,JMC
Harold: Please don't log in during the day on datadiscs. This is a
scarce resource.
␈ CC: ber;les

∂20-JAN-75 1554 ESS,JMC
Letter to Zohar.
␈ CC: :TASK.

∂20-JAN-75 0207 ESS,JMC
paper on cars

␈ CC: :TASK.

∂20-JAN-75 0019 ESS,JMC
analog.le1

␈ CC: :TASK.

∂20-JAN-75 0017 ESS,JMC
call up about ticket

␈ CC: :TASK.

∂19-JAN-75 1354 ESS,JMC
call hutchinson

␈ CC: :TASK.

∂19-JAN-75 1351 ESS,JMC
bah
␈ CC: :FOO.

∂19-JAN-75 0330 ESS,JMC
Is it possible to mail to file that doesn't have an extension?
␈ CC: bh

∂19-JAN-75 0328 ESS,JMC

SABBATICAL
␈ CC: :TASK

∂19-JAN-75 0326 ESS,JMC
SABBATICAL
␈ CC: :TASK

∂18-JAN-75 2013 ESS,JMC
Thanks for the ccpy of you letter to Bethe. A letter dissenting
is on its way via campus mail, but it is on file as lederb.le1[let,jmc],
and a letter to Bethe - previously started - is bethe.le1[let,jmc].
␈ CC: lederberg%ISIA

∂18-JAN-75 1844 ESS,JMC
I have just been badly bitten by a bug in E. I wanted to go up several lines
so I typed ↑ several times in quick succession. This got me into a state
where I was in IOWQ, but E was completely unresponsive to commands. I tried
↑C and reenter, but was told there was no reenter address - a defect I
suppose. Then I tried CONT after which E was responsive so naturally I
tried double bucky dot. This resulted in "FATAL ERROR -- BUG IN WRITE CODE
HALT AT USER 406024. I then tried to edit the file and was tol that the
"DIRECTORY INVALID OR UNDESIRED. KEEP OLD ONE AS PART OF TEXT?" I told
it "n" and it showed me the file most of which was gone, ALAS.
␈ CC: als

∂18-JAN-75 0018 ESS,JMC
Please decorate bethe.le1.
␈ CC: paw

∂17-JAN-75 2242 ESS,JMC AT TTY15 2242
Is it reasonable to hopeto go to press on Monday?
␈ CC: AJT

∂17-JAN-75 1222 ESS,JMC
A final test.
␈ CC: MARKOWITZ%MIT-MULTICS

∂17-JAN-75 1156 ESS,JMC
To Peter Weiner's group.
Peter:
This concerns your inquiry to Col. Russell. I don't know him, but
I think we will get a better result if you tell him what the ARPA
community would like in the way of network access and the possibility of
keeping some non-ARPA financed installations on the net. If he has to
make it all up out of his head, the result may be unsatisfactory, aad
you may have to work much harder to change it than if you say what the
community needs in the first place.

Second point: I am currently trying to organize a local co-op
to rent and operate an IBM 3850 mass storage: 7 users at about $3500
per month each would get 5 billion bytes. The preliminary mode of
operation would be FTP if the net were used or more likely FTP like
for Bay Area users, i.e. users would transfer whole files to and from
the facility. Additional ARPAnet interest is solicited. Storage cost
between $.75 and $1.00 per million characters per month which would go
down to $.35 if the demand grew to allow a maximum system.
␈ CC: Corby%MIT-MULTICS;Markowitz%MIT-MULTICS

∂17-JAN-75 1152 ESS,JMC
This is just a test.
␈ CC: Markowitz%MIT-MULTICS

∂17-JAN-75 1151 ESS,JMC
Peter:
This concerns your inquiry to Col. Russell. I don't know him, but
I think we will get a better result if you tell him what the ARPA
community would like in the way of network access and the possibility of
keeping some non-ARPA financed installations on the net. If he has to
make it all up out of his head, the result may be unsatisfactory, aad
you may have to work much harder to change it than if you say what the
community needs in the first place.

∂17-JAN-75 1131 ESS,JMC
Peter:
This concerns your inquiry to Col. Russell. I don't know him, but
I think we will get a better result if you tell him what the ARPA
community would like in the way of network access and the possibility of
keeping some non-ARPA financed installations on the net. If he has to
make it all up out of his head, the result may be unsatisfactory, aad
you may have to work much harder to change it than if you say what the
community needs in the first place.

∂16-JAN-75 1816 ESS,JMC
please decorate oxford.le1[let,jmc]
␈ CC: paw

∂16-JAN-75 1718 ESS,JMC
Craig:
I don't recall McClellan. There was some guy I thought was from
Washington who discussed it with me, but, as I recollect, it was his move
when we stopped talking. The A.P. wire itself costs us $15.00 per month
and last year A.P. paid for the salary of the guy who worked on it, but
this year ARPA has that honor. We are glad to have other ARPAnet use and
can easily set it up for occasional use, but A.P. considers that we can
use it only for experimental purposes, and other users would have to
negotiate with them. If you think this is vague, you are right. If an
application looks to me as though it would require negotiation with A.P.
and the New York Times news service, I will be glad to help get things
started. The best thing would be for McClellan to send me a note saying
what he wants to do.

There are now seven potential users of the 3850, and the cost of
an essentially minimal system seems to be about $22K per month. Before
the first meeting of the potential subscribers, I would like any information
you have about FTP2 and about the possibilities for higher speed network
transmission, because it looks like present network speeds or even the
mythical 50K bits per second won't do.
␈ CC: fields%ISIA

∂15-JAN-75 1728 ESS,JMC
Please decorate andrei.le4[let,jmc]
␈ CC: paw

∂15-JAN-75 0305 ESS,JMC
don't forget mccloskey
␈ CC: jmc

∂15-JAN-75 0304 ESS,JMC
Please decorate andrei.le4[let,jmc].
␈ CC: paw

∂15-JAN-75 0220 ESS,JMC
SEND OFF SUM.CAR[CUR,JMC]
␈ CC: JMC

∂15-JAN-75 0113 ESS,JMC
I would like someone, and it probably should be you, to write a proposal
to DECUS that DECUS adopt an ARPANET like message and FTP protocol
using the telephone system. It would be desirable that the protocol
be as compatible with ARPAnet as possible with any reductions necessary
for practicality. It would require a telephone dialer for output and
only suitable external telephone numbers for input. It should contain
as few eccentricities of PDP-10 as possible so that DECUS could later
propose it to other user groups.
Maybe Bell Labs would be interested for the thing itself and because it
might be could for AT&T, and I know you are a strong believer in the
proposition that what is good for AT&T is good for the country.
␈ CC: BH

∂15-JAN-75 0102 ESS,JMC
Thanks for your comments on my lecture. I haven't now anything
concrete to say about your work, but I would be glad to listen
to you for an hour and then comment. How about Friday afternoon?

1. The choices that I have made so far in abstract syntax do not
commit me to use LISP data structures, and I don't intend to
confine myself to them. I have not thought much about abstract
data structures, and I consider them part of the semantics. I
certainly agree that they are desirable.

2. mkebinop is a reasonable way to do it, but since in applications
I have considered so far, including all those put on the board, it
would lengthen the formualas although it would reduce the number
of axioms.

3. I don't see the point of the suggestion concerning the definition
of value(e,β). We can't do the semantics without introducing +
in the metalanguage so in order to finish the definition I would
still have to treat the case of a sum of numerals.
4. In the sense that I want to use, x is not inevitable in that
expression. It is need not be for the purpose of reducing
expressions to strong canonical form. Of course, it would reduce
that expression to x. However, if we want the fact that 0=0 is
true to be taken into account, where do we draw the line in knowing
at "syntax time" that an expression will always give the same
truth value. If we want the definition of inevitability to take
into account all constant expressions, we could define it that way,
but then the concept of inevitable variable would be non-effective,
because in general it is undecideable whether an expression is
constant.

Thanks again for the comments, and if you feel like commenting
on future lectures, please do so, and I'll try to respond. I
will mention the mkebinop idea which has come up before.
␈ CC: dbx

∂14-JAN-75 0742 ESS,JMC
An improved protocol would certainly help, because most
traners are protocol limited today, and I agree that a separate
fast protocol is appropriate. However, for many applications the
50 kbaud is too slow, even if it could be attained.
␈ CC: fields%ISIA

∂13-JAN-75 1502 ESS,JMC
Craig:
As you know, I turned down your suggestion of trying to organize
a 3850 co-op. Well, I have thought better of it, and have now expressions
of interest from the Lederberg and Suppes PDP-10s at Stanford and from
Bert Raphael at SRI-AI. The idea would be to rent a minimal 3850 and
attach it to our 3330 and PDP-10. Faster transmission than the present
ARPAnet would probably be required for many of the contemplated applications.
and this might limit the main usefulness to nearby sites. However, if the
ARPAnet were to improve its EFFECTIVE speed of file transfer, then ... .
I will also ask PARC when they return my call. When we have more information
I will send you another message unless the whole thing falls flat. We wouldn't
get delivery until 1976 no matter how fast we act, but after that it could
be up almost immediately. It occurs to me that ARPA might want a share for
its mini-users apart from the shares that the participating ARPA projects
would buy out of their contracts. I don't know how much a share would cost -
perhaps $50K-$75K per year but will know soon. Nothing fancy is contemplated.
Files could be FTPed, and if there were five or seven users, a share would be
seven or five BILLION bytes.

I will be glad to attend the RAND meeting.
␈ CC: fields%ISIA

∂13-JAN-75 1017 ESS,JMC
I now have an idea of how to put vision into FOL in the spirit of FOL.
It involves being able to verify that a scene satisfies a visual pattern,
as distinct from finding the pattern in the scene. It is therefore
epistemological rather than heuristic as is FOL in general. I want to
peddle this idea to you and to Tom Binford.
␈ CC: ajt

∂13-JAN-75 0847 ESS,JMC
I need to circularize dept saying I propose to appoint Manna.
␈ CC: paw

∂11-JAN-75 1805 ESS,JMC
The letter is to Professor Alphonse Juilland.
␈ CC: paw

∂11-JAN-75 1804 ESS,JMC
That's juilla.le1[let,jmc].
␈ CC: paw

∂11-JAN-75 1804 ESS,JMC
Please decorate juillan.le1 as a letter. It started as a memo.
␈ CC: paw

∂10-JAN-75 1956 ESS,JMC
Let me show you my calculations.
␈ CC: mg

∂10-JAN-75 1949 ESS,JMC
finish off fordha.le4
␈ CC: jmc

∂09-JAN-75 2301 ESS,JMC
The reason there is no NILSON at SRI-AI is that the person you want is NILSSON.
␈ CC: PHW%AI

∂06-JAN-75 2312 ESS,JMC
read task and calend
␈ CC: jmc

∂06-JAN-75 2044 ESS,JMC
Les did a monumental job of getting out that proposal. There are somm
rough spots, and there probably will be some afterthoughts, but soon.
␈ CC: licklider%ISIA

∂06-JAN-75 2040 ESS,JMC
Yes, I will be glad to be a member for whatever I can do before March 25 or
after June 30; between those dates I will be in Japan. I have no problem
with non-disclosure. The major issure that occurs to me is how to get a
good successor to the PDP-10. I suggest that we have a telephone conference
as soon as you have a draft agenda. Also please send a list of the network
addresses of the participants to all.
␈ CC: weiner%RAND-RCC

∂06-JAN-75 1124 ESS,JMC
The preceeding was to Winston, copy to you,because Minsker asksyou to.
␈ CC: raphael%SRI-AI

∂06-JAN-75 1108 ESS,JMC
As you have probably heard by now, Jack Minker is trying to organize
a panel on cybernetics and AI at 4IJCAI involving (Benjamin ?) Lerner,
a Soviet Jewish cyberneticist who has been unsuccessfully trying to
emigrate for several years. As he will no doubt cheerfully tell you,
Minker's motives are political. I have agreed to be on the panel if it
is held. In my opinion, you should take special care to get the
scientific appropriateness of the panel evaluated objectively. If Minker
succeeds in organizing a scientifically appropriate panel, it should be
included in the program, if not, not. My agreement to participate represents
more of a "put up" to Minker than a desire to discuss cybernetics and AI,
since I have no clear idea of what Soviets like Lerner mean by it. Therefore,
I consider myself of not much help in evaluating Minker's proposed panel.
$
␈ CC: phw%AI;raphael%SRI-AI

∂05-JAN-75 2326 ESS,JMC
A narrow escape from having to type it in again.
␈ CC: wd

∂05-JAN-75 2322 ESS,JMC
What we are contemplating is not clear even to us, although it is more like
supplementing and eventually replacing the Datadiscs by IC memory.
␈ CC: tk%AI

∂04-JAN-75 1603 ESS,JMC
I looked over your report to the U.N., and I am quite impressed with
what you have accomplished. It would be nice if you could write some
kind of research report or mini-paper before you leave. In any case,
if they ask me, I will say that you have been about the most useful
of our short term foreign visitors. If you could arrange to stay a bit
longer, we would be glad to have you.
␈ CC: dbx

∂04-JAN-75 0240 ESS,JMC
Please decorate peters.le1[let,jmc]
␈ CC: paw

∂03-JAN-75 1809 ESS,JMC
please decorate femano.le1[let,jmc].
␈ CC: paw

∂03-JAN-75 0406 ESS,JMC AT TTY15 0406
What if everyone ran finger at once?
␈ CC: bgb;les

∂03-JAN-75 0142 ESS,JMC
Yes.
␈ CC: les

∂02-JAN-75 2053 ESS,JMC
read task
␈ CC: jmc

∂02-JAN-75 2014 ESS,JMC
The problems you raise are mostly genuine, but they don't strike me
as the most immediate problems at this time.
␈ CC: wd

∂02-JAN-75 2008 ESS,JMC
Bill Pitts and Stan Kugell are the other people who are interested. Pitts
has quite a few good ideas. I would like someone to try to get some publicity,
e.g. Peninsula Electronic News and Grid.

Pitts thinks it can be done for less than $75.
More after I reread your opus. I was interrupted and forgot what
you said.
PUB,JMC is entirely devoted to this project.
␈ CC: wd

∂02-JAN-75 2002 ESS,JMC
Well, I have nothing yet. There will be a handout, but I haven't started
on it, because I have been tied up with the ARPA proposal. I shall have
to start out the class on Tuesday with old material. I will have to
get someone else to take the class on Thursday, maybe Bill Glassmire
talking on FOL, because I have to be in Rochest?r, N.Y. Please get a copy
of the new FOL manual as soon as you can. I may want to get some of
my old papers reprinted, but I haven't decided that yet.

Perhaps you could make a list of all the CS206 material we have on files.
I suppose it is all in PUB form by now; it can make new chapters when I
get to writing some transitional stuff.

Oh, by the way, we need a writeup of the LISP program for producing
blackboard LISP for PUB.
␈ CC: nxl

∂02-JAN-75 1954 ESS,JMC
Yes, I would like to see the PLASMA manual. I need to see some non-trivial
examples in order to convince myself there is something there even though
the idea that ACTORs sending each other messages would make a powerful
computing system seems plausible to me.
␈ CC: tw

∂02-JAN-75 0347 ESS,JMC
The letter aiorg.pln[cur,jmc] should be sent to Alistair Holden with a note
saying that it is in response to his inquiry.

␈ CC: paw
λ