perm filename DEC75.IN[LET,JMC] blob
sn#195392 filedate 1976-01-02 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00059 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00006 00002 ∂02-DEC-75 0958 HJS @ NBST
C00010 00003
C00011 00004 ∂04-DEC-75 0814 LES
C00014 00005 ∂04-DEC-75 1234 BG
C00015 00006 ∂04-DEC-75 1543 AJT
C00016 00007 ∂05-DEC-75 0740 SUZ via CMU
C00017 00008 ∂06-DEC-75 0450 LES Computer Accounting Files
C00018 00009 ∂06-DEC-75 1000 SUZ via CMU
C00019 00010 ∂06-DEC-75 1828 SGK
C00020 00011 ∂06-DEC-75 1844 TAG
C00021 00012 ∂06-DEC-75 1937 REG
C00022 00013 ∂06-DEC-75 2001 RCC
C00023 00014 ∂07-DEC-75 1133 LES PARRY, et al
C00027 00015 ∂07-DEC-75 1250 LES
C00029 00016 ∂07-DEC-75 1230 FTP: host CCA
C00031 00017 ∂07-DEC-75 1424 FTP: host UTAH
C00035 00018 ∂08-DEC-75 0554 TAG
C00036 00019 ∂08-DEC-75 1045 TOB
C00038 00020 ∂08-DEC-75 1141 BG
C00039 00021 ∂08-DEC-75 1752 FTP: host AI
C00040 00022 ∂08-DEC-75 2352 JFR
C00041 00023 ∂09-DEC-75 1006 LES Pony Services
C00042 00024 ∂10-DEC-75 0202 ME RAID in E's upper segment
C00043 00025 ∂10-DEC-75 1003 TOB meeting with Richard Liu
C00045 00026 ∂10-DEC-75 1507 LES Arm Controller Development
C00048 00027 ∂11-DEC-75 1629 FTP: host SRI
C00051 00028 ∂12-DEC-75 1317 FTP: host AI
C00053 00029 ∂12-DEC-75 2302 REG
C00054 00030 ∂13-DEC-75 0203 100 : ED RICE, HONEYWELL -- MULTICS via MTRT PARRY
C00055 00031 ∂13-DEC-75 1400 LES GFF Account
C00058 00032 ∂13-DEC-75 1533 AJT my committee
C00059 00033 ∂14-DEC-75 1825 LES Foonly Revival
C00063 00034 ∂14-DEC-75 2053 TJW thesis
C00064 00035 ∂15-DEC-75 2057 GFF via SRI SCCS
C00065 00036 ∂15-DEC-75 2356 TOB
C00068 00037
C00069 00038 ∂16-DEC-75 1357 LES Bellnet
C00070 00039 ∂16-DEC-75 1427 LES About the Bellnet name
C00071 00040 ∂16-DEC-75 1548 REF UCI LISP manual
C00072 00041 ∂17-DEC-75 1018 LES Dialnet
C00073 00042 ∂18-DEC-75 0242 RWW P&P
C00074 00043 ∂18-DEC-75 0243 RWW
C00075 00044 ∂18-DEC-75 1059 CRF via ISI dialnet
C00077 00045 ∂19-DEC-75 0335 PMF
C00078 00046 ∂19-DEC-75 0809 FTP:ALLEN NEWELL(A310AN02)@CMUA
C00083 00047 ∂19-DEC-75 1327 FTP:CERF at USC-ISI
C00084 00048 ∂19-DEC-75 1658 AJT Thanks
C00085 00049 ∂19-DEC-75 1616 100 : I-SPY@NSA via RTGT PARRY
C00086 00050 ∂20-DEC-75 1053 FTP:ALLEN NEWELL(A310AN02)@CMUA
C00098 00051 ∂22-DEC-75 0215 100 : PMF JOES CHRISTMAS THING
C00099 00052 ∂21-DEC-75 1548 100 : Charles H. Underwood via MTRT Parry
C00100 00053 ∂23-DEC-75 1328 LES
C00104 00054 ∂25-DEC-75 2016 REF primer
C00105 00055 ∂27-DEC-75 0159 LES
C00108 00056 ∂28-DEC-75 0857 FTP:FEIGENBAUM at USC-ISI
C00115 00057 ∂31-DEC-75 1907 TAG
C00117 00058 ∂31-DEC-75 1911 TAG
C00119 00059 ∂01-JAN-76 2042 TAG via AMET
C00122 ENDMK
C⊗;
∂02-DEC-75 0958 HJS @ NBST
John,
I have not run big jobs since October and I have no plans to do so.
The reasons for running those big jobs at that time were detailed
in a message I sent you at the end of October. I believe that the
claims about competing with the drawing program are unfounded.
Thanks.
Hanan,
∂02-DEC-75 0317 LES
I disagree with your assessment of the MAIL changes.
When finished it will greatly improve message sorting, make it possible
to respond to messages without having to make notes to yourself,
and will require knowing less than you do now, since most everything
will be done by E.
∂01-DEC-75 1209 S,LES Spending the Music Hardware Money
According to the Sponsored Projects Office (Simmons), we can't
formally sell the KA-10 to the Music Project. Therefore, in order to
gobble the money, it will be necessary to make up an equipment budget
for things that we want and get them to approve it.
I suspect that we will not be able to predict exactly how we wish to
spend it all right now, but apparently it will be necessary to devise
a plan and then request approval for deviations as needed.
∂26-NOV-75 1640 MA,JAM
Loren has tickets for "Song and Dance" for us in the following proportions:
DGL 2
JWG 1
JC 4 (has to pay for 2)
JMG 2
KS 1
JMC 2
LES 2
JAM 2
No more than this, please. Saturday, December 6, 8PM (show up 1/2 hr
earlier), tickets will be at the box office under the names whose
initials are above, De Anza college, Flint Center. If anyone is not
going to use their ticket, please tell Loren so he can give them to
someone else.
∂02-DEC-75 2008 DCL
VERIFICATION GROUP MEETING
Thursday Dec.4, 3.oopm
John McCarthy: Practical and Theoretical Considerations in
Programming Language Design.
Thursday Dec 11, 3.00pm
Mike Farmwald: What the Pascal Really Does
∂02-DEC-75 2025 DEW
I put a revised copy of my current stuff in your Wilkins file. The
example is on the last few pages. Dave
∂04-DEC-75 0814 LES
Don Woods is a random Computer Science student.
We let the SRI-ARC crew do the Network manual as a courtesy, since they didn't
have any very good alternatives. This turned out to be an amazingly bad idea
for reasons other than the one you mention and will not be repeated.
The largest lossage associated with this project has been the
consumption of local manpower (mainly Brian Harvey and Brian McCune)
in setting up and debugging all that Pub junk. This was supposedly
not part of the original deal, but was undertaken by Harvey on the
basis of personal friendship with Feinler.
A source of annoyence to me is that Harvey's Pub code is atrocious
and I keep getting asked to find and fix the bugs that BPM can't cope
with.
∂04-DEC-75 1234 BG
The next meeting of the seminar on the axiomatization of knowledge and actions
will be on Monday, December 8, at 4:00 p.m. Ivar Tonisson will speak on
relationships between philosophical logic and artificial intelligence.
∂04-DEC-75 1543 AJT
Although I realise that you have other things to do as well, I'm
just a wee bit disappointed that you haven't got very far in
reading my opus. My plan at the moment is to leave on Dec. 20
By then I would like to have sufficient criticism to go ahead
with the next draft. I will return sometime next year to take the
oral if/when you and the others think I'm ready for it. Please,
please try to read a bit more soon, even if you don't like it: it
cannot be improved until you like it.
Thanks. Arthur
∂05-DEC-75 0740 SUZ via CMU
I will try to login early in the morning. nori
∂06-DEC-75 0450 LES Computer Accounting Files
I received a slightly garbled message that sounded like you wanted to know
where the computer time accounting files are stored. They are located in
[ACT,SYS]. The cumulative year-to-date data is in a text file called YTD.
The cumulative data at the end of earlier months is in similar files named
<first three characters of month name>. Data for earlier years is in files
called YEAR.<last two digits of year>. There are also daily text files
going back a month and binary source files also going back a month.
Now, was that the right question?
∂06-DEC-75 1000 SUZ via CMU
Yes, I will finish in a week. Also I successfully transfered the verifier
here, so I no longer has to use as often as November. Thank you very much
any way for giving me the oppotunity to access SAIL system. Have
a nice Christmeas. Nori
∂06-DEC-75 1828 SGK
I am sorry that TVLOG managed to use so much time in my name. I have
disabled the program, and will examine less resource consuming alternatives.
∂06-DEC-75 1844 TAG
WHO THE HELL IS INCLUDED IN THE UBIQUITOUS GROUP "YOU" ? AND WHO IS
RESPONSIBLE FOR THE SIZE OF OUR BIG HAPPY TIME-SHARING FAMILY, ANYWAY?
∂06-DEC-75 1937 REG
PARRY will be flushed in the next system, hopfully Sunday.
∂06-DEC-75 2001 RCC
WE PUT THE LITERARY CRYPT HACK IN SYS:, HAVING CONVERTED FROM
TENEX TO SAIL MONITOR CALLS. IT IS CALLED LITCRY. HAVE FUN.
∂07-DEC-75 1133 LES PARRY, et al
I do not understand why your note of yesterday morning discusses PARRY.
You have asked REG to remove the "not logged in" privelege for PARRY
and I have notified BLF of its demise.
I realized a long time ago that you had a peculiar fixation on that
program that gets in the way of rational discussion. I would rather
flush it than continue talking about it. Since you are still talking
about it, however, I will put in my two bits' worth (hopefully for
the last time).
You say that "... .5% of a tight resource is non-trivial ...".
I would say that is damn near trivial, especially considering that it
is supposedly contributing to a dissertation project of a Computer
Science graduate student (BLF). If you want to cut marginal users,
consider SGK, who in November consumed over twice as many K minutes
and 5 times as many CPU minutes as ALL [100,100] users, including
the PARRY users.
You say "Of course, if we could know that it really only operated
when no-one else wanted any resources ... " We have known that for
some time. PARRY refuses to be run by unlogged users during prime time
or whenever the number of jobs is above some number, like 20. If
you don't know that, it isn't because you haven't been told. BLF
explained it to you when he put in that feature several years ago
and I have repeated it to you at least a half dozen times. It
appears that you have your receiver turned off.
On another issue, when we went over the November Bureaucracy listing,
you jumped on the fact that SUZ is by far the biggest user and you
said you were going to send him a message about that. I pointed
out that he is finishing his thesis work and that it is entirely
reasonable that he should be using a relatively large amount of time.
I also suggested that if you were going to send him a message, you
should think about why you are doing it and what he should do about
it.
I later heard from DCL that SUZ had received some kind of ominous
message from you and was both upset and perplexed about what to do.
My advice to him was "forget it". I was tempted to add "... McCarthy
has his head wedged".
In summary, while I believe that we currently have an overload problem,
your approach to solving it comes across as mildly capricious. I am
willing to follow through on any policy that I understand, but I don't
understand where you are going.
∂07-DEC-75 1250 LES
∂07-DEC-75 1240 FTP: host CCA
Date: 07 DEC 1975 1540-EST
From: TOM at CCA
Subject: CCA COMPUTERS OFF THE AIR FOR ONE WEEK
To: DISTRIBUTION-LIST:
CCA IS GETTING READY TO INSTALL AN AMPEX TERRABIT MEMORY
(LARGE-SCALE TERTIARY STORE). SITE MODIFICATIONS AND
ENHANCEMENTS WILL TAKE PLACE THE WEEK OF DECEMBER 8 AND
THE CCA HOSTS (DATACOMPUTER AND SEISMIC INPUT PROCESSOR)
WILL BE OFF THE AIR DURING THE SITE WORK.
REGULAR SERVICE WILL BE RESUMED THE WEEK OF DECEMBER 15.
PLEASE DO NOT SEND URGENT NETWORK MESSAGES TO CCA THE
WEEK OF DECEMBER 8. IF YOU WISH TO COMMUNICATE WITH US,
USE THE OLDFASHIONED APPROACH AND DIAL 617-491-3670.
--THOMAS MARILL/CCA
-------
∂07-DEC-75 1230 FTP: host CCA
Date: 07 DEC 1975 1529-EST
From: TOM at CCA
Subject: CCA COMPUTERS OFF THE AIR FOR ONE WEEK
To: DISTRIBUTION-LIST:
CCA IS GETTING READY TO INSTALL AN AMPEX TERRABIT MEMORY
(LARGE-SCALE TERTIARY STORE). SITE MODIFICATIONS AND
ENHANCEMENTS WILL TAKE PLACE THE WEEK OF DECEMBER 8 AND
THE CCA HOSTS (DATACOMPUTER AND SEISMIC INPUT PROCESSOR)
WILL BE OFF THE AIR DURING THE SITE WORK.
REGULAR SERVICE WILL BE RESUMED THE WEEK OF DECEMBER 15.
PLEASE DO NOT SEND URGENT NETWORK MESSAGES TO CCA THE
WEEK OF DECEMBER 8. IF YOU WISH TO COMMUNICATE WITH US,
USE THE OLDFASHIONED APPROACH AND DIAL 617-491-3670.
--THOMAS MARILL/CCA
-------
∂07-DEC-75 1424 FTP: host UTAH
Date: 7 DEC 1975 1525-MST
From: HEARN at UTAH-10
Subject: MISC. REDUCE QUESTIONS
To: JMC at SU-AI
cc: GRISS, HEARN
DEAR PROF. MCCARTHY:
TONY IS ON SABBATICAL IN FRANCE (AT CNRS, MARSEILLE), AND
I AM "HOLDING THE FORT".
1) I WOULD GUESS THAT THE TENEX VERSION OF REDUCE
WOULD BE EXTREMELY EASY TO TRANSFER TO SUPPES. ONW COULD SIMPLY
FTP A VERSION FROM EITHER ISIA, OR A NEWER VERSION FROM UTAH-10
IF YOU PROVIDED ME WITH ACCESS TO THAT MACHINE, I COULD
TRY AND GET IT GOING FOR YOU.
2) THERE IS A MANUAL, BOTH FOR REDUCE AND A SYMBOLIC MODE
PRIMER. BOTH ARE ONLINE AT UTAH, AND COULD BE FTP'D.
FILES ARE:[UTAH-10]<REDUCE>REDUCE.MAN,
[UTAH-10]<REDUCE>RLISP.MAN
IF YOU WISH, I CAN MAIL YOU HARD COPY AS WELL.?
3) THE DECUS VERSION OF REDUCE NEEDS TO BE UPDATED, IT
IS A LITTLE OIUT OF DATE. 1973 SOUNDS ABOUT RIGHT.
4) WE HAVE BEEN WORKING ON IMPROVING THE 360 AND PDP-10
COMPILERS, AND WOULD HAVE TO DO SOME COMPUTATION TO DETERMINE
BIT DENSITY. THE CURRENT 360 LISP COMPILER THAT WE USE IS ESSENTIALLY
THE SAME AS ONE ON THE PDP-10 NOW, WHICH IS A LILLTE BETTER THAN
DIFFIES. WE ANTICIPATE ABOUT 30 PERCENT IMPROVEMENT IN CODE DENSITY
OVER THE OLDER LISP 360 COMPILERE. I WILL RUN SOME
TESTS WHEN I GET A CHANCE, AND RELAY TO YOU.
5) WE DO NOT YET HAVE INFIN ITE PRECISION REALS. OVERALL
WE HAVE IMPROVED SUBSTITUTUON METHODS, AND SOME ALGEBRAIC ALGS.
MAJOR NEW THINGS ARE "PORTABLE LISP COMPILER", WRITTEN IN REDUCE
FOR "ALL" MACHINES; AND A VERSION OF REDUCE THAT USES MODE ANALYSIS
IDEAS (LIKE PASCAL) WITH USER DEFEINED DATA STRUCTURES, GENERIC
FUNCTIONS AND COERCION.
IF I CAN PROVIDE ANY MORE INFO, I WILL
BE HAPPY TO DO SO,
SINCERELY,
MARTIN GRISS.
-------
∂08-DEC-75 0554 TAG
THANK YOU FOR THE MORE CIVIL NOTE ON COMPUTER USAGE. THE BULK OF MY
DOUBLOON CONSUMPTION IS THE DRAWING SYSTEM PLOT PROGRAM, USED IN CONJUNCTION
WITH THE ZONKER MEMORY DOCUMENTATION, WHICH REQUIRED MUCH REDRAWING TO BE MADE
TO MAKE SENSE. THE OLD, INEFFICIENT PLOTTING SYSTEM WHICH DRAGS XGP PLOTS
THROUGH THE "CALCOMP KNOTHOLE" IS ABOUT TO BE REPLACED WITH A SYSTEM WHICH
USES SYSTEM XGP VECTOR SERVICE, WRITTEN BY TOVAR, SO THERE IS HOPE.
∂08-DEC-75 1045 TOB
I would like to arrange a meeting with you,
Dick Liu and me this week. Dick is interested
in talking about supporting a student. When
is convenient?
Tom
∂08-DEC-75 1141 BG
Ivar Tonisson will speak at 4:00 this afternoon [Monday] in the seminar on
axiomatization of knowledge and action.
∂08-DEC-75 1752 FTP: host AI
Date: 8 DEC 1975 2051-EST
From: MOORE at MIT-AI
To: jmc at SU-AI
Thanks for arranging a place for us to stay. I will conntact RWW right away.
I would be happy to talk at your seminar. Jan. 5 is fine.
-------
∂08-DEC-75 2352 JFR
My usage of doubloons will decrease.
Point to ponder: The E upper segment is currently 28K, but only 16K is code
for E; the other 12K is RAID and symbol table. It should be possible to
reduce the upper by this 12K without harming the normal use of E.
∂09-DEC-75 1006 LES Pony Services
To: JMC
CC: HVA
I talked to both SGK and PAW about the possibility of taking over the
Prancing Pony on a profit-making basis. Both declined on the grounds
that the potential return isn't worth the hassle.
I think that what we ought to do is drop all those things that are neither
delivered to us nor can be bought on a once-a-month trip. This would
include rolls, hot dogs, beer, and canned goods. This would cut the
labor to less than half what it is now.
∂10-DEC-75 0202 ME RAID in E's upper segment
To: JMC, REG, JFR
The system version of E no longer has RAID in its upper seg. It's 12K smaller.
∂10-DEC-75 1003 TOB meeting with Richard Liu
When can we meet with Richard Liu?
Tom
∂10-DEC-75 1507 LES Arm Controller Development
To: BES, TAG, TOB, VDS
CC: JMC, JBR, REG, TED
This is my understanding of the conclusions we reached in a
review of arm controller development needs this morning.
We need two copies of a new controller. It should be modular,
with (at least) 3 kinds of card, each separately interfaced to a
Unibus. This will permit the controller to run either under the
PDP-11/45 or an LSI-11 (not both at the same time). The cards that
we have identified are as follows.
MOTOR CONTROL (possibly 2 cards each) - 7 needed for each controller.
Accepts a digital command and mode (velocity or force servo) and
produces a motor control signal, based on speed and torque
sensors (the latter probably analog). It will not provide
position sensing except through digital commands.
POSITION SENSING - 7 needed. Accepts either analog pot signal or
optical encoder output and continually produces digital position.
This position, as seen on the bus, will be no more than
50 microseconds old.
MULTIPLEXED A/D - 1 per controller. Accepts multiplexed analog
signals from strain gauges, etc., and produces digital value
on the bus.
The next step is that Gafford will produce a fairly detailed specification
by January 15. Scheinman will examine the tradeoffs on optical encoders vs.
pots for the various joints and will make recommendations.
After review, the design and construction will be undertaken, with
projected completion by June. Who will do the construction is
currently an open question.
∂11-DEC-75 1629 FTP: host SRI
Date: 11 DEC 1975 1628-PST
From: WALKER at SRI-AI
Subject: PLEASE POST
To: Kay at PARC, TW at SU-AI, JMC at SU-AI, LES at SU-AI,
To: Feigenbaum at SUMEX-AIM
On Algorithmic Natural Language Analysis and Understanding
Vladimir Cherniavsky
Institute for Theoretical Computer Science
Technical University, Berlin
A modification of the Godel theorem extends to natural
language understanding and syntax analysis. There exists a
natural boundary dividing human understanding from any algorithm
expected to understand natural language. We will never have a
model either in the form of an algorithm or in the form of a
theory for syntax analysis and understanding.
Monday, 15 December 1975, 3 pm
Building 30, Conference Room K-2042
Artificial Intelligence Center
Stanford Research Institute
-------
∂12-DEC-75 1317 FTP: host AI
Date: 12 DEC 1975 1617-EST
From: MOORE at MIT-AI
To: JMC at SU-AI
IT SEEMS I NEED A LETTER FROM YOU TO THE MIT COMMITTEE ON GRADUATE
SCHOOL POLICY. THE LETTER SHOULD CONTAIN A STATEMENT BY YOU AGREEING
TO HELP SUPERVISE MY THESIS (MARVIN WILL BE MY OFFICIAL SUPERVISOR
AS FAR AS MIT IS CONCERNED), AND ATTESTING TO THE "EDUCATIONAL
SUPERIORITY" OF MY DOING MY THESIS RESEARCH AT SAIL. MARVIN IS
WRITING A SIMILAR LETTER, SO IF YOU SENT YOURS TO HIM, EVERYTHING
COULD GO TO THE COMMITTEE TOGETHER. THANKS VERY MUCH.
ROBERT MOORE
-------
∂12-DEC-75 2302 REG
∂12-DEC-75 2251 JMC
Say again why Goodfellow is good for us
I decline to state.
∂13-DEC-75 0203 100 : ED RICE, HONEYWELL -- MULTICS via MTRT PARRY
TRIED TO DEMO PARRY AT 0300 MST -- AFTER LOGIN BANNERS, TYPED
'PARRY' --GOT BACK THE LINE
PARRY?
IIF THIS FINE DEMO IS STILL AVAILABLE AT ALL, COULD YOU MAIL
INSTRUCTIONS TO ED RICE, 7900 WESTPARK DRIVE, MCLEAN, VA
-
ED RICE
∂13-DEC-75 1400 LES GFF Account
In response to your question about GFF, I am attaching his reply to my
question a week ago. I do not consider his arguments very compelling.
∂09-DEC-75 0128 GFF via SRI Re: Computer Utilization
In response to your message sent 07-Dec-75 2341
My BIG hours are accounted for by a few runs a TECH2 in the
wee hours of the morning. I guess your accounting program isn't
like ours at SRI and tells WHEN the time was used, and used a
different time-sace for different times of day. However, I
have since FTP'd TECH2 over to my machine and am running it there
now, so I sould no longer (I hope) appear on your 22 most wanted
lists.
I am waitting for your Datameda displays to come in and then
I will assist in bringing up the SRI version of TVEDIT
that runs on them, or at least clue someone at SAIL
in on its workings and how it differences from yours, since ours
is a combo of SAIl and FAIL.
Lastly, McCune got this account set up for me as a kindness
for the accounts we have gooten him on our machine, and we also
hope to get a few of the SAIL people who will
have datamedias at home availble for a possable try out on
our packet radio network we are going to have available in the
bay area (roughly like the ALOHAnet but much more superior).
∂13-DEC-75 1533 AJT my committee
I'm wondering, after talking to Shepard, whether it might not be
better (now that the actual balance of material in my thesis is
known) that somone like Bob Smith (assuming he is willing and can spare
the time) would be a better judge of some of the material than Herb Clark
(who will be away until the summer anyway). What do you think? I'm meeting
with Smith on Wed. to hear his comments...
a.
p.s. Tues morning or late afternoon would be best for meeting with RWW.
∂14-DEC-75 1825 LES Foonly Revival
To: JMC
CC: DWP, REG
Your message contains two false premises, but the conclusion may be correct.
It also ends in midsentence with "... you will have to tell".
You say that "Since this Lab is making no use of the Foonly effort, it is
in the public domain ...". That is not the way something gets in the
public domain. I believe that Foonly is NOT in the public domain, but
that question can be checked.
You say that "We must receive comparable benefits for comparable contributions."
Perhaps you forgot that the direct benefit that we received from DEC was
nothing. There was a goodwill effect that eventually bore fruit (or at least
may, if we can get the DEC bureaucracy unwedged).
I think that the right approach to the question of what to ask for
is to examine "What is it worth?". That question has several facits,
including what it cost us, what would it cost them to duplicate,
and what might others be willing to pay.
∂14-DEC-75 2053 TJW thesis
John,
I was planning to write one more paper on the logic verification problem
for this quarter but have changed my plans somewhat. Yesterday (Saturday)
I was hit by a car while comming here on my motorcycle. My left foot is
broken and I have decided to go home for a couple of weeks to convalesce.
I should be back by the first week in January and will have to work out
arrangements for getting too and from the lab then (Hopefully, I will be
able to drive my manual shift car with a cast on by then). Have a nice
Christmas.
Todd
∂15-DEC-75 2057 GFF via SRI SCCS
I have an artical from the L.A. Times, that talks about
SCCS (southern California Computer Society), a group of random people
that are doing the HOTER.BAY thing you tried to get off the ground while
back. If you are interested, I'll Xerox and send/bring it to
SAIL for you to read over. its a nice sized artical.
[Geoff Goodfellow]
SRI-AI.
∂15-DEC-75 2356 TOB
I have a proposal for a collaboration with
Liu, based on a discussion with him. The
outline of the discussion follows. I propose
that we include him in the proposal for 1
student and 10% salary, and that if we have
the money, that we support a student until
the new contract, (continuing support
contingent on the contract providing the
extra money) providing that we can agree
on a research topic for that student.
Liu
OUR SUGGESTIONS FOR AREAS OF COLLABORATION
analysis of assemblies
conceptual analysis
analysis in AL
design for assembly
redesign for automated assembly
simulation of assembly
library for assembly
functional
AL
collision avoidance
CAD-representation
modelling assembly
simulation-feasibility of assembly
LIU'S SUGGESTIONS FOR AREAS FOR COLLABORATION
design classification
automatic parts classification
2d-3d and 3d-2d
engineering drawings--3d models
simulation language for manufacturing
For NSF proposal
10% and 1 student
Hutchison may be here next year on sabbatical
U. of Wis Milwaukee, 1/3 on Liu's contract
STUDENTS:
Ron Sandretti (ME-design)
research interest: classification, robotics
very inventive
Tim Tight (ME-design)
research interest: classification, robotics
very inventive
David Tai (IE with EE masters)
research interest: CAM
extensive programming experience
---------------
research interest: NC machining
worked with minicomputers
∂16-DEC-75 1357 LES Bellnet
I left a revised draft of the Bellnet proposal on your terminal last night.
Before we send that off, we should probably learn more about some related
systems: RSEXEC in TENEX and DECNET. I will try to round up the documentation.
∂16-DEC-75 1427 LES About the Bellnet name
While "Bellnet" has a nice ring, it has a definite Ma Bell bias and
might even produce some proprietary grumbles from AT&T.
An unimaginative alternative would be "PhoNet". We could then refer
to the whole thing as "The Phonetic System". Gag.
∂16-DEC-75 1548 REF UCI LISP manual
I have the source code for the abbreviated UCI LISP manual.
It's in XOFF, which looks like it's similiar to PUB. It is the
file LISP.XOF[PAP,REF]. Would you like
1} A copy of this file.
2} A fix to this file to run in pub, and a copy of that result.
3} Just to request a copy of the technical report from CMU.
bob
∂17-DEC-75 1018 LES Dialnet
OK, I stuck your paragraph in the "Protocols" section and sent a
copy to Cerf, soliciting remarks and/or participation interest.
I'll put in the DECUS references later.
∂18-DEC-75 0242 RWW P&P
You can use quant to do that for the time being
∂18-DEC-75 0243 RWW
do you have an example of only wanting to do one? Id like to see it.
∂18-DEC-75 1059 CRF via ISI dialnet
I just had a chat with Paul Heller of EDUCOM who says they
are interested in ways of letting their members communicate
in the way John has been suggesting. As you may know,
Educom tried to get on the ARPANET but failed, and Stanford
is amember of the board of governors of the EDUCOM organization.
Gene Franklin is the contact. It isn't clear just how the Dialnet
idea and the EDUCOM activities can be coordinated but
it seems worth exploring a little.
Vint
∂19-DEC-75 0335 PMF
I borrowed Dahl,Dijkstra,Hoare Structured Programming.
∂19-DEC-75 0809 FTP:ALLEN NEWELL(A310AN02)@CMUA
From: ALLEN NEWELL(A310AN02)@CMUA
Date: 19 Dec 1975 1106 EST
Subject: Talk with Atkinson on national center
To: DNLC.DIS
- - - -
I have just talked with Dick Atkinson at NSF about the DNL.
(1) He was gloomy about funding of basic research generally.
(2) He appeared really worried that ARPA would use the existence
of such a proposal as a way of decreasing substantially its
committment to basic AI, and if, then, NSF could do little, AI
would be caught off base, half way between 2nd and 3rd.
(3) He felt that the proposal should wait until at least Feb 76
before surfacing. This was based in part the lack of attention
that it would get in NSF prior to that time, in part on feelings
in (2) above.
(4) He reiterated his concern about getting an appropriate
funding base for AI, and some of the flavor of the conversation
was one in the hand is worth two in the bush.
(5) I did not go into any of the details of the proposal -- we
remained at the next level up.
(6) Nothing surfaced about a rumor that Nils had mentioned --
that NSF was going to disband their centers, so that our
proposal would be moot. However, we did not get to that item
directly, and the conversation was somewhat rushed.
(7) I agreed to talk with him again in late January.
*******
I have received no feedback from any of you on the partial draft
that I put out on 16 Nov (with the exception of a few promises
of feedback to come). Correction: Nils mentioned that he
thought it was awfully complex; Lick mentioned that he was glad
it was proceeding. But that's all. Do I infer from this that
you all feel ho-hum about the enterprise and that we should just
let it drift. I take the message from Dick Atkinson as one of a
dozen slow-down - hurry-up messages we will receive as this
thing proceeds. If we are smart, we will continue to put the
proposal together and to get informal reactions to it and the
ideas it contains. For myself, I will continue to try to write
more prose. I would appreciate some feedback either on the
content or on the general issues.
-------
∂19-DEC-75 1327 FTP:CERF at USC-ISI
Date: 19 DEC 1975 1326-PST
From: CERF at USC-ISI
Subject: Re: dialnet and educom
To: JMC at SU-AI
In response to your message sent 19 DEC 1975 0738-PST
john:
EDUCOM doesn't really have anetwork -- just paper organization, so your
criterion about being able to comunicate even if you are not
"on a network" would not be violated in these discussions. I prefer
mail to be sent to ISI since it is my primary mailbox. Thanks,
and see you next Wednesday.
Vint
-------
∂19-DEC-75 1658 AJT Thanks
Thanks very much for your hospitality. I've enjoyed my time here very much,
and look forward to coming back to visit sometime soon.
regards, Arthur
∂19-DEC-75 1616 100 : I-SPY@NSA via RTGT PARRY
WHERE IS PARRY HIDING OUT? THE PARRY COMMAND NO LONGER
WORKS, AND THE
1,BLF DIRECTORY IS PROTECTED. MUST HAVE PARRY IMMEDIATELY
FOR REASONS OF NATIONAL SECURITY.
∂20-DEC-75 1053 FTP:ALLEN NEWELL(A310AN02)@CMUA
From: ALLEN NEWELL(A310AN02)@CMUA
Date: 20 Dec 1975 1349 EST
Subject: John: Did I already send this? If so, pardon the repeat. A.N.
To: JMC@SU-AI
- - - -
From: ALLEN NEWELL(A310AN02)@CMUA
Date: 19 Dec 1975 2026 EST
Subject: response to jmc 19 dec 75
To: JMC@SU-AI
- - - -
John: I too had/have some misgivings about my proposal. I think the
problem is how to shape it in a way that makes it some thing in which
we can all live with and find worth the additional support it gives
us. I let the proposal write iteself, in terms of being dicatate by
a sort of inner logic. But one of the reasons I have not circulated
it at all is that it needs to pass review by those of us who control
(to some degree, at least) the laboratories which would play a major
role in such a proposal.
I look forward to some comments. A.N.
-------
-------
∂22-DEC-75 0215 100 : PMF JOES CHRISTMAS THING
JOE WANTS YOU AND SARAH TO COME. CALL 28-2683. (JOE ZINGHEIM)
∂21-DEC-75 1548 100 : Charles H. Underwood via MTRT Parry
We would like to see the PARRY program
brought back to guest use. The program is not
transportable to other systems, so if we are to use
it , it must be run on yours. The program is a good
demo of AI for the layman and
as such is fairly rare.
Chuck (CHU@MIT-AI
∂23-DEC-75 1328 LES
∂23-DEC-75 1322 FTP:RUSSELL at USC-ISI
Date: 23 DEC 1975 1321-PST
From: RUSSELL at USC-ISI
Subject: Stanford-AI ARPA Order
To: les at SU-AI
cc: Carlson, Carlstrom, Blue, Russell
Les,
The Director signed the Stanford AI ARPA order today.
However, he has directed additional effort on focusing the
program toward relevant problems and laying out useful goals.
Dave Carlstrom and Bill Carlson will be in touch with a detailed
list of what is required. It is important to note that funding
past 30 June 76 will require progress on the issues raised by the Director.
Other than that, Merry Christmas
Dave
-------
∂25-DEC-75 2016 REF primer
As promised, the most rgcent version of the primer is
in your mailbox. I've added additional sections to (I hope)
contend with your critism. Any other suggestions would be
appreciated.
thanks
bob
∂27-DEC-75 0159 LES
.<< Here is a test file for italicizing single words.
. It apparently works most of the time: in the example below,
. the words "brown" and "picked" are italicized correctly, but in the
. latter case, the text response doesn't find a space until after the
. following "a", so it gets italicized too (i.e. it works except for the
. last word in a line, where you get two words).
.>>
.device xgp;
.require "basker.pub[sub,sys]" source;
.turn on "%";
.at "⊗"word" " ⊂ ("%2word%* ") ⊃;
How now ⊗brown cow.
Peter Piper ⊗picked
a peck of pickled peppers.
∂28-DEC-75 0857 FTP:FEIGENBAUM at USC-ISI
Date: 28 DEC 1975 0857-PST
From: FEIGENBAUM at USC-ISI
Subject: ARPA RENEWAL
To: EARNEST at SU-AI
cc: MCCARTHY at SU-AI, GREEN at SU-AI, WINOGRAD at SU-AI
LES, THANKS FOR THE NEWS ABOUT THE ARPA RENEWAL...AND CONGRATULATIONS
TO YOU,JOHN, AND EVERYBODY ON THE SUCCESSFUL RENEWAL!!!!THIS WAS THE
BEST NEWS IN A LONG TIME...FITTING SEASONAL GIFT. IF YOU LOOK ON THE
NEWS AS ONE WOULD LOOK AT TEA LEAVES, WHAT DOES IT SAY?
INCIDENTALLY, THE NEWS BECOMES EVERY SHARPER THAT HEILMEIER WILL BE GOING
TO RCA AT END OF JUNE.
HAPPY NEW YEAR.
ED
P.S. I HAVE A PREPUBLICATION COPY OF WEIZENBAUM'S BOOK. MIT-AI AND
CMU-AI ARE ROASTED BUT NOT SU-AI.
-------
∂31-DEC-75 1907 TAG
AS I BELIEVE I EXPLAINED SOME TIME AGO, THE THIRD IMLAC HAS A
REAL LIVE HARDWARE INTERMITTENT. REPAIRING IT WILL REQUIRE MY BECOMING MORE
FAMILIAR WITH THE PDS1 PROCESSOR, AND WRITING SOME DIAGNOSTICS, AS THE IMLAC
DIAGNOSTICS THAT WERE SUPPLIED WITH THE MACHINES WILL NOT RUN DUE TO QUAM'S
HARDWARE MODIFICATIONS TO THE MACHINE. I AM ONLY SLIGHTLY MORE ENTHUSIASTIC
ABOUT REPAIRING THE THIRD IMLAC THAN I AM ABOUT POKING MYSELF IN THE EYE WITH
A SHARP STICK. THEY ARE A PAIN TO WORK ON, AND I HAVE TWO OTHER MAJOR PROJECTS
TO WORK ON, EACH WITH A HELL OF A LOT MORE MONEY AND TIME TIED UP IN THEM
THAN THE IMLACS ARE WORTH.
∂31-DEC-75 1911 TAG
BESIDES, THE TERMINAL SERVICE PROVIDED TO POSSESSORS OF DATAMEDIAS
WILL GREATLY SURPASS IMLAC CAPABILITIES, BEFORE SIX MONTHS HAVE GONE BY.
∂01-JAN-76 2042 TAG via AMET
I WILL MAKE THE APPROPRIATE ADJUSTMENTS TO YOUR IMLAC
BEFORE FRIDAY EVENING, JAN. 2. WITH REGARD TO THE IMLACS, AND WRITING THEM OFF,
THE SECOND NOTE I SENT YOU, WHICH WAS DELAYED BECAUSE YOU WERE READING YOUR
MAL WHEN I SENT IT, MAKES MY POINT. REGARDLESS OF WHAT WE MAY HAVE SPENT
FOR THEIR PURCHASE, THEY HAVE BEEN A MAINTENANCE HEADACHE, AND THEY WON'T
BE AS DESIRABLE A TERMINAL AS THE DATAMEDIAS, AFTER THE LATTER AND
THEIR SOFTWARE SUPPORT ARE INSTALLED. IN FACT, IF WE WERE TO KEEP USING THE
IMLACS, WE SHOULD FIND SOMEONE TO PROGRAM THEM SO THEY CAN AT LEAST
EMULATE THE DATAMEDIAS. THE IMLACS SUFFER FROM A DISEASE COMMON TO
SEVERAL OTHER PIECES OF HARDWARE HANGING AROUND THE NECKS OF THE MANTENANCE
STAFF, I.E., THEY HAVE THE SMALL-
SERIAL NUMBER, FIRST OF THEIR KIND DISEASE.