Date: Thu, 8 Aug 1996 01:29:22 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "keith reid-green" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Scientific American: calculating census history.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <847614029961.LTK.013@cpsr.org>
X-UIDL: 839517089.017


Sender: "keith reid-green" 
Subject: re: Scientific American

Peter da Silva wrote "Add Scientific American to the list of sources for
our community memory."

I remember the article on fluid computers in SciAm.  Also, I wrote a piece
for Scientific American entitled "The History of Census Tabulation" in the
Feb, 1989 edition, and still have a few offprints of the article.  If
anybody wants a copy, send your address directly to me at
KReid-Green@ets.org and I will mail you one--offer subject to availability.

Keith Reid-Green
Educational Testing Service
KReid-Green@ets.org
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 01:37:56 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Kip Crosby, CHAC" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Orignal Community Memory & PCC
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1352192119474.LTK.013@cpsr.org>
X-UIDL: 839517089.028


Sender: "Kip Crosby, CHAC" 
Subject: Re: CPSR-HISTORY digest 49

>Sender: "Laurence I. Press" 
>Subject: Community Memory
>I visited the CM project in the early 70s.  Lee Felsenstein and others
>had an SDS timesharing system in a converted industrial building in
>San Francisco....They had ttys
>in a few places in the Bay Area.
>They had a lot of trouble keeping the TSS up as I recall.

See ANALYTICAL ENGINE 3.1 (Nov. '95) for more on this.  Our best information
is that CM's SDS 940 was scrapped in 1988, ah, sigh.  As the custodians of
the last (?) remaining SDS 930, we would dearly love to acquire the rackful
that made a 930 into a 940; but we have scant hope, it was a scarce piece at
best.

That 940 was donated to Resource One by Transamerica Leasing after it
finished a stint at SRI, where it had been the control computer for Shakey
the Robot.  Lee says that Shakey was "shaky" because the 940's DMA channel
was wired to the wrong voltage, and when Resource One got it they diagnosed
and rewired it.

>I think a lot of the inspiration came from Bob Albrecht with his
>People's Computer Center in Menlo Park.  PCC offered walk-in computer
>access and a news paper.  (In fact Jim Warren, Dennis Allison and Bob,
>started Dr. Dobbs at PCC).

PCC was also involved with (was the parent of?) a company called Dymax which
published, among other things, Bob's _My Computer Likes Me When I Speak In
Basic._ ....on page 61 of which it says "Please recycle this book."  I'm
just as glad somebody didn't.

>Bob's example inspired me to put TTYs in
>some schools and a public library in LA -- mostly games, BASIC
>classes, car pool matching (remember the gas crisis? -- I wrote a
>program which searched concentrically out from Thomas Map book
>coordinates to find people with nearby destinations and homes), etc.

Mind telling us some more about that?

>Sender: John Oliver 
>Subject: 2-D text editing and interactive games
>Sometime around 1967 or 1968 the Astronomy department at UCLA purchased
>some abolutely wonderful "TV Typewriters"....When the "Interrupt" key was
pressed, >the entire contents of the core was shipped downstairs to a 360
(either mod 90
>or mod 91 ... I forget), and swapped with whatever the channel adaptor had
>waiting.

If that was the same 360 that ended up as the pet of the UCLA Computer Club,
it was a Model 91.
__________________________________________
Kip Crosby                 engine@chac.org
      http://www.chac.org/index.html
Computer History Association of California
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 01:43:18 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Nelson Winkless 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> People's Computer "Something"
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <2134250437797.LTK.013@cpsr.org>
X-UIDL: 839517089.023


Sender: Nelson Winkless 
Subject: CM> People's Computer Something

We've seen references here to People's Computer Club, People's Computer
Company, and in one of Larry Press's recent postings, People's Computer
Center...all, I think, talking about the same organization. Did the
organization really have more than one name going?

It was known to me as People's Computer Company, and before first visiting their
upstairs office on El Camino Real in Menlo Park (must have been in 1976), I
always pictured the staff in quilted uniforms, marching with fists upraised.
Actually, the place was full of pleasant people, having fun, apart from the
usual economic stress.

Their tabloid-format publication was a vigorous journal featuring lots of
pictures of dragons. Indeed, Bob Albrecht, whose formal position was never
quite clear, but who seemed to be the leader of PCC, was widely known as
Dragon Bob Albrecht. Bob published -- with a co-author?--a little book
called "My Computer Likes Me When I Speak in BASIC," which sold at least
100,000 copies, I believe, and would have been classed formally as a
best-seller in the regular book trade. I published a couple of articles in
PCC under names I have forgotten, one of which was illustrated by my
daughter, who probably still has clippings.

My impression was that PCC was involved early in efforts to get kids off the
streets and onto computers, making a grassroots effort to promote general
computer awareness and competence in our brave new era.

Does somebody have a less muddled recollection of what PCC was about?

--Nels
-----------------------------------------------------------------------------
Nelson Winkless                   Email: correspo@swcp.com
ABQ Communications Corporation    Voice: 505-897-0822
P.O. Box 1432                     Fax:   505-898-6525
Corrales NM 87048 USA             Website: http://www.swcp.com/correspo
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 01:48:00 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Michael S. Hart" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Early interactive games.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <94532378579.LTK.013@cpsr.org>
X-UIDL: 839517089.021


Sender: "Michael S. Hart" 
Subject: Re: CM> Early interactive games.


The Tic-Tac-Toe game most people remember around here
was at the Museum of Science and Industry in Chicago,
and my recollection is also that it was part of "Bell
Labs" exhibit, which was always quite nice.

Of course, the friendly operators have been replaced,
by robots, etc. and even when you do get a human now,
they can't "route you through Milwaukee" when all the
connections to O'hare are jammed from downtown.

Michael S. Hart
The Original Cybrarian
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 01:52:54 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: ac959@freenet.carleton.ca (Peter Martin)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "toy" computers: Heath-Kit 1963.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1462221388373.LTK.013@cpsr.org>
X-UIDL: 839517089.030


Sender: ac959@freenet.carleton.ca (Peter Martin)
Subject: Re: CM> "toy" computers

>
>
>Sender: lauren@vortex.com (Lauren Weinstein)
>Subject: "toy" computers
>
>Greetings.  Ah yes, the "toy" computers.  I also had (indeed, still
>have) the book explaining how to build your own digital computer
>from wooden dowels, paper clips, and other assorted "found around the
>house" items.  I know exactly where that book sits; I'll try dig it up.

Heath Kit sold an almost-toy computer kit. I was given one as an Xmas
present in 1963 or 64.

My recollections are vague, but I recall it was a slab of pegboard,
several circular "dials", a few flashlight bulbs, and the requisite
battery holder and wires.

I put it all together and it worked as (only as, I think) a simple
arithmetic calculator. Didn't even play tic-tac-toe!

Perhaps others on this list remember the thing?

(BTW, poor Heath Kit. Great company for hobbyists that fell on evil days
when chips replaced tubes. I wonder if they're still around.)

--
Peter Martin            *****           Garden Books of Sandy Hill
#1 -131 Osgoode St      *****           P.O. Box 20559
Ottawa ON K1N 6S5       *****           Ottawa ON K1N 1A3
(613) 237-4180          *****           (613) 565-2595
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 01:56:56 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: John R Levine 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> 1D vs. 2D editing
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <2053880373614.LTK.013@cpsr.org>
X-UIDL: 839517089.037


Sender: John R Levine 
Subject: 1D vs. 2D editing

> ps -- a bit pedantic perhaps, but I am not sure "2-d editing" is a
> good term, because you are still editing a 1-d text string which is
> displayed in 2-d with word wrap.  You cannot select an arbitrary
> rectangle with a word processor.

Funny you should mention that.

I'm aware of two largely independent lines of development of of screen
text editors, one with a 2D "deck of cards" modem and the other with a 1D
"papertape model".  The 2D model started with Ned Irons work at IDA in
Princeton, and moved with him to Rand (Rand Editor), Yale, and Interactive
Systems in Santa Monica.  All of these editors really treated the text as
a quarter plane, and supported rectangular cut and paste.  This model
made it easy to speed up remote editing by putting a lot of the editing
functions in a terminal.  At Yale in the late 1970s we had an editor
running on our 11/45 Unix with the front end an 11/05 with a dozen home
brew bitmap screens and keyboards.  Worked great, modulo hardware
glitches in screen RAM and keyboards.  At Interactive in about 1980 we
had the INtext, a Perkin-Elmer terminal with custom ROM and keycaps to
support INed, a descendant of the Rand editor.  The INtext supported
full screen interactive editing over line at a time Telenet connections a
big win financially compared to character at a time connections.

The 1D model started with scope Teco at MIT and evolved fairly directly
into emacs and indirectly into vi.  For reasons having little to do with
the merits of the underlying models, emacs and vi became very popular
largely because they were easy to customize to the flock of random
terminals found in most university computer science environments.

Word processors have for the most part developed separately, with their
own models of text.  Word Perfect is paper tape with markup codes, MS
Word is a tree of document, paragraph, character, with styles and formats
applied at each level.

Regards,
John Levine, johnl@iecc.com, Trumansburg NY
Primary Perpetrator of "The Internet for Dummies"
and Information Superhighwayman wanna-be
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:01:12 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Weston Triemstra 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Early interactive games.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <920072952871.LTK.013@cpsr.org>
X-UIDL: 839517089.034


Sender: Weston Triemstra 
Subject: CM> Early interactive games.

Les Earnest writes:
    I vaguely recall that someone at Bell Labs built a relay computer that
    played tic-tac-toe sometime in the late 1940s or early 1950s.  The
    TX-0 computer at MIT also had a tic-tac-toe game when I started
    playing with it in 1959.  It displayed the board on its CRT and you
    selected moves by pointing with a lite pen.

       Do recall deciding that if you went first you could not lose (unless
       you made a mistake) and if you went second, you could not win.

    I don't understand this.  If both sides play correctly it always ends
    in a draw.

It's true that if both sides play correctly it always ends in a draw
with tic-tac-toe, however if you add another dimension ... I wrote a 3-D
tic-tac-toe game in BASIC on a commodore pet 40 in 1987 where if the
computer went first it always won, it was for my highschool computer
science class. I can mail the moves to anyone interested, the manoevre
requires grabbing the centre cube first.

........................................................................
                                            n e t SIGN communications
    weston triemstra                              110 west hastings
    weston@netsign.com                            vancouver, bc v6b 1g8
                         "the future was now"     canada
........................................................................
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:05:06 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "David E. Anderson" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Community Memory, Resource 1.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1429428056105.LTK.013@cpsr.org>
X-UIDL: 839517089.029


Sender: "David E. Anderson" 
Subject: re: Community Memory

The facility at the warehouse was Resource 1, inside of Project 1, the
warehouse.  Loft living was unusual then, cyber-hippies even more so.  Project
Artaud was a more arts-oriented place which pre-dated Project 1 and Project 2
followed both.

I remember studying SNOBOL at Resource 1.  There were some junior high or high
school kids in the class with me - Once I overheard them saying they wouldn't
tell adults what they were doing (programming) because it would scare them...
They're probably millionaires now ;-)

One of the teletype terminals of Community Memory was also installed in a room
at the collective house I lived in, Dwight House, home of "The Village" in
Berkeley. The system was newsgroup-like, as I too remember.  The idea was to
create a shared high-tech infrastructure with some of the flavor of the food
coops that were very active at the time, community oriented, volunteer-based
and decentralized.

Dave
--
David E. Anderson/New Media Support/Oracle Corporation 415-506-6307 fax: 7809
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:09:31 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Mike O'Brien" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> early display-orieneted editors, "expensive typewriter."
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <2012064695364.LTK.013@cpsr.org>
X-UIDL: 839517089.032


Sender: "Mike O'Brien" 
Subject: Re: CM> early display-orieneted editors, "expensive typewriter."

Larry Press says:

> ps -- a bit pedantic perhaps, but I am not sure "2-d editing" is a
> good term, because you are still editing a 1-d text string which is
> displayed in 2-d with word wrap.  You cannot select an arbitrary
> rectangle with a word processor.  Perhaps terms like "display" vs
> "line-oriented" are better.

        Well, then there's the Rand editor, first programmed up at Rand
by Walt Bilofsky, with Peter Weiner's connivance, which was based on
Ned Irons' ideas from Yale.  Still in use today, the Rand editor is a
"white-space" editor which does indeed let you select an arbitrary
rectangle, and do pretty much whatever you want with it, including
drawing an ASCII box around it.  I'm using the Rand editor to write
this message.  Rand historically hot-wired the keyboards of all its
terminals to make editor commands a single keypress of a labeled key;
now we editor users make do with the "standard keyboard" of control
characters.  However, we found the learning curve of computer-naive
users to be very easy, given the hot-wired keyboards and the editor's
support for moving the cursor wherever you want, regardless of the
location of the end of the line.  It also makes typing in tabular
data a breeze.  You can open up a space in a single column.

Mike O'Brien
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:13:41 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Suzanne M. Johnson" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> TVEDIT
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1309391616619.LTK.013@cpsr.org>
X-UIDL: 839517089.035


Sender: "Suzanne M. Johnson" 
Subject: TVEDIT

Regarding the editor known as TVEDIT.  I still consider it a model for other
"word processors" to live up to...and I've not yet found one that does.  I
certainly have never seen another editor or word processor thanked by an
author and acknowledged as contributing greatly to making a book possible.

In the book "Godel, Escher, Bach:  An Eternal Golden Braid", by Douglas R.
Hofstadter, Vintage Books Edition, 1980, the following is contained in the
"Words of Thanks", (pg xx):
        "This brings me to Pentii Kanerva, the author of the text-editing
program to which this book owes its existence.  I have said to many people
that it would have taken me twice as long to write my book if I hadn't been
able to use "TV-Edit", that graceful program which is so simple in spirit
that only Pentii could have written it.  It is also thanks to Pentii that I
was able to do something which very few authors have ever done: typeset my
own book.  He has been a major force in the development of computer
typesetting at IMSSS.  Equally important to me, however, is Pentii's rare
quality:  his sense of style.  If my book looks good, to Pentii is due much
of the credit."

        I do have TVEDIT documentation, and other related
materials..including (I think in storage) a videotape made  of the Context
system.  Context was a timeshared computer at Stanford (early '80s) running
Tenex or Tops-20 that was designed to support text creation across the
breadth of the Stanford community. TVEDIT was a very optimized editor, using
carefully selected "dumb" terminals.  The number of users of this type that
one computer could support made it an extremely cost effective way to
provide computer cycles for text preparation...and, I'll bet it still would
be.

                Suzanne
__________________________________________________
Suzanne M. Johnson           Sunnyvale, California
              johnson@rahul.net
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:21:50 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "A. Padgett Peterson P.E. Information Security" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <822855273109.LTK.013@cpsr.org>
X-UIDL: 839517089.036


Sender: "A. Padgett Peterson P.E. Information Security"

Subject: Keypunchers and cards

Dan rote:
>The other predominate width in the late 60's was the line printer width of 132
>positions.  Print position 1 (on IBM's printers anyway) was reserved for paper
>control.  A blank moved to the next print line, a 0 doubled spaced, and a 1
>ejected the next page.

Think that was a line printer standard or at least all the DEC printers I
saw around them followed it also.

Recall programming many keypunches (forget the designation but this was the
unit that had a small drum in the middle that was programmable. You took
a punch card and ran it through typing various caharacters on the line
(surely someone here remembers the codes) and put it on the drum so that
as a new card fed in from the hopper on the right, it would automatically
register on col 2. When programming in FORTRAN the statement number went
in 2-6 and the statement started in col 7.

For some reason the statement had to stop on 78 and not 80 but I do not
recall why (this was a few years ago). Do know that as each field
was complete you just hit a key and it would automatically index to the
next programmed field.

Also recall that it could be set to automatically print a character or series
on the card but do not think it could autoincrement.

As memory serves, once managed an all-time best of 600 cards in an hour.
Since that was one every 10 seconds it must have had a *lot* of very short
statements.

                                        Warmly,
                                                Padgett

ps for years I carried a "Collins Computer Diary", a small daily calendar
   that was about 2 1/2"x5". In the front there was a history/glossary
   of computer terms and tables of everything from Baudot to EBCDIC. Very
   handy and always used to find a source in the UK to send me one.
   Unfortunately in 1994 it was "decontented" (only ASCII table), in 1995
   it became quite garish, and for 1996 it was discontinued for poor sales
   (no wonder). Just a bit of trivia.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:25:32 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: LESPEA@muze.com (Leslie Pearson)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Where Wizards Stay Up Late: book info.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1841519093331.LTK.013@cpsr.org>
X-UIDL: 839517089.025


Sender: LESPEA@muze.com (Leslie Pearson)
Subject: Where Wizards Stay Up Late


Where Wizards Stay Up Late: The Story Behind The Creation of The Internet
Katie Hafner and Matthew Lyons
ISBN: 0-684-81201-0
Trade Cloth
Simon and Schuster Trade 8/96 $24.00

Leslie Pearson
(lespea@muze.com)
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Thu, 8 Aug 1996 02:29:59 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: peter@baileynm.com (Peter da Silva)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Community Memory & PCC.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1463759121598.LTK.013@cpsr.org>
X-UIDL: 839517089.027


Sender: peter@baileynm.com (Peter da Silva)
Subject: Re: CM> Community Memory.

> Any PCC recollections out there?

I never ran into PCC, but a guy at UCB (I forget his name now) wrote a copy
of the PCC's "public caves" system, and I was inspired to use that as the
model for a BBS I ran for several years. It was sort of like today's "MUDs",
albeit one person at a time (one phone line). You could build rooms (dig
caves) in any of 10 directions (an 8-arm compass, up, and down), and each room
had a description and a chain of messages that were "written on the walls".
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:07:36 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Wesley J. Miller ((8-444) 919-254-9774)" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine Questions on Machine History.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <88110959855.LTK.013@cpsr.org>
X-UIDL: 839600992.026


Sender: "Wesley J. Miller ((8-444) 919-254-9774)" 
Subject: Machine History

I'm a short-timer in the world of computing; I started with FORTRAN-66 on punch
cards about 1978 using WATFIV and FORTRAN-H.

As I read the recollections of folks who came into computing before and about
the same time as I, I am remembering all those stories I heard and questions I
only ever got partially answered.  So, here is a list of questions that I hope
will start a few memories flowing and get some history nailed down for all of
us short timers.

1. The first "personal computer" I used was a Datapoint 1134.  I know the
last digit represented the number of floppy drives installed.  I also
remember that the machine had 16K of memory on four 4K memory cards that
plugged into the processor cards on which I was told there was an 8080 -- not
the chip but a hardwired-from-discretes equivalent which Datapoint used
because Intel couldn't get the 8080 out in time.  Questions:  Was it in fact
an "8080"?  Was it Datapoint that drove Intel to develop the 8080?

2.  It's easy to see that EBCDIC came from punch card encoding.  Where did
ASCII come from?  Why did someone choose to rewrite the coding standard when
both EBCDIC and ASCII seem perfectly capable of doing the same thing?

3.  Just why is it that some folks say IBM never got time-sharing right?
They've sold a LOT of TSO, after all. (No, this is NOT a debate on
market-share = got-it-right)

4.  I personally think TSO is about the most user-hostile environment/op-sys
on earth.  I first used IBM's CMS version 2 as an op-sys about 1982.  I still
use it (even as I type this note).  Yes it's old and character only but it
is SO much easier to use than TSO.  What's the history of CMS?  Why did it
never replace TSO?  Unix?

5.  Yes, I use UNIX.  Yes, I like it--yes, I hate it.  Where did the
let's-make-it-so-cryptic-no-one-can-intuitively-figure-it-out mentality
and naming conventions come from?  TYPE ... | FiND .... sure seems
a lot more obvious to me than CAT ... | GREP ...

6.  IBM numbers bits from left to right; the MSB is bit 0.  Who decided to
do this dumb thing?  Same is true of byte and character string numbering.

7.  Who decided that token-ring and ethernet addresses are sent in opposite
orders?  LSB-first vs. MSB-first?

8.  Why are people trying to shoehorn FORTRAN into an array handling language?
Why aren't they using APL/2 instead?

9.  And finally, for now, who designed and why do we use the file system used
by DOS, MAC, Unix and I don't know what else?  Specifically, these file systems
treat all text files as one long string with records indicated by wasting
the two characters 0D and 0A.  Mainframes and even my dear departed Datapoint
1134 used a full blown file system where records are stored as records.  Each
record either points to the next/last record or has a length indicator.  No
characters are wasted.  Same question really applies to C and DOS ASCII-Z
strings.  Who decided 00h wasn't a datum?  Ever try to pick apart non-text
data and not be able to use those nice routines that think the 00's aren't
data?

Wesley J. Miller

[Moderator's Note: In answering these questions, I urge you to remember to
keep your posts grounded in people, rather than de-contextualized
speculation -- i.e. tell us stories you know of first-hand, or head from
someone who was involved in these decisions.]
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:13:34 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Nelson Winkless 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Adam Osborne, Hot Pigs, Cool Ice.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1978130094413.LTK.013@cpsr.org>
X-UIDL: 839600992.027


Sender: Nelson Winkless 
Subject: CM> Adam Osborne

Just a quick image--

At the 1977 National Computer Conference in Dallas, National
Semiconductor threw a splendid reception on the top floor of
a downtown bank building. The affair featured huge ice sculptures
and two (count them, two) roast suckling pigs on delicacy-laden
tables at opposite ends on the hall.

As we entered the venue, Dallas belles tied brightly colored
western bandannas around our necks to give the proper cowboy
flavor to our appearance. Since most of us were wearing suits,
the effect was a trifle odd.

Adam Osborne, a tall, spare fellow, was wearing a dark,
pinstriped suit, and he was a distinctive figure, visible from
any part of the room in his stripes and bright neckerchief.

Actually, Adam wasn't alone in pinstripes. His fellow Englishman
John Peers ("the Monty Python of the Computer Industry") of
Logical Machine Company was also in pinstripes, looking just as
peculiar, if not quite as tall as Adam.

Apart from some pleasantries delivered by Charlie Sporck of
National the evening featured the spectacular crash of one of the
ice sculptures, its base presumably melting faster than expected,
owing to exposure to hot pig. One of Byte Shop founder Paul Terrell's
numerous brothers also sampled the wares at Mr. Sporck's bar
so diligently, that he lost his way, and wandered harmlessly out of
the building into the late-night Dallas streets. He couldn't explain
who he was, or what he was doing, but a cop, noticing the bright rag
around his neck, followed a trail of similarly marked people, and
brought the dazed wanderer back to the party.

That was the first time I saw Adam. The image sticks firmly in the
memory.

--Nels Winkless

-----------------------------------------------------------------------------
Nelson Winkless                   Email: correspo@swcp.com
ABQ Communications Corporation    Voice: 505-897-0822
P.O. Box 1432                     Fax:   505-898-6525
Corrales NM 87048 USA             Website: http://www.swcp.com/correspo
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:18:34 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Josh Hodas 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Early interactive games
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1080365102713.LTK.013@cpsr.org>
X-UIDL: 839600992.046


Sender: Josh Hodas 
Subject: Re: CM> Early interactive games

> > Sender: Les Earnest 
> > Subject: CM> Early interactive games.
> >
> > Padgett Peterson writes:
> >    In 1956 or possibly early 1957 I visited the Franklin Institute in
> >    Philadelphia and one of the exhibits was a "computer" that played
> >    tic-tac-toe. My memory says it was a Univac but that is becoming
> >    increasingly unreliable. What I do recall is that it was big and
> >    the display was a tic-tac-toe matrix with illuminated "X"s and "O"s.

This brought back fond memories for me of a different initiation.
It seems to me that my irst contact with computers must have
been in the IBM building in New York. I cannot say for sure at
what age this happened. I would suspect somewhere between the ages
of 6 and 10, which would be roughly 1970-1974.

They used to have a display set up with a cluster of TTY's on which
you could play twenty-questions with the computer. I think there
may also have been a tic-tac-toe setup similar to the one Padgett
described as well.

I have very fond memories of these visits (I think we went almost
any time we were in that part of Manhattan), and I think
they may well have had alot to do with my eventual career choices
(though I was not the sort of kid who immediately went home and tried
to figure out how computers worked and built tic-tac-toe players
from matchboxes).

(Of course my other fond memory of that period was visits to the
Monsanto Mill showroom where they had a demonstration of fiber/
cloth manufacturing. Perhaps if computers hadn't worked out
I'd have gone into fabric production...)

The mention of TTYs brings up another thing. Last year in my intro
CS class I was going through the standard diagram of the prototypical
computer (you know the one: input unit ----> computer ----> output unit)
and how what was meant by input and output had changed over the years.
It suddenly became clear to me that the students had no idea what
a printing terminal was and couldn't even conceive of a non-glass TTY.

Josh
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:22:59 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Brad Wieners 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Book Alert: Where Wizards Stay Up Late.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1002790160978.LTK.013@cpsr.org>
X-UIDL: 839600992.038


Sender: Brad Wieners 
Subject: Re: CM> Book Alert: Where Wizards Stay Up Late.

As far as I know _Wizards_ is due in stores no later than Septmeber 1.
(Official pub date is Aug. 20).  If you're going to review it, you might
hit Kerri Kennedy at S & S up for a review copy.  (212) 698 7537

_Wizards_ also receives a favorable notice from Stewart Brand in the Sept.
issue of _Wired_.

Having also read a preprint copy, my only complaint is that it could have
been a better _story_.  Hafner and Lyon manage to fill us in on the
employment histories of an impressive number of people, (so much so I felt
like I needed a flow chart to keep track of them all), but few complete
characters emerge. But my desires as a reader aside, I have little doubt it
will be a sourcework on the Net.

BW
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:28:25 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Mike O'Brien" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Early interactive games, EDSAC simulator for Mac/PC.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <880454593304.LTK.013@cpsr.org>
X-UIDL: 839600992.041


Sender: "Mike O'Brien" 
Subject: Re: CM> Early interactive games, EDSAC simulator for Mac/PC.

> Sender: Les Earnest 
> Subject: CM> Early interactive games.
> I vaguely recall that someone at Bell Labs built a relay computer that
> played tic-tac-toe sometime in the late 1940s or early 1950s.  The
> TX-0 computer at MIT also had a tic-tac-toe game when I started
> playing with it in 1959.  It displayed the board on its CRT and you
> selected moves by pointing with a lite pen.

        Such a machine graced the "demo hall" of the Buhl Planetarium in
Pittsburgh throughout the '50s and '60s, at least.  It was one of the
seminal influences of my childhood.  It was housed in a plastic case so
you could see all the telephone switching relays firing.  It's the first
computer I ever consciously used, and the last one which played
Tic-Tac-Toe slower than I did.  What's odd is that most of the kerchunking
of the relays (you could hear groups of them firing in distinct phases)
occurred AFTER the machine made its move.  I remember being puzzled about
that at the time, and I guess I still am.  Could it be that it was
precomputing?

        Pittsburgh, at least at that time, lacked a science museum (it
still may not have one) but the Buhl filled the bill rather well.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:36:21 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: wandd@netcom.com (Dirk H van Nouhuys)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> J.C.R. Licklider and Doug Engelbart
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1498916295780.LTK.013@cpsr.org>
X-UIDL: 839600992.047


Sender: wandd@netcom.com (Dirk H van Nouhuys)
Subject: Re: J.C.R. Licklider and Doug Engelbart

I wrote:
      It's been mentioned here before but it bears repeating that several of the
      features we associate with interactive use of personal computers were
      developed first in a timesharing environment by the Doug Engelbart group
      at SRI in the late 60's and early 70's. These include the mouse,
hypertext,
      and two-dimensional editing. (This use of the screen two-dimensionally (as
      opposed to the previous standard, line by line) did not involve
graphics or
      menus in the Engelbart group.)

Les Earnest commented
   While Engelbart's group did do some pioneering work, they were not the
   first to do two-dimensional editing.  A display-oriented timesharing
   system called Zeus was developed at Stanford University by John
   McCarthy in the 1964-66 period based on a DEC PDP-1 computer with 8 or
   so Philco CRT displays attached and included a 2-D text editor called
   TVEDIT.  I'm sure that Engelbart's group knew of that work inasmuch as
   they were located just two miles away and visited occasionally.

I should first explain that I worked in the Engelbart group (The
Augmentation Research Center to use its formal name in later days) from
 1971 through its move to Tymshare in 1978 and until 1980.

I've been cudgeling my brain about this. Les is right. I have seen the
statement I made above, I think in print, but it is wrong, and my memory
is we did not make it at the time. I didn't see the Stanford system until
Suppes' time


I wrote:
      Note that machines with a tiny fraction of the power of your desktop
      computer were "supporting" 30 or more people. Response was horrendous.

Les commented:
   Not necessarily.  Our DEC-10 timesharing system could happily carry a
   load of 30 or more people doing a mixture of editing and computing,
   with fraction-of-a-second response times for editing.

I don't doubt it. But they were not running Augment, which was a Big
System, vaguely comparable to Microsoft Office, but in some ways more
 powerful though lacking spreadsheet or database functions. My memory is
 that at that time it was around 400,000 lines of code. It was slow.

By the way, it was later called Augment, but at first called NLS, which,
strangely, stood for On-line System, since an OLS existed at the time.


I wrote:
      Later on in the 70's all the work
      in Menlo Park was being done on machines at BBN in Boston or ISI in Los
      Angeles through the (then called)ARPANet, partly to save money and
      partly to prove it was possible. It was barely possible.  [. . .]

Les commented:
   I find this remark puzzling, in as much as SRI had substantial computer
   facilities from the 1960s on.  Perhaps a few people there used
   computers elsewhere via the ARPAnet, but not anyone I knew.

It didn't make sense to many of us either, but it was done for the reasons I
mentioned. I'm not sure whether Les refers to SRI machines or
Augmentation Research Center machines. ARC was always psycho-socially
rather isolated from the rest of SRI and depended on its own machines.


 I wrote:
      Engelbart was fully aware of the problems of time sharing for interactive
      work with the computer power and network bandwidth of the time and
      would have preferred to work on what would now be called personal
      computers, but was dragged into it by Licklider and others who were
      interested in the development of time sharing and controlled his funding.

Les Commented:
   This remark makes no sense to me -- at the time that ARPA started
   funding Engelbart's group and for about another decade there was no
   such thing as a personal computer, hence it is rather unlikely that
   Lick dragged him away from working on that nonexistent device.

By personal computer I meant one that an individual could use
interactively, for example a PDP-8. Engelbart foresaw the day when
 machines that could run his system would be available to at least some
 individuals and would have liked  to work toward that.

ARC was required to make copious reports to ARPA, the Rome Air
Development Center, and the National Science Foundation, where most of
this information is spelled out in detail. I even have some in boxes
somewhere. Information about ARC, including it's copious e-mail
archives, were housed in a collection at Stanford in the  care of Henry
Lowood the last time I heard. I joined after the celebrated demo, but I still
know people that were presnt; I'll try to get some one to write about it for
this group. I believe Engelbart himself is still active through Stanford.

[Moderator's Note:  Perhaps someone on this list knows Doug well enough to
ask him to contribute to this thread -- his relationship with ARPA,
Licklider and ARCs mission.  It would be terrific to read Doug's thoughts
on all this.]
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:49:22 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Chris.P.Burton" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Noughts and Crosses - Tic Tac Toe (MENACE)
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1909394650561.LTK.013@cpsr.org>
X-UIDL: 839600992.044


Sender: "Chris.P.Burton" 
Subject: Noughts and Crosses - Tic Tac Toe

In message <584865613924.LTK.013@cpsr.org> cpsr-history@Sunnyside.COM writes:
>
> Sender: Ross 
> Subject: Re: CM> Early interactive games.
>
> > Sender: Les Earnest 
> > Subject: CM> Early interactive games.
> >
> > Padgett Peterson writes:
> >    In 1956 or possibly early 1957 I visited the Franklin Institute in
> >    Philadelphia and one of the exhibits was a "computer" that played
> >    tic-tac-toe. My memory says it was a Univac but that is becoming
> >    increasingly unreliable. What I do recall is that it was big and
> >    the display was a tic-tac-toe matrix with illuminated "X"s and "O"s.
> >
> > I vaguely recall that someone at Bell Labs built a relay computer that
> > played tic-tac-toe sometime in the late 1940s or early 1950s.  The
> > TX-0 computer at MIT also had a tic-tac-toe game when I started
> > playing with it in 1959.  It displayed the board on its CRT and you
> > selected moves by pointing with a lite pen.

I built an electromechanical N&C machine in 1953 while at University.
The fashion then (as it is now?) was to give everything an acronym
so"MENACE" stood for Mechanical/Elecrical Noughts And Crosses Engine.
As a confirmed hoarder, thank goodness I have kept it, in working order
all these years - it is currently on display at the Computer
Conservation Society exhibition at Bletchley Park in the UK.

I know many other such engines were built at the time. Donald Davies
for one in the UK, and others in the USA I was aware of. Any other
amateur builders of the time want to speak up?

[... interesting EDSAC material snipped...]

--
Chris P. Burton,  Wern Ddu, Llansilin, Oswestry, Shropshire, SY10 9BN, UK
Tel+fax: +44 1691 791274    A member of The Computer Conservation Society
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 01:56:12 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Paul Ceruzzi 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Timesharing & History (cont.)
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1718011834971.LTK.013@cpsr.org>
X-UIDL: 839600992.040


Sender: Paul Ceruzzi 
Subject: Timesharing & History (cont.)


I do not claim that Liklider, Fano, et al, were not visionaries; I only
say that the time-sharing model of computing they had was flawed. That,
among other things, helps us understand why the announcement of a $400
bag of parts, the Altair, in 1975 caused such a furor. People really
wanted a system they could call their own and do with as they wished,
even it it meant using it as a doorstop (which was about all you could
do with an Altair unless you were really talented with a soldering iron
& sweep oscilloscope).

See my recent article in _History & Technology_, 13 (1996), 1-31.

I'll go out on a limb & predict that the $500 so-called "Internet
Appliance" will fail for this same reason.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Fri, 9 Aug 1996 02:02:46 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: SteveElde@aol.com
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> IBNUC
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <172648034364.LTK.013@cpsr.org>
X-UIDL: 839600992.037


Sender: SteveElde@aol.com
Subject: IBNUC

Does anybody know about IBNUC?  This is a contraction of "IBM Nucleus", and
the way I heard it described, it was what we now call ROM.  Any information
will be appreciated.

Steve Elder
steveelde@aol.com
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:04:30 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Peter H. Salus" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> early display-orieneted editors, "expensive typewriter."
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <247034076265.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.027


Sender: "Peter H. Salus" 
Subject: Re: CM> early display-orieneted editors, "expensive typewriter."


As a pedantic historian, I feel it necessary to carry
Mike O'Brien's paragraph about the Rand Editor a step
further.

Ned Irons created the screen editor and command
interface for the CDC 6600 at the IDA in Princeton
in 1967 (IDASYS).  Irons went to Yale in 1969,
where he wrote a new (similar) editor on the
PDP-10 in 1971 (Yale Editor).  When Peter Weiner, head of
CS at Yale, left to become head of IS at Rand, he
contracted Walt Bilofsky to implement an editor
called ``re'' (1974).  The first Rand Editor was in C
on UNIX v5 (later v6).  There were extensions in 77
(when it was renamed ``ned'' -- New Editor *and*
Ned Irons) by Bilofsky and in 78 (when Bruce Borden
wrote FFIO and Dave Crocker the FFIO library into ned).
In 1979, Dave Yost went to work for Rand and rewrote most
of ned.  This went further in 1980 with the aid of
Rick Kiessig and Mark Horton.  Yost left Rand in 81, but
continued on the editor, which was renamed the Grand
Editor, while employed by Davis Polk and Wardwell in NY.
In 1988 he published ``The Grand Editor Book.''  I
ran the Grand Editor, v.3.1, at USENIX in 1988.

Peter H. Salus
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:10:36 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: John Oliver 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <833394932635.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.039


Sender: John Oliver 
Subject: Re: CM> Keypunchers and cards

At 02:22 AM 8/8/96 -0700, you wrote:
 ...
>Recall programming many keypunches (forget the designation but this was the
>unit that had a small drum in the middle that was programmable. You took
>a punch card and ran it through typing various caharacters on the line
>(surely someone here remembers the codes) and put it on the drum so that
>as a new card fed in from the hopper on the right, it would automatically
>register on col 2. When programming in FORTRAN the statement number went
>in 2-6 and the statement started in col 7.
I spent many happy (?) hours programming the 026 and later the 029
keypunches ... the alternate program feature allowed a great deal of
flexability in the use of the skip, autodup, char/numeric shift, and other
instructions available.

Minor note: I believe that all Fortran card formats used 1-5 as the
statement number, 6 as the continuation number, statement in 7-72, and card
number in 73-80.
--
John Oliver http://www.astro.ufl.edu/~oliver
 ... keeper of the [Netscape Navigator] UFAQ
     http://www.astro.ufl.edu/~oliver/faq/
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:14:04 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: baird@VV.COM (Jonathan B. Baird)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "toy" computers: Heath-Kit 1963.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <15308418843.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.029


Sender: baird@vv.com (Jonathan B. Baird)
Subject: Re: CM> "toy" computers: Heath-Kit 1963.

>(BTW, poor Heath Kit. Great company for hobbyists that fell on evil days
>when chips replaced tubes. I wonder if they're still around.)

        A quick check of http://www.heathkit.com says that they are...

                                Jonathan
 _
 \\\   _   /            Jonathan B. Baird
  \\\ //  /             baird@vv.com
   \\//  /              http://www.vv.com/
    \/ \/ .COM          http://www.vv.com/~baird
-----------------------------------------------------------------------
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:18:29 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "William F. Anderson" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "toy" computers: Heath-Kit 1963.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1721735400109.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.046


Sender: "William F. Anderson" 
Subject: Re: CM> "toy" computers: Heath-Kit 1963.

I remember the kit, but I remember much earlier than 1963. By that time, I
had built a vacuum tube calculator. As I remember, it was made out of
masonite plug board with masonite dials. I built my first Heathkit radio in
1957, when I was in the sixth grade. That same year I made a bet with my
Father that I could get my amateur radio license by the end of the year. I
won, and he bought my first Heathkit transmitter. Heathkit was bought out
by another company and I am not sure if they are still in exsistence.

Bill Anderson
Author, Consultant
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:21:33 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: thvv@best.com (Tom Van Vleck)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> editors and tic-tac-toe
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1457397553292.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.049


Sender: thvv@best.com (Tom Van Vleck)
Subject: editors and tic-tac-toe

I think it was Joe Ossanna who programmed tic-tac-toe for
Ken Thonpson's QED editor on CTSS, about 1967.  Might have been
Doug McIlroy though.  About the same time, Bill Gosper worte a
program in TECO to compute pi, using continued fractions.
Clearly both of these editors were Turing-equivalent, in principle
able to compute any function.  What other editors and command languages
of that era were able to do so?

Regarding Emacs: Bernie Greenberg's "Mother of all Emacs papers" is
available online at
  http://www.best.com/~thvv/mepap.html
it describes the history and implementation of Emacs in the late 70s.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:25:02 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Julian Reitman" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Drum Memories
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <772357872501.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.053


Sender: "Julian Reitman" 
Subject: Re: Drum Memories

The recent comments on the IBM 650, its drum memory and why not disks
brought back experiences with the first drum in real time commercial
use. The TeleRgister "Reservisor" drum. This unit, actually two drums
coupled together, each writing and reading the same data, went into
service in 1952. The information density was 20 BITs per inch and the
synchronizing was from engraved tracks separate from the magnetic
media. These were large units. One was on display at the Smithsonian
for some time.
Maybe Paul Ceruzzi has the physical discription, but these were large
units, around twenty inches in diameter and three feet long.
All inforamtion was redundant and during their life they were not
uncoupled. Finally, the drums were very reliable compared to the
associated tube circuits.

Julian Reitman
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:28:50 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "keith reid-green" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <972347300729.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.057


Sender: "keith reid-green" 
Subject: re: CM> Keypunchers and cards

>Padgett Peterson said, "Recall programming many keypunches (forget the
designation >but this was the
>unit that had a small drum in the middle that was programmable. You took
>a punch card and ran it through typing various caharacters on the line
>(surely someone here remembers the codes) and put it on the drum so that
>as a new card fed in from the hopper on the right, it would automatically
>register on col 2. When programming in FORTRAN the statement number went
>in 2-6 and the statement started in col 7.
>
>For some reason the statement had to stop on 78 and not 80 but I do not
>recall why (this was a few years ago). Do know that as each field
>was complete you just hit a key and it would automatically index to the
>next programmed field."

Two of the keypunch machines that were programmable via a punched card on a
drum were the IBM 026 and the IBM 029.  Regarding the FORTRAN statement, it
ended on column 72 for two reasons.  The first is that columns 73 through
80 were for card sequence numbers--if anybody dropped a deck they could put
it back together by running a sorter on cols 73-80.  The real reason was
that the IBM 704, for which fortran was originally written, had a card
reader that was designed to load object programs.  A word was 36 bits long,
so you could get two words on one row in columns 1-72.  The card reader
couldn't read all 80 columns, only 1-72.  Somebody wrote a program that
would convert the row-wise reading process back into columns, so cards that
were keypunched column-wise, e.g., fortran statements or assembler language
programs, could be read from the card reader, but only the first 72
columns.

Keith Reid-Green
Educational Testing Service
KReid-Green@ets.org
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:33:00 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Siebuhr, Bryan" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <2035367413979.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.060


Sender: "Siebuhr, Bryan" 
Subject: Keypunchers and cards


PADGET wrote - -
Recall programming many keypunches (forget the designation but this was
the
unit that had a small drum in the middle that was programmable. You took
a punch card and ran it through typing various caharacters on the line
(surely someone here remembers the codes) and put it on the drum so that
as a new card fed in from the hopper on the right, it would automatically
register on col 2. When programming in FORTRAN the statement number went
in 2-6 and the statement started in col 7.

For some reason the statement had to stop on 78 and not 80 but I do not
recall why (this was a few years ago). Do know that as each field
was complete you just hit a key and it would automatically index to the
next programmed field.

Also recall that it could be set to automatically print a character or
series
on the card but do not think it could autoincrement.
 ------------
In the mid-70's I was programming in Fortran on an XDS (as in Xerox Data
Systems) 930.  All of the code was punched onto 80 column cards with
Fortran statement numbers in cols 2-6, statements in cols 7-72, and a
sequence number in cols 73-80.  The cardpunch was an IBM 029 (?) and
could be programmed to autoincrement the sequence number.  Sometime the
programs would fill several card boxes at 2000 cards per box.

The presence of the sequence number saved many long hours that could have
been lost when a box of cards was dropped and cards went everywhich way.
 It didn't matter, you just picked up the cards, read them into a file
and sorted based on the sequence number before compiling.

Another advantage of having the sequence number was in updating code.
 Invariabley, after feeding cards into the hopper for 15 minutes and
compiling, there were errors.  Instead of having to re-read all the
previous cards, you could punch a small deck with deletes and adds based
on the sequence number, run this small deck against your original file
with a sort/merge, and then compile the result.

Bryan Siebuhr
Systems Programmer
The Evergreen State College
siebuhr@elwha.evergreen.edu
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:39:28 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Austin Meredith 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> TV-EDIT, "Electric Pencil."
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1399218414366.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.041


Sender: Austin Meredith 
Subject: Re: TV-EDIT

Despite what Douglas R. Hofstadter wrote in his book _Godel, Escher, Bach:
An Eternal Golden Braid_:

> I have said to many people that it would have taken me twice as long
> to write my book if I hadn't been able to use "TV-Edit",...

when I myself asked him, during that period, for his recommendation of a
word processor to use to write a book, I remember quite clearly that he
recommended not TV-EDIT but instead something I was never able to find,
called "Electric Pencil."

\s\ Austin Meredith , "Stack of the Artist of Kouroo" Project
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:45:18 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: rsaha@hqgate.orbital.com
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "toy" computers: Heath-Kit 1963.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1684218303717.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.058


Sender: rsaha@hqgate.orbital.com
Subject: Re: CM> "toy" computers: Heath-Kit 1963.

     A note on Heath-Kit.  I bought two Heath-Kits back in 1985. Both were
     also labeled Zenith.  ( I don't know if Zenith bought them, or if they
     always owned Heath-Kit).  One was a "Clapper" (Clap on-Clap off) type
     gizmo. The other was an alarm system that rang if you broke a light
     beam.  Both worked moderatly well until my dog found them...  There
     was only one place in St. Louis (where I grew up) that sold Heath-Kit
     at that time.

     Rahul

     --
     Rahul A. Saha
     Systems Engineer
     Taurus Launch Vehicle
     Orbital Sciences Corporation
     rsaha@orbital.com
 ______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:50:42 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "L.R. Zeitlin" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> 80 char line, early programming memories.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <145204161441.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.051


Sender: "L.R. Zeitlin" 
Subject: Re: CM> 80 char line, early programming memories.

      Louis Rader, President of GE computers in the late 60's and early
70's had been a Univac executive. When he was recruited to head up GE he
brought a number of engineers and executives with him. That probably is the
source of the 96 hole GE punch card.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 01:56:55 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: rab@well.com (Bob Bickford)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "toy" computers, also tvedit, LIFE program.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <39710598414.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.059


Sender: rab@well.com (Bob Bickford)
Subject: "toy" computers, also tvedit


On the subject of toy computers, I have two items:

1. My college roommate, Mike Iddings, told me several times that he had
once "built a computer out of paper clips".  I think he tried to explain
it to me once, but lacking requisite materials he just had to wave his
hands a lot.  Does anyone know or recall the details?  I assume that there
must have been some kind of flip-flop element involved?

2. Speaking of flip-flops, I checked out a library book several times in
my high school freshman year, the title of which I no longer recall, but
it was about building your own computer.  I had been teaching myself about
electronics for a couple of years at that point, but still found some of
the discussion in the book a bit dense.  Eventually, I figured it all out:
the book was instructions for building a series of flip-flop based adders,
and using a telephone dial as the input device.  When I realized that the
long list of complicated instructions (the authors went into laborious
detail on everything down to soldering technique) and even longer list of
parts would only build a machine that added two numbers and displayed the
result in binary (lamps), when what I wanted to do was build a machine to
do the fascinating LIFE algorithm that I'd just read about in Scientific
American, well, I was a bit disappointed.  Within a year or two I learned
about TTL and integrated circuits, and recovered my enthusiam.


Regarding TV-Edit, I know I used a program by that name on the Intel
"MDS-230" microprocessor development system in 1979.  Does anyone know if
this is the same thing as has been discussed here (well, a port obviously,
if anything)?  I know I liked it a lot, nearly as much as I liked the
original CP/M WordStar.

--
Bob Bickford          rab@well.com
"Professional Wild-Eyed Visionary"
Coordinator, National Libertarian Party's Anti-CDA Campaign
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Sat, 10 Aug 1996 02:00:29 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Fred Cisin 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <753131636628.LTK.013@cpsr.org>
X-Status: 
X-UIDL: 839776397.061


Sender: Fred Cisin 
Subject: Re: CM> Keypunchers and cards

> Recall programming many keypunches (forget the designation but this was the
> unit that had a small drum in the middle that was programmable. You took
> a punch card and ran it through typing various caharacters on the line
> (surely someone here remembers the codes) and put it on the drum so that
> as a new card fed in from the hopper on the right, it would automatically
> register on col 2.

Those were the 26 series (old, rounded, black) and 29 series (sleek
modern, beige).  There were a few deviant models, such as verifiers,
interpreters, etc.


> When programming in FORTRAN the statement number went
> in 2-6 and the statement started in col 7.

In FORTRAN, a 'C' in column 1 meant that it was a comment.  But I guess
that REAL programmers didn't comment :-)    Any code in column 6 meant
that the card was a continuation of the previous card.


> For some reason the statement had to stop on 78 and not 80 but I do not
> recall why (this was a few years ago).

Actually 72.   That was mostly so that you could put a series of
consecutive numbers in the final columns.  Then, WHEN one of the
operators dropped your deck, you could use an 82 sorter to put it back in
order.  But most people just put a diagonal line on the deck of cards
with a magic marker - any deviation in the line meant that the cards were
out of order.  Any revisions to the program then required a new line,
until the deck was so marked up that you had to make a duplicate copy.


> Also recall that it could be set to automatically print a character or series
> on the card but do not think it could autoincrement.

SOME (but not most) could. And there were more bizarre goodies available.
At GSFC (Goddard Space Flight Center), I spent an unpleasantly large
amount of time operating a Gerber Data Digitizer (aka "that damned
etch-a-sketch") interfaced to a punch.  Turn the knobs to align the
crosshairs, hit the footpedal, and the punch punched a pair of 3 digit
coordinates.


> As memory serves, once managed an all-time best of 600 cards in an hour.
> Since that was one every 10 seconds it must have had a *lot* of very short
> statements.

In what number base?  If I did 10 seconds per card, then I got 6 cards a
minute, or 360 per hour.  But then, I was never very fast on the keypunch.

--
Fred Cisin       (510) 436-2663   Computer Information Systems Department
Merritt College         12500 Campus Drive      Oakland,  CA  94619
RE: Peralta district internet connection..  "Mir zul nor leben su zane."
                                              ("May we live so long.")
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 01:08:34 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "William F. Anderson" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine Questions on Machine History.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <472554111313.LTK.013@cpsr.org>
X-UIDL: 839859838.029


Sender: "William F. Anderson" 
Subject: Re: CM> Nine Questions on Machine History.

I will try and answer your questions.

Bill Anderson
Author, Consultant

----------
> From: Wesley J. Miller ((8-444) 919-254-9774) 
> To: Multiple recipients of list cpsr-history@cpsr.org

> Subject: CM> Nine Questions on Machine History.
> Date: Friday, August 09, 1996 2:06 AM
>
>
> Sender: "Wesley J. Miller ((8-444) 919-254-9774)" 
> Subject: Machine History
>
> I'm a short-timer in the world of computing; I started with FORTRAN-66 on
punch
> cards about 1978 using WATFIV and FORTRAN-H.

Other than building a vacuum tube calculator in high school, my real
computer experience started in 1969. Coourtesy of the USAF, I started of in
AUTODIN working on RCA Spectra 70 Model 55. We first learned to program in
microcode with LIDMEN. Programs were punched on paper tape, and eight
deletes constituted a new program. Assembler was a high level language
compared to microcoding. My first real high level language was JOVIAL.

>
> As I read the recollections of folks who came into computing before and
about
> the same time as I, I am remembering all those stories I heard and
questions I
> only ever got partially answered.  So, here is a list of questions that I
hope
> will start a few memories flowing and get some history nailed down for
all of
> us short timers.
>
> 1. The first "personal computer" I used was a Datapoint 1134.  I know the
> last digit represented the number of floppy drives installed.  I also
> remember that the machine had 16K of memory on four 4K memory cards that
> plugged into the processor cards on which I was told there was an 8080 --
not
> the chip but a hardwired-from-discretes equivalent which Datapoint used
> because Intel couldn't get the 8080 out in time.  Questions:  Was it in
fact
> an "8080"?  Was it Datapoint that drove Intel to develop the 8080?
>
The 8080 was developed as a CRT controller for Datapoint was not accepted,
so Intel pushed as a general purpose microprocessor. As I mentioned before
on this list, my first home computer was an IMSAI 8080. While it normally
came with 2K or 4K of memory, I bought to 8K Dutronic memory boards. Of
course, in those days, everything was a kit. I remember that there were
1900 solder connections on each memory board.

If I remember correctly, the Z80 was designed to be a washing machine
controller.

> 2.  It's easy to see that EBCDIC came from punch card encoding.  Where
did
> ASCII come from?  Why did someone choose to rewrite the coding standard
when
> both EBCDIC and ASCII seem perfectly capable of doing the same thing?
>
In the 60's, there were multiple Holerith codes and even EBCDIC had
variations. Plus, there were other code sets such as Field Data Code.
Communications between different varieties of hardware required translation
tables. This creates a nightmare when you are trying to build a network as
diverse as AUTODIN. So, ASCII came into being as a standard communications
code between devices. Devices could use whatever internal code set they
desired as long as they communicated in ASCII. I was really surprised when
I got my IMSAI and saw ASCII as internal code set.

> 3.  Just why is it that some folks say IBM never got time-sharing right?
> They've sold a LOT of TSO, after all. (No, this is NOT a debate on
> market-share = got-it-right)
>
I can't answer this one. However, it does remind my of a little story. When
I was in the Air Force, I went to the IBM 370 product announcement at the
State Department. After this long presentation on the new features of the
370, the saleman, who was the moderator, got up and thanked for coming to
see the product announcement for the "new IBM system 360."

> 4.  I personally think TSO is about the most user-hostile
environment/op-sys
> on earth.  I first used IBM's CMS version 2 as an op-sys about 1982.  I
still
> use it (even as I type this note).  Yes it's old and character only but
it
> is SO much easier to use than TSO.  What's the history of CMS?  Why did
it
> never replace TSO?  Unix?

I hated TSO. When I was at Four Phase systems, they used another package
for which I forget the name at the moment. I do remember that when you type
help, you got the message "the great pumpkin helps those who help
themselves."

>
> 5.  Yes, I use UNIX.  Yes, I like it--yes, I hate it.  Where did the
> let's-make-it-so-cryptic-no-one-can-intuitively-figure-it-out mentality
> and naming conventions come from?  TYPE ... | FiND .... sure seems
> a lot more obvious to me than CAT ... | GREP ...
>
You have to thing of cat as a filter and then the name makes sense for it
is an abreviation of concatenate. I forgot what grep stands for.

> 6.  IBM numbers bits from left to right; the MSB is bit 0.  Who decided
to
> do this dumb thing?  Same is true of byte and character string numbering.

I don't know, but it does keep you on your toes.
>
> 7.  Who decided that token-ring and ethernet addresses are sent in
opposite
> orders?  LSB-first vs. MSB-first?
>
This comes from those machines that do reverse byte ordering.

> 8.  Why are people trying to shoehorn FORTRAN into an array handling
language?
> Why aren't they using APL/2 instead?
>
I never really got involved with FORTRAN or APL.

> 9.  And finally, for now, who designed and why do we use the file system
used
> by DOS, MAC, Unix and I don't know what else?  Specifically, these file
systems
> treat all text files as one long string with records indicated by wasting
> the two characters 0D and 0A.
In UNIX only the OA (linefeed). The idea being that text files are free
format files without any inherent structure. This allows any structure to
be imposed on the file by the application. The OA was used only in text
files that could be displayed or printed as is. What created constant
confusion is the use of OD and OA in DOS. By the time DOS appeared, most
devices had could deal with adding a carriage return to the sole linefeed.
Although, DOS is technically more correct as a linefeed is supposed to move
down one line and the carriage return takes you to the beginning of the
line. If you entered only a carriage return on a true ASCII device, you
would overtype the same line.

>  Mainframes and even my dear departed Datapoint
> 1134 used a full blown file system where records are stored as records.
Each
> record either points to the next/last record or has a length indicator.
No
> characters are wasted.  Same question really applies to C and DOS ASCII-Z
> strings.  Who decided 00h wasn't a datum?  Ever try to pick apart
non-text
> data and not be able to use those nice routines that think the 00's
aren't
> data?
According to ASCII 00h is a null character and is valid data. Sending nulls
was used to introduce delays when you where trying to communicate to a slow
device.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 01:33:33 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Michael Seadle 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine Questions on Machine History.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1993021061809.LTK.013@cpsr.org>
X-UIDL: 839859838.030


Sender: Michael Seadle 
Subject: Re: CM> Nine Questions on Machine History.


Regarding the history of VM:

Melinda Varian of Princeton wrote an excellent and substantial history of
VM, which she presented at a SHARE conference sometime in the 1980s.  I
don't think she's ever published it, but she might be persuaded to put it
on her home page on the Web.

She is also a major figure in the history of VM.  What VM systems
programmer has not memorized her "What your Mother Never Told You about
VM Maintenance"?

--Michael Seadle
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 01:50:51 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Lawrence Zeitlin 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Decimal computers, IBM 7072, Licklider.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1707765345839.LTK.013@cpsr.org>
X-UIDL: 839859838.031


Sender: Lawrence Zeitlin 
Subject: Re: CM> ecimal computers, IBM 7072.

I worked on a decimal computer, the Bendix G15, in the late 50's at the RCA Air
borne Systems Lab. As I recall, the computer used 10 step switching vacuum tube
s and output a display on Nixie tubes. Programming was done by inserting pins i
n cribbage boardlike plastic panels which could then be hooked together in a lo
ng chain. The research project I was involved with dealt with the control dynam
ics of the proposed F108 fighter plane which got cancelled in favor of guided m
issles. J.C.R. Licklider was a consultant on that project and I got to know him
 well in the pre-ARPA days.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:26:03 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Les Earnest 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Licklider, Timesharing, and Popular History
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1927422201961.LTK.013@cpsr.org>
X-UIDL: 839859838.048


Sender: Les Earnest 
Subject: CM> Licklider, Timesharing, and Popular History

Responding to an earlier claim that advocacy of a "computer utility"
by unnamed "early time-sharing advocates" had impeded progress, I
remarked:
   > Thanks, but I don't buy this argument.  Who was the unnamed person who:
   >    "thought of one or a few central computers (one even proposed siting
   >    it in Kansas City), accessed through a jack in the wall just like
   >    you get electric power?"
   > Doesn't sound like any timesharing pioneer that I know of.

John Cowan responded:
   A version of this scheme was proposed in John Kemeney's book
   *Man And The Computer*, which was (especially in its later chapters)
   an idealized design of a universal timesharing utility, with
   big iron in major cities.

I confess that haven't read that book, but I would be surprised if it
held up progress as much as a microsecond.  Also, though Kemeny did do
some pioneering work in developing a multi-user BASIC system, it was
not a general purpose timesharing system.

In response to the same remark, Tom Van Vleck wrote:
   Well, Corbato & Vyssotsky wrote, in "Introduction and Overview of the
   Multics System" (1965 FJCC) (http://www.best.com/~thvv/fjcc1.html)

   One of the overall design goals is to create a computing system which is
   capable of meeting almost all of the present and near-future requirements of
   a large computer utility. Such systems must run continuously and reliably
   7 days a week, 24 hours a day in a way similar to telephone or power
systems..."

   The idea of computing delivered via a wall jack was suggested often
   at Project MAC in the 60s.

Not only that, it became a reality by the late 60s in the form of
commercial timesharing systems that were accessible by modem.
However, I doubt that anyone who had built a timesharing system would
claim that "one or a few central processors" could meet the nation's
computing needs.

Thus I stand by my earlier rejection of the hypothesis that the
concept of timesharing somehow "impeded progress."  In fact, it
greatly enhanced productivity of programmers and data entry people
and made both display-based interaction and networking practical
for the first time.

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:32:51 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Les Earnest 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> 80 Character Line
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <751912376370.LTK.013@cpsr.org>
X-UIDL: 839859838.040


Sender: Les Earnest 
Subject: CM> 80 Character Line

Woody Franke writes:
   On the topic of 80 character lines, that was the standard when I started
   programming the IBM 1401 in 1964.  But I seem to remember that Univac (who
   was then either Sperry or Sperry-Univac) had a card reader and keypunch
   standard that had round hole punches instead of the IBM rectangle hole
   punch and, because of the space the round hole took, had a standard 70
   characters on a punch card.  Is my memory correct?

As I recall the Sperry-Rand cards had 45 characters across each card in
each of two fields, one on the upper part of the card and one lower,
for a total of 90 characters per card.  The punches were indeed
circular, in contrast with the IBM punches which were tall and thin.
The Sperry equipment used the same card stock as IBM but they were
much less successful commercially and eventually faded away.

Thus, when computer terminals were created, the initial standard was
80 characters per line, to accomodate IBM's "unit record" standard.
IBM insisted throughout that period that their cards be called
"Hollerith cards" rather than "IBM cards" out of fear that the latter
would become a generic term and would cause them to lose "IBM" as a
trademark.  Mr. Hollerith was the inventor who created punched cards
and associated equipment for use in an early U.S. Census.

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:37:34 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "Peter D. Junger" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> TV-EDIT, "Electric Pencil."
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <62598987653.LTK.013@cpsr.org>
X-UIDL: 839859838.036


Sender: "Peter D. Junger" 
Subject: Re: CM> TV-EDIT, "Electric Pencil."

Austin Meredith writes:

: Despite what Douglas R. Hofstadter wrote in his book _Godel, Escher, Bach:
: An Eternal Golden Braid_:
:
: > I have said to many people that it would have taken me twice as long
: > to write my book if I hadn't been able to use "TV-Edit",...
:
: when I myself asked him, during that period, for his recommendation of a
: word processor to use to write a book, I remember quite clearly that he
: recommended not TV-EDIT but instead something I was never able to find,
: called "Electric Pencil."

Electric Pencil was a word processor for CP/M microcomputers.  When I
bought my NorthStar Horizion with 56K of memory Word Star had just
come out.  At the time I was told that ``Word Star is a superset of
Electric Pencil''.  As I recall what a colleague who did use Electric
Pencil said, it would only handle as much text as one could get in
memory, so every time you did a couple of pages you had to save it and
start a new file, or at least a new section of a file[

--
Peter D. Junger--Case Western Reserve University Law School--Cleveland, OH
Internet:  junger@pdj2-ra.f-remote.cwru.edu    junger@samsara.law.cwru.edu
                     URL:  http://samsara.law.cwru.edu
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:42:07 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "A. Padgett Peterson P.E. Information Security" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Heathkit
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <598957771926.LTK.013@cpsr.org>
X-UIDL: 839859838.035


Sender: "A. Padgett Peterson P.E. Information Security"

Subject: Heathkit


>(BTW, poor Heath Kit. Great company for hobbyists that fell on evil days
>when chips replaced tubes. I wonder if they're still around.)

Just part of Zenith. Sometime between 1982 and 1985 it became "Heath-Zenith"
and what is now "Zenith Data Systems" is located on the Heathkit property.

I *think* the Zenith Z-100 "semicompatable" PC was the first joint product.
Later that portion of Zenith was sold offshore.

Have an application here for the "Heath/Zenith User's Group" (HUG) from
about 1986 (have the details here *somewhere* but other than Z-159 and Z-200
(kit 5) manuals, cannot find it. Not unusual.)

Just as a comment, while many decried the Zenith computers, for anti-virus
reseach and exploring how a PC booted it is unequalled since the ROM BIOS
contains a complete DEBUG-like debugger (my little Columbias also have
a ROM debugger but the commands are nothing like DEBUG's).

                                                Warmly,
                                                        Padgett
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:46:56 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Nelson Winkless 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Thundering keypunches, punch cards roadkill.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1511997661215.LTK.013@cpsr.org>
X-UIDL: 839869011.002


Sender: Nelson Winkless 
Subject:

>...the sequence number saved many long hours that could have
>been lost when a box of cards was dropped...   --Bryan Siebuhr

Nobody has mentioned the thunder of the doggoned keypunches.

Back in the mid-sixties, the Thomas Bede Foundation in Los Altos CA
did a series of projects developing self-organizing systems for Sandia
Corporation. The nice redwood-and-glass building we were in at 745 Distel
Drive was rather loosely constructed. If you got up next to the ceiling
at night, you could peek through gaps above the partitions large enough
so you could see light in somebody else's office at the far end of the
building. The point of this description is that the IBM keypunch machine
we had in pretty steady use for several programmers could be heard in
every office in the building. It thumped away hour after hour, getting on
the nerves of the real estate people, the architects, the sales reps, the
clinical psychologist, and all other tenants. They didn't enjoy the onset
this age of computers, and complained regularly.

And with respect to spilled cards --

Our folks didn't have a computer, just a computer courier service, an
outfit called CEIR that rented available time on computers all around the
Bay Area, and would run their client's cards wherever a system was available.
(Time was available mostly at night, of course.) Our stuff ran on a system up
at Shell in the City, for example, somewhere in San Jose, and somewhere else
in Walnut Creek, I think. (Later, we used the 6600 at CDC's center on Hanover
in Stanford Industrial Park, and it was remarkable to be able to see the
actual computer through the glass walls.)

We felt that the CEIR service was really good when we could get two shots
at a computer every day. The programmers would frantically get their stacks
of cards ready for the pickups, and then stand around nervously, waiting for
results when the courier returned. The courier would typically come and go
with a couple of IBM Card boxes full of the labor of half a dozen people.

One morning as I turned on to Distel from El Camino, I saw what looked
rather like snow in the intersection. IBM cards were scattered all over the
street,
and spread far into the field across the way (in whose middle we
occasionally dumped a gallon or two of MEK at night, having no other
chemical disposal method...what did we know in 1965?).

As I reached the office, I found two people weeping. One was the programmer
who had failed to close the boxes securely with rubber bands. The other was
the CEIR courier, who had put the boxes on top of his station wagon briefly,
while
unlocking the door, and had then driven off with the boxes still on the
roof...until making the hard turn at the corner.

The seguence numbers might have been helpful before the cards had been run
over in heavy traffic, but...

Does anybody still make Christmas wreaths out of cleverly curled and stapled
IBM cards?

--Nels

-----------------------------------------------------------------------------
Nelson Winkless                   Email: correspo@swcp.com
ABQ Communications Corporation    Voice: 505-897-0822
P.O. Box 1432                     Fax:   505-898-6525
Corrales NM 87048 USA             Website: http://www.swcp.com/correspo
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:51:27 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Jay Hosler 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <424871352486.LTK.013@cpsr.org>
X-UIDL: 839859838.052


Sender: Jay Hosler 
Subject: Re: CM> Keypunchers and cards

>
>
> Sender: Fred Cisin 
> Subject: Re: CM> Keypunchers and cards
>
> > Recall programming many keypunches (forget the designation but this was the
> > unit that had a small drum in the middle that was programmable. You took
> > a punch card and ran it through typing various caharacters on the line
> > (surely someone here remembers the codes) and put it on the drum so that
> > as a new card fed in from the hopper on the right, it would automatically
> > register on col 2.
>
> Those were the 26 series (old, rounded, black) and 29 series (sleek
> modern, beige).  There were a few deviant models, such as verifiers,
> interpreters, etc.
>
During my tenure at SDC (late 1958 - early 1961) the 026 keypunches were
converted from the old 9-mil card thickness to the new 6-mil thickness.  (I
don't believe any 029 keypunches adjusted to the old 9-mil standard).  The
9-mil and 6-mil cards were kept in distinct racks; each keypunch had a sign
revealing whether it was adjusted for 6-mil or 9-mil cards.  The penalty for
error was a jam.  Two 6-mil cards will fit, or rather jam, in the read throat
of a 9-mil keypunch or card reader, and 9-mil cards jam 6-mil throats.

To edit a program deck, one punched (or had punched) the needed replacement
cards, went to the the drawer containing the deck, and modified the deck
manually, guided by sequence numbers, listings, and the card contents printed
at the top border of each card.  It was all too easy to include a 9-mil card
in a 6-mil deck inadvertently.  The resulting jam often left the deck in
considerable disarray.

> Actually 72.   That was mostly so that you could put a series of
> consecutive numbers in the final columns.  Then, WHEN one of the
> operators dropped your deck, you could use an 82 sorter to put it back in
> order.  But most people just put a diagonal line on the deck of cards
> with a magic marker - any deviation in the line meant that the cards were
> out of order.  Any revisions to the program then required a new line,
> until the deck was so marked up that you had to make a duplicate copy.

At SDC, the normal procedure was to get a new listing after each modification
had been made, and, after a couple of mods, to repunch and resequence the deck
using an interpreting card duplicator.  Card decks of the size we used at SDC
-- 3000-15000 cards -- were normally removed from their trays only by
operators.  We always sequence-numbered decks; using just a diagonal line
would have unreasonably jeopardized the deck.

By 1965, when I was at CSC, the use of cards was limited to original input and
edit decks; the program files themselves were kept on tape.  By 1970 or so I
had worked with my last card deck.

BTW, the armies of 20-drawer card files that lined every office and computer
room have mostly passed to the scrap pile.  I am lucky enough to have found
one in a surplus store; it makes a wonderful tool cabinet, as well as a
nostalgic link to a technology I hope never to enjoy again.

Jay
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 02:56:11 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: "christopher f. chiesa" 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Keypunchers and cards, programming in high school.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <24010985771.LTK.013@cpsr.org>
X-UIDL: 839860918.000


Sender: "christopher f. chiesa" 
Subject: Re: CM> Keypunchers and cards


It's absolutely FASCINATING to read about all the cool things one could
evidently do with the 026 and 029 keypunches.  Despite apparently being
one of the "youngsters" here :-) , I can relate, because I was among one
of the LAST groups of users of keypunches, cards, and "batch" computing,
in a couple of math courses in the tenth and eleventh grade (1979-81).

The computer was an IBM 1130 with, I'm told, 4K of magnetic core memory.
I know that the system had "a disk," but had no concept of even WONDERING
about the type or capacity thereof, so never asked about, or knew, such
details.  Similarly, I have no idea what type of keypunches we used, or
what their capabilities might have been: the only thing we students were
allowed, or taught how, to do was "type onto cards" and occasionally
duplicate all or part of a previously-punched (or mis-punched) card.
These keypunches had a rotating drum visible through a window in the
center of the top of the unit, between the "blank card supply" bin (on the
right) and the "punched card receptacle" bin (on the left), but we never
knew or were told what that drum was for.  Likewise there were lots and
lots of mysterious buttons and switches about which we were told only,
"don't touch 'em."  Oh, and these punches also printed a visible,
human-readable copy of the punched data, along the top of each card they
punched.

I once discovered the "overpunch" key and worked out what characters/keys
I needed to "overpunch" in order to produce a card whose holes spelled
out my name, like this:

      ***** *   * ***** * *****
      *     *   * *   * * *
      *     ***** ***** * *****
      *     *   * *  *  *     *
      ***** *   * *   * * *****

I vaguely recall getting into some kind of trouble just for PUNCHING a
card like this, even though I never attempted to load it into the
computer.

I have a lot of other memories associated with that computer.  Every now
and then the system needed to be rebooted, and I remember that the first
card was "more holes than card," which I found pretty surprising/amazing.

The lights on the console and on the line printer were small incandescent
bulbs behind/under carefully-shaped, -polished, and -labeled chunks of
colored, frosted glass; the bulbs would cause the glass chunks to glow
softly in their respective colors; it was quite pretty.

One time I wrote a version of Conway's "Game of Life," which took
something like five MINUTES to compute-and-print one "generation," and
which made my classmates groan whenever they saw me approaching the
card-hopper with my deck.  :-)  I actually had to ask for special
assistance from our computer teacher, to implement a "save current
generation to cards" feature, so that I could break long Life sessions
into shorter batches so the other kids could get some real work done! :-)
I still have the source code, only slightly modified over the years to run
on other IBM systems (RPI, 1982 - IBM 3083? 3033?  I don't remember;
running "MTS," the Michigan Terminal system, worthy of an entire posting
here in its own right) and later on a VAX/VMS system.  If anbody wants a
copy...

Our instructor, of course, could do quite amazing (to us :-) things with
that system.  Often he would drop in a mere three or four program cards
and cause a game or line-printer-picture-printing operation to occur.  We
had no idea how he was able to do that with only three or four cards, but
in retrospect, obviously he was running something "from the disk."

About a year after I graduated from high school, the school finally
retired the IBM 1130 and replaced it with a PR1ME system of some type,
which was terminal-based, interactive, and timeshared.  I was actually
offered the 1130 myself, to take home and install in my parents' basement,
but my parents didn't go for the idea, so I didn't get the computer.
Ironically, a little more than a year after THAT, I acquired an Atari 800
"personal computer" whose high-resolution screen display alone used twice
as much RAM as the entire magnetic-core memory of the 1130.  How things
change!

Chris Chiesa
  lvt-cfc@cyber1.servtech.com
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:05:31 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: SteveElde@aol.com
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Early interactive games.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1948767304330.LTK.013@cpsr.org>
X-UIDL: 839859838.034


Sender: SteveElde@aol.com
Subject: Re: MH> Early interactive games.

Michael S. Hart writes:

>The Tic-Tac-Toe game most people remember around here
>was at the Museum of Science and Industry in Chicago,
>and my recollection is also that it was part of "Bell
>Labs" exhibit, which was always quite nice.

As I remember, you used a telephone dial to enter your moves.  The machine
always took the first move, but it didn't always take the center square, so
it was possible to beat the machine or play to a draw.

Steve Elder
steveelde@aol.com
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:11:46 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Michael Froomkin 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Toy computers -- cardboard
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <381372388705.LTK.013@cpsr.org>
X-UIDL: 839859838.046


Sender: Michael Froomkin 
Subject: Toy computers -- cardboard

In grade school, in about 1973 give or take a year or two, we were treated
to an educational experiment.  It was a little cardboard device, not much
smaller than a sheet of paper, that simulated a computer.  It had a little
"bug" (a lady bug?) that you moved from hole to hole to simulate the
location of memory currently being accessed, and you wrote in the various
rectangles next to the holes to simulate the very simple, assembler-like,
instructions/data that could be in a memory location. It had a cardboard
strip that you pulled through a little window that simulated the boot
strip or the card instructions, I don't remember which.  It taught you the
concepts of programming, at a very basic level.  You could simulate
addition, subtraction and I think simple multiplication although that was
very hard.

I can't remember what the thing was called, but I remember enjoying it; I
think it taught me a great deal too.  I suppose the experiment was not
considered a success, since I've never seen one since.

I'd love to know what it was called, and how I would get one for nostalgia
value ...and possible deployment on my children...

[This message may have been dictated with Dragon Dictate 2.01.
Please be alert for unintentional word substitutions.]

A. Michael Froomkin        | +1 (305) 284-4285; +1 (305) 284-6506 (fax)
Associate Professor of Law |
U. Miami School of Law     | froomkin@law.miami.edu
P.O. Box 248087            | http://www.law.miami.edu/~froomkin
Coral Gables, FL 33124 USA | It's hot here!  And humid!
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:16:36 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Fred Cisin 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> "Electric Pencil."
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <72613389993.LTK.013@cpsr.org>
X-UIDL: 839859838.041


Sender: Fred Cisin 
Subject: Re: CM> "Electric Pencil."

On Sat, 10 Aug 1996, Austin Meredith wrote:

> recommended not TV-EDIT but instead something I was never able to find,
> called "Electric Pencil."

"Electric Pencil" was by Michael Shrayer.  It originally came out before
CP/M did.  For a while they also provided a disk format conversion
program to convert between "Electric Pencil" disk format and CP/M disk
format.  It faded out when Wordstar achieved market dominance.

"Electric Pencil" made a brief comeback on the TRS-80 model 1, inspite of
the fact that it required some minor hardware mods. (addition of an extra
key to serve as a CTRL key, and cuts and jumpers to enable lower case.)
It faded out when Radio Shack's "SCRIPSIT" achieved market dominance.

"Electric Pencil" came back one last time on the PC.  The final attempt
was marketed by H.C. "Harv" Pennington, best know for his book "TRS-80
Disk And Other Mysteries".  Didn't get off the ground.  I never even
broke the shrink-wrap of my copy.

--
Fred Cisin       (510) 436-2663   Computer Information Systems Department
Merritt College         12500 Campus Drive      Oakland,  CA  94619
RE: Peralta district internet connection..  "Mir zul nor leben su zane."
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:21:40 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Les Earnest 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Shaky and the Stanford Cart
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <848477329299.LTK.013@cpsr.org>
X-UIDL: 839859838.049


Sender: Les Earnest 
Subject: CM> Shaky and the Stanford Cart

Kip Crosby writes:
   That 940 was donated to Resource One by Transamerica Leasing after it
   finished a stint at SRI, where it had been the control computer for Shakey
   the Robot.  Lee says that Shakey was "shaky" because the 940's DMA channel
   was wired to the wrong voltage, and when Resource One got it they diagnosed
   and rewired it.

Actually, Shaky acquired its name from the rather shaky manner in
which it moved about, which was the result of a mechanical stability
problem.

The Stanford A.I. Lab had a computer controlled electric cart in the
same era (late 1960s to 1980) that was somewhat more stable.  It was
linked to the central computer by two-way radio, allowing the computer
to receive television images from its camera and to give steering and
speed commands.  It was programmed to steer along roads around the lab
and generally did so successfully, though on one occasion it was found
wandering off campus by a staff member who was coming to work.  He got
out of his car and waved his hands in front of the TV camera to alert
the person who was running the experiment to the fact that the cart
had apparently made a wrong turn onto Arastradero Road.  Fortunately
it escaped with its "life."

The Cart was built originally by some mechanical engineers at Stanford
who had been asked by NASA to look into the possibility of controlling
a lunar roving vehicle from Earth taking into account the round-trip
communication delay, which is a couple of seconds.  NASA was planning
at that time to send an unmanned vehicle to the moon but later decided
that humans would make a better show.

Trying to steer a lunar vehicle from Earth by using the just TV image
doesn't work at any reasonable speed -- it breaks into oscillation and
the driver loses control.  However, the ME group was able to develop
an analog computer that presented a cursor on the display at the point
where the vehicle would be when a change in direction or speed could
take effect, and this worked well.

Hans Moravec also used the Stanford Cart to do a PhD dissertation on
visually recognizing and steering around obstacles using stereoscopic
imagery.

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:26:46 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Alan Bawden 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> editors and tic-tac-toe
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <2133790532605.LTK.013@cpsr.org>
X-UIDL: 839859838.044


Sender: Alan Bawden 
Subject: CM> editors and tic-tac-toe

   Date: Sat, 10 Aug 1996 01:20:51 -0700
   From: Tom Van Vleck 
   Subject: editors and tic-tac-toe
   ...
   Clearly both of these editors were Turing-equivalent, in principle
   able to compute any function.  What other editors and command languages
   of that era were able to do so?

Sometime in the late 1970's MIT's Student Information Processing Board
(SIPB) granted me $50 worth of computer time on MIT-Multics to investigate
this very question for the Multics `qedx' editor.  qedx had extremely
limited control structures -- I believe that all there was was an
unconditional subroutine call (the "b" command I think), and errors that
would cause the currently executing macro to terminate prematurely.  These
limitations were somewhat offset by the fact that you could write
self-modifying code, but it was still difficult to construct an actual
-loop-.  I was able to cobble together a simple Turing machine emulator
using the subroutine call command to drive the main loop.  In theory, a
subroutine call as the last command in a macro could have been implemented
as a branch, consuming no additional stack.  Experimentation revealed that
the actual Multics implementation of qedx lacked this optimization.  (I
never actually looked at the qedx source code.)

So the answer was that qedx -was- Turing-equivalent, although as a
practical matter you would quickly run out of stack.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:31:19 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: Bob Bickford 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine questions, four answers.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <317062186878.LTK.013@cpsr.org>
X-UIDL: 839869011.000


Sender: Bob Bickford 
Subject: Re:  CPSR-HISTORY digest 52


Wesley J. Miller asks several questions.  Here are some comments on four
of them:

5.  (names of UN*X commands)  Most people react to these the way you
did, because most people are not told the derivation of the names.  But
every UN*X command that I've used has a perfectly reasonable derivation
for the name.  CAT is short for "concatenate" and it was called that for
the simple reason that it was written for the purpose of combining inputs;
when there is only one input this fact is obscured.   GREP comes from
"Global Regular Expression Parser", which I won't explain here except to
say that it makes perfect sense if you use VI or similar editors at all.
Similarly for other commands.  Absolutely none of the commands were given
names that were intended to be cryptic; I grant you, they may not be very
intuitive to someone without any experience using the system, but then I
don't think they were intended to make sense to non-computer-science folks.
That UN*X has survived so long and is used so widely must be a great surprise
to the original authors of these commands.

6.  (bit numbering)  The "reversed" bit numbering that you mention also
appears in certain hardware devices, notably certain audio and video DACs.
As recently as 1992 I diagnosed a design error on a friend's project as a
reversal of the bit order due to this problem.  One of the vendors (I think
it was Burr-Brown but am not sure) a long time ago told me that the reversed
bit orders and numbering were from British designers; but a British friend
denied that they prefer different bit numbering.  This is just one of those
cases where both possible approaches have been used in the past, and still
survive to this day, precisely because there is no really overwhelming reason
for one way or the other when you're starting out fresh.  Now that most of
the world numbers with LSB as '0' there is a good reason to adhere to that,
of course.

7.  (token-ring and ethernet addresses orders)  Nobody "decided" that these
should be different; it was simply another case of different designers and
no really overriding reason to pick one over the other.  Yes, yes, I know
*all* of the religious-debate reasons for both sides of this one (I've made
those arguments myself), but the fact of the matter is that either byte order
works just fine.  Of course, it's a pain in the posterior when you have to
write code that deals with both interfaces in the same system (as I have).

9.  (file systems)  This is pretty obviously based on a misconception.  In
particular, the slightly bizarre idea that lines of text are analogous to
"records" and thus that carriage return and/or line-feed are somehow "wasted"
characters because of that.  Of course, text is just text, it's not "records",
and the line boundary characters are in no sense "wasted".  Anyway, the file
systems that stored data as "records" wasted a lot of space as well as a lot
of machine cycles dealing with that structure; this was hidden from not only
the users but from most programmers by the design of the hardware.  It was a
very good thing to leave behind as we moved into micros and had to really be
efficient with bits and cycles.  Just think of file systems as byte streams
and forget about records.

As for using NULL as a string-terminator character, well, if you're going to
use *any* byte as a signal that the text string has ended (contrary to the
old Pascal practice of giving a length byte, which has its own problems)
then what better value than the "nothing here!" value of NULL?  If you're
looking at binary data, you need tools to deal with that and then the
strings-end-with-null issue won't matter to you.  (If you're trying to use C
string-handling routines to deal with binary data, well, STOP THAT!!  Wrong
tools!  You're using a set of screwdrivers on hex drive fasteners!)

--
Bob Bickford     rab@well.com
Former Web-master, The Atlantis Project (Oceania) Web Site
http://www.oceania.org/
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:36:03 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: John R Levine 
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine Questions on Machine History.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1587717481162.LTK.013@cpsr.org>
X-UIDL: 839869011.004


Sender: John R Levine 
Subject: Re: CM> Nine Questions on Machine History.

A few comments.

> 4.  I personally think TSO is about the most user-hostile environment/op-sys
> on earth.  I first used IBM's CMS version 2 as an op-sys about 1982.  I still
> use it (even as I type this note).  Yes it's old and character only but it
> is SO much easier to use than TSO.  What's the history of CMS?  Why did it
> never replace TSO?  Unix?

As I understand it, CMS was originally a single-user debugging written for
a 360/40 at an IBM Lab in Cambridge, Mass.  (The C was for Cambridge.)
CP/67 was originally a separate project at Yorktown experimenting with
virtual machines.  Programming a virtual machine from the virtual switches
wasn't much easier than programming a real machine from the real switches,
so they heard about CMS and loaded it up on CP.  CP was much faster than
the IBM competition (TSS, mostly) because CP was a small project run by a
small, coherent, group of people at Yorktown, while TSS was the project
that conclusively proved that adding extra people to a project that's late
makes it later.  TSS, which I used a fair amount around 1970, also showed
why you need iterative design -- some parts of the TSS design were quite
nice, others were hopeless, so in practice the parts of the system that
the developers used worked, and the other parts didn't.  In later years,
CP evolved into VM/370 with a large support staff and has started to
suffer some heat death problems, while TSS was cut back to a tiny
dedicated skeleton support staff who apparently fixed much of what was
wrong with TSS, too late to do much good.

TSO was and is a bolt-on to the batch OS/360 and its successors.
Unpleasant though it is, you're stuck with it if you need to use MVS
interactively, and MVS is where most of the major IBM applications live.

> 8.  Why are people trying to shoehorn FORTRAN into an array handling language?
> Why aren't they using APL/2 instead?

Partly because there's a zillion lines of well-debugged numerical
libraries in Fortran already, partly because people believe (quite
possibly correctly) that large Fortran programs run faster than APL.

> 9.  And finally, for now, who designed and why do we use the file
> system used by DOS, MAC, Unix and I don't know what else?
> Specifically, these file systems treat all text files as one long
> string with records indicated by wasting the two characters 0D and 0A.
> Mainframes and even my dear departed Datapoint 1134 used a full blown
> file system where records are stored as records.

In the CACM paper on Unix, Thompson and Ritchie commented that on the
mainframe systems with which they were familiar, there was more space
taken up by traling blanks in card image files than in the complete
directory and inode overhead in Unix.  They made the deliberate decision
to move the record management stuff out of the file system into
application libraries, so the kernel just handles byte streams.  The
CR/LF line terminator used in DOS comes from CP/M which cosmetically
resembles TOPS-10.  The Model 33 and 35 Teletypes commonly used as
terminals on PDP-10s used CR/LF to start a new line, so the file format
made it possible to dump the file on the terminal and have it look nice.
At Bell Labs they had newfangled Model 37 Teletypes which treated the LF
character as a newline, so that's what you got in Unix.  Why the Mac uses
just CR, I have no idea.

> Same question really applies to C and DOS ASCII-Z
> strings.  Who decided 00h wasn't a datum?

I first met zero-terminated ASCII strings on the PDP-6.  The instruction
set made it easy to sequence through a byte string and test for zero.  It
was considerably harder to move a byte pointer ahead N bytes.  (Indeed,
that's one of the few new instructions added in later PDP-10 models.) On
the PDP-11, same answer, easier to test for zero than to count bytes,
since a single MOVB (r1)+,r2 instruction would load the byte pointed to by
r1 into r2, increment r1 to point to the next byte, and set the condition
codes so you could immediately do a conditional branch on zero.

On 360s there are instructions to move N bytes, so it's not surprising
that their file system uses fixed length and variable-with-byte-count
records.

Regards,
John Levine, johnl@iecc.com, Trumansburg NY
Primary Perpetrator of "The Internet for Dummies"
and Information Superhighwayman wanna-be
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date: Mon, 12 Aug 1996 03:40:54 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: peter@baileynm.com (Peter da Silva)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> Nine Questions on Machine History, & pointers to docs.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <1913880782453.LTK.013@cpsr.org>
X-UIDL: 839859838.047


Sender: peter@baileynm.com (Peter da Silva)
Subject: Re: CM> Nine Questions on Machine History.

> 5.  Yes, I use UNIX.  Yes, I like it--yes, I hate it.  Where did the
> let's-make-it-so-cryptic-no-one-can-intuitively-figure-it-out mentality
> and naming conventions come from?  TYPE ... | FiND .... sure seems
> a lot more obvious to me than CAT ... | GREP ...

The problem is that all these names are arbitrary. Even the ones you think
are intuitive. For example, UNIX has a "find" command that does something
different from DOS find. Type can mean "show me the type of this file" as
well as "display this file". In fact, wouldn't "display" be more obvious
since so far as I know nobody *ever* hooked a teletypewriter up to DOS.

> 9.  And finally, for now, who designed and why do we use the file system used
> by DOS, MAC, Unix and I don't know what else?  Specifically, these file
>systems
> treat all text files as one long string with records indicated by wasting
> the two characters 0D and 0A.

Or the one character 0A.

The reason UNIX text files are the way they are is a reaction to all the
typed (there's that word again) files in mainframe operating systems.

For both these questions, and many others, you can feed your Intuition gland
by reading Kernighan and *:

        Kernighan and Ritchie: The C Programming Language.
        Kernighan and Pike: The UNIX Programming Environment.
        Kernighan and Plauger: Software Tools (the original, not 'in Pascal').

Also, the early UNIX papers printed in the Bell System Technical Journal in
the mid-70s. I want to say July/August '76. Check your local (paper) library.

It would be maximally cool if whoever had the copyright on these papers (Lucent
would that be, now?) would publish them electronically.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@cpsr.org that reads:
                       SET CPSR-HISTORY MAIL DIGEST
______________________________________________________________________
Date:         Tue, 13 Aug 1996 20:00:11 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> We moved.  And this is the formal WELCOME.
To: Multiple recipients of list CYHIST 
X-UIDL: 839981274.000

We have changed listserver addresses for Community Memory.  If you get this
message, you have been successfully placed on the new listserver.

Odds are that you will receive this message before the formal FAREWELL
message from Community Memory at cpsr-history@cpsr.org.

As I mentioned, we decided to move the list a new listserver to better
handle the volume and flow of posts.  If you are reading this message *then
you have been transfered to the new address*.  What follows is the formal
WELCOME message you would have seen had you subscribed manually to
Community Memory at its new address, which is cyhist@sjuvm.stjohns.edu.
The name change was required because St. John's Listserv program requires
that names be no more than six characters long.

CPSR is very grateful that St. John's has made their listserver available
for this project.

Here is the official WELCOME message.  Please save it for future reference.

Thanks,
db

-------
Your subscription to CYHIST, a.k.a. "Community Memory: Discussion List on
the History of Cyberspace" succeeded.  Thanks for subscribing. When
posting, please keep in mind the following guidelines:

        o focus your post on a personal story; if you speculate, ask
questions, or simply confirm someone else's post without contributing
something of your own experience to the list, your post is less likely to
be forwarded.
        o Do not offer to sell products in your post, or promote any
commercial activity.
        o Be courteous to your fellow list-members.

Think of this list as a collective conversation, with a mission:
distinguishing the history of cyberspace so that we can all be
well-informed concerning the impact digital technology is having on our
lives, institutions and cultures.  This is why *specific* facts and
memories should appear in your posts.

This list is moderated to ensure focus and topicality by David S. Bennahum
(davidsol@panix.com).

Traffic on this list may be high, sometimes reaching 12 or more messages a
day.  You can receive this list in DIGEST form, meaning all the day's posts
are bundled into one message and set to you with an index.  To recieve the
list in digest form, send a note to listserv@sjuvm.stjohns.edu reading

        SET CYHIST DIGEST

This list has a web site where posts will be archived and made searchable
to the entire Internet community.
http://www.reach.com/matrix/community-memory.html

If you wish to signoff the list, send a note to listserv@sjuvm.stjohns.edu
reading

        UNSUBSCRIBE CYHIST

Please *save this note* so that you will remember all this.

Thanks for subscribing; your participation makes a difference.
Date: Wed, 14 Aug 1996 01:06:46 -0700
Reply-To: cpsr-history@Sunnyside.COM
Originator: cpsr-history@cpsr.org
Sender: listserv-reply-errors@Sunnyside.COM
Precedence: bulk
From: davidsol@panix.com (David S. Bennahum)
To: "Multiple recipients of list cpsr-history@cpsr.org" 
Subject: CM> We have switched listservers. Formal FAREWELL.
X-Listprocessor-Version: 9.1 -- List Server by Sunnyside Computing, Inc.
X-Comment:  Discussion of history of computing
X-Info:  For listserv info write to listserv@cpsr.org with message HELP
X-Message-Id: <19233586516.LTK.013@cpsr.org>
X-UIDL: 840029587.032

This is the last message you will recieve from cpsr-history@cpsr.org.

The list has been transfered to a new location:

        cyhist@sjuvm.stjohns.edu

Messages should appear from CYHIST in the next 24 hours, including a formal
Welcome Message which will contain descriptions of how to sign off from
CYHIST, get a digest version, and so on.  It is even possible that you will
get a message from CYHIST before seeing this one.

If you do not receive a message from CYHIST in the next 24 hours, do not
despair.  Send a note to listserv@sjuvm.stjohns.edu reading:

        SUBSCRIBE CYHIST yourfirstname yourlastname

There is a small chance that the most recent subscribers may need to do
this (last 3 days).

[Do not send subscribe requests to cyhist@sjuvm.stjohns.edu.  That address
is only for *posting*.]


The old listserver will be unplugged soon; in the meantime please do not
post to it.  Post to cyhist@sjuvm.stjohns.edu starting now.

If you have any questions, send them to me at davidsol@panix.com

Best,
db
Date:         Wed, 14 Aug 1996 19:07:32 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Sun Workstations
To: Multiple recipients of list CYHIST 
X-UIDL: 840064784.000

>X-Sender: davidsol@panix.com
>Mime-Version: 1.0
>X-URL: http://www.reach.com/matrix/
>X-INFO: To subscribe to MEME, send a message to listserv@sjuvm.stjohns.edu that
> reads subscribe meme  
>X-INFO: To subscribe to Community Memory, send a message to listserv@cpsr.org
> that reads subscribe cpsr-history  
>Date: Wed, 14 Aug 1996 10:14:27 -0500
>To: cyhist@SJUVM.STJOHNS.EDU
>From: Les Earnest 
>Subject: CM> Sun Workstations
>X-UIDL: 840032786.000
>
>This message was  originally submitted by les@STEAM.STANFORD.EDU  to the CYHIST
>list at SJUVM.STJOHNS.EDU. If  you simply forward it back to  the list, using a
>mail command  that generates "Resent-" fields  (ask your local user  support or
>consult  the documentation  of  your mail  program  if in  doubt),  it will  be
>distributed  and  the  explanations  you   are  now  reading  will  be  removed
>automatically. If on the other hand you edit the contributions you receive into
>a digest, you will have to  remove this paragraph manually. Finally, you should
>be able  to contact  the author  of this  message by  using the  normal "reply"
>function of your mail program.
>
>----------------- Message requiring your approval (66 lines) ------------------
>
>Sender: Les Earnest 
>Subject: CM> Sun Workstations
>
>William F. Anderson writes:
>   [.  .  .]   Although I
>   don't know the full details, someone on the list should know about the
>   first Sun Workstation development woes. I remember that they had a problem
>   with not having enough memory to load the kernel. They managed to shrink
>   the kernel enought that it would load, but left little memory for running
>   any applications. I heard this story from an engineer at Sun, so it needs
>   to be verified.
>
>The first SUN Workstation was developed in the Stanford Artificial
>Intelligence Lab in the 1980-81 period, principally by grad student
>Andreas Bechtolsheim, with kibbutzing by several others including me.
>As you may be aware, "SUN" originally stood for "Stanford University
>Network."  That SUN ran vanilla BSD Unix, which easily fitted into the
>memory.  We were debating whether to use 16k memory chips, which were
>standard at the time, or the new 64k chips, whose availability was
>less certain.  64k chips were chosen and turned out to be available at
>reasonable prices quite soon.  That gave us plenty of room.
>
>There may have been a memory crisis after Sun Microsystems was formed
>a couple of years later to develop and market commercial workstations
>and they began elaborating the operating system, but I was not aware
>of it.  I do recall, however, that Sun had some serious power supply
>and software problems in their early workstations.
>
>I had founded Imagen Corporation in the meantime to manufacture laser
>printing systems.  We were using the SUN computer card as a raster
>image generator in our system, with no problems.  However, in 1983
>when Sun Microsystems moved their dozen or so employees into a new
>facility just 50 feet from ours, we worked out a barter arrangement
>under which we swapped our laser printers for their workstations.
>Unfortunately the workstations turned out to be unreliable and we
>couldn't get the Sun guys to walk across the parking lot to fix them
>-- they were too busy creating the next version.
>
>When we started negotiating to get some diskless Sun-2 workstations,
>we were puzzled by Sun's recommendation that we use our Sun-1's
>as file servers instead of buying new ones.  When I pressed on this
>issue they finally admitted that their new file servers didn't work!
>We ended up buying a bunch of workstations from Apollo.
>
>In talking with other early Sun workstation users, I found that many
>of them had major problems, but they really liked the idea of an open
>architecture and found ways to cope with the problems.  The curious
>thing was that no one complained publicly.  This enabled Sun to keep
>selling their marginal products while they got their act together,
>eventually becoming a runaway success.  If anyone had complained
>publicly about their early products it might have ended badly.
>
>        -Les Earnest
>______________________________________________________________________
>            Posted by David S. Bennahum (davidsol@panix.com)
>                    Moderator: Community Memory
>            http://www.reach.com/matrix/community-memory.html
>         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
> Materials may be reposted in their *entirety* for non-commercial use.
> Want to receive the day's postings bundled together into one message?
> It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
>                       SET CYHIST DIGEST
>______________________________________________________________________
>
Date:         Wed, 14 Aug 1996 19:24:59 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> H2000e anyone?
To: Multiple recipients of list CYHIST 
X-UIDL: 840066072.000


Sender: Michael Froomkin 
Subject: H2000e anyone?

If memory serves, the computer I learned to program on in highschool back
in 1977/78 was an H2000e. I think the H stood for "Honeywell".  I think
the "e" meant it had extra memory, maybe as much as 32K!  I'd love to look
at the technical specs if anyone knows where information on this can be
found.

[begin nostalgia]

My school didn't actually own the machine - we dialed up to another
school, and timeshared.  Access was through an old, old, old teletype
machine [an RCA 33?  I'm not sure] with rolls of yellow paper and input
via paper tape (great confetti!).  [When I graduated to punch cards on my
first job they seemed a lot more sensible since you could change one card
without redoing the tape.] No one on the faculty knew how to program much,
so we went downtown and bought self-programmed texts on BASIC.  A simple
tic tac toe program made a big tape. The monopoly program we wrote was
several rolls -- and the prgram was too big to fit in memory so it had to
have many modules that were chained..  I think it spent more time passing
data from module to module to module than actually doing anything.
Eventually this was judged too slow and we switched to a commercial
service (tymshare? I forget). This was more fun as our account did not
have rights to all the services. In principle, any way.

Now my old school has a flashy website and an entire room full of
terminals that used to be the chemistry lab.

I am too young to be a geezer.  Right?

[This message may have been dictated with Dragon Dictate 2.01.
Please be alert for unintentional word substitutions.]

A. Michael Froomkin        | +1 (305) 284-4285; +1 (305) 284-6506 (fax)
Associate Professor of Law |
U. Miami School of Law     | froomkin@law.miami.edu
P.O. Box 248087            | http://www.law.miami.edu/~froomkin
Coral Gables, FL 33124 USA | It's hot here!  And humid!
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Wed, 14 Aug 1996 23:04:34 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Toy computers, answer: CARDIAC.
To: Multiple recipients of list CYHIST 
X-UIDL: 840118559.003

Sender: "Michael R. Williams" 
Subject: Re: CM> Toy computers -- cardboard

At 03:09 AM 8/12/96 -0700, you wrote:
>
>Sender: Michael Froomkin 
>Subject: Toy computers -- cardboard
>
>In grade school, in about 1973 give or take a year or two, we were treated
>to an educational experiment.  It was a little cardboard device, not much
>smaller than a sheet of paper, that simulated a computer.

[~snip~]

>I can't remember what the thing was called, but I remember enjoying it; I
>think it taught me a great deal too.  I suppose the experiment was not
>considered a success, since I've never seen one since.
>
>I'd love to know what it was called, and how I would get one for nostalgia
>value ...and possible deployment on my children...

The device was developed by Bell Telephone Laboratories and was distributed
by the Bell system as an educational aid.  It was called CARDIAC (a
CARDboard Illustrative Aid to Computation).  It had a number of movable
strips (4 actually) that could be used to simulate a simple CPU (one to set
the result of the "Test Accumulator Contents - TAC" instruction, one to set
the op-Code, and two for the addresses of the operands).  The instruction
book contained 16 sections describing everything from what the CARDIAC was
(and was not) to how to develop subroutines for double precision floating
point arithmetic.
  You are correct that it did have a "ladybug" which was moved from memory
location to location to act as the program counter.

Yes, I still have one - No you can't have it.  It is fun!

Mike Williams

---------------------------------------------------
Dr. Michael R. Williams
Editor-in-Chief, Annals of the History of Computing
Department of Computer Science
University of Calgary
Calgary, Alberta
Canada      T2N 1N4

Ph:  (403) 220-6781
Fax: (403) 284-4707
email: williams@cpsc.ucalgary.ca
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Wed, 14 Aug 1996 23:20:35 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Messages delayed.
To: Multiple recipients of list CYHIST 
X-UIDL: 840118559.005

I am having trouble configuring the listserver so that the From: field on
messages reflects the actual address of the post's author, rather than
mine.

If I do not resolve this by tomorrow (Thursday), I will send them out
anyway with the From: header containing my address.  Eventually this silly
problem will be solved.

I'm sorry for this minor hiccup in our transition, which otherwise seems to
have gone off allright.

Feel free to post new messages in the meantime, they are being successfully
forwarded to me.

Best,
db
Date:         Thu, 15 Aug 1996 11:08:09 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> How to set digest mode.
To: Multiple recipients of list CYHIST 
X-UIDL: 840121768.000

Those of you who were getting Community Memory in digest mode must request
this feature again.  I was unable to transfer that setting to the new
listserver.  Getting digest mode is simple.  Send a command to:

        listserv@sjuvm.stjohns.edu

Reading:

        SET CYHIST DIGEST

...You should get confirmation of this in the mail.

best,
db
Date:         Thu, 15 Aug 1996 11:10:31 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Kemeny, Basic, and Timesharing
To: Multiple recipients of list CYHIST 
X-UIDL: 840121889.000

Sender: John R Levine 
Subject: Kemeny, Basic, and Timesharing

> Also, though Kemeny did do some pioneering work in developing a
> multi-user BASIC system, it was not a general purpose timesharing system.

Kemeny and Kurtz' work eveolved into DTSS which by 1975 was an entirely
general purpose time-sharing system.  You could write your programs in
Basic (which was compiled into machine code before running), Fortran,
Cobol, Trac, XPL, PL/I, and a couple of other languages.  Internally they
had a very sophisticated system with multiple processes and fully
asynchronous system calls.  Terminals ran through a Datanet 30 front end
which scanned the TTY lines one bit at a time and presented lines of text
to the terminal interface program on the GE 635 mainframe.  The terminal
monitor then created a "com file", similar to a Unix socket except that it
had a master and slave end, and the master end could respond to all of the
possible slave end system calls to create virtual devices, for each
terminal.

When you logged in, your terminal's com file was passed to SIMON, the
simple monitor which read the familiar line numbered input that most
Basic environments recreated.  When you typed RUN, it would sort and
merge the lines you'd typed into your source file, then create a new
process with the appropriate compiler and pass that file and your
terminal com file to it.  The compiler would create object code and run
it.  You could save object code if you wanted, but the compilers were all
so fast that nobody bothered to do so.  (At least not until the ultimate
pig, PL/I, came along.)  DTSS had other goodies like background batch
facilities that emulated GECOS, the vendor's operating system.

This system easily ran 100 users with good response on a computer about
the performance of a KA10 PDP-10.  All the development and maintainance
was done by undergraduates who were unaware that they weren't qualified to
do that sort of work, so they just did it.

Returning to Kemeny, I heard some of his lectures and he did indeed forsee
wide access to central time-sharing.  Like many of us into the mid-70s, he
failed to forsee that microprocessors would make centralized computing
utilities irrelevant, and make the remaining such utilities metamorphose
into (in effect) database servers.

Regards,
John Levine, johnl@iecc.com, Trumansburg NY
Primary Perpetrator of "The Internet for Dummies"
and Information Superhighwayman wanna-be
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:14:31 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Spacewar and other interactive games
To: Multiple recipients of list CYHIST 
X-UIDL: 840122199.000

Sender: Les Earnest 
Subject: Spacewar and other interactive games

In a discussion of bit-mapped displays at UCLA in the late 1960s, John
Oliver writes:
   I must confess that one of the most interesting uses of these
   terminals involved "Space War" ...

Computer games have been providing entertainment for users and
prestige or financial rewards for some of their creators for about 50
years now, but their development has also induced substantial advances
in technology that are not widely appreciated.  For example, as John
McCarthy is fond of saying, "Computer chess is the drosophila of
artificial intelligence," in that it has facilitated the testing of
analytical processes in a simpilified environment, much like the
breeding of fruit flies by geneticists.

While AI researchers are often serious about computer games, most
early game developers did it just for fun. In the summer of 1965 when
I visited Prof. Jack Dennis at MIT, who was my former thesis
supervisor, the first thing he showed me was Spacewar running on their
new PDP-1.  He mentioned that this elegant game had been invented by a
group of students led by Steve Russell.  When I went to Stanford a few
months later to become executive officer of SAIL I found that Steve
Russell had moved there as a staff "system wizard."

When our PDP-6 computer arrived in June 1966 it came with a number of
Model 33 Teletypes.  We had wanted to use displays instead but
encountered substantial delays in getting them.  You couldn't buy
displays "off the shelf" in those days -- you had to draw up a
specification and get someone to design a system that would meet it.

We tried to interest Philco in bidding, given that they had built some
displays in 1964 that worked pretty well in the Zeus timesharing
system at Stanford, but Philco was no longer convinced that displays
would "catch on" as part of computer systems.  After another
manufacturer contracted to build the displays, then defaulted, we
finally got Information International Inc. (III) to build some.

The first thing that Steve Russell and a group of students did after
the displays were delivered was to bring up a version of Spacewar and
build a set of four-button boxes that were used to control the
spaceships.  However, we discovered that Spacewar didn't run well in a
heavily loaded timesharing environment -- the spaceships would move,
rotate and shoot in an erratic manner.

In order to get around this problem we added a real time feature to
the operating system, called "Spacewar Mode."  Whereas conventional
timesharing gives sequential time-slices to all active processes,
which means that the intervals at which a given process runs is rather
variable, a program running in Spacewar Mode was guaranteed service at
whatever regular interval it specified.  Needless to say, abusive use
of Spacewar mode could bring the system to its knees, but that was
never a problem --if the system seemed to be behaving badly we could
run a display program that identified all prosesses that were running
in Spacewar Mode and could swiftly inhibit any misconduct.

Spacewar Mode turned out to be a very useful feature for a number of
purposes besides playing the game.  For example, we could do
peripheral hardware debugging while providing regular timesharing
service by having a program in Spacewar Mode poking at regular
intervals at the device being debugged, so that the resulting signals
could be displayed on an oscilloscope and analyzed.

Another application of Spacewar Mode was in synthesizing music, which
helped our computer music group, led by John Chowning, to become the
world's leading developers of music synthesizer technology.  They made
quite a bundle for Stanford and for themselves when they later
patented this technology and licensed it to Yamaha.

Throughout the 1960s McCarthy periodically worked on the chess program
that he had begun at MIT.  Around 1966 he challenged a group at a
physics research center in Moscow, who had also developed such a
program, to several matches in which moves were to be exchanged by
telegram, with each message consisting of four decimal digits
representing moves in international chess notation.  The Russians
clobbered us, but they also got into trouble with the KGB for
exchanging coded messages with foreign agents.

Our Russian correspondents eventually got off the hook by proving to
the KGB investigators that all of the telegraph messages represented
legal chess moves.  If the KGB had been more knowledgeable about
information theory, of course, they would have realized that it is
possible to exchange a lot of secret information by encoding it as
legal chess moves!

It should be noted that Russian authorities do not have a monopoly on
this kind of ignorance -- in the U.S. the Reagan, Bush and Clinton
administrations have consistently pursued restrictive encryption
policies that are based on the absurd premise that they can recognize
secret communications when they see them, provided that they can get
access to the "official" cryptographic keys through an escrow scheme.
Little do they understand!

We have already discussed Art Samuel's Checkers program a bit.
One of his students at SAIL wrote a Go program that played a decent
game, communicating with the user in terms of board coordinates.
Another student then did a nice display hack that showed the Go board
on our bit-mapped displays and communicated with the Go program
through a "pseudo-teletype," which was an early form of inter-process
communication on the DEC-10.

Some of the projects at SAIL that were aimed at answering serious
questions produced programs that were fairly entertaining and were
treated as games.  One such was PARRY, a conversational model of a
paranoid patient that was developed by Ken Colby's group and that
passed a kind of Turing test with psychiatrists.  PARRY was
parameterized so that one could vary its degree of paranoia and
agitation.  A psychiatrist friend of mine could be entertained for
hours at a time by PARRY, though she also sometimes got rather upset.

A grad student named Don Woods developed a number of games, the most
popular of which turned out to be Adventure.  It has many modern
descendants in the form of MUDs and similar interactive programs.
As far as I know Don never got a nickle out of any of that, though he
eventually took enough time off from game development to get his PhD.

The game of Spacewar enjoyed sustained popularity throughout the 25
year life of the SAIL computer, particularly late at night and on
weekends.  Various features were added, such as the dangerous
"hyperspace" maneuver, which in terms of modern Star Trek terminology
would be called "cloaking."  After connecting to ARPAnet around 1970,
SAIL became a distribution center for free programs of all kinds for
DEC-10 and DEC-20 systems, including many games.

At some point we held the "First Intergallactic Spacewar Contest" at
the invitation of Stewart Brand, who was writing an article about
interactive computing for Rolling Stone magazine.  Many participants
ended up with numbed fingers that night as a result of frantic
attempts to push buttons ever faster.  Bruce Baumgart came in first,
winning a year's subscription to Rolling Stone.

Also in the early 1970s a couple of our grad students, Bill Pitts and
Phil Petit, decided to explore the commercial possibilities of
Spacewar.  They bought two PDP-11's and some coin acceptors, built
interfaces and button boxes and reprogrammed Spacewar in PDP-11
assembly language.  They put one system in the Stanford Coffee House
and another in an off-campus bolling alley, allowing up to five
players to engage in a Spacewar match for 25 cents each.  It proved to
be quite popular, but not financially rewarding for Pitts and Petit
because of the cost of amortizing the computer system.

Meanwhile, an entrepreneur named Nolan Bushnell, who had been exposed
to Spacewarin college, began pursuing the same idea in a slightly
different way by forming a small company called Atari to develop
special purpose hardware for games that could manufactured
inexpensively.  The first one he developed was Spacewar, but he
decided that it was too complicated to be understood by the general
public, so he sold that design to another company and developed a much
simpler game called Pong.  Pong turned out to be wildly popular,
making him a multi-millionaire.

Of course, Atari took a dive later on, sucking down a lot of
Time-Warner's money with it, and Nintendo eventually ate their lunch.

In summary, computer games have enriched many people's lives, both
psychically and financially, and have contributed much to the field of
artificial intelligence and other computer-related technologies.  Of
course, there is also a dark side.  Many people are addicted to
computer games and spend a great deal of time on them that might be
used more productively.  (Mea culpa!)  On the other, as a hedonist
friend assures me, time spent in pursuit of pleasure is not wasted!

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:14:43 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Punchcards as Christmas cards.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122199.001

Sender: Shelly Julien 
Subject: RE: CM> Thundering keypunches, punch cards roadkill.

Nelson Winkless  writes:

Does anybody still make Christmas wreaths out of cleverly curled and stapled
IBM cards?

Not anymore, but we used to make those and little stars you could hang
for on the tree.  Now we just paint old Windows 95 beta CD-ROMs.  Muy most
vivid memory from college in the late 70s was sitting on the floor in the
Computer Center waiting for my turn at the keypunch to retype the cards in
my deck that had typos, missed periods, etc.  My astonishment when I got to
my first "real" job and sat down at a timesharing terminal -- "what, I don't
have to punch cards anymore?" -- was nearly boundless.  If someone had told
me during those keypunch days that one day I'd spend this much time
*willingly* typing at a computer terminal I would have sent them to the
campus psychiatrist!
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:14:49 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Datapoint CTO responds.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.001

Sender: "Wesley J. Miller ((8-444) 919-254-9774)" 
Subject: CPSR-HISTORY digest 56

Ref:  Your note of Mon, 12 Aug 1996 03:46:44 -0700

Regarding the nine questions I posted last week, I received the following
correspondence from Vic Poor who graciously granted me permission to
add these remarks to CPSR-History.

Many, many thanks to Vic.

Wesley

 ================== Referenced Note Follows ==================================



A friend of my who has been following the cpsr-history list server sent me
the following quote he extracted from a message of yours and suggested that
I send you a comment on it:

" 1. The first "personal computer" I used was a Datapoint 1134.  I know the
last digit represented the number of floppy drives installed.  I also
remember that the machine had 16K of memory on four 4K memory cards that
plugged into the processor cards on which I was told there was an 8080 --
not
the chip but a hardwired-from-discretes equivalent which Datapoint used
because Intel couldn't get the 8080 out in time.  Questions:  Was it in
fact
an "8080"?  Was it Datapoint that drove Intel to develop the 8080? "

I was chief technical officer for Datapoint from 1969 until I retired in
1984 and can probably shed light on some of the happening in those days but
I must admit my memory is filling with more interesting things.

The original Datapoint computer architecture was what became the Intel
8008. The instruction set, register layout, and all that originated with
Harry Pyle and myself in 1969. Our original idea was a pure bit-serial
one-bit-wide bus that would work effectively with the then available MOS
shift registers as main memory and could be fitted onto a single 16-pin(!)
chip.  (Semiconductor RAMs were not yet known - everything else was using
core memories in those days.) The appeal of the MOS shift registers was
that a whole processor, CRT, main memory, and tape decks could be fitted
onto the footprint of a Selectric typewriter.  Datapoint contracted with
Intel to make the chip.

Why the architecture was what it was is an interesting story in itself but
enough to say that if Harry and I had any idea of what it would ultimately
be used for it would have been very different!

Intel improved on the design and produced the same instruction set and
register layout using an 8-bit-wide bus and if my memory serves it ended up
on an 18-pin chip.  The only problem was that it was late in development
and Datapoint had already finished its product (the 2200) using MSI TTL for
the processor design although using the same architecture.  Intel wanted to
be paid for the work so there was a meeting between Bob Noyce of Intel and
Phil Ray the CEO of Datapoint.  I was not at the meeting so I don't know
what all went on (it was friendly) but the end result was that Datapoint
gave Intel the design and Intel put it in their catalog as the 8008. The
rest is history.

The 8080 was an upgrade to the 8008 that came after the rather striking and
unexpected acceptance of the 8008  "computer on a chip".

There is still more to the story.  Datapoint ordered the development of the
chip from Intel in December 1969.  Around April of 1970 Datapoint also
ordered the same design from Texas Instruments as a second source. T.I.
actually delivered their design before Intel but the noise margin on the
T.I. part was so poor that it could not be used commercially.  T.I. was
really gung ho to produce a chip computer but Intel was not too excited
about the idea. They wanted memory business where there was REAL volume,
not processors where there was only one chip in each box.  I still remember
the conversations with Gordon Moore and Bob Noyce on that score. In the end
I think they only took on the computer chip to keep Datapoint's memory
business.  Oh my! If any of us had only known.

The 1134 that you used was a later generation machine that had a somewhat
richer instruction set than the 8080 but still used MSI TTL for the
processor simply because in those days you could produce a TTL design and
get it to market faster at a reasonable cost (that is, what was reasonable
at that time) than you could get comparable chip processor.  The first chip
processor that Datapoint used in any machine was the Z-80 but that was
years later.  And of course by the time the 1100 series were designed MOS
RAMs were available and that made a world of difference too.

I hope this helps answer your questions.  I will try to get hooked up to
the cpsr-history list server and follow the dialog.

Regards, Vic Poor
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:14:58 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Toy computers: Crackerjack calculator.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.000

Sender: "keith reid-green" 
Subject: re: toy computers

The niftiest Crackerjack toy I ever saw (for the youths in the audience,
Crackerjack is a brand name for caramel-coated popcorn and there
was--is?--a toy or game inside, usually plastic and never worth more than
one cent) was a three wheel calculator on the style of the original Pascal
adding machine.  It was a one-bit adder--to add 3 you had to push the lever
three times--but it displayed in decimal and could be reset to 000.  I had
it for years before some low life swine stole it from my office.

Keith Reid-Green
Educational Testing Service
KReid-Green@ets.org
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:04 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Punch cards falling.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.003

Sender: "keith reid-green" 
Subject: re: Keypunching and sequence numbers

I had a thing about sequence numbers in cards and never used them.  I
figure I must have used about a million cards.  That would have been 8
million keystrokes and another 8 million to verify, and it just didn't seem
worth the money to me.  I know it was possible to sequence a deck on a
reproducing punch, but I still couldn't be bothered.

I can't say I never dropped a deck.  I had about a 2,000 card program for a
1401--I know it was 2,000 cards because it just fit in one card box--and
one afternoon in New York City I was rushing to get an up-to-date listing
of the program so I could look at it on the train on the way home.  I put
the cards on top of the 1403 printer to go around the back to get the
fanfold paper folding properly as it came out, and the printer ran out of
paper.  Who remembers what happened on late model 1403's when the paper ran
out?  The lid lifted automatically.  My cards fell on my head, because I
was still stooped over to mess with the paper, and this had the effect of
shuffling the deck pretty thoroughly.  So on the way home that night I
spent about an hour reorganizing the cards by comparing them with the
listing.  And I still think punching sequence numbers was a waste of time.
So was commuting, but I did it for about three years.

Keith Reid-Green
KReid-Green@ets.org
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:11 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Punch cards flying, nicknames for computers, origins?
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.004

Sender: Jeff Buxton/CAM/Lotus 
Subject: Re: CM> Thundering keypunches, punch cards roadkill.

Nelson Winkless writes:

   "...And with respect to spilled cards --...As I reached the office, I found
two people weeping.
One was the programmer who had failed to close the boxes securely with rubber
bands..."

I have my own anecdote to add on the subject of weeping programmers and card
sequence numbers.

While an undergraduate @ Rensselaer Polytechic Institute in 1978 I had the good
fortune to learn FORTRAN
programming using punched cards on Godot, the campus IBM 360 system (as in
Waiting for...) and IBM 026 and
029 cardpunches; incidentally quite a step down from high school, where I had a
pair of Wang 2200 PC's
(BASIC, CRTs, attached printers, cassette tapes, etc.) to play with.

The entrance to the computer room was down a half-flight of very short stairs;
that is, the riser for each
stair was of a non-standard height, and each stair was a different height; in
addition, each stair had a
pronounced "safety" strip (or "lip") on the edge.  All of this made these
stairs tricky to negotiate, especially
if you were toting a box or two of punched cards; I myself had taken many a
spill there.  The next relevant fact
was that the 029s in the computer room were _not_ set to autosequence the cards
when punched.  Put the two of
these together, and you get the following tragic incident:

One fine morning towards the end of the semester I was stumbling out of the
computer room as a red-eyed,
wild-haired, unshaven grad student emerged from what was obviously a grueling
battle with Godot and attempted
to negotiate the stairs laden with no fewer than nine boxes of cards.  I held
the door for him.  Perhaps this
unexpected courtesy confused him, but the next thing I knew we were both
surrounded by a flurry of IBM cards;
the stairs had claimed another victim.  All nine boxes had gone flying, and the
cards were a jumbled mess in
the stairwell.  Worse yet, the student was on the floor crying: "Oh $&#* !
There goes my master's thesis.
I just got it to work, and I don't have a listing of the latest changes."  The
cards had no sequence numbers.

I have no idea who this grad student was, nor what happened to him subsequent
to this disaster. All I have to
offer is this cautionary tale from the dark ages of computing.


------------------------------------------------------------------
Nicknames for machines:

I've noticed individual machines tended to acquire names, like Godot,
especially at university sites.  Does
anyone have any especially apropos ones to submit to the list ?  Godot, for
example, is apropos, so named
because it so slow: it was an IBM 360 attempting to serve an entire campus, and
it ran Michigan Terminal System
(MTS).

A historical view of the names given to machines (or systems) may prove an
interesting study in themselves,
as they may indicate changing attitudes toward technology and/or systems.

As if Godot wasn't bad enough, when RPI purchased an IBM 3033 to replace Godot,
a contest to name the new machine
was held, if I recall correctly the winning name was Merlin (Machine
Electronically Reducing Logic Into Nonsense).


------------------------------------------------------------------
Unusual Locations for machines:

I recall my first day on campus wondering why the unused former chapel building
(really more like a small
cathedral, which had last housed the pathetic library, the state of which
caused RPI to lose its accreditation)
was not the computer room.  Someone must have heard: MERLIN was housed in the
chapel right where the altar used
to be, mounted behind glass on a raised platform.  Either it was someone's idea
of a joke, or the position it
occupied of apparent worship was some sort of Freudian slip.

Again, the location of a machine in what was once a house of worship says
something about the beliefs of the
people who put it there, as may other locations.  I'd be curious if anyone
would care to take this up.


---JJB
-----------------------------------------------------------------------------
Jeff Buxton                    Email: jeff_buxton@crd.lotus.com
Lotus Development Corporation  Voice: 617-693-7632
1 Rogers Street                Fax:   617-693-5522
Cambridge, MA 02142 USA        Website: , but see
http://www.components.lotus.com/
All opinions and errors are solely mine.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:22 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> TV-Edit/Intel; and those wreaths
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.005

Sender: Suzanne Johnson 
Subject: TV-Edit/Intel; and those wreaths

>Regarding TV-Edit, I know I used a program by that name on the Intel
>"MDS-230" microprocessor development system in 1979.  Does anyone know if
>this is the same thing as has been discussed here (well, a port obviously,
>if anything)?  I know I liked it a lot, nearly as much as I liked the
>original CP/M WordStar.

The TV-Edit on the Intel "blue boxes" was not a relative of the timesharing
TVEDIT from IMSSS.  Those were interesting times at Intel.  I was
responsible for engineering computing support at Intel at that time, and
Intel senior management tried to regulate computer usage by insisting that
engineers share time-sharing terminals unless their job required at least 5
or more hours of computer use a day...then they were classified as a
"frequent user" and could have a terminal of their own.  Many engineers
(primarily circuit designers) found that  they could acquire one of the
development systems and use that for a good part of their work (which
required a good deal of text-editing..hence there was internal demand for a
good editor.).   At the time, there really was a feeling on the part of
senior folks that: "we designed chips with a slide rule, why is so much
computer power now required".  Engineers however were quite resourceful in
getting around these early limitations.

>Does anybody still make Christmas wreaths out of cleverly curled and stapled
>IBM cards?
>
I would if I could still find the cards..I recall preferring well punched
cards, the overall effect was more like evergreens.  Great when sprayed
silver or gold.

________________________________________________________________________
Suzanne M. Johnson        johnson@rahul.net        Sunnyvale, CA
 ------still speaking only for myself, no longer an Intel employee------
Date:         Thu, 15 Aug 1996 11:15:28 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Early Computer Memories
To: Multiple recipients of list CYHIST 
X-UIDL: 840138197.024

Sender: "Woody Franke" 
Subject: Early Computer Memories

The discussions on keypunching, card decks, early printers, and drums have
brought back many old memories.  Most are good memories and non-theoretical.
Such as:
I remember attending the IBM pavillion at the 1964-65 New York City World's
Fair and, as a beginner programmer, being amazed by the Russian language
keypunch that produced Russian cards.  This was just after the Bay of Pigs and
about the time of the movie "The Russians are Coming,...".
I remember IBM refering to their punch cards as Hollerith cards.
I seem to remember that IBM had a "Service Bureau" branch that did ADP work
for other companies.  And that the government ruled that IBM had to divest
themselves of that company.  Is that correct?
I remember CDC sueing IBM and a counter suit.  I was working for CDC at the
time on an Air Force contract using IBM equipment.  We couldn't even get any
IBM manuals.  This was 1969-70.
I remember when top dollar for programmers was $12000 a year (1964)and systems
analysts got $15000.  And keypunch operator was a job of the future for
female secretaries.  I worked at a bank that acted as a service bureau doing
payroll, oil dividends, and general ADP work for private companies that
couldn't afford a computer.  We had keypunch operators and keypunch verifiers
who verified the card a second time for errors.
I remember getting a maximum of an hour a day on the computer to test code.
Usually we got time at night.  One course at the University of Houston in
Fortran or Cobol was a Tuesday/Thursday evening course.  U of H didn't have
any computers at the time.  We keypunched our programs and submitted them on
Tuesday.  They were remotely loaded onto Texas A&M's computer miles away.  If
we were lucky, we had a result back on Thursday.  Usually not.
I remember that sorting large databases (several thousand punched card
sequential files were not really called databases where I worked in 1964) took
a minimum of four tape drives (pre disk drives where I worked).  One day
someone from the bank magazine came in to take a photo of our computer system
while we were in the middle of a large, long sort.  The flash from the camera
made all four tape drives rewind, causing an abend.
I remember computer programs (provided by the IBM CEs) that would create
music on the IBM printer.  Anchors Away was one of the favorites.
I also remember a computer program called Edith in which a printout of Edith
would take her from a cocktail dress, to bikini to nothing depending on the
set switches on the 1401 console.  Take about racy!
Enough dribble.  I hope this didn't bore anyone.
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:36 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Bit-mapped displays and mouse droppings
To: Multiple recipients of list CYHIST 
X-UIDL: 840122560.006

Sender: Les Earnest 
Subject: Bit-mapped displays and mouse droppings

John Oliver writes:
   Sometime around 1967 or 1968 the Astronomy department at UCLA purchased some
   abolutely wonderful "TV Typewriters" whose proper name escapes me.  A box
   about the size of a large home mail box contained enough core memory to
   store 16 40 character lines of text. A keyboard and a standard TV set were
   connected to the box.  Everything typed showed up on the screen, and there
   were cursor keys, delete, etc.  [. . .]
   The display was bit mapped and one of the
   most powerful functions was the ability to switch to graphics mode and
   preview the output of a CalComp plot on the screen instead of waiting for
   hard copy.

I'm surprised by this account.  Core memory was still so expensive in
the late '60s that the cost of building bit-mapped displays in this
way was prohibitively expensive if multiple terminals were to be
provided.  That was why the fanciest CRT displays generally used
random-access vector and text drawing through the 1960s.

The first commercially viable bit-mapped display systems that I know
of used the Data Disc video generator system, which first became
available around 1970 and used digital disk memory to synthesize
multiple video channels.  We used one to build a 64 terminal display
system at SAIL in 1971, so that we could put one on each person's desk
instead of forcing them to come to the Display room.  This system
inevitably came to be called "Data Risc" by our irreverent staff.
I designed and built video and audio switches as part of this system,
which allowed users to choose between computer displays, commercial
television or laboratory sources or to interleave them under keyboard
or program control.

Sometime around 1982, when the Data Disc began to get flakey, a grad
student named Ted Selker built a semiconductor memory replacement and
also put television receiver channel selection under computer control.
Ted later went on to fame and fortune at IBM by, among other things,
inventing the red button in the middle of their PC keyboard that
replaces the mouse.

For what it's worth, I'm not much of a mouse fan, though I'm now
forced to use one a lot.  While it is rather easy to learn to use, in
many applications it slows you down a lot by forcing you to move your
hand back and forth between the keyboard and the mouse.  Mine also
picks up grunge and stops working rather often, forcing me to clean
the damn thing.  My dirty keyboard, on the other hand, "Takes a
licking and keeps on ticking."

Other schemes for selecting from a list of elements or for placing a
cursor at a certain point on the screen are often much faster than
using a mouse once your fingers learn the key strokes needed to do
them.  The "Mousecateers" have won for now, but I have faith that they
will be superseded in the long run.

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:45 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> F-108 and bit order.
To: Multiple recipients of list CYHIST 
X-UIDL: 840138197.026

Sender: Les Earnest 
Subject: CM> F-108 and bit order

Lawrence Zeitlin writes:
   [. . .]  The research project I was involved with dealt with the
   control dynamics of the proposed F108 fighter plane which got cancelled
   in favor of guided missles. J.C.R. Licklider was a consultant on that
   project and I got to know him well in the pre-ARPA days.

Actually, the F-108 went into production and thousands were built --
that was the stubby-winged interceptor that crashed a lot.

Inconsistencies in the way bits are numbered, as mentioned here
recently, led to an embarrassing hardware design error in the late
1950s, when my group at MIT Lincoln Lab had the responsibility for
making the F-108 and several other manned interceptors operate in the
SAGE air defense system.  This was accomplished initially by having
Intercept Directors sitting in front of their CRT geographic displays
giving directions to pilots by radio.  However, the Air Force decided
to improve this communication link by building ground-to-air digital
data links that would permit the SAGE computer to send commands
directly to a display in the cockpit showing the desired heading,
speed and altitude.  The same link was to also directly control the
IM-99 (Bomarc) ground-to-air missile.

The data fields needed to convey guidance information were carefully
specified and two different companies were chosen to build the digital
transmitter and receiver.  When the system was first tested, however,
it was discovered that even though both groups had constructed their
equipment to extract the right number of bits from each command, the
"specifications" had neglected to state whether the first bit sent was
the high- or low-order bit.  Sure enough, the two manufacturers made
different assumptions, which eventually caused that system to cost a
quite bit more than was expected.

Another curious incident happened when a manufacturer of flight
simulators (I think it was Link) came to us to find out how to
simulate SAGE guidance commands so that they could build a simulator
for a new McDonnell interceptor that was then under development.  We
told them how we generated the guidance commands, but in the course of
the discussion they mentioned a problem that they had encountered
while testing a prelimianry version of their flight simulater.  It
seemed that whenever the plane got into a certain attitude with the
nose up, the simulation showed that it would rotate further upward,
which would lead to the embarrassing result that the wings would tear
off.

They subsequently tried to find out what was wrong with their
simulation by discussing this problem with McDonnell engineers.
McDonnell ran their own simulations and confirmed that the problem did
not lie in the simulation -- the "pop-up" maneuver would happen in the
real airplane.  They added a warning buzzer to try to keep the pilot
away from the situation in which it would occur.

        -Les Earnest
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:15:54 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> early interactive games: Checkers.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122651.000

Sender: Philippe Chartier 
Subject: CM> early interactive games: Checkers.


Les Earnest writes :

>I believe that Art Samuel began his work on a checkers program at the
>University of Illinois, before he went to IBM.  He reportedly got the
>logic instructions added to the 704 after demonstrating that they
>would be good for the checkers program.

>Art joined us at the Stanford Artificial Intelligence Lab immediately
>after retiring from IBM in 1966 and continued the development of his
>checkers program there through the 1970s, using our DEC-10 system.
>His program came close to defeating the then-reigning world champion,
>who was quite interested in the project, but the champion died before
>the program got to beat him.  Art then decided to retire himself.

Lately I read an article on computer games (or rather on game-playing
computers) in the June edition of Discover Magazine and I was wondering
if the "then-reigning world champion" you mention is the same referred to
in the article, i.e. Marion Tinsley.

Chinook, the checkers program in the article, was designed in 1988 (or
1989, there's a bit of confusion if you read closely : inspired by Deep
Thought, Schaeffer began working on Chinook in 1989, but Tinsley first
played with Chinook in 1988... quite an impressive piece of vaporware
;-), but it also says that Tinsley held the title for more than 40
years...

If we're talking about the same champion, I wonder how many checkers
program projects were "fueled" by  Tinsley and lost their momentum when
he died...

I include an excerpt of the article "Silicon Gambit" which can be found
at :
http://www.dc.enews.com/magazines/discover/magtxt/060196-1.html

Philippe Chartier
philippe@dra.com

 --------------
CHECKERS

After seeing Garry Kasparov beat Deep Blue's immediate predecessor, Deep
Thought,
back in 1989, Jonathan Schaeffer, a computer science professor at the
University of
Alberta in Edmonton, resolved to do for checkers what his colleagues at
IBM were doing
for chess. He set to work on Chinook, a checkers-playing program, and
made it his goal to
take on the world's best checkers player.

At the time, this title had belonged for more than 40 years to Marion
Tinsley, a
mathematics professor at Florida A&M University in Tallahassee. Indeed,
he had beaten
so many of his opponents so easily that most checkers experts considered
him a
one-of-a-kind genius. Tinsley first played Chinook in 1988, and all four
games were drawn.

He won a rematch in 1990, one game to Chinook's none, with 13 games
ending in a draw.
Still, Tinsley was sympathetic to Schaeffer's efforts and expressed
interest in the
technology behind Chinook. Schaeffer went back and reworked the program
entirely, and
in 1992 a rematch was played. Tinsley won, four games to two, with 33
draws. Schaeffer
spent another two years tinkering with Chinook, and in 1994 he brought
the program to
the Computer Museum in Boston to play Tinsley once again. It was to be
the last time.
Tinsley and Chinook played six games, each of them ending in a draw, when
Tinsley
resigned the match for health reasons. The next week he was diagnosed
with cancer. He
died a year later.

Chinook has already beaten the current champion, Ron King, seven times
without
dropping a match, with another nine matches ending in draws. The logical
next step for
Schaeffer is to solve the game once and for all. Chinook already has
about half a trillion
board positions solved, squirreled away in its database. This means that
when Chinook is
playing a game and it sees one of these half trillion boards, it knows it
can play a flawless
game from then on. "We play a lot of games where on move five of the
game, Chinook
gets to one of the positions in its database and announces it's a draw,"
says Schaeffer.
"Then it's a matter of whether or not the human makes a mistake." To
solve the game,
Chinook needs to increase its database of board positions by a factor of
a billion, for a
total of half a billion trillion board positions, which would constitute
all of the possible
positions in the game.

Schaeffer believes that this computational task is by no means out of
reach. It would take
only a year or two on a network of small computers, or perhaps a few days
or weeks on a
supercomputer. "It's hardly worth the trouble when in a few years I'll be
able to buy a
personal computer that can do it," he says. The real problem, though, is
that Chinook's
quasi victory over Tinsley has left Schaeffer unsatisfied. "To be honest,
when Tinsley
died, I lost all motivation," he says. "This guy was just about a perfect
player, and I
wanted to beat him."
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:16:03 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Keypunchers and cards
To: Multiple recipients of list CYHIST 
X-UIDL: 840122651.001

Sender: "Mike O'Brien" 
Subject: Re: CM> Keypunchers and cards

> Those were the 26 series (old, rounded, black) and 29 series (sleek
> modern, beige).  There were a few deviant models, such as verifiers,
> interpreters, etc.

My first "real" computer experience was with the Michigan Terminal System
running on tandem IBM/360 Model 67 processors at the University of Michigan.
Although they pioneered the use of remote terminals all around the campus,
most of the undergraduates (at least) haunted the computer center and did
things the old-fashioned way, with cards.

I got to be pretty good with an 029 keypunch.  There was one lonely 026
around; they told us to stay the $#*$#@$ away from it.  Apparently it
was needed either for back-compatibility for something-or-other or (more
likely) some tenured person insisted it be kept around so he (or she
)didn't have
to learn how to deal with an 029.

The 029 had a space in the top center where you could put a drum, about three
or four inches in diameter and four or five inches high, on a spindle.  You
would first wrap a specially punched card around the drum.  There was one
thing you were NEVER supposed to do, although it was easy to forget and
do it.  If you did it, the little star wheels that read that card (they
looked like little silver spurs about 3/8" across) would get ripped off
their arms and fly around the inside of the keypunch.  I never did the
Bad Thing, but I can now no longer remember what it was.  It might have been
something as simple as unlocking and removing the drum without first
locking the reader arms in the raised position.

The 029 had a "multipunch" capability, which could put any combination of
punches onto a card.  I only ever needed this capability to make up the card
that I put around that drum.  I could never see any rhyme or reason to the
codes that were used, but in those days I was having too much fun to
insist on either rhyme or reason.

The drums themselves, being removable, would have been removed and converted to
beercan holders or some such, so you had to check those out at the operator's
window.  There weren't very many students, even in the nerd squad, who were
willing to go through the pain of learning how to use a drum card, but I did.
Having made that sacrifice, I have never been tempted to go through the
latter-day equivalent of this rite of passage (i.e. I don't use EMACS).

The drum card acted as a map of the 80 columns of the cards you were
punching; I used mine for punching FORTRAN statements.  I set up the drum
card to skip columns 1-6.  If I needed a statement label or a continuation
character I backspaced to put one in.  Columns 7-72 were left alone.
Columns 73-76 were set to "duplicate", and columns 77-80 were set to
"numeric increment".  Thus, the last eight columns were set up to be of the
form PROG0010, PROG0020, PROG0030 and so forth.  We didn't tend to write
programs longer than 1,000 statements, at least not in those days, and the
interval of ten index numbers let us insert bug fixes that could still be
kept in sequence.  One huge bonus was that you couldn't forget and write a
statement out past column 72.  As soon as you hit column 72, BRAAAAAAP! the
sequence number was filled out, the card kicked out, and the next card was
inserted and spaced over to column 6.  Time to backspace and hit a
continuation character.

I remember the verifier.  I hated the damn thing because it looked exactly like
a keypunch.  It WAS a keypunch, just wired differently.  Instead of actually
punching cards, it just read each input card as you typed stuff on the keyboard.
As long as what you typed matched what was on the card, great.  As soon as there
was a mismatch, the whole machine locked up.  I cannot tell you how many times
I walked into the computer center when an assignment was due, saw what I thought
was the only free keypunch, jammed some blank cards in the hopper, skipped
two cards
to get the thing loaded and ready to go, spaced over to column 6, typed a
character and ka-THUNK!  Hey, it's the verifier!  Let's squirt the guy in the
face with water while we're at it.

Mike O'Brien
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:16:12 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Datapoint.
To: Multiple recipients of list CYHIST 
X-UIDL: 840122651.002

Sender: "Walt Crawford"   
Subject: Re: Nine Questions, more on Question 1

Re Nine Questions: some notes on #1 (the Datapoint)...

My recollection of chip history is roughly as already recounted,
except that I thought it was the 8008, not the 8080. In any case,
the single-chip design was too slow for Datapoint's purposes.
They used TTL circuitry instead, at least at first.

So in a bizarre way you can (a) credit Datapoint as one of the
true founders of PCs, and (b) blame Datapoint for the peculiar
native instruction set of 8080s and most later Intel CPUs.

I spent years running Datapoint systems (later ones used Z80s);
the operating system was a superior product, making the hardware
look a lot better than it deserved. For example, indexes were
supported at the OS level to the extent that you could add a new
fully-defined, instantaneous-access index (ISAM) by simply adding
one statement in Databus, an odd combination of high-level and
low-level language. Recompile the program, start running it, and
*existing databases* suddenly had new, complete, indexes!

With a colleague, I wrote a 24-hour control
system for an 8-terminal Datapoint timesharing system (using one
Z80 CPU and 128K RAM, I think); because of the elegance of the OS
and language, it worked beautifully for years.

Datapoint also created ARC, "attached resource computing,"
which turned into ARCNet, and was for a time the dominant
local network technology in some businesses. Unfortunately,
Datapoint didn't originally think of it as a local network:
as the name suggests, they thought of it as a way to incrementally
increase the power of a "computer" by attaching more computers...
also unfortunately, they failed to submit ARC for IEEE
standardization. It was a token-style network.

Datapoint had some glory years and was used behind the scenes
in many large businesses, before fading away [I assume].
You could build serious real-time databases with trivial
hardware, and they worked. For a while, I could almost assume
that if you asked a corporate computer person to look around
long enough, you'd find a Datapoint time-sharing or ARC system
somewhere--probably in accounting or shipping or the like.

walt crawford, br.wcc@rlg.org


______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:16:21 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Keypunchers, cards and printers.
To: Multiple recipients of list CYHIST 
X-UIDL: 840138197.027

Sender: CGeerdts@idrc.ca
Subject: Keypunchers, cards and printers

Following the nostalgic anecdotes on card punchers.

I was at the University of Cape Town in 1980. Although terminals were
available on the Sperry Univac mainframe, there were not enough for the first
year hordes, so were reserved for students in their second year. First year
students learned Fortran with cards.

Picture the room, hot from all the power-hungry equipment, with some students
cursing at the card punchers, and others hovering around the card-reader/line
printer. At peak times, it could take about forty-five minutes to run one
job, and at 11.45 pm the operator came and switched off - the pressure was
always on.

The cards were read, and often jammed. It took some skill to open the reader,
follow the long track inside to the jammed card, and reclose. The machine was
then very fussy about re-activation - switches two and four had to be first
switched off, and then switched on. Any mistakes, and the operator had to be
called down to reset. Often he would announce that the machine was broken,
and we would all go home, regardless of the status of our assignments. When a
newer card-reader was installed mid-year which was faster, easier to reset,
and less card-hungry, we thought the 21st Century had arrived!

Cards were all stacked into one pile, and a weight placed on top. If the card
pile was messed up, you could have up to five students fighting for their
cards from the same mess on the floor. We khokied colours onto the sides of
the cards, and could then tell where out job was in the reader.

The header card was crucial: it contained a number of instructions and
setting switches, and one mistake meant your job was thrown out.

The long delays were from the printer. It printed a whole page with the job
id, and a whole page with the final usage statistics for the job. A mistake
meant three pages of useless printout. The printer used 132 simultaneous
rotating drums, which were struck with a hammer when the correct letter came
past. A solid line meant all 132 hammers hit simultanously and very loudly,
and there were other signature tones that one's ear learnt to tell whether or
not one's job was succesful. One could hear when a new job started, as the
job id was printed in large letters on one page, and when a job was aborted,
the last page was full of zeroes, which also produced a loud clack. Using all
'1's for one's job id meant a repetitive clack, so you could hear when your
job was emerging from the printer.

Printing statistical graphs produced reams of paper, which slowed down the
queues somewhat, and one was always very pleased when the job before one's
own produced some loud clacks, meaning premature abortion. Equally pleasing
was the rapid clacking of the hyphens used for gridlines, meaning one's graph
was successful. It could take five to ten runs just to get a printout, even
before one worried about the correctness of the curves.

Horror of horrors was the endless loop. This produced twenty pages or more of
fast-clacking but useless printout until the CPU time allocation was used up.
Usually the embarrased programmer responsible made straight for the card
punchers, rather than run the gauntlet past the angry queue to fetch the
useless print job. Reams of unclaimed printout used to litter the floor.

Two years later the machine was de-commissioned. I was pleased to see it go,
but also pleased to have found a love for computers in spite of the untold
late nights, anxiety and frustration.

Christopher Geerdts
cgeerdts@idrc.ca
______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date:         Thu, 15 Aug 1996 11:16:30 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> programming in high school.
To: Multiple recipients of list CYHIST 
X-UIDL: 840138197.030

Sender: cobbjw@ornl.gov (John W. Cobb)
Subject: Re: CM> Keypunchers and cards, programming in high school.

>christopher f. chiesa comments:
>It's absolutely FASCINATING to read about all the cool things one could
>evidently do with the 026 and 029 keypunches.  Despite apparently being
>one of the "youngsters" here :-) , I can relate, because I was among one
>of the LAST groups of users of keypunches, cards, and "batch" computing,
>in a couple of math courses in the tenth and eleventh grade (1979-81).

Well, I guess we must be about the same age.

>The computer was an IBM 1130 with, I'm told, 4K of magnetic core memory.
>I know that the system had "a disk," but had no concept of even WONDERING
>about the type or capacity thereof, so never asked about, or knew, such
>details.

Wow, me too. At my high-school (in Nashville, TN) at the time we also used
an old 1130. Considering how old the machine was at that time, and how few
1130's remained, I'm betting there is a good chance we must have been
classmates. ;>

In my case, the local school system had (at considerable expense, I
believe) purchased a used 1130 about 10-15 years earlier as a demonstration
of their committment to education. It was located at a single high-school
and "computer-math" classes from all over the city would visit the machine
to run their codes. Each school had one or two card-punches at their
school. This class was my first  real introduction to computers (aside from
the various "build a computer out of pegboard and flashlight-bulb"-type
project books.

While the course was personally interesting and exciting, the "coolest"
aspect of the class was that we (i.e. the students) were actually allowed
to leave campus and drive to the other school that physically housed the
1130, often without supervision -- a true measure of independance for a
high-schooler. Since our school's computer-math course was the final
period, we often stayed much longer than the alotted 1 hour.

A friend and I used the computer for our science-fair projects that year.
He developed a program to give a 2D line-printer plot of how the night sky
would look on a given night at a given time looking in a certain direction.
My project was a measure of the efficiency of prime number generators.
Since the fortran compiler at our disposal only caontained 16 bit (15
number + sign) integers, I had to limit myself to primes < ~32,000.

We had a tough time finishing our projects because the machine kept dying.
IBM farmed out its maintanence to a third party -- Xerox I believe. The
local repair technician had never been trained to service 1130's and had a
real tough time keeping it running. One afternoon when he was there, we got
a peek into the cabinet and we were horrified to find that it looked like
most of the logic was contained in discrete components. That is, there were
huge number of three-legged transistors directly soldered into the PCB
boards and almost no integrated circuits. No wonder the guy was always
scratching his head! He finally saved himself some grief when he found an
old abandoned 1130 from somewhere else and carted it to the high-school,
stored it in the janitorial closet, and cannibalized boards from it when
the main machine crashed.

It is really amazing, as I look back on it, how much I learned from this
class. In fact, it was the only place in my formal education where I had
exposure to machine language. Our first semester was spent learning basic
computer architecture and machine language programming. One of the best
learning tools was a simulated machine language called "BC1" which I
believe stood for "Binary Computing 1". It was a simulated machine
language. I believe it was produced by SRI. Does anyone else have a
recollection of BC1 or other SRI computer-teaching tools. As I seem to
remember, SRI had a very large array of teaching materials, not only for
technical subjects but for self-paced instruction in more orthodox
educational subjects like language arts and mathematics.

When we moved to high-level languages, we used fortran. However, it was a
really old version, even before Fortran IV. Many of the control structures
in Fortran today were not yet available. Most people don't realize how
significant an advance Fortran IV was from its many variad predecessors. I
seem to remember that one of the favorite (and at the time, most advanced)
control structures was the: "on X goto (1,2,3)" which was a branch on the
value of X to go to statement label 1,2, or three based on whether X was
negative, zero, or positive, respetively.

The 1130 itself was, even 20 years ago, a peek into computer history. It's
footprint was about 50 square feet, including the line printer. The memory
was so limited that almost everything interesting was contained on the disk
that was manually mounted. The housing was about 15 inches in diameter. It
looked like a thick, plastic hub-cap that one mounted, by hand, onto one of
the computer sub-systems. The users' programs were loaded into the card
reader, and once loaded, were run by flipping a switch on the console. I
only knew the 1130 as an out-of-date dinosaur that we were lucky enough to
have because someone surplused it and our school system got it cheap. I
would be interested in hearing from "our elders" about the start of its
career. When was the 1130 first developed? For waht market? How widespread
was its use?

By the time I was in high-school early, mass market personal computers were
just beginning to catch on. I remember sitting in the computer lab, waiting
for my turn to run my deck. To pass the time, I would play a game on the
TRS80's in the lab (with their tape-recorder storge systems). I still find
it ironic to compare the computing power of the 1130 with the dozen or so
TRS80's scattered about the lab.

On the whole, however, I wouldn't trade anything for the experience of
programming a 3x3 matrix inversion program in a machine language. The
understanding of the machine and the appreciation for the difficulty are
things that still serve me well, even today.

>Similarly, I have no idea what type of keypunches we used, or
>what their capabilities might have been:

One of our favorite things was to take the small pieces of card-stock that
were punched-out and use them as confetti at football games and the like.
There always seemed to be an overly adequate supply.

-john .w cobb

John W. Cobb                                    cobbjw@ornl.gov
Office of Computing and Network Management      V. 423.576.5439
Oak Ridge National Laboratory                   F. 423.241.5722
MS-6486                       "Generally, I don't order meatloaf
Oak Ridge, TN 37831-6486       in a restuarant that sells bait."

______________________________________________________________________
            Posted by David S. Bennahum (davidsol@panix.com)
                    Moderator: Community Memory
            http://www.reach.com/matrix/community-memory.html
         A CPSR Project -- http://www.cpsr.org -- cpsr@cpsr.org
 Materials may be reposted in their *entirety* for non-commercial use.
 Want to receive the day's postings bundled together into one message?
 It's easy.  Send a note to listserv@sjuvm.stjohns.edu that reads:
                       SET CYHIST DIGEST
______________________________________________________________________
Date: Thu, 15 Aug 1996 15:00:59 -0400
X-Sender: grevinf@nyslgti.gen.ny.us
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
To: CYHIST@SJUVM.STJOHNS.EDU
From: grevinf@nyslgti.gen.ny.us (Frederic J. Grevin)
Subject: CM> Spacewar
X-UIDL: 840138197.055

This  message  was originally  submitted  by  grevinf@NYSLGTI.GEN.NY.US to  the
CYHIST list  at SJUVM.STJOHNS.EDU. If you  simply forward it back  to the list,
using  a mail  command that  generates "Resent-"  fields (ask  your local  user
support or consult the documentation of your mail program if in doubt), it will
be  distributed and  the  explanations  you are  now  reading  will be  removed
automatically. If on the other hand you edit the contributions you receive into
a digest, you will have to  remove this paragraph manually. Finally, you should
be able  to contact  the author  of this  message by  using the  normal "reply"
function of your mail program.

----------------- Message requiring your approval (47 lines) ------------------
Comment on Les Earnest's posting:
>Also in the early 1970s a couple of our grad students, Bill Pitts and
>Phil Petit, decided to explore the commercial possibilities of
>Spacewar.  They bought two PDP-11's and some coin acceptors, built
>interfaces and button boxes and reprogrammed Spacewar in PDP-11
>assembly language.  They put one system in the Stanford Coffee House
>and another in an off-campus bolling alley, allowing up to five
>players to engage in a Spacewar match for 25 cents each.  It proved to
>be quite popular, but not financially rewarding for Pitts and Petit
>because of the cost of amortizing the computer system.

>Meanwhile, an entrepreneur named Nolan Bushnell, who had been exposed
>to Spacewarin college, began pursuing the same idea in a slightly
>different way by forming a small company called Atari to develop
>special purpose hardware for games that could manufactured
>inexpensively.  The first one he developed was Spacewar, but he
>decided that it was too complicated to be understood by the general
>public, so he sold that design to another company and developed a much
>simpler game called Pong.  Pong turned out to be wildly popular,
>making him a multi-millionaire.

In the late 1970s, the popular and raffish student bar, the "West End" (on
Broadway, across from Columbia University's main campus), had several Space
War machines. They were wildly popular with persons of all types and ages (I
remember waiting in line to use them), and the West End probably made a fortune.

The game was sophisticated for arcade games of the time:  the "spaceships"
would run off the right edge of the screen and reappear on the left, and you
could "spray" fire while on the move. It wasn't easy to play well, and it
could be addictive (confession of former addict). I would gladly pay for a
version that could run on a DOS or Windows machine ("former" addict? Well....).

I regret to say that I don't know whether they were Pitts & Petit's machine
or Bushnell's.

Fred.
Fred Grevin
Micrographics Coordinator
New York City Department of Finance
(The opinions expressed in this message do not necessarily reflect the
policy of the NYC Dept. of Finance)
31 Chambers Street  (room 201)
New York  NY 10007
grevinf@nysLgti.gen.ny.us
   tel:  212 788-8519
   fax:  212 788-8521
Date:         Thu, 15 Aug 1996 16:16:55 -0500
Reply-To: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
Sender: "CYHIST  Community Memory: Discussion list on the History of
              Cyberspace" 
From: "David S. Bennahum" 
Subject:      CM> Moderator: List Problems.
To: Multiple recipients of list CYHIST 
X-UIDL: 840140306.000

Many of you emailed me in the last few hours saying you could not set the
list to digest mode.  The list was "locked" and I did not know this.  The
list is now "unlocked" and that digest command should now work.

Some of you have found that sending messages to the list, or commands like
UNSUBSCRIBE does not work either -- yet you are receiving messages.  I am
stumped by this one, and trying to solve the problem.

Until these problems are resolved, I will keep the flow of messages low, so
as not to swamp people.

best,
db