All, it's time to nudge the conversation away from debugging 2017 OOM issues
and the pre-UNIX history of the Arpanet.
We've been able to recover quite a deal of UNIX artifacts in the past two
decades, but what artifacts (in your opinion) are still out there that
we should try and unearth? Remember that the 50th anniversary is coming up
in 2019.
Here's my list:
- more PDP-7 source code: the shell, the rest of the utilities
- more 1st Edition source code: the rest of the utilities
- ditto the missing bits of 3rd, 4th and 5th Editions
- the Phil Foglio artwork that became a Usenix t-shirt (Armando, any ideas?)
- more details on who was Ted Bashkow, and the story behind his (+ others?)
analysis of the 1st Edition kernel at http://www.tuhs.org/Archive/Distributions/Research/Dennis_v1/PreliminaryUni…
- a firm date on the day that Ken added pipes to the kernel :)
What else should be we looking for? What physical artifacts (drawings,
artwork etc.) have we missed that should be sought after?
Cheers, Warren
Hi.
What's the difference between spell.old and spell in the V10 sources?
I'm guessing spell.old is the original version for the small-address-space
PDP-11 and that plain spell takes better advantage of the roomier vax...
Thanks,
Arnold
>> The choice of "# " and "> " interests me. Because roots prompt of
>> "hash" has a side effect of making a cut-paste over it a comment in
>> most shells.
>
> "#" as the root prompt predates # as the comment in the Bourne shell,
> not to mention predating copy/paste entirely. (My understanding is that
> the do-nothing command, ":" was used for comments. Talk about minimalist!)
>
> Same point for ">", since copy/paste didn't exist in the late 70s when
> Bourne was doing the shell, it wasn't an issue.
As early as V5 the (thompson) shell prompts were “#” and “%”, and “:” for
a label. As the goto command exists in V4 (there is a man page for it), I
would assume that those characters were used in V4 as well. So it would
seem to go back to 1974.
In the V7 (bourne) shell the default non-root prompt is “$”. Goto is
dropped at this point.
Don’t know when or where “>” was first used on Unix as a prompt character
(on my boxes it still is “$”).
Paul
> We've been able to recover quite a deal of UNIX artifacts in the past two
> decades, but what artifacts (in your opinion) are still out there that
> we should try and unearth? Remember that the 50th anniversary is coming up
> in 2019.
I’d be interested in anything on Spider/Datakit networking in V4-V7.
(at them moment the trail starts at V8, with just a few hints in earlier
source materials, and the bits that Noel found).
My thinking is that there were two main lines of early networking development
on Unix (and I realise that this gross simplification excludes many other
worthy projects):
1. The “sockets” lineage from UoI NCP Unix -> BBN NCP Unix -> BBN TCP Unix
-> 4.1a BSD -> 4.2 BSD
2. The “device” lineage from Spider -> Datakit -> UUCP -> streams
-> STREAMS
In the first lineage there is much material available, in the second very
little. This is probably because Datakit was AT&T confidential at the time.
Warren Toomey <wkt(a)tuhs.org> asks on Wed, 6 Dec 2017 08:21:13 +1000:
>> - more details on who was Ted Bashkow, and the story behind his (+ others?)
I found a short obituary at
http://engineering.columbia.edu/web/newsletter/spring_2010/memoriam
which is, in full:
>> ...
>> Theodore R. Bashkow Dr. Theodore R. Bashkow, professor emeritus of
>> electricial engineering and computer science, died Dec. 23, 2009, at
>> his home in Katonah, N.Y. See PDF version
>>
>> He was born in St. Louis, Mo., and attended Washington University,
>> where he received his BS degree in mechanical engineering. He went on
>> to receive his master’s and doctorate degrees at Stanford
>> University. He served in the U.S. Air Force as a first lieutenant
>> during World War II from 1943 to 1945.
>>
>> While in the Air Force, he served as maintenance officer and helped to
>> stage the Enola Gay. In the 1950s, while at Bell Labs, Professor
>> Bashkow became well known for his development of a new method for
>> analyzing linear electrical networks, Professor Bashkow’s A matrix. He
>> also became involved with digital computers. He joined the faculty of
>> the Columbia Electrical Engineering Department in 1958 and helped
>> transform the Electrical Engineering Department into the Department of
>> Electrical Engineering and Computer Science.
>>
>> When, in 1979, this department was divided into the Electrical
>> Engineering and Computer Science departments, Bashkow became one of
>> the founding faculty members of Computer Science. He taught courses in
>> digital logic, computer organization, and computer programming. He did
>> research on parallel processing. In collaboration with Herbert
>> Sullivan, he pioneered a new approach to that subject through the
>> development of CHoPP, Columbia Homogeneous Parallel Processor, a
>> large-scale, homogeneous, fully distributed parallel machine. A number
>> of Columbia graduate students and a junior faculty member, David
>> Klappholz, were also involved at various stages.
>>
>> In 1980, the Computer Science Department instituted an annual award in
>> his honor, the Theodore R. Bashkow Award. Among his many affiliations,
>> Professor Bashkow was an active member of IEEE, ACM, and Sigma Xi
>> organizations.
>> ...
He is apparently not in Wikipedia.
I then searched our local bibliography archives and found this
publication-title summary (Bashkow is an uncommon name, so I didn't
attempt to disambiguate the reported articles):
MariaDB [bibtex]> select filename, label, substr(title,1,80) from bibtab where (author like '%Bashkow%') order by year, filename;
+-------------------------+--------------------+----------------------------------------------------------------------------------+
| filename | label | substr(title,1,80) |
+-------------------------+--------------------+----------------------------------------------------------------------------------+
| jacm.bib | Bashkow:1958:CPR | A ``Curve Plotting'' Routine for the Inverse Laplace Transform of Rational Funct |
| ieeetranscomput.bib | Bashkow:1963:RDA | R63-106 The D 825 Automatic Operating and Scheduling Program |
| ieeetranscomput.bib | Bashkow:1963:C | Contributors |
| ieeetranscomput.bib | Bashkow:1963:PSD | A Programming System for Detection and Diagnosis of Machine Malfunctions |
| ieeetranscomput.bib | Bashkow:1964:SCA | A Sequential Circuit for Algebraic Statement Translation |
| fortran1.bib | Bashkow:1967:SDF | System Design of a FORTRAN Machine |
| ieeetranscomput.bib | Bashkow:1967:SDF | System Design of a FORTRAN Machine |
| ieeetranscomput1970.bib | Bashkow:1971:BSS | B71-6 System Structure in Data, Programs, and Computers |
| ieeetranscomput1970.bib | Bashkow:1971:BIC | B71-2 Introduction to Computer Organization |
| ieeetranscomput1970.bib | Bashkow:1973:CRO | Comment on Review of Operating Systems Survey |
| ovr.bib | Sullivan77b | A Large Scale, Homogeneous, Fully Distributed Parallel Machine |
| ovr.bib | Sullivan77a | A Large Scale Homogeneous Fully Distributed Parallel Machine |
| sigarch.bib | Sullivan:1977:LSHb | A Large Scale, Homogenous, Fully Distributed Parallel Machine, II |
| sigarch.bib | Sullivan:1977:LSHa | A large scale, homogeneous, fully distributed parallel machine, I |
| ieeetranscomput1980.bib | Ghafoor:1989:BFT | Bisectional Fault-Tolerant Communication Architecture for Supercomputer Systems |
| super.bib | Ghafoor:1989:BFT | Bisectional Fault-Tolerant Communication Architecture for Supercomputer Systems |
| ieeetranscomput1990.bib | Ghafoor:1991:SOG | A study of odd graphs as fault-tolerant interconnection networks |
+-------------------------+--------------------+----------------------------------------------------------------------------------+
17 rows in set (2.67 sec)
-------------------------------------------------------------------------------
- Nelson H. F. Beebe Tel: +1 801 581 5254 -
- University of Utah FAX: +1 801 581 4148 -
- Department of Mathematics, 110 LCB Internet e-mail: beebe(a)math.utah.edu -
- 155 S 1400 E RM 233 beebe(a)acm.org beebe(a)computer.org -
- Salt Lake City, UT 84112-0090, USA URL: http://www.math.utah.edu/~beebe/ -
-------------------------------------------------------------------------------
> From: Jon Forrest
> LBL has never been part of UC Berkeley. It's (always?) been a
> Department of Energy laboratory managed by the Univ.
Actually, I think if you go back far enough (1930's), it was part of UCB,
back when Lawrence first started it.
And of course the DoE didn't exist until 1977, so during the early ARPANET
era if would have been under the AEC, and then I assume the Energy Research
and Development Administration after 1974 (I assume it didn't go with the NRC
when the AEC was split up).
Noel
This page:
http://www2.lbl.gov/Publications/75th/files/exhibit.html
mentions 3 links between LBL and Arpanet:
- In 1974, the Lab’s CDC 6600 became the first online supercomputer when it was connected to ARPANET, the Internet’s predecessor.
- In 1986, when the Internet was on the verge of collapse from congestion, a Berkeley Lab researcher, Van Jacobson, co-developed the congestion control algorithms that allowed the Internet to keep growing.
- In 1995, Jacobson and Steven McCanne developed MBone, the first successful software for multiparty audio and video conferencing over the Internet.
I don’t think anybody was thinking you wilfully misrepresented things,
it is always interesting to hear about strands of history that might
have been missed earlier.
It would be helpful to better understand the time period, people
involved and the scope of work.
I’m a bit confused as to what time period you are referring to: I
think you are referring to the initial development of Arpanet, i.e.
the second half of the sixties. Is that correct?
There is a page here with some info on events in that period and it
may have missed some interesting development work done at LBL:
https://www.livinginternet.com/i/ii_roberts.htm
Paul
>
> Message: 7
> Date: Mon, 4 Dec 2017 19:46:21 -0800
> From: Deborah Scherrer <dscherrer(a)solar.stanford.edu>
> To: tuhs(a)minnie.tuhs.org
> Subject: Re: [TUHS] ARPAnet now 4 nodes
> Message-ID: <8254fc85-12e6-4730-8f14-faf060ad6a70(a)solar.stanford.edu>
> Content-Type: text/plain; charset=windows-1252; format=flowed
>
> Yes, Van Jacobson was involved. Great guy. So sorry you feel the need
> to think I am lying. Why would I make up this stuff? I was a
> teeny tiny piece of it. Doesn't affect my career one way or other. I
> don't care what you believe, but this really did happen.
>
> D
> From: Deborah Scherrer
> I don't know about the historical record. But everything I said is true,
> based on my own personal experience. ... I was there, this happened. If
> people didn't write it down, I don't know why.
FWIW, I was actually at many of those meetings. (You can find my name in a lot
of those Meeting Notes.) Nobody from LBL, or UCB in general, was involved -
and the Meeting Notes (which, you will note, are quite detailed) indicate the
same thing.
(Later on, of course, Van Jacobson of LBL did some imporant work on TCP
congestion control, but that was in '87 or so - I can't instantly lay my hands
on my copy of Van's famous e-mail, to get a more exact date - some years after
the full-scale deployment of TCP/IP in January, 1983.)
> Why would I misrepresent?
Perhaps you are conflating several different things in your memory? Human
memory is very fallible, which is why historians prefer contemporary documents
(and even those sometimes have errors). Here:
http://www.chiappa.net/~jnc/nontech/tmlotus.html
is a mildly amusing example (from a completely different arena) of all that.
Noel
