29 Jan
2022
29 Jan
'22
6:03 a.m.
Not really on topic for Unix, but historical and relevant to this
conversation and I think not well known.
In 1978 I was an exchange student working at EIR (now the Paul Scherrer
Institute), using the CDC 7600 (I think that was the model) at ETH, where
Wirth was a professor. EIR is about 30km from Zürich, and EIR had a remote
job entry system.
The computing environment was very odd, and I asked about it. Two things I
learned:
1) The University had bought a CDC machine instead of an IBM one, somewhat
against advice, because CDC, being a smaller company, did not have the
wherewithal to translate their manuals. IBM's manuals came in German and
were all but incomprehensible as they avoided the accepted terms of art
known and used even by a German-speaking programmer. The CDC manuals, being
in English, were easier to understand, especially when considering the
nuance and precision necessary to learn the correct interpretation of the
description of a computer's execution. The Swiss, being polyglots, handled
English manuals just fine.
2) The operating system's I/O model was bizarre, but it was also unique. It
was a version of NOS locally modified, partly (if I remember right) in
support of the remote execution setup. The peculiar sequence of cards
necessary to terminate the input was due to the local changes to NOS made
at ETH. This was the system Pascal was created for, and a consequence of
the design is the idiosyncratic way input worked in early Pascal, which
made little sense to almost anyone, was seriously hard to recreate on Unix,
but worked naturally if using punch cards on, and only on, the ETH 7600.
-rob
On Sat, Jan 29, 2022 at 10:32 AM Will Senn <will.senn(a)gmail.com> wrote:
On 1/28/22 5:18 PM, Dan Cross wrote:
On Fri, Jan 28, 2022 at 6:09 PM Will Senn <will.senn(a)gmail.com> wrote:
I'm reading in, Kernighan &
Plauger's 1981 edition of Software Tools in
Pascal and in the book, the author's mention Bill Joy's Pascal and Andy
Tanenbaum's as being rock solid. So, a few related questions:
1. What edition of UNIX were they likely to be using?
I'm afraid I can't speak to your 2nd and 3rd questions, but I can offer
what I think is a reasonable guess about the first.
One of the neat things about Unix and Unix-adjacent books of that era is
that very often the copyright page held some information about the
production of the book itself. I just so happened to have a copy of,
"Software Tools in Pascal" sitting on my desk, and it says, "This books
as
set in Times Roman and Courier by the authors, using a Mergenthaler
Linotron 202 phototypesetter driven by a PDP-11/70 running the Unix
operating system."
Given the PDP-11 and the date (1981) one may reasonably conclude that it
was running 7th Edition. I imagine the pascal was Joy's, from Berkeley.
- Dan C.
Great hint. 20 seconds after I hit send on the original email, I came
across this:
http://www.lysator.liu.se/c/bwk-on-pascal.html
Where Brian Kernighan talks about the challenges they faced porting the
ratfor examples into pascal. He explains that:
The programs were first written in that dialect of Pascal supported by the
Pascal interpreter pi provided by the University of California at Berkeley. The
language is close to the nominal standard of Jensen and Wirth,(6
<http://www.lysator.liu.se/c/bwk-on-pascal.html#lit-6>) with good
diagnostics and careful run-time checking. Since then, the programs have
also been run, unchanged except for new libraries of primitives, on four
other systems: an interpreter from the Free University of Amsterdam
(hereinafter referred to as VU, for Vrije Universiteit), a VAX version of
the Berkeley system (a true compiler), a compiler purveyed by Whitesmiths,
Ltd., and UCSD Pascal on a Z80. All but the last of these Pascal systems
are written in C.
So, you were right about it being Joy's pi.
Thanks,
Will