Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!boulder!unicads!les
From: les@unicads.UUCP (Les Milash)
Newsgroups: comp.sys.transputer
Subject: Re: Transputer scheduler and documentation
Message-ID: <348@unicads.UUCP>
Date: 4 Apr 89 17:27:23 GMT
References: <8904032222.AA02110@tcgould.TN.CORNELL.EDU>
Reply-To: les@sirius.UUCP (Les Milash)
Organization: Unicad  Boulder, CO
Lines: 36

In article <8904032222.AA02110@tcgould.TN.CORNELL.EDU> ADRIAN@VAX.OXFORD.AC.UK writes:
>Even then, if you read between the lines, you could infer a good deal.

YES YES I agree.  I was out there trying to reverse-engineer the 242
trying to figure what was in it from the cycle-counts from that red
spiral-bound book, remember that one?  I would have been buying those
suckers with every spare cent if they could only multiply.  (I always
wished the T212's had a real multiplier on it, you know, a 1-cycle
multiplier like in a DSP.  now THAT'd be a killer chip; a scalable
DSP.  (the niche is STILL OPEN, and I'd STILL buy them)

>
>I also think that anyone working with transputers will benefit enormously
>by some appreciation of CSP (Communicating Sequential Processes)...

I THINK SO TOO.  I think ANYONE will benefit enormously, (but then I'm a nerd).
You start to see breakfast as a function that maps {coffee, electricity,
a hungover wire-wrapper, and a clean kitchen} onto {coffee grounds, waste
heat, an employable programmer, and a dirty kitchen}.  It'll be good for you
even if you never see a T800 or OCCAM.  It'll taint any infatuation with C;
the "machismo" of trying to get correct results from a language with partly
unspecified semantics will seem more like "rambismo".

>one of the theoretical underpinnings for occam and the transputer. There is
>a vast literature of papers, but start with another Prentice Hill Book :
>"Communicating Sequential Processes" by C A R Hoare (who invented CSP along
>with so much else!). Sorry, I don't have the reference to hand; published
>about 3 years ago.

I HAVE A COPY RIGHT HERE ON THE SHRINE and here is the reference:
CAR Hoare, "Communicating Sequential Processes", Prentice Hall _International_,
1985.  does this stuff "ISBN 0-13-153271-5" help?

Also interesting is to read CARHoare's ACM Turing Award Lecture; as i recall
Mr. Hoare botched his first job pretty well :-), and got interested in 
correctness as a result of that.