Re: Paper on transactional version control

On Wed, 2012-04-11 at 15:44 +1000, Richard Walker wrote: 
> I have a vague (but strong) memory of some years
> ago reading a paper that explained the concepts and benefits
> of transactional version control, and that this
> was somehow part of the inspiration for the
> way that Aegis works.

One of the inspirations was

        Babbich, W.A, (1986) "Software Configuration Management,
        Coordination for Team Productivity"

It is the source of much of the terminology used in the Aegis commands,
and error messages, and documentation.  In reading Babbich today, you
have to remember what software development was like in the early 1980s,
when compilers did almost no static analysis, memory was in short
supply, a *huge* disk was 300MB, and the ANSI C Standard was still 5+
years away.  Conversely, some of the problems it describes are still
with us.

I first saw such a process in action, between 1986 and 1989, when I was
working for Prime Computers, in Canberra, Australia.  It was run
manually, but the process featured full code reviews, continuous
integration, and automated testing.  My objective in writing Aegis was
to capture this process, and yet relieve the humans of the book keeping
required.  I foolishly though that giving good process away would mean
that good process would beat me to my next job; I'm still hoping.

> It may have mentioned ClearCase, but my
> memory suggests it predated that (1990s?) and
> was more on concepts
> than a particular implementation.

I had read about ClearCase by 1991, but it wasn't until 2001-ish that I
actually used it.  Had a laugh at the "unco" command.  Once I went to a
seminar put on by Rational, and they told me that Aegis was in the
training manual as one of "the competition", the trainer was a but
flustered when he found out who was in the audience.


-- 
Regards
Peter Miller <[email protected]>
/\/\*        http://miller.emu.id.au/pmiller/

PGP public key ID: 1024D/D0EDB64D
fingerprint = AD0A C5DF C426 4F03 5D53  2BDB 18D8 A4E2 D0ED B64D
See http://www.keyserver.net or any PGP keyserver for public key.

"Curiosity was framed. Ignorance killed the cat."