| Re: Version Control Strategies For Block-Based Forths? |
|
 |
|
 |
|
 |
|
 |
Group: comp.lang.forth · Group Profile
Author: John DotyJohn Doty Date: Sep 19, 2008 13:44
John Passaniti wrote:
> Part of what I'm trying to see (aside from my usual interest in computer
> history) is what kinds of trade-offs different approaches took, and in
> particular approaches used to support multiple developers instead of a
> lone developer.
I don't recall any serious multiple developer issues with the boxes of
floppies approach. But it was rather different in scale from a modern
project. I have a printout of one of the last revisions done that way:
it was 33 blocks, and the blocks contained lots of white space. That was
about all that would fit in the little 1802. Something like 6-8 people
made serious contributions to this code. I can't think of any recent
project I've worked on with so many contributing to so little code.
The RCS approach had the well-known multiple developer malady of RCS:
somebody would check out a source file, begin to work on it, and then
get distracted by some higher priority issue and leave it locked, so the
archive admin (me, grumble) was always having to break locks. But now
there are better alternatives: too many and every project wants a
different one, arrrgggh!
> All this is being used as research for a new project at
> work with some unusual requirements that might lend itself to
> block-style development directly on the target.
You deny the application matters, and then you tease us ;-)
--
John Doty, Noqsi Aerospace, Ltd.
http://www.noqsi.com/
--
The axiomatic method of mathematics is one of the great achievements of
our culture. However, it is only a method. Whereas the facts of
mathematics once discovered will never change, the method by which these
facts are verified has changed many times in the past, and it would be
foolhardy to expect that changes will not occur again at some future
date. - Gian-Carlo Rota
| 
