Fwd: alternate version control (was Re: [IMC-Tech] is this what we need instead of CVS)

[Paul Riismandel]

   FYI:
>Delivered-To: mediageek-org-listgeek at mediageek.org
>Delivered-To: imc-tech at indymedia.org
>From: Jeremy Kahn <jeremy_kahn at yahoo.com>
>Reply-To: jeremy_kahn at yahoo.com
>Subject: alternate version control (was Re: [IMC-Tech] is this what we 
>need instead of CVS)
>To: Miles Thompson <miles at cyberelves.com>, imc-tech at indymedia.org
>Date: Wed, 6 Feb 2002 12:42:17 -0800 (PST)
>
>Miles wrote:
> > Perhaps its a way out of the multi-versioned, semi
> > CVS
> > semi not in CVS chaos that constitutes 'Active' and
> > its
> > immediate children right now.
>
>Miles, I think you've put your finger on the real
>problem -- it's not that CVS is particularly weak,
>it's that it's not being used consistently.
>
>Full disclosure -- I'm a big fan of CVS.  It's well
>tested, it's maintained, there's lots of tools that
>already plug into it, including emacs and the latest
>vi versions. It also works just about anywhere, even
>POSIX-non-compliant systems like the NT box from which
>I write this email.
>
>However, I contend that the real problem is not CVS
>but that we're NOT using CVS -- or any other version
>control (VC) system -- regularly.
>
>The version control problems we're struggling with are
>not particularly new to the software world. In fact,
>the distributed nature of the Active problem is not
>nearly as distributed as trying, for example, to keep
>tens of thousands of people in sync with the latest
>version of "patch" or "gcc", for example.
>
>Most of the Active work done is done on a handful of
>machines (okay, less than fifty) worldwide. If each of
>these development efforts were on a branch within CVS,
>then merging among and across them would not be very
>hard, and we could quickly reduce the number of active
>branches to three or four at the outside.
>
>But when much of the work is done on independent
>machines without any VC checkin, then merging is a
>manual process, and most of us find ourselves fixing
>the same problems instead of being able to share
>innovation and cleverness.
>
>If using arch or subversion would solve this problem,
>I'd be all for it.  But as the Cederqvist CVS manual
>says somewhere (I quote loosely): "source control is
>not a substitute for coordination among the
>developers." We really need to get it all in source
>control -- ANY source control -- and start to talk
>about consolidating the branches.
>
>There was an earlier thread about just this from Kevin
>O'Donnell, and I'm all for the idea of consolidating
>codebases. I think we would be fine, though, to stick
>with CVS once we can consolidate the various code
>directions.
>
>Although learning a new VC system might be fun (well,
>for a certain flavor of geek like me), I think most
>people would rather not learn any more about VC than
>they really have to -- and learning a new one might
>just turn some people even farther off.
>
>Sorry to seem so negative. I am glad to see other
>people thinking about this problem, which is, in my
>opinion, spiraling out of control. I feel, though,
>that changing VC tools fails to identify the real
>problem -- the lack of consistency in using those
>tools.
>
>My 2c.
>
>--jeremy
>
>Apologies for the irritating ad below; please snip in
>any response.
>
>__________________________________________________
>Do You Yahoo!?
>Send FREE Valentine eCards with Yahoo! Greetings!
>http://greetings.yahoo.com
>_______________________________________________
>imc-tech mailing list
>imc-tech at lists.indymedia.org
>http://lists.indymedia.org/mailman/listinfo/imc-tech