autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: conversion to git


From: Russ Allbery
Subject: Re: conversion to git
Date: Wed, 19 Sep 2007 21:01:29 -0700
User-agent: Gnus/5.110006 (No Gnus v0.6) Emacs/21.4 (gnu/linux)

Bob Friesenhahn <address@hidden> writes:

> I have looked at Subversion and find its installation to be rather
> unwieldly, requiring many additional packages to be installed of
> particular versions.  Subversion seems to use a rather exotic
> implementation rather than a fairly simple one like CVS.  It seems that
> GIT has the advantage in this regard.  I tried 'arch' but it did not
> compile on my machine at the time so I never looked back.

> Maintainers of portable software need portable version control systems.
> The version control system client needs to put up no fuss when compiled
> on a "bare" machine (often a porting target, for extra challenge) so
> that it is possible to get going right away.

This just isn't going to matter enough to really make a difference, I
think.  I don't disagree with you, but despite the fact that all the new
version control systems are more complex to compile than CVS (heck,
Subversion depends on APR, which has caused no end of problems), that
isn't even slowing migration down.

I'm pretty comfortable at this point saying that CVS is basically dead.
The corpse is going to keep twitching for a while, just like you still
find people using RCS and SCCS, but it's the only version control system
in widespread use in the free software world that can't version
directories, can't deal with renames, doesn't have file properties, and
doesn't have atomic commits.

The distributed VCS debate is still ongoing, and it's not clear to me that
everyone is going to like that model and go all the way to Git (or
Mercurial or Bazaar), but everyone I see out there using CVS for major
projects is at least seriously considering Subversion even if they're
sticking with the same model.  We're converting all of our internal
repositories at Stanford (where we don't need or care about the
distributed model) to Subversion.

I think in practice adoption drives portability more than the other way
around.  I don't think CVS became popular because it was portable; I think
people put a lot of work into getting it to run on a lot of platforms
because it was popular.  Subversion, being one of the first of the next
generation, has the portability lead at the moment.

You're right that all these new systems are exotic or complicated compared
to CVS.  This is for many of the same reasons that CVS is exotic and
complicated compared to RCS.

-- 
Russ Allbery (address@hidden)             <http://www.eyrie.org/~eagle/>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]