|--==> "MB" == Martin Buchholz <martin(a)xemacs.org> writes:
>>>>>"SY" == Steve Youngs
<youngs(a)xemacs.org> writes:
SY> I'm definitely open to suggestions
for improvements.
MB> To keep things sane, it makes sense to have an unstable branch of the
MB> packages. Developers could check out the unstable branch using
Actually, I think this is a bad idea. There are probably lots of
reasons why, but the one that sticks out in my mind is that it would
be an administrative nightmare.
We currently have 88, yep *Eighty-eight* packages, I keep two copies
of the xemacs-packages CVS source tree locally, one for
testing/building and a clean copy from which I do all CVS
commits/updates etc. That's 264 (if you include
cvs.xemacs.org)
packages that I keep abreast of.
And you want to double that number by having an unstable branch. Man,
my wife already thinks I love the computer more than her. :-)
In a perfect world (well one of my versions of a perfect world, at
least), each and every package would have an active maintainer. That
maintainer would keep their development tree somewhere else, such as
SourceForge, and only commit tested stable versions of their code to
our CVS.
But alas, it's not a perfect world...
And if I ever get some time to do it, I'm going to start posting lists
of packages that have no maintainer and try to encourage adoption.
MB> The hard part of this is committer education. There are far more
MB> xemacs packages committers than core committers, at least potentially.
Another reason for stricter controls on who gets commit privileges.
But that's a can of worms that I don't wish to re-open. Especially
since everyone has been so good lately. :-)
--
|---<Steve Youngs>---------------<GnuPG KeyID: 9E7E2820>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|