Re: mutt development status
On Tuesday, 12 July 2005 at 11:33, John J. Foster wrote:
> On Mon, Jul 11, 2005 at 04:06:10PM +0200, Oswald Buddenhagen wrote:
> > so, thomas, unless you're satisfied with this situation, i think it
> > would be wisest to officially resign from mutt maintenance and hand over
> > to the mutt-ng guy (or somebody else, fwiw), possibly offering yourself
> > as a co-maintainer/consultant. sure, that would mean that some things not
> > to your liking are merged, and possibly the code quality suffers short
> > term, but in the longer run this step _would_ be to mutt's benefit.
>
> This conversation so far has been much needed, but it seems to me that
> there are several people who ought to be participating, but are instead
> MIA. It does absolutley /no good/ for bozos like me, or any of you fine
> hackers out there, to talk about any of this.
>
> Thomas, Brendan, where are you? What are /your/ feelings here? Join the
> conversation.
First of all, let me say that I don't consider myself as part of a
"core team." Thomas is the maintainer and makes all decisions about
features or behavioural changes (outside of the IMAP domain, where I
take more of an executive role). All I commit to the main tree are
obvious fixes and optimisations. Anyway...
I kind of had the impression that this thread was an attempt to
rekindle the fire that the last mutt-ng fork thread started (because
Oswald's list of complaints were fairly abstract, eg "no patches
generated"), so I didn't feel the urge to rush in. But here's some
off-the-cuff thoughts on the state of development:
1. mutt is a pretty good mailreader. That means that as time goes on,
patches are more and more likely to be a little outside mutt's core
domain (nntp, smtp) or a bit froofy (sidebar patch) or abstruse
(guile bindings). Personally I'm more liberal than tlr about a
number of these, and would probably tend to include a patch if most
mainstream distributions felt the need to apply it. But I would
expect the rate of acceptance for non-bugfix patches to be lower
nowadays than it used to be.
2. forks are bad. They just double (or triple or n-ple) the work of
bug-fixers, patch-writers, and distribution maintainers, not to
mention the poor user who has to choose his favourite puppy. If the
alpha project is getting old and grey, some newcomer can challenge
for dominance, but two or more long-term coexisting forks is a
nuisance (eg emacs vs xemacs, though some may disagree about that
one :)).
3. Thomas and I are obviously too busy, but finding a trustworthy
additional maintainer isn't easy. Most people who code for mutt
(myself included) have their pet projects (eg new mail handling :))
but not global expertise in the codebase, mail standards, and
operating systems that you need as a maintainer (rather than a
patch-writer). I didn't think the mutt-ng author displayed enough
of those credentials, and I can't say I'm too surprised to see that
it's already withering a little.
4. A BTS and a 1.5.10 followed soon after by a 1.6 would be
nice. Either GnuPG or sourceforge seem like they'd be suitable and
easy to start using. As I mentioned at the top, I'm not the one to
pull the trigger though.
-b