Re: policy summary (new packages without man pages)
>>"Owen" == Owen Dunn <owend@chiark.greenend.org.uk> writes:
>> What I'm really trying to say is that what we should expect from a
>> maintainer is _commitment_.
Owen> As I've understood it, the only thing that has really been
Owen> required in the past of a maintainer (beyond any initial
Owen> packaging) is to act as a clearing house for bug reports and
Owen> feature requests. If they have time to investigate and fix all
Owen> reported bugs themselves, keep in communication with the
Owen> upstream maintainer, &c., so much the better.
I think I tend to disagree here. I really expect more of a
debian developer than a glorified bureaucrat -- we are trying to
create the best free distribution, and that often entails making the
package *better* than what is created upstream, and then reporting
the improvements back upstream.
Owen> Many people working on Debian are students and have lots of
Owen> time; some of them (and God only knows how) have the time
Owen> actively to maintain many packages and still keep up with the
Owen> mailing lists. Many, though, have full-time jobs, copious
Owen> other interests, a bevy of other free software projects, and so
Owen> on, and weeks may go by where they are genuinely too busy with
Owen> other things to contribute to Debian. If you feel that's a
Owen> lack of commitment and that they shouldn't be maintaining
Owen> packages, I disagree.
If their lack of time is lowering the quality of the
distribution, I think they should get rid of some (all) their
packages till they have more time to devote to the project.
Membership in Debian is not an entitlement; a certain amount
of commitment is definitely expected. Speaking as one who has not had
a whole lot of time recently, I still thik that quality of Debian is
more important than letting folks who are busy just neglect doing
what thier packages require.
We already have an NMU mechanism. What we need is a place
where busy people can log their lack of time in the short term, and
ask for NMU's.
Owen> (a) Often highly competent people are very busy with many things, the
Owen> relative priorities of which only they can know. This does not
Owen> mean they're not committed to Debian, or that they're not a great
Owen> asset to the Project when they are able to contribute.
If they lack time in the short term, they can still give up
their packages (maybe to the QA team), and take 'em back when they
get more time.
Owen> (b) We have mechanisms for other developers to rectify problems in
Owen> other maintainers' packages if the maintainer isn't active.
Yes, and we need to have these mechanisms utilized more often.
Owen> In the case in point, I think it's more important (in terms of
Owen> benefit to Debian) that we have a package at all than that we
Owen> have the `perfect' package;
Though ``perfect'' is a troll, I still reject that quantity is
better than quality argument. The social contract does not speak of
the biggest distribution, it speaks of the best.
Owen> the work of a maintainer on a package is still useful work even
Owen> if that maintainer happens not to speak fluent nroff, or not to
Learning pod2man is not onerous. You do not need to know nroff
fluently to make a man package.
If a developer can't learn pod2man to provide documentation
for his package, I strongly suggest he does not have the time,
motivation, or the competence to maintain a package.
Owen> have the time to write a good manual page. We should accept
Owen> that work into the distribution, where other developers
Owen> (through the BTS and mailing lists) can help fix any problems.
Owen> This benefits Debian by encouraging the developer, and benefits
Owen> the world with the new package; I don't see how rejecting that
Owen> work benefits anyone.
Quality control always benefits the project. Quality is indeed
to be preferred to mediocre quantity.
manoj
--
/earth is 98% full ... please delete anyone you can.
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Reply to: