Re: Sponsorship in Debian (Re: Ubuntu-originated packages in Debian (Re: ubuntu keyring?))
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi Neil (and Piotr who said similar things),
On 20.05.2011 11:22, Neil Williams wrote:
> It's also very frustrating for the sponsor and the hassle involved in
> filtering out some truly abysmal packaging efforts from the list of
> sponsorship requests is one reason why I have not returned to
> sponsoring after an enforced break some time ago.
I know. I review packages myself (although my opinion as non-DD is not
necessarily effectual). I'm absolutely with you, if you claim some
packages are not worth to be considered to be sponsored while being in
such a bad shape.
Problem is, there /are/ some packages which are of good quality and
there /are/ people willing to help Debian as prospective developers, but
they are scared off sooner or later, because they get no resonance for
their efforts as no one cares and you slowly starvate their efforts.
The whole point is to find a procedure to help those people and still
prevent DDs to be overwhelmed by people whom didn't ever read policy.
This is where things have to improve as this would be a win-win
situation for both parties.
> There really is no point asking for a sponsor until the package is
> ready for upload and that means lintian clean, fixes obvious bugs
> listed in the BTS, pbuilder compliant and something which is useful to
> the rest of Debian (i.e. not your pet upstream one-user project).
True. On the other hand debian-mentors is and should be a place where it
is allowed to make mistakes. You can't expect everyone to be born as
Debian developer knowing everything.
You should neither expect from your sponsoree to make no mistakes at
all. This is, what your role as mentor is for: To assist people, being
new to Debian and help them once they are caught in pitfalls you can
help to come over.
However this isn't such a problem in reality. There /are/ experienced
people willing to help for packaging/technical questions. Just take a
look on d-mentors in IRC or mailing list, and you will see, there is
hardly a technical question remaining unanswered. This /does/ change as
soon as you ask for sponsoring where you have good chances not to get
any reaction at all, regardless the quality of your package.
> Honestly, I feel that Debian can do without the majority of NEW packages
> awaiting sponsoring - I'd only really be interested in sponsoring
> ITA bugs, not ITP and for packages with a maintainer, the sponsoring /
> upload needs to happen with that maintainer or that team.
Fair enough. On the other hand we are all volunteers and therefore we
all help, where we do care for. Its not about taking over a package for
the sake of being advocated one day, but because you love to work on a
given package you do use daily (ideally).
> which goes two ways - it is just as demotivating for the DD.
Again, fair enough. Bug again: You can't expect to get more helping
hands for Debian as a whole if you scare people off by not giving them a
chance though. Of course not everyone deserves a chance (yet?), but that
should be communicated to prospective maintainers as early as possible.
Its just not fair writing on the Wiki or on similar places about the
sponsoring procedure, encouraging people to read _a lot_ of
documentation (New Maintainer's guide, policy, developer reference,
dozens of resources in the Wiki, ...), letting them work for weeks on
their package(s) and then just ignore them, regardless of their quality
of work, just because DDs made bad experiences with other, totally
unrelated candidates.
I'm absolutely with you if you tell me, not everyone deserves this
attention, but for the sake of Debians philosophy you should really make
sure, you don't scare off people whom do.
> Fly-by sponsoring is *not* helpful to Debian.
True. It neither does for the candidate.
What are alternatives though, if no one else feels sympathy for your
problems and your former sponsors don't react? Again, my case: I'd love
to improve my package's quality (i.e. upload new upstream releases, fix
bugs, port new architectures), in fact I did all of that. I just can't
spread my improvements, as I can't upload, so I spend my time updating
my package on mentors /while/ being in seek for sponsors and waiting to
someone pushing it to the archive.
> I won't return to sponsoring until I am reasonably confident that I can
> allocate enough time to a particular set of packages and their
> maintainer(s) to get each of those maintainers through NM (or DM but
> I'm less keen on DM personally).
Unfortunately this is what most DDs seem to think.
> If the maintainer isn't interested in becoming a DD or DM, then Debian
> shouldn't be interested in sponsoring packages for that "maintainer".
I agree. In fact I proposed to start DM application by the day one is
seeking for a sponsor (i.e. he files a RFS). This didn't find a larger
audience though so I threw away my idea.
I would be even fine if you would /require/ a prospective developer to
proof his skills before being allowed to file an ITA/ITP (e.g. similar
to the NM process).
> The emphasis of sponsoring seems wrong to me currently - it is *not
> about the packages*, it should be about the people.
Partly true. Its about the people /and/ its about the packages they care
for. You would perhaps neither invest so much love in packages you don't
care for, and Debian most likely does not need DDs who just love to see
the "DD" in front of their name. Its more about people being
enthusiastic to help Debian for things they /do/ care, and if that's
some random Gnome theme, for god's sake, let them do if they obey all
rules (and I personally /don't/ think one needs to package every theme).
- --
with kind regards,
Arno Töll
GnuPG Key-ID: 0x9D80F36D
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/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=Cf3x
-----END PGP SIGNATURE-----
Reply to: