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

Re: Developers vs Uploaders

On Thu, Mar 15, 2007 at 01:02:15PM +0000, MJ Ray wrote:
> The sponsors who have enough time to make such silly claims should be
> spending more time checking their sponsorees packages!  I wonder if
> some of them are sponsoring because they think NM is needlessly slow,
> rather than because they want to do a proper education/mentoring task.

  Well at first it is. One of my main sponsoree is Fathi[0]. I sponsor
him for quite a long time now, I'd say a year at least. The beginning of
our relationship was indeed really a teacher/student one. I reviewed his
first packages with a lot of care, discussed with him many details,
helped him to understand some debian packaging tricks, and so on.

  But for now something like 6 months, I've nothing to say wrt his
packages. I merely do cosmetics remarks, he knows his stuff. He _is_ a
goddamn very valuable Debian Contributor: very good packages, excellent
senses of his responsibilities. Sadly, he is stuck in NM[1] because the
process is a tad tedious, that he finds the whole process repulsive, and
that I'm reliable enough to upload his packages in a timely fashion
(those are guesses, he can correct me if I'm wrong).

  So for the last 6 month, my sponsoring work is mostly beeing a build
and upload machine. That's quite a waste of my time (since my reviews
have never triggered any valuable flaw), and of his (because he has to
ask me for the reviews, wait for my answers, and cannot work as fast as
he could because of that).

  Here, there is no way of fast-tracking him, because you know,
fast-tracking is kept allot for prominent people, and ... well, I guess
you never heard of him before. So he's stuck in NM, has no right to
upload his packages alone, and NM is clearly inadequate for him, whereas
he is skilled, knows how to limit himself, and only work on things he
know he will able to cope with later, and knows when there is a thing he
does not understand/controls correctly, and go ask questions to people
that know, to avoid mistakes. His understanding of debian's philosophy
is what we expect from NM's too. He's just ready. but no, he continues
beeing my sponsoree, for the time being, and IMHO for still a long time,
because he cares way much more about the quality of his packages, rather
than having his @debian.org address.

  This kind of people is the kind of developers I would be glad to meet
more often in our project, and we are completely unable to make them
full DD in a decent amount of time. That's a shame.

  At least, with the "DM" proposal, he would not depend upon me anymore,
except for the introduction of NEW packages. That would allow me to take
new sponsoree, that would need the "teaching" part I quite like to do,
and that Fathi no longer needs for months.

> Sponsoring is probably something that needs more guidance.  Each
> sponsor has their own habits and there's little to say what's the
> best. Personally, I think the sponsor should appear in Uploaders
> @debian.org to be obvious if it goes wrong, but I think others
> disagreed last time I wrote that.

  Well the sponsor already appears on DDPO, so it's not hidden. I'd
rather create a Sponsor: field than an Uploader, or we need to rework
Maintainer field.

  I mean, the current semantics is:

Mantainer: <-- the person in charge (often an ML when it's a team)
Uploaders: <-- the detailed list of co-maintainers or subset of the team
               really in charge of the package.

  I dont really see why a sponsor shall be in Uploaders, I sponsor a lot
of package I never use, and I don't want to be in either of the fields.
Though a Sponsor: field could be of use, yes.

  [0] http://qa.debian.org/developer.php?login=fboudra@free.fr
  [1] https://nm.debian.org/nmstatus.php?email=fboudra%40free.fr
·O·  Pierre Habouzit
··O                                                madcoder@debian.org
OOO                                                http://www.madism.org

Attachment: pgpFkx6mC0zUU.pgp
Description: PGP signature

Reply to: