Re: New-Maintainer
On Jan 03, Nate Duehr wrote:
> I've been spending some time reading carefully through numerous Debian
> documents tonight and looking at the website. I'm one of the many who
> has been waiting patiently for new-developers to start up again, and
> have voiced some displeasure from time to time on a couple of lists, but
> I understand that doing that doesn't really accomplish anything.
[...]
> As far as #1 goes, that was done by the Project Leader and he's well
> within his Debian Constitutional rights to do it.
I don't think the DPL himself chose to halt taking new maintainers.
My interpretation of what Joey and James have said is that they both
had concerns about NM that weren't being addressed, and they stopped
taking applications because of that.
> #2 is a little bit trickier, but still gives appearances of being "okay"
> on the outside. When tied to #3 however, it has grave implications.
>
> #3 means that those of us either being sponsored, or as in my own case,
> one who's already contacted a developer and asked to be sponsored, are
> in a precarious state. We can contribute, but we have no voting powers,
> no say in what happens to the overall project, and (gasp!) NO FREEDOM
> within the project. Somehow this seems like the one thing that is being
> glossed over in the setup of this pseudo-maintainers group via the
> sponsorship program. We do NOT have the right to have a say in the
> project as a whole. This is the NUMBER ONE reason I'm not going to
> pursue my sponsorship any further. I don't WANT to be involved in a
> project that won't have me. (Unlike Groucho Marx... :) )
Sponsorship is sort of an *ad hoc* system to try to harness the
interest in maintaining packages. It is not intended as a long-term
replacement for being a maintainer, but it could form the basis of
some sort of apprenticeship once we begin taking applications again.
I believe Wichert's proposal includes an "internship" period that
would be similar (see below).
> #4 I could be wrong on this one, but I think it does mean that bug reports
> go to the wrong person, and it's not very easy for the actual package
> maintainer to work within the bounds of the BTS and other Quality
> Assurance programs. Instead, bugs go to the uploader (again, I think)
> and the normal system isn't used. This certainly can't lead to anything
> good as far as package quality goes. It sure is going to be strange for
> people to get responses from the uploader who says, "Hey, let me get you
> in touch with the right person to fix this." If this is true, it's just
> not smart to be doing these sponsorships. Sure, people being the good
> folks that they are in Debian, this one has a chance of working itself
> out, but why run the risk? Sponsored maintainers also have no access
> to the various machines owned by SPI or any of the tools or abilities
> that access to those machines might allow.
The bug reports will go to the actual maintainer, not the sponsor, if
the control file is set up properly. The sponsored party is listed as
the "Maintainer" for dpkg purposes; they just can't upload the package
themselves.
> #5 and #6 Folks, if you're being sponsored, you really should get on this
> webpage. Perhaps if the TRUE size of the list of sponsored packagers
> were known, it would be more important to someone. Whoever that may be.
> You never know, there could end up a LARGE percentage of the developers
> doing work for Debian doing it as sponsorships, and there'd be no
> evidence of it anywhere. SIGN UP!
No argument here.
The page is: http://www.internatif.org/bortzmeyer/debian/sponsor/
> #7 Probably the most controversial thing I'll say here, is that I feel
> that if the Debian project is truly OPEN, then the developers reference
> should reflect the TRUE state of the project. Chapter 2.2 should have a
> note in it about the current state of things, and the above URL
> reference for those interested in being sponsored with a disclaimer that
> the rest of Chapter 2.2 will again soon be valid. Of course, coming
> from me that's just an opinion, but any developer could make the
> request to have it added and start the formal process. Or the
> maintainer might read this and just decide to do it himself?
Well, since what we have now is an "ad hoc" procedure, I don't think
it belongs in the reference. Perhaps someone could write up a
document about "How do I contribute to Debian without becoming a
maintainer", maybe?
> Why have Debian's developers allowed new-maintainers to be closed for
> so long. Many are willing to blame the leadership, but in my
> reading of the Constitution, it would appear that ANY developer with a
> good idea on how to fix new-maintainers could have proposed it at ANY
> time and if passed by a large enough margin, it would have been done.
> Of course, here in the real world, you'd need to work with the
> new-maintainers crew to keep them somewhat happy or you'd be finding a
> new crew to take up your newly proposed "fix", but that's just normal
> politics. Welcome to dealing with people!
[...more...]
I think the reason is that there are very few people who are trusted
with this job; it is after all very sensitive: not just security-wise,
but also politically. Another reason may be that it puts the new
maintainer people in an awkward position; they've certainly gotten a
lot of flamage (including some from me) that has been over-the-top,
and I can't see anyone wanting to put themselves in that position...
Anyway, this is perhaps all moot since the "word" is that we're close
to reopening the new maintainer queue, though I haven't heard anything
about whether or not Joey and/or James will be participating, and/or
who else will be involved. The plan is to follow the procedure
outlined in this mail message:
http://www.debian.org/Lists-Archives/debian-project-9910/msg00003.html
It may be amended somewhat by now; talk to Wichert if you have any
particular suggestions.
Chris
--
=============================================================================
| Chris Lawrence | It's 2/3 of a beltway... |
| <quango@watervalley.net> | http://www.lordsutch.com/tn385/ |
| | |
| Debian Developer | Are you tired of politics as usual? |
| http://www.debian.org/ | http://www.lp.org/ |
=============================================================================
Reply to: