On Sun, Mar 05, 2006 at 06:06:18AM +0100, Josselin Mouette wrote:
> 1. One of your major goals is to release more often. After being
> involved in the longest release cycle ever, what are your
> proposals to make this cycle shorter?
First, I wouldn't say that "release more often" is one of my major goals;
in my platform I cited "release early, release often" in aid of the goal
that things should *happen* rather than being put on hold until they're
done perfectly, and that affects Debian not just for stable releases,
but also for uploads to unstable, development of policy, consideration
of important issues like the GFDL or trademark policies, resolution of
tasks through the technical committee, and otherwise.
As far as stable releases are concerned, my approach will continue to be
to support the release team as best I can; I have complete confidence in
Andy and Steve's ability to manage the release process for a December
conclusion, and have absolutely no intention to interfere in their
decisions in any way.
I do think it would be interesting for the project to embrace the d-i beta
releases and the testing-security support and turn those into regular
"mini-releases", without many of the standards we expect of stable,
but in a form that's still useful.
> 2. Everyone has his own character. However, a representative has to
> be cautious to avoid compromising the project as a whole. Were
> you elected, would you make efforts to stop being contemptuous
> in public communication, at least when acting as DPL?
When acting as DPL, certainly -- and I would hope the issue wouldn't
come up in the first place. If you've got any examples of any of
my communications signed of as "release manager" that you think are
contemptuous I'd be interested in discussing them with you, to make sure
I can avoid anything similar.
As far as "contemptuous" is concerned, I don't really think singling
me out like that is particularly fair. Our current DPL posted "Frank
Carmickle and Marco Paganini must die" to d-d-a about six months
before being elected, eg; and plenty of other people in the project act
similarly. Personally, I do rate getting things done more important than
being nice about it -- and I shouldn't think you'll find it hard to find
examples where I'm not nice -- but it really does frustrate me when we
can't have both.
Let me summarise it this way: I'd absolutely _love_ to have people helping
me be more fun to get along with, just so long as others are willing to
make the same commitment. I don't really care whether that's friendly
advice, or backed up by some threats, or something else -- just whether
other people do it too.
In my defense, I have stood up for people in the past in ways I think
have reflected well on Debian, such as adding a moderating voice in the
Nexenta conflict [0].
> 3. If accusations are made against during the DPL term, do you
> intend to abandon this position, like it happened in the past
> for the Release Manager position?
I don't think I "abandoned" release management: I left the leadership
position in Colin's and Steve's hands, which I think were perfectly
capable, and continued (and continue) to be involved in assisting the
RMs, primarily at a technical level in helping with transitions, but
also on things like the architecture release qualification process.
The reason I stepped down as release manager wasn't because of any
accusations per se, but rather because I felt that I lacked support
in continuing to act as release manager. That impression was built up
both by the level of disgruntlement from both sides of the debate [1],
and the lack of interest in either supporting my actions in response
to the changes, or overruling them by either the DPL or the technical
committee [2]. That this was then followed up by a proposal for a GR to
force amd64 not only into the archive but also into sarge convinced me
that I wasn't likely to be able to much good in continuing.
I'm not terribly happy with what happened, but I'm not remotely ashamed
of handing the RM position over to Colin and Steve. I think the LWN
article [3] following Steve's note that I'd stepped down was my only
public comment on the matter.
If it's not already obvious, the above's why I think that support from
the DPL (and others) is worth stressing, both in my platform last year,
and in applauding Branden's response to the sarge release last year.
The DPL is in a different position though; support for the DPL comes
in the form of a vote each year. If you win, you've got support, if you
don't, you don't. In theory, the DPL team concept should be another leg
in ensuring the DPL is active and confident in carrying through on his
(or perhaps someday, her) goals. In practice, I'm not sure that worked
last year, so maybe not.
I suspect that it has at least worked in the other direction, that is
I'd expect simply being named as part of the DPL team have given Steve
and Andreas more confidence in helping the project than they might have
had otherwise.
Cheers,
aj
[0] http://lwn.net/Articles/159248/
[1] With those in favour of the changes telling me I had been violating
the social contract and incoherent in claiming "software" only meant
"programs", and was acting against freedom. Meanwhile those against
the changes would claim that I'd been derelict in my responsibilities
by not making these concerns clear enough before the vote.
[2] That might be unfair; the GR happened at the end of April, and
I authorised the ctte to consider it within the next couple of days,
debate amongst the committee then stopped at the 4th of May without
any resolution, being overtaken by the md5sum issue, and not being
restarted until the 1st of June, at which point I was adopting the
small target strategy of "do whatever you like, and please don't
blame me for it anymore".
[3] http://lwn.net/Articles/97528/
Attachment:
signature.asc
Description: Digital signature