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

ajt's anti-rebuttal

Hey *,

So as per last-year, I think it's interesting to accompany my rebuttal
[0] with an "anti-rebuttal" and summarise the things from the other
platforms that I like. Given there are so many candidates and some of
them list whole swathes of things I like, I figure it makes more sense to
do it in a separate email than make my platform a million pages long. I
also thought it might be interesting to do it from the point-of-view of
what I've already done to support some of these ideas, rather than just
make promises of what I might do if elected.

From Gustavo's platform:

  * Bug tracking system features

Like Gustavo, I've had a bunch of wishlists for the BTS; and thanks to
the support of Darren Benham back in the day, I've had the opportunity to
make many of them happen including bug archiving, dynamically generated
bug pages, tags and usertags and other neat stuff. I presented a talk
at debconf 5 giving an introduction to hacking on the BTS, and I'm very
please to have had the good fortune of being able to help other people
get involved, including Colin Watson (version tracking), Don Armstrong
(version graphs), Pascal Hakim (bug subscriptions), Blars Blarson
(spam prevention) and Steinar H. Gunderson (version tracking for the RC
bug list).

  * New developers

In late 2005, I volunteered as an AM, and was lucky enough to have
Holger Levsen as my first applicant, whom I'm glad to say is now
a developer. There's definitely some things I'm not happy with --
particularly how long it took overall, and that Holger spent quite so
long hamstrung in how he could contribute -- but overall I think it was
a rewarding experience for both of us.

  * More official events

I was lucky enough to be at the first Debconf (DC0 in Bordeaux,
France, in Y2k), and to be in a position to support James Bromberger's
inspiration to start the trend of miniconfs at linux.conf.au with a
Debian miniconf in 2002, and I've given talks at both those events
as well as many of their successors, though happily I've so far been
able to avoid actually organising one myself. More recently I've tried
to provide some moral support to other groups setting up local Debian
conferences and representing Debian in other fora, too. Unfortunately,
though, we're still a ways off having an official DebConf in Australia.

  * Debian publicity

At the press BOF at DebConf 6, I made an effort to get some of the
concerns and frustrations people had with the way Debian was being
promoted (or, rather, wasn't being promoted) focussed in a positive
direction -- and that ended up being to create a parallel forum for
communicating Debian's achievement in the form of an open debian-publicity
list. Since then we've also had Debian Times established and a get
together in Germany for some further planning on how to best promote
Debian. I've also tried to be accessible to press folks looking to
do Debian-related stories, although I've been less successful at that
recently than I'd hoped; and I've also tried to encourage the press to
talk to other people in the project than just the DPL, though with very
limited success so far.

From Sam's platform

  * More teams / Uploaders field

The Uploaders field that Sam raves about was James Troup's creation
back in 2001, when looking for a way to be able to accurately work out
whether an upload was an NMU or not so that dak could work out whether
to close a bug or just tag it fixed [1]. I've no idea if I was involved
in that change, but I was contributing to dak in the same period.

  * Bigger teams

While I tend to see more downsides than upsides to forcing people into
existing teams, I definitely see benefits to having more people working
on problems. Apart from the wonderful increases in participation in the
bugs team, I've also worked on expanding the release team, which was just
Richard Braakman when I volunteered to help prior to the potato release,
and then was just me until Joey Schulze took over managing updates to
the stable release a few months later. Eventually I made the effort to
recruit and train some release assistants in 2003, who took over from me
in 2004, and that format has continued since then. Joey passed on the
stable release baton last year, and there's now a team behind stable
release management too.

  * Sexier BTS

Sam thinks our BTS is "ugly and hardly usable" because of the wnpp
pseudopackage. I can't say I disagree about that particular package, but
that's more because I simply don't think a pseudopackage is a useful way
of tracking those issues -- using usertags and assigning them directly
to the package concerned could be one approach, and using an entirely
different tracking system might be another. In any event, we've already
had some major efforts at making the BTS prettier by Erinn Clark and
myself a bit over a year ago, and more useful by the entire debbugs team
on a regular basis.

  * Sexier distribution / Packages-xx.gz files

Translated packages files have been a wishlist for years now -- from
at least back in 2002, according to my mail logs. That discussion
(including Michael Bramer, Jason Gunthorpe and myself) decided on the
Packages-xx format. That was then brought up again by Michael Vogt in
October '05, and after some more discussion resulted in an upload of
apt to experimental with support for translated descriptions, and some
test translations for unstable included in the archive in May. AFAICS, that
doesn't seem to have actually been announced at any point, though maybe
I'm just not looking for the right thing. The draft was:


From Joey Hess's anti-platform: 

(C'mon: you couldn't possibly expect me to resist an anti-rebuttal for an
anti-platform could you?)

  * Constantly usable testing

Personally, this has been my goal for testing ever since the idea began
[2], and I've been trying to do what I can to help keep that happening
since, most notably I guess in helping the RMs maintain testing and
helping the testing-security team be more able to work with the stable
security team.

  * Double the archive sync frequency again

Well, I helped get it doubled the first time. :)

Hrm, I dunno, maybe vague promises were a better idea. Using "I" all
the time gets old, and ultimately all these sorts of things are going
to end up getting done by someone who feels inspired enough to put in
their own time, not by fiat from whoever's DPL.


[0] Yes, I've only just gotten around to writing it now. I'm slack, I know.

[1] http://lists.debian.org/debian-tetex-maint/2001/06/msg00140.html

[2] http://lists.debian.org/debian-devel/1998/05/msg01695.html

Attachment: signature.asc
Description: Digital signature

Reply to: