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

Re: Bundled votes and the secretary

On Mon, Dec 15, 2008 at 03:49:14PM +0000, Ean Schuessler wrote:
> ----- "Pierre Habouzit" <madcoder@debian.org> wrote:
> > The point is, the secretary chooses interpretations that suits his own
> > proposals to the vote. Explain to me how the "release lenny" options
> > need [3:1] supermajority where the very same vote didn't need it in the
> > past ?
> From a rigorous perspective, the release Etch vote should have been
> 3:1. If we are worth our salt we should not be allowing DFSG
> violations past "testing" and developers should be aggressive about
> filing bugs on errant packages. I can understand the necessity of
> providing certain users non-free drivers to help them get their
> equipment going. Serious users should be selecting equipment that
> won't have install problems. Last time I checked, this was a
> distribution for serious users (that also happens to want to be
> friendly to people just getting started). I fail to understand how
> serious Debian Developers arrive at a point where enforcing the DFSG
> is an exercise for "zealots".

I disagree. What would be 3:1 (to date) is to decide that such bugs
aren't RC. The funding documents don't enforce the release team to
release without a single known DFSG-related issue, unless I'm deeply
mistaken. A $suite-ignore tag is _NOT_ the same as downgrading the
severity of a bug. It's acknowledging it's a serious issue, but that we
shall not wait for it to be solved to release.

I don't say that DFSG freeness is a secondary issue. What I'm saying is
that when:
  * we see DFSG freeness issue that need quite a long time to resolve
    properly and would else delay the release too much,
  * there is no sign of foul play from the maintainer (IOW those bugs
    have not been sneaked into testing, but have just been detected
    after the migration to testing),
Then I fail to see why deciding to release with such bugs needs a 3:1
vote. It's merely pragmatism.

What would be quite arguable, would be the Kernel Team and Release Team
deciding they just don't care about DFSG issues at all. It's not the
case, the firmware front is better at each release. It's just that new
firmwares pop up every kernel release, and there have been new firmwares
that have not been spotted. That's all. The former would be foul play,
and is condemnable. The latter is just a bug, from a specific kind, but
in the end just a bug.

You're welcome to start a GR that we shall not release with a single
known DFSG violation and write it in the constitution (which is kind of
what the 1 vote is supposed to mean in the current vote). But the
project has voted 3 times on similar issues, and it doesn't seem this
opinion is shared by more than a few.
·O·  Pierre Habouzit
··O                                                madcoder@debian.org
OOO                                                http://www.madism.org

Attachment: pgpOoBkYNpAkk.pgp
Description: PGP signature

Reply to: