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

Re: First call for votes for the Lenny release GR

* Wouter Verhelst (wouter@debian.org) [081231 21:55]:
> On Tue, Dec 30, 2008 at 07:31:10PM +0100, Andreas Barth wrote:
> > I still think we should have someone not the DPL (e.g. the secretary) for
> > the "first call" on intepretation of the constitution, and then have an
> > appeal instance which makes the final decision if necessary.
> ... that could also make sense, I guess. But I'm not so sure it helps:
> Whether we do this kind of thing or not, of course the secretary (or
> anyone, really) is going to be interpreting the constitution on a
> personal level, thereby assuming that his/her interpretation is correct.
> Usually that's not going to be a problem, so we don't need the body.
> Only when there's a conflict are people going to invoke that body.

Well, sometimes it boils down to "can we do it that way?", and
everyone is happy (enough) with the answer. I don't mind if we do the
appeal court in a way that people can go directly to the appeal court
in case we expect things to escalate anyways, but I'd like to have a
lean way in case we just need an authoritative-enough answer. (And I'd
like to see some requirements like "we need Q/2 developers to appeal"
to prevent easy stuff to escalate in case only one developer is

> > This will keep all the "trivial" cases of the appeal instance. (Which
> > doesn't conflict with anything you said.)
> I don't think trivial cases are going to be much of a problem. In any
> case, I was thinking of a voting procedure for this body where the few
> voters would only be allowed to vote "yes" or "no", plus perhaps a
> rationale; we don't need a full condorcet system for interpretation,
> methinks (but I have an open mind on that one...)

A full condorcet doesn't hurt however, see the tech ctte. (Though we
have one bug in case we need qualified majorities, as the constitution
doesn't require 1:n, but 1:n+1, which is an issue in small groups -
but we should fix that anyways.)


Reply to: