On Sun, Mar 20, 2011 at 4:09 PM, Holger Levsen <hol...@layer-acht.org> wrote: > How does this comitee decide? (I think we are getting there, I just want to > point out that "the comitee decides" is just slightly better than "a decision > is made". Both are too unspecific and lack details, like > timelines+timelimits.)
As said elsewhere, it should come down (at worst) to a non-private vote using Debian's standard procedures. Since it's typical to have two bids, the voting can be carried out immediately on IRC. Beyond that, I don't think we can really define precisely how people should decide their votes: if we can come up with clear agreed criteria, those criteria should already be used in the stage where we try to reach a decision by consensus. So if things reach a vote, it's because whatever criteria we had thought of already turned out not to be clear enough. The point of the vote is then to make a decision between the different interpretations of our criteria. We already tried in the past to go to the extreme of clearly defined criteria, ordered weighted options, and numerical scores for each point, but that didn't really solve any problems, and made some of the arguments worse, as teams felt they had to argue about the precise numerical scores or about the precise weighting of different points. If the bids are close, this doesn't really get us anywhere though, as we've shown we *can't* agree on this precise level (and if they're not close, it's a waste of time). Instead I argue we should discuss the priority list points more informally, seeing what is good and bad, then openly discuss people's thoughts on the bids' overall merits, trying to reach consensus as in past years. Then, as in past years, we may need a vote, but now that possibility is defined in advance, rather than us making up the procedure as we go along. Of course, since the vote will only make a difference if the bids are a similar standard, anyone who don't like the idea can just think of it as a random number generator that gives *some* decision between similar bids. ;) -- Moray _______________________________________________ Debconf-team mailing list Debconf-team@lists.debconf.org http://lists.debconf.org/mailman/listinfo/debconf-team
