On Wed, Feb 09, 2011 at 05:07:25PM +0100, Ana Guerrero wrote: > > OTOH, on a complex project, I know about the level of involvement some > > people have on some topics. I don't know about many others. And yes, I > > do let other voters' votes influence mine. > > Yeah, but you never know if other's know about that person better than you.
I'm not really convinced about the "I can not see other raters opinion" option but I could live with this. However, your pro argument is definitely wrong. Strictly logical speaking if you say "never" one counter example is sufficient to prove you wrong: I personally know that Zack is involved in Ocaml team and I have no idea at all who does a good job there or not. The person I should vote on claims to be a member of Ocaml team. So I would definitely look at Zacks opinion about this person and I think this is a reasonable (but not necessary step to do). So there are cases where I know that other persons will know better than me and there are lots of cases where I even have the competence to decide who knows the person better than me (for instance I would not look so deeply at your (Ana) vote than on the vote of Zack - it would be different if the applicant would claim to be a KDE packager). > > Probably we should rather make explicit that if you don't feel > > confident about rating somebody, you should refrain from voting (for > > or against). Then, a vote of 0 would mean "nobody knows this person", > > a negative vote would mean "he is mostly a liability", and we would > > just sponsor the positive levels. > > Yup. If there is an explicite way to say "neutral - no vote at all" I'm fine with this. But this should be distinct from vote 0 (which could exactly mean 0 and nothing else). > On a related note, maybe the travel sponsorship commitee wants to start > already working at the best method for ranking this years before > they get the list of candidates? Yes. Kind regards Andreas. -- http://fam-tille.de _______________________________________________ Debconf-team mailing list Debconf-team@lists.debconf.org http://lists.debconf.org/mailman/listinfo/debconf-team