Re: GNOME Foundation elections!

[Bart Decrem <bart eazel com>]

Hi Daniel,

The slates idea is rather old by now.  It was part of drafts of the charter
way back in early July.  We've had a number of  discussions about it on this
list (see, for example, this thread:
http://mail.gnome.org/archives/foundation-list/2000-July/msg00309.html) and I
definitely felt there was consensus around that idea.

The main reason to have slates, instead of electing individuals, is to end up
with a team for the board that makes sense and has a number of checks and
balances.  For example, some people may feel that it's important to have
representation from Europe, US, and other regions.  Others will want
representatives of the core GNOME building blocks (GTK, Bonobo....).  We will
want to balance out corporate influences and make sure there's critical mass
of people who aren't beholden to a corporate employer.  We'll want to have a
few consensus builders, and some folks who like to argue about things.  If we
elect a bunch of individuals, we are going to end up with a really random
collection of people and it's going to be very hard to have the board
function effectively.    Maybe it helps to thing of this less as a Congress
for GNOME and more as an executive branch that really needs to function
effectively as a team.

More comments below...


Daniel Veillard wrote:

> > - October 16 (Monday): Deadline for submitting slates for the
> > elections.  Deadline for making changes to the charter.
>    Ouch only 2/3 weeks from this preannouncement to have all
> the slates constitued.

I was trying to keep things moving forward at a reaonsably fast speed (note
that it'll still take 1.5 months to get the elections done), but if the
timeline is too compressed, or should be arranged differently, that's fine
with me.


>
>    Why slates too ? Can't this just be an election of individuals
> especially with so little experience and notice. It's a congress
> why not have a list of runners and let people pick X names from
> the list and declare elected the X most cited persons ? this is
> direct democracy and limits the political aspects.
>
> > 4.  Putting together slates.
> > As outlined in the charter, registered voters will elect a slate of
> > board members and there is a process by which members can submit
> > slates.  Hopefully there'll be a number of slates from which
> > GNOME hackers can choose. I believe that the GNOME Steering Committee
> > will also submit a slate for the board.  Hopefully, slates will be
> > submitted with a note explaining why the proposed board composition will
> > best meet our needs.
> >
> > 5.  Election mechanics.
> > There are 2 ways to have the elections:
> > - "the basics": we set up a mailman list, auto-subscribe all the
> > registered voters to the list, limit posts to the list to members of the
> > list, and then ask everybody to send email to the list address (let's
> > call it vote gnome org) with a slate number as the subject field.  When
> > the election is closed, we can then do a quick check to make people
> > didn't double-vote, and tally the results.  This is very simple, very
> > open (anyone can inspect the ballots), and no machinery is required;
>
>   yes there is a machinery, at another level, how do you plan to
> get the slates made ??? Surprizingly the Steering Committee will
> submit a slate, sounds a bit fishy to me, sorry !

Slates are put together by people who feel strongly about this.  For example,
perhaps you want to put together a slate that you feel would best represent
the GNOME community.  Since the Steering Committee has been working as a team
for a while now, dealing with the same issues that the Board will deal with,
it makes a lot of sense to me that they would submit a slate (I think I have
proposed that on this list in the past).  Of course, if their slate doesn't
reflect the desires of the hackers, it won't get elected.  Nothing fishy
about that.


> > - "the fancy way": people have expressed the desire to have a voting
> > process where one could vote for more than one slate, with a first and
> > second preference.  To explain this: take the US presidential
> > elections.  Wouldn't it be great if people could vote for Ralph Nader
> > (or whoever) and, if Nader doesn't get enough votes, then your vote goes
> > to Gore or Bush.  This encourages people to state their real voting
> > preferences as opposed to voting 'strategically', but it requires more
> > complicated machinery. If someone wants to take the lead to identify
> > software that can help with this, or come up with a manageable way of
> > tallying the votes under such a system, then we can go this route.
> > Otherwise, I recommend sticking with the basics for this election
> > cycle.   We would need to make a decision on this over the next week or
> > so, since it might affect what kind of slates are put together.
> >
> > So rather than having a heated debate about this, if someone is really
> > excited about going the fancy way, please find software or propose a
> > concrete way to do this within the proposed timeline.
>
>   I hate this slate thing. Basically if there is 4 candidates I would
> like to see elected, but they are on different slates I can only vote
> for 1 (or 2 in the second case) of them. Damn that's frustrating espcially
> if X is like 10. Or do you expect all the "big names" to be in the same
> slate ;-)
>   You also didn't explained one of the crucial aspects of the vote,
> Suppose Slate A has 200 votes and Slate B 100 votes. Is the resulting
> board made of 2/3 of slate A and 1/3 or Slate B (which requires ordering
> in the slates then) or Slate A is declared winner alone. in the second
> case and assuming the Board propose a slate, then I feel it's simply
> not fair. All bets would immediately go to this slate, drop the elections
> in that case and autoelect the board we will all loose less time.
>

I don't think merging slates based on election results makes sense for the
reasons described above.  Personally, I am in favor or the simpler election
process ('the basics').


>
>   So I think that there is a lot of thing still in grey areas, and
> a 2 week deadline sounds really really short !
>
> Daniel
>
> --
> Daniel Veillard w3 org | W3C, INRIA Rhone-Alpes  | Today's Bookmarks :
> Tel : +33 476 615 257  | 655, avenue de l'Europe | Linux XML libxml WWW
> Fax : +33 476 615 207  | 38330 Montbonnot FRANCE | Gnome rpm2html rpmfind
>  http://www.w3.org/People/all#veillard%40w3.org  | RPM badminton Kaffe