Re: changed charter, new elections proposal

[Bart Decrem <bart eazel com>]

I think we should do as Allen suggests: let's keep things very simple this
year, and then reflect on how things went and tweak the system for next year.
Frankly, I suspect our problem in this election will be to get enough
candidates to nominate themselves and to get decent voter turnout.  Keeping
things simple, well, keeps them simple, and encourages voter turnout.

So I think we should just:
- let everyone vote for up to 11 candidates
- not publish interim election results or hold more than one round of
elections
- keep the mailing list archive closed until the end of the election
- if there's tied votes for the last slot, the candidates whose votes are tied
should have the right to ask for a run-off election, but they should also have
the right to allow the already-elected board members to resolve the tie (since
that would simplify thing - I am not looking forward to the idea of run-off
elections between the  two candidates who had the 11th highest number of votes
- that will not be an exciting event).  In otherworse, unless they all agree
to let the other board members decide, there'll be a run-off.
- the charter has provisions for dealing with cases where more than half of
the elected members are affiliated with a company.  Basically, only the 5
folks who work for the company that have the highest number of votes would be
considered.  Let's leave it as is.  Sure, as Maciej points out, it's a high
treshhold, but it's an emergency provision, so I think it's fine.
- I disagree with Russell's proposal below.  Setting aside specific slots for
specific types of representation would be a mess that's an order of magnitude
bigger than having slates.

No-one has commented on the proposed timeline.  Let me know if you have
concerns about that.  Meanwhile, I'm going to start writing up election
documents.

Also, I know there's a membership committee that was created by the steering
committee to oversee the members registration process.
Havoc: who's on that?
I propose that we turn it into an election committee that would oversee
membership registration but that would also monitor the election, announce the
results, and propose resolutions to election logistics issues that we bump
into.
Russell, can I volunteer you to join that committee?

Bart

Russell Steinthal wrote:

> On Thu, 28 Sep 2000 16:40:21 PDT, Bart Decrem wrote:
>
> >Hi everyone,
> >
> >I just modified the charter, removing the slates system.  I think this
> >is by popular demand :)
> >
> >Check it out at http://www.decrem.com/gnomefdn.html.  There's a change
> >log at the end.
> >
> >Here's a new proposal for how to hold elections.
> >
> >1.  Proposed election process
> >------------------
> >- The charter provides for 7 to 15 board seats.  For this election, I
> >propose that we have 11 board members.  I picked this number because
> >it's odd, and large enough to allow a broad mix of people, but hopefully
> >still manageable.
> >- I don't think we should designate specific slots for specific types of
> >representatives.  If we try to set aside slots for "documentation
> >person" or whatever, I think we're going to end up with a logistical
> >nightmare and flame wars that make the debate about slates look like a
> >joy.  It would get hopelessly complicated.  The reasons for this have
> >been discussed at length on this list in recent days.  People should
> >just vote for people that they think will best represent GNOME in all
> >its diversity.
> >- I propose that candidates nominate themselves by sending email to a
> >list to be created by Havoc with a few sentences about their involvement
> >with GNOME and what they would add to the board.  There'd be a two-week
> >period for people to nominate themselves, and then we'd publish a list
> >of candidates (see timeline).
> >- I propose that everyone be allowed to vote for up to 11 candidates.
> >So everyone gets to elect their dream-team.
> >- Havoc will write a script to help tally the votes.  Votes would be
> >held by sending email to vote gnome org (or whatever).  Havoc would come
> >up with a ballot format that could be processed by his script.
>
> Although I am certain it will add to the complexity of the process, I think
> we may need to consider allocating board slots for specific comp
> onents of the project/region; if we don't, we're not going to achieve
> the desired result.
>
> I think it is quite likely that the top vote-getters in any given
> poll are going to be hackers, probably even the current steering
> committee members (should they stand for election).  They are simply
> the most prominent members of the project, and their names are most
> known by the members of the project.  Even if there were a concerted
> effort to elect a documentation representative, or a translator, I
> doubt they would have the name recognition needed to crack the upper
> tier of vote-getters.  If there are more than one of a given type in
> the election, the chances of their votes being fragmented seem high,
> and they are not going to have a huge margin to work with.
>
> This is one of the consequences of a direct democracy- even if
> everyone who reads this list actively decides to consciously vote for
> a certain number of non-hackers, I wonder whether that sentiment will
> filter through to the rest of the membership.  The slate system,
> while flawed, would correct for that by essentially allowing the
> "active" membership to pre-screen candidates through the slate
> creation process.  (Yes, that is anti-democratic, but it would have a
> desirable result.)
>
> I think that geographic diversity is likely to be more happenstance:
> Miguel is certainly going to be elected, but does he count as from the
> U.S. (where he works) or Mexico?  Trying to mount a campaign to elect
> a European (for example) will, I think, run into the same vote
> fragmentation problem I mentioned earlier, unless one of top hacker
> -vote-getters is from Europe.
>
> In that light, I think that if representational diversity is
> important, we may need to bite the bullet and devise a "seat" system:
> that's only feasible, however, if we make some compromises like
> reserving seats for "non-code contributors" rather than documenters,
> translators, user helpers, etc. all separately; similarly for,
> "non-Americans" (or non-U.S.'ns, as the case may be), rather than
> seats for Europe, Asia, Africa, etc.  Otherwise, we're not going to
> have enough available board slots for at-large voting, or we're going
> to have a board that is so large as to be ungainly.
>
> Just my thoughts...
>
> -Russell
>
>
> --
> Russell Steinthal               Columbia Law School, Class of 2002
> <rms39 columbia edu>            Columbia College, Class of 1999
> <steintr nj org>                UNIX System Administrator, nj.org