Re: [council] Rules for Convention-Style Voting

[Amadeu Abril i Abril <Amadeu@nominalia.com>]

Raul Echeberria wrote:
> 
> Amadeu:
> 
> I see your point and I agree with you, but I prefer three votes for each NC
> member in all of the three rounds.
> 
Raul,

There will most probably be more than three rounds...the experiment
goes on until support is coialized around someone getting moe than
half the votes. And then you go gain for the second one.

Ialso first thought about keeping the three votesbut....then it is
nearly impossible to get the simplicity and clear proceeding of a
convention-style system. It is ismposible to get someone elcted with
just half plus one votes, as there are multiple votes for each one.

One exemaple: if three votes are cast, with no preˇfetrematil wieght,
then the toatl number would be 57 votes (19 memvers per three votes).
Majorituy of NC members voting for one nominee mesans 10 NC members
(10 votes). But as each NC member has three votes, we could find
ourselves with four candidates having more than 10 votes.....and the
election would be perfelcy unreasonable. Of we ask the majority of
votes, ie. 29, this would simply impossible to achieve: as each  NC
member has to cast the votes for three differnet reps, then the
maiumum is 19.... No way out.

If we use preferences to solve th9is impasse (one vote for the thrid
candiate, two for the second and three for the first one) then the
toal is 76. Majorituy is then 39- Please note that this means that the
majority of the NC voting for a nominee in first palce eauals 3' votes
(10 x 3). Not enough. So this sysmtem implies supermajority rule....
Is we say that half the NC voting in faovur of one reas the first
candiate meeets the requirement, then the requirment is 3' and, again,
we find ourselves with multiumple cnadidates being elected in a single
round in a "blind" way. Exactly what we were trying to prevent.

Multple votes in all the rounds are beter dealt with systems much more
complesx,close to the one proposed by Dennis. With all the problems
they might bring into it.

This was the reason of my proposed amenment: we use multiple vote once
or twice, so as to identify the nominees with real supprot. Then, with
this group of mnominees retained, we go on to try to get the election
as simple and transparent as apossible: we go voring with one vote, by
ronds in order to get aggregate support for one candidate. We
eliminate the nomineees form the same region and start over, and a
third time for the third regional slot if necessary.

I porposed not to eliminate the candidates "form a certain point" (say
the second round) so, for instnace, for the folliwng reason:

* Imagine that you support Alejandro, but you see that after three
rounds he only got thrree votes. OK, he is "in" but now there are two
candidates getting many more votes: say Dennis and Jonahtan. If we
eliminate "all" candiates with fewer votes form this moment on, you
will keep voting Alejandro, I will keep voting say Nii, and nothing
will happen. Or perhaps all LA candidates will be eliminated...while
we are in fact electing the first seat,m that, let'ˇs say, eventually
got to a Norrth Americanb.
 Under my proposal, when you see that the vote is really getting to be
decided bettween two Eyuropeans or one European and one North
American, you might drop your support for Alejandro, choose the one
amone the most voted cnadidates you prefer best, accelerating the
process of election and being ablot to use your voice to the election
of "htat" first or second seat, while still being sure that your move
has NOT prevneted your preferred coandidate to be voted by you and
others in the next round, for the next seat.

If we don't do somehting like that, nobody will ever change the vote
between rounds until his candidate gets eliminated (which is the
regular urle) but then it will be more difficult to knwo how to select
candidates for other rgions.

Just my contribution ;-))

Amadeu