[MG] Votespace social map - layout stability

[Michael Allan]

Thomas von der Elbe wrote:
> Ok. But in the rarer cases, where a voter reaches the top20 of his 
> candidates voters he would replace the former 20. voter, right? ...

Yes.

> ... BTW, will we have a place for "others" in the circle too,
> i.e. voter 21, voter 22, ... and one for "me" if I'm not among the
> top20?

Definitely for "you".  I'm not sure about the "others", yet.  They're
unimportant by definition, so maybe they can wait till we have
something running.

> Another thing: I feel, it might be better to give the positions not
> according to the name of the voter. The only advantage I see, the
> easy recognition across different polls, will probably not work many
> times, because the places will be taken. Then people might even give
> up to look for him, because they think, he always has to be the
> "south-guy". If we give this up, we can give the first voters always
> the positions rectangular to the tangente of the circle, which looks
> more beautiful imo. And if they grow in number, they open up to half
> a circle, like a flower. :-) Also it looks more like real people: if
> they want to talk to their candidate, they will all try to be in
> front of his face, as long as there is room there...

More here: http://zelea.com/var/cache/irc/_v/2011-04-16/Nmetagov.html

Consider: http://zelea.com/project/votorola/a/crossforum/vote/_/3.xht
Thomas is suggesting that An's initial voters would always be
positioned where Ar. and Ur. are.  You'd never see a voter in Fr's
position till the whole arc was filled.

Staggered fanouts will still be common, I think, even if we use
Thomas's rule.  Most won't show absolute gaps like Fr's fanout.  Most
will be filled with mosquito voters, like An's, but the *substantial*
voters (Fr., Go., Ea. etc.) will still end up in staggered positions.
The only difference will be that *initial* voters aren't placed in
awkward, lonely positions.  So I think it's a good rule.

(See all those mosquitoes around An.?  They're too small for even
Fr. to pay attention to.  But Fr's own voters, and Go's, Ea's, etc.,
will go hunting amongst them, trying to pick up what they can.  They
won't passively wait for the voters to come and find *them*.  Their
hunting toolkits may include general visual displays, like these, but
maybe also specialized ones.)

> ... And then I would stick to the rule, The_User has proposed: if a
> voter withdraws his vote, leave the position open for his return
> until all others are taken. What do you think?

I can easily store the last vote shifter and his old position in the
poll's runtime state.  In most cases that'll prevent an accidental
vote shift (away and back) from changing the position.  You'll get the
old position back.  Later we can improve on that, if need be.

-- 
Michael Allan

Toronto, +1 416-699-9528
http://zelea.com/



Originally posted to the mailing list of the Metagovernment Project:
http://metagovernment.org/mailman/listinfo/start_metagovernment.org