[EM] language/framing quibble

[Kristofer Munsterhjelm]

Raph Frank wrote:
> On 9/10/08, Kristofer Munsterhjelm <km-elmet at broadpark.no> wrote:
>>  If duplicate votes don't count, then there'll be a natural incentive to
>> pick friends instead of central party figures. All campaigning would do
>> would be to give whichever candidate's being promoted a lot of votes, which
>> is no better than the candidate in question getting a single vote.
> 
> That is interesting.  You could still vote for someone who is 'semi' famous.
> 

Also, if the voters don't know about this, the famous people will be 
elected anyway. In that manner, it's not stable. What could happen in 
reality is that there are some "honest" voters and some tactical voters, 
and thus there'll be a subset that almost never changes for successive 
final assemblies.

> Another option would be to pick 200 people for a 100 person
> legislature, but include an ordering.  Everyone who is nominated makes
> it into the 2nd round.  However, your position in the queue is
> determined by the position of the first person to order you.  For
> example,
> 
> person 1 nominates person A
> person 2 nominates person B
> person 3 nominates person C
> person 4 nominates person B
> person 5 nominates person D
> 
> The order for the 2nd round is then
> 
> A
> B
> C
> D
> 
> and the total length of the list is shorter.

I guess this could be iterated until you find a stable state. The first 
approximation uses a random order of positions for the other candidates 
(to answer who is person 1). Then the list is reordered and the 
algorithm is run again.