[EM] A different "strategyproofness" notion

[Jonathan Lundell]

On Oct 19, 2009, at 4:34 PM, . wrote:

> er, not "if they are the frontrunners", but "if they are PERCEIVED  
> to be the front-runners", which was the whole point of the  
> experiment. :)

Right. Strategy must be linked to knowledge (or at least conjecture)  
about the behavior of other voters. A voter with zero knowledge (of  
other voters) ought to vote like a dictator, regardless of method. I  
think. That may not apply to approval, since there isn't really a  
strategy-free approval vote.

>
> On Mon, Oct 19, 2009 at 16:33, . <thebrokenladder at gmail.com> wrote:
> no, this is EXACTLY the strategy my simulation uses. if X and Y are  
> the frontrunners, and you prefer X to Y, then you give X and  
> everyone you like better than X a "10", and Y and everyone you like  
> less than Y a "0", and everyone else a sincere normalized utility.
>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20091019/eec29a0d/attachment-0003.htm>