[EM] Re: WV methods fail FBC with 3 candidates

[Kevin Venzke]

Mike,

--- MIKE OSSIPOFF <nkklrp at hotmail.com> a écrit:
> Thanks for settling that question. Of course I've just found your posting 
> and haven't had the opportunity to thoroughly check th example out yet, but 
> you probably have, and it's probably a genuine FBC failure example for WV.

I definitely invite you (and anyone) to check it out thoroughly. There could
be a mistake.

> How did you find that example? 

I guessed how the failure might occur, and then wrote a computer simulation
to find such a failure.

> I'd  thought that I'd demonstrated for myself 
> that it was impossible for a 3-candidate WV FBC failure example to exist. 
> Probably the example is genune and there was an error in my demonstration. 
> Well, it's probably that I was only looking at one way of failing, and it 
> can happen in a way different from what had occurred to me.

I hate to bring this up again, but this is why I consider FBC hard to
use. I have a new Condorcet-like method which I *believe* satisfies FBC,
but I can only prove some things that are implied by FBC compliance; I
can't prove FBC compliance directly.

Kevin Venzke



	

	
		
_____________________________________________________________________________ 
Découvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mails, photos et vidéos ! 
Créez votre Yahoo! Mail sur http://fr.mail.yahoo.com