[EM] A method satisfying Minimal Defense and much Later-no-harm

[Kevin Venzke]

Russ,

 --- Russ Paielli <rpaielli at mail.arc.nasa.gov> a écrit : 
> Kevin,
> 
> I think your proposed method is too complicated, and I doubt the public 
> would ever accept a method that starts with "random ballot."

I don't think the method is more complicated than e.g. Schulze's method,
especially if you don't use DSC or IRV for the ranking.

Also, you don't need to use Random Ballot. You can use any method that 
satisfies Later-no-harm. I just think the properties of the method are easier
to demonstrate if you use a simple method like Random Ballot.

If you use FPP, the method can be defined very simply: "Elect the candidate
with the most first preferences who has a majority-strength beatpath to
every candidate who dominates him."

> However, I searched back in the EM archives, and I see you had suggested 
> a method back in 2003 that I think has great potential. It is very 
> similar to "Condorcet-Approval hybrid" I proposed a couple of days ago. 
> I'll post a message about it soon and try to find out why you have 
> apparently abandoned the idea.

I look forward to your message, although I hope the method you're referring
to is not the one from the "Approval-Condorcet hybrid encouraging truncation"
thread. That was silliness.

Kevin Venzke



	

	
		
Découvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! 
Créez votre Yahoo! Mail sur http://fr.mail.yahoo.com/