[EM] IFNOP Method (was Re: Question about Condorcet methods)

[Juho]

On Oct 18, 2006, at 23:55 , raphfrk at netscape.net wrote:

> I assume you mean the "later no harm" property?

Yes, but also any other variants that fall between IRV and Condorcet.

> In effect, this means that you cannot look at a later choice on a  
> ballot until you are sure one of the following 3 conditions is true
>
> a) the candidate at the current choice is already elected
> b) the candidate at the current choice cannot be elected
> c) the later choices can in no way affect the election of the  
> current choice

Maybe:
c) the later choices can in no way REDUCE THE PROBABILITY OF election  
of the current choice

IRV/Condorcet hybrids could also not have a "current choice" in IRV  
style but maybe something more Condorcet like. (don't know, not  
proposing that such nice methods can be found)

Juho Laatu


> c) folds in on a) and b) as if later choices cannot affect election/ 
> elimination of the current candidate then you already know if the  
> candidate is elected or eliminated.
>
> In effect, the process has to be:
>
> 1) Look at all first choices
> 2) elect and/or eliminate some candidates based on current choice  
> total
> 3) reweight ballots and recompute totals using the highest  
> candidate on each ballot still undecided
> 4) goto 2) unless all seats filled
>
> Basically, all you will have is a list of candidates and a total  
> for each candidate.  You don't know voter rankings as you are not  
> allowed to look at them.  How can you determine who is elected or  
> eliminated ?  IRV seems the only reasonable way of doing it.  The  
> only possible other piece of info is the rankings of eliminated and  
> elected candidates, but I don't see how useful they would be.
>
> Maybe asset voting could be used.  After the round, each candidate  
> can give some/all his votes to other candidates.  Any candidate  
> above the quota gets elected, and also candidates can resign.  Each  
> ballot is then rescaled based on what percentage was "spent" by its  
> current holder.  However, if asset voting is used, then there is no  
> point in doing IRV as well.  I guess it could be used as a deadlock  
> breaker or something.
>
> Raphfrk
> --------------------
> Interesting site
> "what if anyone could modify the laws"
>
> www.wikocracy.com
>
>
> -----Original Message-----
> From: juho4880 at yahoo.co.uk
> To: davek at clarityconnect.com
> Cc: mrouse1 at mrouse.com; election-methods at electorama.com
> Sent: Tue, 17 Oct 2006 9:24 PM
> Subject: Re: [EM] IFNOP Method (was Re: Question about Condorcet  
> methods)
>
> Maybe there is some potential in doing the IRV style "never  
> considering all the given opinions" in some better way. I don't  
> have any opinion yet on if this is that case but maybe something  
> can be found. Juho Laatu
> Check Out the new free AIM(R) Mail -- 2 GB of storage and industry- 
> leading spam and email virus protection.
> ----
> election-methods mailing list - see http://electorama.com/em for  
> list info

Send instant messages to your online friends http://uk.messenger.yahoo.com