[EM] Re: The "Turkey" problem and limited ranks

[Kevin Venzke]

Dave,

 --- Dave Ketchum <davek at clarityconnect.com> a écrit :
> On Tue, 20 May 2003 10:53:51 -0700 (PDT) Forest Simmons wrote:
> >>The point of obtaining priorities is that we may be able to avoid a
> >>situation where the strict-ranking CW polarizes people more than the
> >>three-rank CW would.  Neither candidate need be a turkey.  The situation,
> >>quite simplified, looks like:
> >>55: A>B|  (A preferred to B, but given same rank)
> >>45: B|A   (B preferred to A and placed in a higher rank)
> >>
> > 
> > It would be nice to have B win in this case because the choice is between
> > 55 mildly disappointed voters and 45 devastated voters.
> 
> Seems to me you are assuming more than you know.

This is my fault, I should've been clearer.  The 55 voters' A>B is their
sincere preference, but it would not be indicated on the ballot.  They
are putting A and B in the same rank and no difference can be discerned
in preference.

> Agreed that those who prefer A are not giving A a major preference over B 
> (else they would assign different ranks).
> 
> However, those who prefer B are not necessarily saying "major preference" 
> - could be B barely deserved the rank it got and A barely missed deserving 
> the same rank - thus that the 45 might be barely disappointed by losing, 
> rather than devastated.

It seems that you're saying that "devastated" was too strong of a word.
(It probably was, since I don't believe we'd even decided how many ranks
there were.)  I'm not sure if you're also saying that even if we had ranks,
we wouldn't be able to believe they meant anything.

> I continue to prefer Condorcet because I can rate each x as better or 
> worse than the adjacent y, without getting swamped in trying to be 
> understood by the vote counter as to how much better I rate my preferred 
> candidate.

I can appreciate the desire for an expressive ballot.

I don't think voters would be overwhelmed by having to divide all the
candidates into 2-4 groups.  I think they might well find it less taxing
than a strict ordering.

> One complication that seems addable to Condorcet is to be able to rank two 
> or more candidates as liked equally - that they be ranked normally as 
> better and worse than other candidates, but I do not care which among tied 
> candidates win.

I think that's already part of Condorcet.  At least, I don't see the
advantage in not permitting this.  That would mean you couldn't truncate.

> And I dislike plain Approval because, as a voter, I want to be able to 
> declare x as preferred over y, while declaring both to be acceptable.

What method permits you to declare X>Y while declaring both to be
acceptable?  I thought you were advocating plain Condorcet?

Kevin Venzke
stepjak at yahoo.fr


___________________________________________________________
Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français !
Yahoo! Mail : http://fr.mail.yahoo.com