Differences of sincerity definitions

[David Catchpole]

On Sun, 3 Dec 2000, Bart Ingles wrote:

>
>
> LAYTON Craig wrote:
> >
> > Sorry in my rather hasty example, there are supposed to be 5 candidates,
> > with the voters preferences being:
> >
> > A=B>C=D>E
> >
> > My question being, how can you (in a relatively simple fashion) allow a
> > voter to vote like this?  What kind of instructions could you use, how would
> > you structure the ballot paper, what would determine a valid vote?
>
>
> I assume you're asking specifically about IRV.  One approach which was
> kicked around a few years ago in this list (some of the members
> apparently tried to interest CVD in it as a way to mitigate some of
> IRV's problems) was to simply count the votes as you would with approval
> voting; in other words each surviving top-choice vote counts as a full
> vote.  The voter would rank your example above as
>
> A: 1
> B: 1
> C: 2
> D: 2
> E: 3 (or unranked).
>
> The vote doesn't transfer to C and D until A and B are both eliminated.
>
> This is probably a slight improvement over normal IRV in some ways, but
> may be slightly more susceptible to pushover strategy.  You can probably
> find it in the list archives; I think it was called IR-1 or something
> like that.

Just plain equal division of the vote doesn't seem that unappealing. Does
anyone want a squiz at my pet STV system, that includes this?

>
>

-------------------------------------------------------------------------------
Guy at hot dog stand:	Yo, Dalai Lama, what'll you have?
The Big DL:		Make me one with everything.
				-Thanks to www.newgrounds.com