[EM] PR-STV with approval based elimination

[Jonathan Lundell]

On Apr 30, 2009, at 10:33 AM, Raph Frank wrote:

> On Thu, Apr 30, 2009 at 5:42 PM, Jonathan Lundell  
> <jlundell at pobox.com> wrote:
>> The problem with these approaches (a problem, anyway) is that they  
>> abandon
>> later-no-harm. That seems a rather high price to pay.
>
> Well, the first suggestion, where the approval ballot is separate from
> the ranked ballot still meets later no harm (with respect to the
> ranked ballot).  Ranking all the candidates is still risk free.

The problem I have with the distinction between the two approaches is  
that they seem equivalent to me. What would motivate a voter to have a  
different approval set in the two cases?
>
> Also, I think later no harm basically means "won't compromise".  I am
> not sure that it is even a desirable criterion for a method to have
> and think that the fact that a method that doesn't meet later no harm
> is a not major issue.

Well, we can disagree about that, no doubt. For me, it's a high  
priority to eliminate or reduce strategic considerations from the  
actual voting act, and LNH eliminates a big one.