[EM] Poll on voting-systems, to inform voters in upcoming enactment-elections

[Kristofer Munsterhjelm]

On 2024-04-12 22:37, Michael Ossipoff wrote:
> Right!! That’s something I wanted to say. I’m removing Schulze from the 
> upper part of my ranking for that reason, & replacing it with 
> Smith//Approval(implicit).
> 
> How about we say to rank in order of overall merit for public 
> proposal…which includes proposability?
> 
> Then the unproposably complex methods could be left unranked or ranked 
> near bottom.
> 
> Or take it a step further & trim the candidate-set to only include 
> proposable methods? But might it be quicker to just let that be a voting 
> judgment, instead of having to do that evaluation as a separate 
> preliminary collective evaluation, which would delay the voting?

I would prefer that the merit question for the poll stays the same: 
"which voting methods do you prefer to which others?", i.e. ranking them 
in preference.

Then it would be up to the individual voter to consider what aspects of 
the method are most important; and anyone who wants to use it to guide 
reform can just screen away the unproposable methods.

After all, we have to do that anyway, because it's pretty much 
impossible to collapse disparate concerns into a single order without 
making some assumptions about which concerns are most important. Would I 
recommend Benham ahead of Schulze? Well, that depends on whether there's 
tons of strategy in the place in question and whether they (and I) can 
accept the nonmonotonicity.

In the absence of any such situational information, any order will be 
imperfect. In any case, if the poll's output ranking ends up being like

Extrinsic Borda-Weighted Landau Intersection > Iterative Refinement 
Keener + Sinkhorn (mean) > Schulze > RP > Approval > IRV,

then it's a simple matter for reformers to just discard everything above 
Schulze (or RP) for a public proposal. In practice, I doubt the exotic 
methods will rank that high anyway.

-km