Meta-Tool

[Michael Allan]

Martin and Friedrich,

Martin Häcker wrote:
> I think that in theory every voting engine where in the end each
> user has one vote and the issues are _exactly_ the same these votes
> could be translated. ... However as soon as this assumption breaks
> in the slightest (different proposals / any difference in the
> voting) this cannot be done anymore.

I share your concern, but I'm looking for a specific example of how it
might fail.  I posted a translation where the proposals (= motions)
and the poll (=~ issue) are not identical (at least not initially):

  A-motion/EN4J8(single transitive) -> B-P/grfin(single transitive)
  http://t.zelea.com/wiki/User:ThomasvonderElbe_GmxDe/Vote_mirroring#Adhocracy_to_Votorola

Also, where the voting methods are not identical:

  C-Mayor-Toronto(IRV) -> B-P/m(single transitive)
  http://t.zelea.com/wiki/User:ThomasvonderElbe_GmxDe/Vote_mirroring#IRV_to_single_transitive

Is there anything wrong with these translations?  I'm trying to find a
well-intentioned translation that goes wrong.

> Especially different proposals or just additional proposals should
> be a very common cause here - therefore I propose that it is far
> more productive to show what other engines are doing, instead of
> trying to mirror the votes.

It's still unclear in my mind, exactly how mirroring can lead to bad
consequences.  I agree with Friedrich, we should probably be extra
clear about this.  We should try to find examples.

-- 
Michael Allan

Toronto, +1 647-436-4521
http://zelea.com/