[EM] Summability

[Kevin Venzke]

Markus,

 --- Markus Schulze <markus.schulze at alumni.tu-berlin.de> a écrit : 
> The task of an election method is to find the best candidate according
> to a given heuristic. The task of an election method is not to help
> voters vote strategically.
> 
> Markus Schulze

I'm having trouble imagining how this principle could be applied.  Perhaps
I'm not understanding the term "strategically."  My thought is that a method's
strategy should be as simple as possible (i.e. require as little information
and calculation as possible) so that informed and uninformed voters basically
play on a level field.  I can't think of a method which does this better than
Condorcet methods.

Considering methods which aren't summable (e.g. IRV, and methods which simulate
repeated approval balloting)...  The strategy is harder, to the point that your
ballot is more effective if you let someone else (your party) tell you how to
vote.  Surely that is not desirable?

Are you talking about making a method so complicated (to the point that it
cannot be summable) that no one can figure out how to cheat it?  That doesn't seem 
possible to me...  I don't think you mean that.

And I don't think I believe that you think IRV has an advantage due to its
non-summability (other than runtime for calculation).  I wonder what you have
in mind...?


Kevin Venzke
stepjak at yahoo.fr


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