Dave Ketchum wrote: (Wed.May19)
Condorcet as simplest - rank best first, use same thinking to rank best of remainder next, and continue until I consider remainder to not deserve ranking. Desirable for the method to permit equal ranking - a simple and understandable enhancement that I will sometimes desire.
Brian Olsen (same day):
Why should voters, in a pure rankings method, only rank candidates that they consider "deserve" ranking,So, is it a problem to instruct a voter in the usage of a ballot? "Choose One" "Mark all choices you find to be acceptable."
or that they "find acceptable"? Why this insistence on confusing rankings with ratings?
This instruction is only appropiate if the method fails Later-no-harm and meets Later-no-help. In that case
truncation is in effect a de-facto approval cutoff, and in my book the method is not a pure rankings method.
Later-no-harm and Later-no-help are incompatible with Condorcet, but there is no reason why the chances
that ranking an extra candidate can help or harm already ranked candidates shouldn't be the same; so that
the voter's best "zero-information strategy" is just to give his full ranking (to the extent that he has one).
Chris Benham
---- Election-methods mailing list - see http://electorama.com/em for list info
