> > The reason is that condorcet has this "problem", > > even with no quorums whatsoever. Martin Schulze's post > > http://lists.debian.org/debian-vote/2003/debian-vote-200305/msg00119.html > > illustrates this principal.
On Sun, May 25, 2003 at 11:39:06PM +0200, Jochen Voss wrote: > Huh? Plain condorcet voting does not drop options. True. Then again, plain condorcet doesn't resolve the case where more than one option is in the schwartz set. Then again, that's not what http://www.mathematik.uni-kl.de/~wwwstoch/voss/comp/quorum.html is talking about. However, we all know that sequential dropping (of weakest defeats) ultimately results in all but one option being dropped from the winning set. Markus Schulze's example (see above url) shows a case where sequential dropping, without any quorum rules whatsoever, produces exactly the kind of behavior you object to on your page under the subtitle "One more vote makes a difference". [And, no, I'm not prepared to discuss the sitation in any more depth, at this time.] -- Raul