[EM] tied clones retraction

[MIKE OSSIPOFF]

EM list--

Right after I sent my most recent posting, it occurred to
me that 2 of my reasons for preferring SSD's choice in the
tied clone sets example weren't really good reasons at all.

The one about lesser-of-2-evils voters, and the one about
the fairness of overruling the fewest voters. Because when
you honor those voters' preferences in one of the clone sets,
and it makes the other clone set win, the voters whom you
honored probably won't thank you. Still, they'll understand
the dropping of the weakest defeat, and won't regard it as
an irrational action of the voting system.

Maybe the tie solved by RandomBallot is better, but for
actual meeting or committee purposes, isn't it a lot more
convenient to avoid a random solution, and wouldn't people
be more impatient with a method that requires random solutions
more? Aside from that, as Norm said, SD, SSD, & Tideman are
more intuitive than Schulze. Is the merit difference in that
tied clone sets situation important enough to outweigh the
greater difficulty of getting Schulze adopted, and the annoyance
that its greater need for random tiebreaking would cause?

I'm only talking about committees & meetings because, as I said,
SD, SSD, & Schulze choose the same winner under many-voter
conditions.

Mike Ossipoff



________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com