[EM] A few poll details

MIKE OSSIPOFF nkklrp at hotmail.com
Fri Apr 20 17:50:30 PDT 2001

When you vote in the voting systems poll, post your ballot, and
also Cc: it directly to Rob LeGrand.

I told how Random Ballot would be implemented, should the final Approval
count return a tie. Of course if it returns a tie, that just means
that 2 or more methods are reported as the collective favorite of the
poll-participants. For that purpose there's need for Random Ballot.

But there could be a situation where it's desired to choose 1 particular
voting system. I don't know why that would be necessary for EM, since
it's agreed that for EM voting, where different people prefer different
methods, it's better to not just use one single method. But suppose
some organization asked EM to recommend _one_ voting system to them.
Then, if we had several tied for best, then the Randon Ballot
choice could immediately be recommended to that organization.

Sure, there could be a case for applying other voting systems to
solve ties, but when that violates ICC, it means that there's incentive
for strategic voting & nominating. And it means that a group of voters
lose their rightful chance to be in a random tie determination.
So the added decisiveness can bring unfairness.

So if there's a final tie, a Random Ballot winner will be returned.

Buried at the bottom of the e-mail that contained the ballot was the

I claim that beginning with the end of the balloting period (which
takes place exactly 2 weeks after the beginning of the balloting period),
there should be 48 hours in which people may trunacate their ballots
, but may not otherwise modify them or vote new ones.

That's an anti-order-reversal precaution. It's good that this poll is
taken especially seriously. But that means that we need precaution
agains offensive order-reversal. That 48 hour period for truncation
furnishes the means to easily thwart offensive order-reversal, for
certain rank-counts which I shouldn't name here, because this isn't
a campaign e-mail. And the order-reversal might not be difficult to
detect, because we have a good idea of people's method preferences, and
the rank-counts that I've referred to give no need for defensive
order-reversal. So, after the balloting, if you think that someone
has offensively order-reversed, especially if it appears to have
succeeded, then truncate your ranking so as to not include that person's
favorite, or whatever alternative won due to the order-reversal.
With the above-mentioned methods, offensive order-reversal backfires
for sure unless its victims vote for the reverser's alternative(s).

Mike Ossipoff

Get your FREE download of MSN Explorer at http://explorer.msn.com

More information about the Election-Methods mailing list