<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>
Because ABucklin is Approval that gives its votes stepwise, it was natural to consider<br>a Conditional Approval that takes away votes stepwise. That's what I referred to yesterday<br>as nested sequential conditionality. <br><br>It seemed to make sense, and was worthwhile to check out, but, upon examination,<br>there didn't seem to be much point in it. There didn't seem to be significant improvement<br>over ordinary Conditional Approval.<br><br>By the way, for MMPO, why not start by disqualifying Condorcet losers?<br><br>Or maybe do the tied-at-top version (that meets Kevin's criterion) of Condorcet, and complete it with MMPO.<br><br>The former of those two suggestions makes for a briefer-defined method, which<br>I'd call Condorcet Loser Disqualification MMPO (CLDMMPO). <br><br>That's the version of MMPO that I'll count in the poll's rank ballotings (unlimited and 3-slot), if there is enough participation for it to<br>be a poll.<br><br>There's enough interest in 3-slot methods at EM to justify a separate 3-slot balloting, and IRV3/AV3 has<br>been discussed here enough to justify a special balloting for it too.<br><br>In general, of course when I proposed this poll, I never intended that it should be entirely up to<br>me how it's conducted, what the ballotings will be, etc. As I said, such decisions are collective<br>matters.<br><br>I'll do a Voter's Choice count, but, that just means that I'm doing a count that I like, as of course<br>can anyone.<br><br>And the various options that I offer for voting in the Approval election are, of course _options_.<br><br>Mike Ossipoff<br><br> </div></body>
</html>