<div>Two things that I should add, partly in answer to possible objections:</div><div> </div><div>1. I'd said that it's the Approval voter's own business how s/he uses here approvals.S/he needn't justify it. Don't concern yourself</div>
<div>with hir motivation. </div><div> </div><div>But then it occurred to me that you might say that the same could be said about Pl/urality: "It's the voter's own business whom s/he designates</div><div>favorite. Don't worry about hir motivation. (such as the matter of whether s/he is favorite-burying)"</div>
<div> </div><div>The difference is that the Plurality voter doesn't have a choice. What you call hir voting power isn't fully hirs. S/he doesn't have the power to choose</div><div>how to rate each candidate. So a comparison to Plurality wouldn't have validity.</div>
<div> </div><div>2. When answering your comments about electing the candidate given operational approvals by the most voters, I should have mentioned</div><div>Approval's other optimizations. </div><div> </div><div>Even if people are voting strategically:</div>
<div> </div><div>Of course, in a u/a election, Approval elects the candidate who is acceptable to the most voters.</div><div> </div><div>In a non-u/a 0-info election, Approval elects the candidate who is above-mean for the most voters.</div>
<div> </div><div>If it's neither of the above, Approval elects the candidate who is better-than-expectation for the most voters.</div><div> </div><div>Consider that last one: Approval's general better-than-expectation strategy says to approve the candidates who</div>
<div>are better than what you expect from the election. </div><div> </div><div>In other words, approve optimistically. If a candidate is exactly what you expect, merit-wise, then it would be reasonable</div><div>to approve hir too. But, in general, you're approving better-than-expectation candidates.</div>
<div> </div><div>So Appreoval elects the candidate who is, to the most voters, better than what they expected from the election.</div><div> </div><div>...the candidate who whose win is an optimistic prospect to the most voters.</div>
<div> </div><div>...the candidate whose win is good (even if at least a little unexpected) news to the most voters.</div><div> </div><div>...the candidate who was, to the most voters, at least a little too good to be true.</div>
<div> </div><div>I know I've discussed that before, but I wanted to go into a little more detail, and it seems relevant, since</div><div>you're claiming that the "most opertional approvals" optimization says nothing about whether one of the</div>
<div>strategies was used in choosing the approvals. </div><div> </div><div>This optimization that I described above is based on strategic voting. All of the Approval strategies are special</div><div>cases of better-than-expectation. </div>
<div> </div><div>Mike Ossipoff</div><div> </div><div> </div>