<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'>
<div id="mpf0_readMsgBodyContainer" class="ReadMsgBody"><div class="SandboxScopeClass ExternalClass PlainTextMessageBody" id="mpf0_MsgContainer"><pre><br>>I admit that that is a mess--when my <br>>optional-conditionality-by-mutuality algorithm definition<br>>is in three widely-separated postings. At least I should re-post the <br>>corrected pseudocode in<br>>one posting. Should have already done that before now. Will within a few days.<br> <br>While there may be value for this in terms of working on improved <br>methods, as to theory, as to possible public implementations, not <br>method that is so complex to explain has a prayer of seeing <br>application outside of specialized societies where they are willing <br>to tolerate that.<br><br>[endquote]<br><br>How many people have seen, or asked to see, the computer program for vote-counting<br>in our current elections? How many people in IRV jurisdictions have seen or asked to see,<br>or understood the count program for IRV?<br><br>People are told how IRV works, but they don't have to see the software. <br><br>AOC conditionality can be described in terms of what it does for the voter.<br><br>A conditional approval isn't counted unless it is reciprocated.<br><br>It can be said in more detail, but a little more wordily:<br><br>Call a ballot's unconditionally-approved candidates its "favorites".<br><br>A ballot on which C is favorite is called a C-favorite ballot.<br><br>For each pair of candidates, C and D, the number of ballots on which D, but not C is favorite,<br>and which conditionally approve C must at least equal the number on which C, but not D is<br>favorite, and which conditionally approve D. Otherwise enough C-but-not-D-favorite ballots' conditional<br>approvals of D are ignored to achieve the above-described parity condition. <br><br>But people will understand that, in examples like the one below, it's good if the voter can<br>make an approval conditional upon reciprocity:<br><br>(If you haven't been on the list lately, you might not have seen this "Approval bad-example":<br><br>Sincere preferences:<br><br>27: A>B<br>24: B>A<br>49: C<br><br>The A voters should approve B, and the B voters should approve A. But what if the A voters<br>approve B, and the B voters don't approve A? Then B will win, and the B voters will have<br>successfully taken advantage of the A voters' co-operativeness and sincerity.<br><br>That's the co-operation/defection problem, or the chicken dilemma.<br><br>If you're an A voter, you'd be glad to hear that you can give a conditional approval to B, an<br>approval that is conditional upon reciprocity.<br><br>So, what AOC does isn't complicated to tell. People would understand why they'd like it.<br><br>In any case, remember that I don't suggest AOC for a first proposal, partly because the simpler<br>plain Approval is simpler, and partly because AOC is to computation-intensive for an easy, convenient<br>handcount. At first, till a count-fraud-proof computer count can be guaranteed, only a handcount<br>is acceptable. The benefits of the best and most sophisticated method are nil if count-fraud<br>changes the result.<br><br>I don't know whether GMAT &/or MMT is suitable for handcounting.<br><br>By the way, though Bucklin was used with a handcount, ER-Bucklin, with the MMC-preserving delay that I spoke<br>of, is incomparably more computation-intensive than ordinary Bucklin, and therefore, almost surely unsuited to<br>a handcount. And, without that delay, you lose MMC compliance.<br><br>You asked about what I meant, regarding that delay:<br><br>Suppose that, at your 3rd rank position, you've ranked 5 candidates. Say that in round N, they get votes from your<br>ballot. The delay provision that I speak of (and which is in the electowiki definition of ER-Bucklin) says that<br>your votes to your 4th ranked candidates won't be given any sooner than they would be if you'd ranked your 5<br>rank-3 candidates in separate consecutive rank positions. In other words, in this example, your 4th ranked<br>candidates don't get their votes from you until round N+5.<br><br>If you'd ranked those candidates in consecutive rank positions, then one of them would get your vote in round N.<br>The 2nd would get a vote in round N+1....and the 5th would get your vote in round N+4. So only in round N+5<br>would your ballot then give to your next candidate. <br><br>As I said, that preserves Mutual-Majority-Criterion compliance, but it greatly increases the labor of a handcount,<br>almost surely making handcount infeasible.<br><br>So then, when you rank 5 candidates at rank 3, receiving your votes in round<br>N,your 4th ranked candidates don't get votes from you until round N+5. At that time, all of your<br>4th-ranked candidates receive your votes.<br><br>So, of the Approval election vote-management options that I've proposed, the only ones suitable for a handcount<br>would be MTA, MCA (ordinary, non-conditional), and maybe GMAT &/or MMT (someone else might be able to answer whether<br>GMAT or MMT would be handcount-suitable). <br><br>At least for now, a handcount is the only reliable way to avoid count-fraud.<br><br>Some people are very worried about fraud on the part of some voters. What we should really be worried about<br>is count-fraud.<br> <br>>u/a election:<br>><br>>u/a stands for unacceptable/acceptable. A u/a election is one in <br>>which there is one or more completely unacceptable<br>>candidates who could win.<br> <br>If write-ins are allowed, this is theoretically possible for all <br>zero-knowledge elections. However, many elections, for many potential <br>voters, do not fall into this category. <br><br>[endquote]<br><br>Of course. It depends on the voter.<br><br>You continue:<br><br>As shown by the fact that <br>they do not vote. Obviously, whatever is the result of the election, <br>it doesn't make enough difference to them to vote. That's actually <br>important. If you somehow entice these voters into participating in <br>an election, without shifting something else, you will be adding <br>noise, and noise probably biased by such factors as media exposure of <br>candidates.<br><br>[endquote]<br><br>If they aren't voting because they're uninformed, then it's better if they <br>don't vote. But they've been misinformed by the media, and that misinformation<br>is a self-fulfilling prophecy, when it makes Plurality results that seem to confirm it.<br>With a better method, such as Approval, they wouldn't be deceived as they now are.<br><br>They aren't voting because they believe that it's a contest between Tweedele-Dee and<br>Tweedle-Dum. The media have told them that the Democrats and the Republicans are<br>"the two choices". Is it any wonder that half of the people don't vote?<br><br>That's the illusion that Approval vote results will dispel.<br> <br>>In such an election, avoiding the election of an unacceptable is <br>>all-important.<br> <br>Sure. However, in situations where that election is reasonably <br>likely, the voter is teetering on the edge of a falling-out with <br>society itself.<br><br>[endquote]<br><br>I don't understand that conclusion. That voter will want to vote so as to<br>minimize the probability of an uncacceptable winning.<br><br><br> <br>> In a u/a election by ABucklin,<br>>it's definitely your best strategy to top-rank all the acceptables, <br>>and not rank any unacceptables.<br> <br>This is a black-and-white analysis. "Unacceptable" has been pushed to <br>the maximum of unacceptability, i.e., if So-and-So is elected, I <br>might as well commit suicide. <br><br>[endquote]<br><br><br><br>I wouldn't say that. I'm speaking of the limiting case in which<br>the utility-differences among set A candidates are negligible compared to the<br>utility difference between set A and set B candidates (whose utility is lower).<br><br>As I said, in my own personal subjective opinion, we have such a situation. Or something<br>so close to it as to be effectively the same thing.<br><br>You continued:<br><br>Given that, if some truly horrible <br>candidate is on the ballot, I'll definitely vote, and I'll vote in <br>such a way as to defeat the candidate, that's the condition of the <br>problem.<br><br>[endquote]<br><br>But you wouldn't necessarily vote unless there is a significant probability that that<br>truly horrible candidate might win. <br><br>People (mistakenly) believe that the winner must surely be a Democrat or Republican.<br><br>Many people (correctly) feel that the difference between those two isn't worth voting on.<br><br>They don't vote. Who can blame them?<br><br>You continued:<br><br>If there is more than one such "totally unacceptable" <br>candidates, each a possible winner, I might really wonder about the <br>society itself.<br><br>[endquote]<br><br>But if one such might win, then you'll want to vote so as to minimize that<br>probability.<br><br>You continued:<br> <br>Note that if I don't consider the unacceptable candidate a possible <br>winner, the whole argument is false. <br><br>[endquote]<br><br>Of course. That's why I made that a condition for a u/a election.<br><br><br>> I consider our<br>>public political elections to be u/a. The Republocrats are the among <br>>the unacceptables, though there are probably<br>>others too. That's just my opinion as a voter, judging by standards <br>>such as dishonesty, corruption, bought-ness, etc.<br> <br>That's your opinion<br><br>[endquote]<br><br>Yes, I emphasized that it's an individual subjective opinion.<br><br>You continued:<br><br>, to which you have a right, but you are rather <br>obviously not typical. At all. <br><br>[endquote]<br><br>I hope that I haven't seemed to imply otherwise.<br><br>But don't be so hard on the public. Don't entirely write-off their<br>judgement. They're misinformed, to the point of completely internalizing<br>what they've been told.<br><br>There was once a book entitled "I've been down so long, it seems like up to me."<br><br>That could be said by our voters. If you've accepted that the Democrat and the Republican<br>are indeed "the two choices", and so you conclude that the politicians are just all corrupt,<br>and if you've accepted that as fact, and have become completely resigned to it, then you<br>might call them "acceptable", because you believe that they're all that can be. Or maybe<br>you consider them unacceptable, because you know that the politicians are hopelessly corrupt,<br>but, since your tv has taught you that they're the only choices, that they and their corruption<br>are inevitable, simply a fact of life, you feel that there's no point trying to vote against them.<br><br>You continued:<br><br>And how you would vote is not <br>particularly relevant, then, to the design of public systems.<br><br>[endquote]<br><br>Oh come on now. If I were designing voting systems for myself, I wouldn't give top billing to FBC.<br><br>I've recently said, for instance, that IRV would be perfectly ok if voters weren't committed to <br>resigned lesser-of-2-evils giveaway.<br><br>FBC is for the public, not for me. They need it badly.<br><br>If everyone were more like me, it would be a completely different ballgame. Then, Beatpath would<br>be a perfectly good method. Even IRV would. They'd both have their own advantages. IRV fails Condorcet,<br>but meets MMC, and doesn't have the co-operation/defection problem. If everyone were more like me, maybe Beatpath<br>would be best. <br><br>If people were sufficiently honest, in fact, maybe 0-100 RV would be best. I'd like RV, if everyone were as<br>honest as I. (My advocacy of strategic RV voting, and my comments about "sucker-voting" are only intended for existing conditions)<br><br>But we needn't concern ourselves with that.<br><br>We must recommend the best voting systems for the voting public_as they are_.<br><br>Approval meets FBC, and does so in the most obvious and transparent way. No one will think that<br>s/he needs to withold a vote from hir favorite in order to fully help hir compromise.<br><br>Approval is the simple, obvious and natural improvement on Plurality. Plurality done right.<br><br>Approval has the most easily-implemented balloting of any method other than Plurality. Its ballot<br>is identical to the Plurality ballot, except that, instead of "Vote for 1", it says "Vote for 1 or more".<br>That's it. Two additional words on the ballot.<br><br>Approval is the most easily handcounted method except for Plurality.<br><br>Can there be any question about what the voting system reform proposal should be?<br><br><br>You continued:<br><br> I would <br>argue that public systems should allow you to express your <br>preferences fairly. If you want to vote by categorizing candidates <br>into two classes, acceptable and unacceptable, voting only for the <br>acceptable candidates, with maximum strength, you should be free to <br>do so, because you are, by definition, willing to accept the loss of <br>choice between the acceptable ones.<br><br>[endquote]<br><br>Yes. A lot of methods allow that. Probably Approval and Bucklin allow<br>the most power to 2-level voting.<br><br>You continued:<br><br>I think, Mike, that you may want your cake and to eat it too. You <br>want to be able to get the "unacceptables" out of the way, first, <br>then you can choose.<br><br>[endquote]<br><br>Wouldn't that be nice. We'd all like to have that cake and eat it too,<br>were it possible.<br><br>But I never said it was possible. I said that, in an ABucklin election,<br>I, along with others who feel as I do, would top-rank all of the acceptables,<br>and not rank any of the unacceptables. But I've always emphasized that we'd do<br>that at the cost of being able to distinguish between the (relatively identical)<br>merits of the the acceptables.<br><br>If, in ABucklin, you vote other than at top-rank, then either 1) You don't consider<br>it a u/a election, as I've defined that term; or 2) You want to vote expressively,<br>adventurously, gamblingly, though it's contrary to your best interest.<br><br>You continued<br><br>Mike, I'll repeat this: most voters will not be thinking like you.<br><br>[end quote]<br><br>Of course not. That's why I claim that FBC is absolutely essential.<br><br>Maybe others wouldn't vote as I would in ABucklin. ABucklin has the flexibility<br>to allow quite different kinds of voting, and that's why it's worth considering (if<br>count-fraud can ever be reliably prevented in computer-counts--because MMC-preserving<br>ABucklin is probably not handcountable).<br><br>I don't claim that everyone agrees with me that our elections are u/a.<br><br>Those who don't think so won't regard Approval strategy (as I do) as a matter of just voting<br>for the acceptables and for no one else. They'll use one of the other implementations of the<br>better-than-expectation strategy. Or maybe they'll use the direct implementation of that strategy,<br>by voting for everyone whom they'd rather appoint to office instead of holding the election.<br><br>I emphasize that even the u/a Approval strategy is one of the implementations of the <br>better-than-expectation strategy. <br><br>The better-than-expectation strategy has various different implementations, according to what kind <br>of information the voter has.<br><br>With no predictive information whatsoever, when you approve all of the above-mean candidates,<br>you're still using an implementation of the better-than-expectation strategy.<br><br>...as you also are if you use a strategy based in information about the two frontrunners, etc.<br><br>You continued:<br><br>We all support voting systems which will allow you to, at least, act <br>clearly to prevent unacceptable results. With Count All the Votes, <br>you know how to vote, and it's simple and easy to understand. I'm <br>merely suggesting that with some better knowledge of the <br>probabilities, you may be able to maximize your expected return.<br><br>[endquote]<br><br>Undeniable.<br><br><br>You continued:<br><br>This <br>black-and-white understanding of what is acceptable leads you to the <br>black-and-white voting. <br><br>[endquote]<br><br>Of course.<br><br>You continued:<br><br>It reduces your real effectiveness in the <br>world.<br><br>[endquote]<br><br>Incorrect.<br><br>It maximizes the effectiveness of my vote. It maximizes my expectation in<br>the election.<br><br>...in terms of my candidate-utilities, but maybe not in terms of your candidate-utilities.<br><br><br>You continued:<br> <br>>The Republocrats are a set, effectively a party, consisting of two <br>>nearly identical subsets called<br>>Democrats and Republicans. Gore Vidal said that we don't have a <br>>two-party system--We have one party with two right wings.<br> <br>Must be true if Gore Vidal said it.<br> <br>[endquote]<br><br>Did I say that it was true because Gore Vidal said it? <br><br>There was a time when people didn't attribute the ideas that they quoted.<br><br>Now it's considered dishonest to quote someone without attribution. That's why<br>I named Gore Vidal. I didn't want to imply that I was the<br>originator of that statement.<br><br>I didn't name him to prove that the statement is correct.<br><br>You continued:<br><br>No, if he was actually making sense, we have one party with a right <br>and a left wing.<br><br>[endquote]<br><br>No, he said "...two right wings", and there's no reason to believe that he<br>meant otherwise.<br><br><br><br><br>Sort of. So we have a one-party system. We are <br>accustomed to thinking of that as a bad thing. Is it? I'm not sure at <br>all. We have one government, and that is really "one party." Imagine <br>the communist regimes, where the Party nominates the candidates, for <br>unopposed elections. Is this undemocratic? Not necessarily. It <br>depends on the Party's nomination process, and whether or not the <br>Party truly represents the people, as it claims. The problem was that <br>it didn't. Or doesn't, as the case may be.<br><br>[endquote]<br><br>I'd say that a 1-party system is undemocratic, based on the etymology of "democracy".<br><br>If there's only one party, one policy platform, for voters to choose among, then the voters don't have a choice,<br>and there's no meaningful sense in which they're governing.<br><br>><br>>At 04:04 PM 3/5/2012, MIKE OSSIPOFF wrote:<br>><br>>You said that achievability, getting from here to there, is more <br>>important than optimality, for<br>>a voting system proposal. Yes, and that's why strongly suggest that <br>>Approval should be the first<br>>proposal. In later proposals, vote-management options could be <br>>added. They include:<br> <br>You continued:<br><br>Count All the Votes. Terminally simple, clear improvement, only bogus <br>arguments against it.<br> <br>Not Perfect. So?<br> <br>Count All the Votes. When it's is presented as Approval, hosts of <br>distracting questions are raised<br><br>[endquote]<br><br>Isn't Count All the Votes another name for Approval?<br><br>I think that when we speak of "voting for" several candidates, we encourage<br>the fallacious "one-person-one-vote" objection.<br><br>We indicate which candidates are acceptable. Or we indicate which candidates we approve of.<br>And the method elects the candidate approved by the most, or acceptable to the most.<br><br>That's a difficult thing to criticize.<br><br>Or maybe first talk about 0-10 RV, and then offer 0-1 RV, to show that Approval is a points<br>system, no less than is 0-10 RV.<br><br>I think that I understand what you're suggesting with Count All The Votes. You're<br>referring to a Plurality election in which overvotes are counted. The trouble with that<br>is that it encourages the objection that Approval is legitimizing and accommodating something that<br>is currently considered not valid. I think that that invites the criticisms that Approval gets.<br><br>We count all the votes because we want to elect the approved-by-most or acceptable-to-most candidate.<br>It's a whole different basis of what the election is. It's necessary to explicitly confront and reject<br>Plurality's assumptons for what we want to do, replacing them with the above introductions.<br><br>Ok, sometimes people will be voting strategically, rather than by "Whom do I approve?" or <br>"Who is acceptable?". In a non-u/a election, strategy doesn't ask about acceptability.<br><br>But it would be difficult to criticize each voter's right to rate each candidate, to vote up/down on each<br>candidate. Whatever strategy the voter might be using, it's hir vote, to use as s/he wishes.<br><br>For each candidate, it really is one-person-one-vote. Just as it is in any other point-system.<br> <br>You continued:<br><br>Voting is better understood as adding weight to a balance, rather <br>than as a "sincere expression of an opinion," though we can design <br>systems that make the adding of the weight relatively easy to deduce <br>from the opinion. But, always, it's an action, not a sentiment. <br><br>[endquote]<br><br>Quite so. <br><br><br><br><br><br>>Approval, though entirely adequate as a destination method, is also <br>>the best route to ABucklin or<br>>SODA or maybe RV.<br>><br>>Or, alternatively, RV could be the first proposal:<br> <br>In NGO elections, quite possibly. In public elections, I'm obviously <br>not opposed, but ... understand that Range can badly fail, because of <br>utility normalization problems. I'm a whole lot happier with Range as <br>an element in a runoff system.<br><br>[endquote]<br><br>In our electorate, RV will be like Approval, but with suckers who<br>are taken advantage of by Approval-strategy voters. I prefer Approval,<br>but I also recognize that RV is Approval. One just must try to make<br>sure that the suckers aren't one's co-factionalists. <br><br>So I don't oppose RV, and would welcome it, because it's Approval.<br><br><br>You continued:<br> <br>Range polling of course, is drastically superior to any other kind. A <br>Range ballot is maximally expressive, with one exception. It doesn't <br>express approval, which is a separate decision. We need either a <br>voter-determined approval cutoff, which is complicated, or we need to <br>predefine the ratings, classifying them into approved and <br>not-approved. We may not need so many unapproved ratings as we do <br>approved ones, so the proposal of using mid-range with a linear range <br>ballot is certainly not the only one.<br><br>[endquote]<br><br>I assume you're referring to a popularity contest.<br><br>Nothing to add to what you've said there. <br><br>Of course public political elections are the <br>important thing. <br><br>And you can't be sure that people won't strategize<br>in a populiarity contest. Maybe you want someone to win the<br>popularity contest. Might you not falsify some rating-feelings<br>to that end? If people are inclined to do that, then Approval would<br>be a more sincere method for a popularity contest.<br><br> <br>>Of course it goes without saying that Approval voting is <br>>automatically a way of voting in RV (you<br>>top-rate your approved candidates and bottom-rate everyone else). <br>>Then, all of the abovementioned<br>>Approval vote-management options could be proposed too.<br> <br>Which is how you would vote in the u/a scenario, right?<br><br>[endquote]<br> <br>Correct. I'd vote Approval-style. Only at 1st rank position.<br> <br> <br>>Approval elects the candidate who is acceptable to the most voters.<br> <br>Not exactly, but close. It elects the candidates who were given the <br>most votes.<br><br>[endquote]<br><br>Ok, sure, I was just interpreting it in that way. Above, in this post,<br>I talked about justification/interpretation that doesn't assume that.<br><br> <br> <br> <br>>...But the runoff would complicate the proposal. You could propose <br>>it later, though.<br>>But when you complicate or elaborate a method, FBC-failure tends to sneak in.<br> <br>Maybe. But if the existing method is top-two runoff, you could be <br>making things worse by, say, going to raw Approval, which often <br>reduces to plurality.<br><br>[endquote]<br><br>No way. Approval is entirely different from Plurality. In Approval, if you<br>only vote for one candidate, that's because s/he's your genuine favorite.<br><br>People voting for one candidate in Approval isn't a failure of the method. There can be<br>good reasons for voting only for one in Approval.<br><br>Of course, people who vote for only one in Approvalwould do fine in a Plurality election.</pre></div></div>To be continued...<br><br>Mike Ossipoff<br><br> </div></body>
</html>