[EM] Typo. Two polls.
In my previous post, I meant to say: "Voters shouldn't have strategic need to abandon their favorite." I accidentally said "should" instead of "shouldn't". Polls: Polls don't seem to be popular among EM's current frequent-posters. Of course neither am I, because I frankly criticize some people's favorite methods, and embarrassingly ask for justifications and mitigating advantages, to outweigh some listed disadvantages. That's ok. I'm not doing this to gain popularity. Proposing a menu of methods won't do. It's necessary to propose _one_ proposal. That means choosing. Choosing requires criticism, to explain why something shouldn't be the choice. Anyway, so a poll proposed by me, then, would be doubly unpopular. For example, my most recent poll got zero participation. I've never considered that a reason to not propose a poll. So I'm going to propose two polls. Regardless of whether or not there will be participation, I do so because I like to, and because I'd like these poll proposals to go into the EM record, for the reason described immediately below. . As I've often pointed out, you can't know what it would be like to use the various voting systems unless you use them--in polls. Only thereby can you know which one(s) you like. And even merely asking a poll's voting question brings some of that voting experience,, even if the reader doesn't actually write out a ballot and post it. Also, posting a poll includes some discussion of the choices between the alternatives. --- I propose a poll that chooses among the Declaration-mentioned voting systems. Approval Score (also called "Range") Majority-Judgment Instant-Runoff Condorcet (in various proposed versions) The Condorcet versions currently being proposed include; Ranked-Pairs(wv) Beatpath(wv) Ranked-Pairs(margins) Beatpath(margins) Improved-Condorcet-Top (ICT) Symmetrical ICT. For one thing, as I said, using the voting systems is the only way to find out what they re like. Also, if we every have better government, there will likely be initiatives or referenda regarding a choice of voting systems, calling for a choice like the one in this poll. Also, the relative merits of these methods is the subject of discussion here, and so polling about that naturally follows. I've argued that most likely the only way we're going to get a better voting system for state and national office will be by electing to government office a party that offers a better voting system. The Greens are the most winnable progressive party, and they offer IRV in their platform. So, suppose the Greens won the presidency and most of Congress, and most of the state legislatures. IRV would soon be the voting system in the U.S. No doubt the Greens would facilitate a much expanded power of initiative, in all states, and even at the national level. Maybe Constitutional amendments regarding voting systems too. In any case, surely there'd be state initiatives or referenda, or even a national initiative or referendum, regarding voting systems. Yes, voting systems are now decided by states, but we don't know that there wouldn't be a Constitutional amendment that would change that, and make it a national issue. Anyway, there'd likely be state initiatives or referenda, or a national initiative or referendum, regarding the choice of a voting system--if there's a lot of interest in that choice. Of course the initial voting system would be IRV, because that's what the GPUS platform offers. No doubt, then, a vote on voting systems would be by IRV. It would be instructive to use IRV for that vote here, then. But, with a different electorate, there might not be much point in using the same voting system. That's why I suggest using all of the Declaration's mentioned voting systems. Though I don't like IRV's FBC failure, with our existing electorate, an official public vote on voting systems could happen only after the election of a Green government. As I said, an electorate that could elect a Green government would be competent to make good use of IRV, and so IRV would be a perfectly satisfactory method under those conditions. Or at least I claim it would. I mention this because others, too, might feel that IRV's merit would be better under the conditions assumed by this voting-systems poll. This poll assumes that the public have just elected a Green government. --- I propose a poll that chooses among the 6 presidential candidates who've been in televised and Internet debates. Jill Stein (GPUS) Rocky Anderson
[EM] Typo. Convenient terms.
In my most recent posting, I said: "I didn't ask to minimize the max s/p. I asked to minimize, over all of the pairs of states, the amount by which one state's s/p differs from that of the other." The meaning is probably clear, but I should add "...maximum...", before the word "amount". So it should say: "I didn't ask to minimize the maximum s/p. I asked to minimize, over all of the pairs of states, the maximum amount by which the s/p of one of those two states differs from that of the other." Also, p, a state's population, is really an unwieldly and impractically large number for these purposes. I've therefore often been speaking of Hare quotas as a measure of population. For the purposes of discussing Sainte-Lague/Webster, or other divisor methods such as d'Hondt, I'd rather use "q" to stand for the quotient resulting from division of the state's population by the final divisor, the divisor that results in the desired number of seats. When not discussing divisor methods, the Hare quota might be the most convenient population measure to speak of. If I'm going to use "q" to refer to a state's number of Hare quotas, I'll say so in advance. Otherwise, in my usage, "q" means the result of dividing the state's population by the final divisor, as described above. That's the number that I've previously called the state's "cps". So, I'll speak of "s/q" instead of s/p. And, instead of the unwieldly "s/q_i", I'll just use "M" standing for "middle", to represent the ideal s/q value for all of the states. I don't know if there will be any more discussion about this, but, if there is, I wanted to say what terms I'd use. Anyway,even if there isn't, I like clarifying these things. Mike Ossipoff Election-Methods mailing list - see http://electorama.com/em for list info
[EM] Typo
I'd said: "But then, why not make that goal explicit? Instead of dividing along the shortest crossing line that halves the district, why not choose the halving division so that it minimizes either the maximum or the sum of the two resulting subregions?" I meant "...minimizes either the maximum or the sum of the _diameters of_ the two resulting subregions. So here is what I meant to say, in its entirety: "But then, why not make that goal explicit? Instead of dividing along the shortest crossing line that halves the district, why not choose the halving division so that it minimizes either the maximum or the sum of the _diameters of_ the two resulting subregions?" Mike Ossipoff Election-Methods mailing list - see http://electorama.com/em for list info
[EM] Typo. Plurality strategy, not Approval strategy, is a difficult problem.
When I said "Approval strategy is a difficult problem", I meant "Plurality strategy is a difficult problem." Let me repeat a little of what I said before: It's difficult because of the difficulty of getting an agreement, among those who want something better than the Democrats, regarding where they will combine their Plurality votes. That's why I suggest just voting sincerely, at least once, to actually find out what people want. In a presidential election, or a specified congressional election, or in specified state elections for specified unimportant offices. Mike Ossipoff Election-Methods mailing list - see http://electorama.com/em for list info
[EM] Typo in Condorcet's Criterion definition
The Condorcet's Criterion definition that I posted today doesn't make any sense, due to a typo. Where I said, "rank X over Y", I should have said "prefer X to Y". Of course, likewise, "rank Y over X" should be replaced with "prefer Y to X". So, the correct definition is: A definition of Condorcet's Criterion: X is collectively preferred to Y if the number of voters who prefer X to Y is greater than the number of voters who prefer Y to X. The "Condorcet candidate" is a candidate who is collectively preferred to each one of the other candidates. If there is a Condorcet candidate, and if everyone votes sincerely, then the Condorcet candidate should win. [end of definiition of Condorcet's Criterion] As I said, I've previously posted a precise definition of sincere voting, which I'll re-post upon request. But, if you don't ask for the definition, then rest assured that the precise definition is consistent with the usual conversational meaning of that term, with which you're familiar. My system of criterial has been criticized for its use of the word "prefer". I've been asked for definition of it. My answer: Take any reasonable dictionary definition, or your own conversational meaning, as the definiiton of "prefer". For the purpose of my critreria, that word's definition doesn't matter. That's because my definition of sincere voting uses "prefer", and specifies a particular manner of marking a ballot, if the voter prefers a certain way. So, saying, in the premise of a criterion, that voters have certain preferences among the candidates, and that those voters vote sincerely, therefore stipulatres, in that criterion-premise, a particular manner in which those voters mark their ballots. Because the definition of "prefer" doesn't matter, it could (for example) just as well be replaced by Lewis Carroll's nonsense verb "outgribe", without changing the meaning of my critreria that use the word "prefer". Now that I've already said so much, I might as well include my criterion-system's definition of sincere voting: A voter votes sincerely iff, on her ballot, s/he doesn't contradict any of her preferences, or fail to vote a preference of hirs that the balloting system in use would have allowed hir to vote in addition to the preferences that s/he actually does vote. [end of definition of sincere voting] To vote a preference for X over Y means voting X over Y, when preferring X to Y. To contadict a preference means voting Y over X, when preferring X to Y. I've previously posted two precise and usable definitions of voting X over Y. For the purposes here, either will do. I'll re-post them upon request. However, if you don't ask for the precise definitions, rest assured that they're consistent with what you mean by "vote X over Y". Why this system of critreria? Because it's the way to define some criteria (such as Condorcet's Criterion) in a way that applies neatly and seamlesslly to all methods. For that reason, some of my critreria refer to preferences. They've been called "preference criteria". But sometimes I've used criteria that are not preference criteria. Not all critereia need that kind of definition in order to be universally seamlessly applicable. There has been much sputtering and anger about preference criteria. But, as I said, they're the only way to specify certain requirements (such as what we mean by Condorcet's Criterion) in a way that is seamlessly and neatly applicable to all methods. My definitions of sincere voting, contradicting a preference, voting a preference, and voting X over Y, are supporting definitions, for the system of preference criteria. Mike Ossipoff Election-Methods mailing list - see http://electorama.com/em for list info
[EM] Typo
When I said "The pairwise-count problems that ICW doesn't get rid of don't really seem problematic to me.", I meant "ICT", and not "ICW". ICW was a typo. So I meant: "The pairwise-count problems that ICT doesn't get rid of don't really seem problematic to me." Mike Ossipoff Election-Methods mailing list - see http://electorama.com/em for list info
[EM] typo
In my alternative definition of voting x over y, in the first sentence, I accidentally wrote "is" when I meant "if". Here is the posting written correctly: Alternative definition of voting x over y: You're voting x over y if switching the names of x and y on your ballot could change the winner from x to y, but could not change the winner from y to x. [end of alternative definition of voting x over y] This avoids the "probably" or the phrase "consistent with more configurations of other voters' ballots". It's simpler and neater. Either definition would do. Of course by this #2 definition, in IRV you never really know whether you're voting x over y or y over x. No problem. My criteria still apply to IRV. A criterion-failure-example-writer can always make up a monotonic example for hir failure example. Election-Methods mailing list - see http://electorama.com/em for list info
