Talk:Instant-runoff voting/Archive 5

Archives: ~ Dec 30, 2007 - Jan 23, 2008

RFC: Introduction includes term detail, single vote, RRONR ref?
Should the introduction of the IRV article include the explanation of why the term "instant runoff" is used, that the process results in a single vote being cast because of its equivalence to Single Transferable Voting and the fact that Robert's Rules of Order, Newly Revised recommends preferential voting for elections by mail and gives IRV as the example? (diff)
 * Yes for the many reasons I have already stated above. MilesAgain (talk) 04:15, 30 December 2007 (UTC)
 * No These are subtle issues with political implications. The "single" in Single transferable vote refers to the usage in Proportional representation, where each voter has one vote, but that vote is passed around and divided up; in many-member STV elections, most voters have indeed cast an effective single vote. There is current legal controversy, ref'd above, over whether or not IRV (which does *not* make every vote effective) satisfies "one person, one vote" or not. Given that the language "single vote" is not at all necessary to describe the method itself and, in fact, it makes it more complicated, I have to ask why this possibly controversial claim must be in the introduction, and the only conclusion I can imagine is that it's propaganda, given that this very question is currently being raised and is the subject of a political campaign. Robert's Rules could be considered to "recommend" "preferential voting," but that's an interpretation of the source; what it does is to note that it's possible to use Preferential voting, which refers to the ballot form, which is used by other methods as well, and then it gives a sequential elimination method, which is almost the same as IRV, as an example, but then it criticizes this specific example, describing a serious flaw (that does not exist with some other preferential methods; but those methods are not currently in common use, whereas IRV is more common). To reduce this complexity to a promotional sound bite in the introduction of the article is likewise POV-pushing. The detail is already present, later in the article. --Abd (talk) 06:35, 30 December 2007 (UTC)
 * Yes/No Based on the diff provided, there seem to be 4 issues presented:
 * 1) The claim that IRV simulates a series of run-off elections. It is my understanding that that statement is true, however if someone is questioning this, then a source should be provided.  Assuming it is true, it would seem to be a reasonable thing to keep in the introduction.
 * 2) The statement about it being a single vote. Quite honestly, it doesn't make a whole lot of sense to me.  I think that it is trying to point out how, at any point in the counting process, each person is only giving support to one candidate at a time.  I think a better way of saying that needs to be developed.  Also, I question the need for that statement to be in the introduction.
 * 3) Use in elections. Something indicating how widespread IRV is (or isn't) seems to be a good candidate for the introduction.  Personally, I don't see too much of a difference between the two ways it is presented in the diff, though personally I think the one on the left sounds better.
 * 4) Roberts rules. I think it is important to consider exactly what RONR says about preferential voting:"While it [preferential voting] is more complicated than other methods of voting in common use and is not a subsitute for the normal procedure of repeated balloting until a majority is obtained, preferential voting is especially useful and fair in an election by mail if it is impractical to take more than one ballot. In such cases it makes possible a more resresentative result than under a rule that a plurality shall elect.  It can be used only if expressly authorized in the bylaws (RONR (10th ed.), p. 411, l. 22-29)."  As RONR does not explicitly endorse or criticize IRV, it would seem to me that (as this is an article on IRV not on preferential voting as a whole), it would be inappropriate to make such a reference in the intro to the article.  Additional, I think the claim that RONR "recommends preferential voting for elections by mail" should be reconsidered.  It seems that it simply states that preferential voting is better than a plurality vote, but does not indicate that it is recommended (in fact RONR seems to recommend using a majority vote system). meamemg (talk) 21:12, 30 December 2007 (UTC)
 * Comment, While meamemg voted "Yes/No," every point he made confirmed what I've been arguing. First, the simulation of a series of runoff elections needs sourcing; I haven't raised it that way, but was actually about to. No runoff election process being used is what IRV does. However, batch elimination of all but the top two vote getters in the first round is, indeed, a simulation of a runoff election. However, the word "series" arguably covers this. My point is that there are substantial and significant differences between IRV and actual runoff elections, so the resemblance to runoffs is debatable and should therefore not be in the introduction. The name "instant runoff voting" is not what this method was and is called elsewhere in the world. It was a political invention by FairVote, a way developed of making the method seem understandable quickly. But it's also, arguably, misleading. Secondly, the single vote language is indeed unnecessary in the introduction, and it is debatable that it should be anywhere else unless balanced by other significant points of view. Thirdly, having a rough description of prevalence in the introduction is good. But it should be balanced. Is there a wave of IRV adoptions in this country? What percentage of elections are actually being handled by IRV? It's a tiny percentage. Growth rates can look spectacular when you start from zero. I have not attempted to remove solid prevalence information from the introduction. Finally, I've quoted that language from Robert's Rules, here, many times, and that is pretty much what is in the article, last time I looked, but not in the introduction. I'd say what I've been arguing has just been confirmed, quite precisely. Thanks all.--Abd (talk) 21:41, 30 December 2007 (UTC)


 * YES as to explanation of IRV's use of a single transferable vote, as this is the FUNDAMENTAL difference between IRV and other preferential voting methods. This section also explains WHY it is called IRV...also important for the introduction. As to the RONR reference, I believe it is appropriate in the introduction (and factually correct), although I am less insistent since I added a section on non-governmental use of IRV in the article that discusses RONR. However, I also still believe that overly-long quote from RONR is inappropriate there, as a summary that I originally used was exactly correct and captured the essence of that overly-long quote. I would propose the RONR sentence for the introduction read as follows...
 * "Robert's Rules of Order, Newly Revised, 10th edition pages 411-414, describes the IRV method of tabulating votes as an example of preferential voting and states that it "makes possible a more representative result than under a rule that a plurality shall elect," and "is especially useful and fair in an election by mail if it is impractical to take more than one ballot." Tbouricius (talk) 13:48, 31 December 2007 (UTC)


 * CommentThis is cherry-picking what is in Robert's Rules to extract what appears to be a recommendation from a context which is much less useful for promotion of IRV, if it is a recommendation of IRV at all-- and it is not. It notes as potentially useful, "preferential voting", then describes IRV as an example, then notes a serious problem with what it has just described. There is no room in the introduction for a brief mention that does not cherry-pick. I tried and it was all taken out by the same editor who just voted. Deeper in the article there is a fuller explanation, and, as I recall, even that may not tell the whole story, I have not reviewed it lately. --Abd (talk) 00:00, 1 January 2008 (UTC)


 * What I would REALLY like to know is what RONR does NOT mean in regards to Preferential voting. That it is if NOT IRV, what else? I can't imagine they mean Borda count, not even Condorcet method since it isn't a complete method by itself. I'm just unable to imagine any other interpretation besides IRV - single-vote counts, bottom-up elimination, i.e. rules automating an exhaustive runoff. If there's special conditions to consider - requirements that exhausted ballots still be counted, that doesn't make it NOT IRV, just a special rule that blocks a winner in such cases.
 * P.S. Existing RONR reference in article at: Instant-runoff_voting Tom Ruen (talk) 07:05, 1 January 2008 (UTC)


 * I see nothing to indicate that RONR would prefer IRV over Borda or over any particular type of Condorcet method. It simply indicates that a method that takes into consideration a voters less preferred choices is preferable over a plurality vote election.meamemg (talk) 18:04, 1 January 2008 (UTC)


 * Comment Roberts Rules, clearly, from its description of Center squeeze, would prefer a method which does not neglect expressed preferences. Ruen's comment about Condorcet methods is strange, he knows better, for, yes, Condorcet refers to the Condorcet criterion, i.e., it's not a method but a standard which methods follow or not, so, presumably, Robert's Rules would prefer a method that satisfies the criterion, and there are quite a few of them, see the article. So, as an example, Robert's Rules, but for one consideration, would prefer, as to performance, Schulze method over IRV. The consideration? It's not in common use, and RONR is descriptive, i.e., does not develop new procedure, it only reports and discusses common existing procedure, and single-winner STV was common enough to describe. They could have described Bucklin voting, which, while it can be argued that it is not Condorcet-compliant -- they would be uncomfortable with that -- doesn't have the center squeeze problem that they are explicitly concerned about. Center-squeeze is a relatively *common* problem, frequently causing trouble in top-two runoff contests. The compromise candidate, preferred by a majority over all other candidates, is eliminated in the first round due to insufficient fanatic support, er, Core support. Even in well-behaved societies, the result can be serious dissent and division, weakening the bonds of voluntary participation, and in poorly-behaved societies, the result can be civil war. It's not a minor problem. --Abd (talk) 20:20, 1 January 2008 (UTC)


 * The reason I'd assume RR isn't talking about Condorcet is because it already states a preference for an exhaustive runoff when people can vote directly, which is a one-vote system, repeated balloting until a majority unify behind a single choice. In practice I accept runoffs are crude and ENCOURAGE coalitions of voters to use whatever power they have. I'd certainly support a Condorcet count via rank-ballots and live voters. BUT BUT BUT BUT BUT BUT BUT I question whether voters will trust the result. It's all wonderful to say A can beat B,C,D,E,F each head-to-head BUT the question is what WOULD voters do if they KNEW the result before they voted? If Condorcet is wonderful, it ought to be TESTED in practice, not just when absentee voting is used, but in direct voting. Why doesn't RR encourage preferential voting over exhaustive runoffs? Tom Ruen (talk) 21:11, 1 January 2008 (UTC)


 * I've never heard the phrase exhaustive runoff before, but I am assuming it means the system where you keep voting (and no candidates are removed from the ballot) until one candidate receives a majority of the votes. The reason the RONR prefers this to any other from of voting is because under RONR all decisions must be made by at least a majority vote of the members.  Under RONR no decision (including elections) should be made until a majority of the members agree that it is the proper decision to be made.  No form of preferential voting ensures that this is the case.  I'm thinking what I wrote here is not very clear, so please let me know if it makes any sense. meamemg (talk) 21:50, 1 January 2008 (UTC)


 * Yes, Meamemg is correct. "Exhaustive runoff" is not what RR recommends. They recommend repeated balloting, with no candidate elimination (runoff implies candidate elimination). What happens with repeated balloting is that candidates make coalitions or agreements and some withdraw. Sometimes it takes a lot of ballots, but, then again, by refusing to compromise, voters are essentially saying that they would continue to try to get a better outcome, over closing the process prematurely. A majority can terminate this process at any time, if they can come to agreement, and RR dislikes that *any* decision be made by less than a majority. Doing nothing is better than minority rule, would be a succinct way to say it. --Abd (talk) 05:38, 2 January 2008 (UTC)


 * Thanks, yes Exhaustive ballot, varied rules but usually(?) means no forced elimination. Why couldn't they recomment Condorcet FIRST, and then go to Exhaustive runoff if Condoret produces no winner? I mean it seems more efficient, if you want a faster process. I mean I watched political endorsement runoffs WITH elimination rules still take 10 hours of voting, WHILE Condorcet would likely clear out the winner in a single vote - WELL IF everyone is satisfied with the discussion before the vote. I SUPPOSE you could AS WELL do a Condorcet count AND require 50%+1 of the ballots on each pair for victory, so truncated preferences could cause no winner, and repeated balloting that way! So why don't they recommend this? Tom Ruen (talk) 22:02, 1 January 2008 (UTC)


 * Ruen's questions have been answered above.... and, whatever, Roberts Rules does not "recommend IRV," unless one neglects portions of the text and interprets the rest as a recommendation. Which violates WP:RS, explicitly. Exhaustive ballot uses candidate elimination, *not* recommended by RR --Abd (talk) 05:38, 2 January 2008 (UTC)


 * I think we can agree to not use the word "recommend" as regards IRV and RONR...but it is correct and balanced to say simply that RONR says preferential voting (with IRV as its example) "is especially useful and fair in an election by mail if it is impractical to take more than one ballot." and that it "makes possible a more representative result than under a rule that a plurality shall elect."


 * As to whether RONR means exclusively IRV...No, we can agree it could mean Bucklin, or some other two-step method (not Borda or Condorcet), since it defines "preferential voting" as "any of a number of methods by which, on a single ballot when there are more than two possible choices, the second or less-preferred choices of voters can be taken into account if no candidate or proposition attains a majority." (emphasis added) This definition excludes most ranked-voting methods, but allows at least contingent vote, alternative vote, and Bucklin. Tbouricius (talk) 22:00, 2 January 2008 (UTC)


 * I replaced "recommends" with "advises using", after studying the dictionary definitions of the terms. It is clearly more neutral, and completely accurate.  We already have the detailed quotes later in the article, and I don't think they are appropriate for the summary introduction. MilesAgain (talk) 22:12, 2 January 2008 (UTC)


 * I don't see how this improves anything. Merriam-Webster defines "advises" as "recommends", so nothing is really changed.  Perhaps the sentence could be replaced with "Roberts Rules of Order, Newly Revised recommends preferential voting in cases where multiple ballots cannot be taken, rather than allowing election by a plurality vote."  meamemg (talk) 23:17, 2 January 2008 (UTC)


 * American Heritage has "To offer advice to" as the first definition. "Suggests" seems too weak to me, but perhaps, that is a good thing for NPOV. MilesAgain (talk) 00:40, 3 January 2008 (UTC)


 * I have no real problem with using the would recommend (in fact I used it in my proposed language). I'm not sure where you got the word suggest from, but I think that works fine too.meamemg (talk) 03:58, 3 January 2008 (UTC)


 * Wow! A fun intepretation - only IRV(+variations) or Bucklin apply to the incomplete description. It is most curious to me to wonder if they meant to allow Bucklin, or if it is merely an accidental technically of an incomplete definition! Of course there's more - apparently also compatible with a sort of two-round Bucklin with second round counting ALL lower preferences, which Approval advocates would love! Fun fun fun! Tom Ruen (talk) 22:31, 2 January 2008 (UTC)


 * RONR does not allow any form of Preferential voting. Any voting method other than a standard majority elects must be authorized by an organizations bylaws. meamemg (talk) 23:10, 2 January 2008 (UTC)


 * Meamemg, by standard majority, I assume you mean one person, one vote. I'm very happy for that position myself, but the debate above appears to be entirely whether or not preferencial voting (as recommended by RONR) can mean anything besides IRV. IS a top-two IRV (contingent vote) is acceptable to RONR? Or what else? Tom Ruen (talk) 23:43, 2 January 2008 (UTC)


 * By standard majority I meant a system where each person gets one vote and a candidate has to receive a majority (more than half) of the votes. As RONR defines preferential voting to be "any of a number of systems" and that "preferential voting has many variations", I would say that, yes, it does include systems other than IRV.  But, RONR will "allow" any system that is authorized by a groups bylaws.  meamemg (talk) 03:56, 3 January 2008 (UTC)


 * (edit conflict) No, that's not what meamemg means. He means that a majority of voters have consented to the election of a candidate. This varies from a simple Yes vote on a motion like, "Resolved, so-and-so is elected to the office of chair," to a majority voting in a vote-for-one election, to an accumulated majority in a preference voting election. I disagree that preferential voting cannot be used without bylaw authorization, but with an important caveat that I've raised many times: a majority must be a majority of ballots cast with votes for eligible candidates, not a majority remaining after some have been eliminated. With that caveat, preference voting is merely an expedited means of discovering a true majority.


 * What does Robert's Rules mean by "preferential voting"? First of all, they are explicit that the method they describe is only one method of many, and then they are explicit that this particular method, that they describe, exhibits a problematic effect which is commonly known as center squeeze. There are other preferential methods which don't exhibit this effect; any Condorcet method, for example. So why did they not describe another method? I'm tired of writing why. There is a reason. It's irrelevant here. They did not describe another method, but they clearly state that they exist, and Tbouricius has named some possibilities. Bouricius is correct that Condorcet does not explicitly satisfy the description, but "preferential voting" usually refers to the ballot rather than to the specific method of analyzing the vote. The parliamentarians who edit Robert's Rules are not generally election methods experts, and their goal was, as far as I can see, merely to note that other methods exist; they were also aware of a serious problem with the method they described, because they note it. It is moot here what the other methods would be. They are not IRV, that's the point.


 * If someone reads our article and thinks that implementing IRV as it was described in this article when we started this is satisfying Robert's Rules of Order, they would be mistaken. Yes, now, we do mention in the article the majority problem, that the winner in the last round of an IRV election may not have a majority vote; it was actually quite a battle to get that it the article, and there was continual attempt to have the word "majority" used to describe this plurality in the last round. "Majority" is a selling point for IRV, and that's why this has been so strongly defended.


 * Ignored by some in this discussion, and in particular by MilesAgain in his efforts to fabricate "compromise language" is that we have been over all this before. Essentially, if the pro-IRV editors -- who are clearly attempting to make the article be effective in promoting the method -- want the mention of Robert's Rules in the introduction to the article, then, too, the *negative* comments that Robert's Rules makes should be there, plus *no* interpretation of the source should be done as part of this. "Recommends" and "advises" are, as noted, the same. There is no necessity that RR be mentioned at all in the introduction. It's an attempt to promote the method by adding the cachet of Robert's Rules to it, and, quite obviously, the same editors do not want the opprobrium that comes with this to be there. Only what they think readers should know, which is biased by POV.


 * Let me remind readers that MilesAgain is an acknowledged sock puppet, and he is clearly working quite boldly to promote IRV with his edits. This article was maintained as a propaganda piece for a long time by a cabal of sock puppets and the Executive Director of FairVote, editing "anonymously" -- he, likewise, acknowledged that (but it was already apparent from the IP address). Tbouricius, from the timing of his entry here, was so reasonably suspected of being a "meat puppet" that he was blocked as such; I helped with the recovery of his access, but we must keep in mind that he is a Conflict of Interest editor. Tom Ruen has listed FairVote as an affiliation; he may not be COI in the same way as Bouricius, but he is an advocate. These editors are now cooperating to make large numbers of edits to the article in pursuit of the original agenda, and restoring Roberts Rules "recommendation" to the introduction is definitely part of that. None of this means that any particular edit is inappropriate, nor that these are not sincere and independent editors in various ways; however, it is becoming increasingly difficult to maintain the move toward NPOV that was accomplished, for a time, with this article, if the continual insistence upon controversial content continues.
 * --Abd (talk) 04:04, 3 January 2008 (UTC)


 * Let me remind readers that Abd is an acknowledged vocal supporter of an alternative voting reform, who thinks IRV is inferior to his favorite method which is not actually used for government elections anywhere. He has admitted that his work on IRV-related articles and talk pages is to bring them in line with his own POV, which is that FairVote, whoever they are, has been conspiring to make IRV look better than his favored alternative. He constantly flaunts WP:CIVILity by insinuating that other users are conspiring to push their POV while openly admitting that his goal is to push his own POV. If he continues with this absurd invective, I intend to report him on WP:WQA. MilesAgain (talk) 09:34, 3 January 2008 (UTC)


 * I should add, Tom Ruen has stayed away from most problematic edits, I do not mean to imply that all the users I mention above are acting improperly, and, as well, it had seemed to me that User:Tbouricius was cooperating well toward the improvement of the article until User:MilesAgain raised the old issues, I had thought resolved by consensus, as if nothing had ever happened. --Abd (talk) 04:16, 3 January 2008 (UTC)

Summary: Robert's Rules in Intro
The editor creating the RFC conflated issues making it more difficult to determine consensus on each one. I'm splitting them for a summary. I've added possible COI or other information relevant to voter POV. The summary is, of course, my opinion; please correct errors. Add new info if relevant. --Abd (talk) 20:27, 4 January 2008 (UTC)

also commenting, without Yes or No: User:Tomruen. known advocate of IRV, "affiliated" with FairVote, no formal COI known. No opinion expressed on Robert's Rules issue.
 * Yes: MilesAgain, sock, unknown master.
 * Yes: Tbouricius, COI editor: paid consultant to FairVote, legislative proponent of IRV.
 * No: Abd known critic of IRV, no affiliation.
 * Yes/ No: meamemg, presumed neutral, however, confirmed Robert's Rules mention was not "recommendation." Meaning of Yes/No not clear.
 * The Yes/No was because the RFC contained multiple issues. On the issue of RONR sentence as it was at the time of the RFC, I am a no. meamemg (talk)
 * What is your opinion of the current version? ("Roberts Rules of Order, Newly Revised suggests using preferential voting when an exhaustive ballot election is infeasible, such as for elections by mail, giving IRV as its example.") MilesAgain (talk) 09:00, 8 January 2008 (UTC)
 * Comment: Editor MilesAgain asked three questions together in this RFC. The very fact that these questions were asked together, and answered Yes or No, was a demonstration of factional affiliation. Absolutely, with variations, I could answer Yes to none, one, two, or three of these questions. However, MilesAgain has never engaged in a process of actually seeking consensus here. First of all, he's raising old issues, there has been a great deal of discussion on this point in the past. Secondly, in his question on Robert's Rules, he incorporates a presumption.: "Should the introduction of the IRV article include ... the fact that Robert's Rules of Order, Newly Revised recommends preferential voting for elections by mail and gives IRV as the example?" So the only question is whether or not the "fact" should be included?


 * NPOV does not merely cover "facts," it covers the manner of their presentation, it must be balanced.


 * Then, when he got a "No" from an editor who has no apparent bias on this question, he made an inconsequential change to the phrase in the introduction and claimed that this satisfies consensus. When this was pointed out, he again changed the wording, and now claims that there is no consensus in the RFC. (Note that previously, same RFC, same situation, he claimed consensus.) Unspoken assumption: therefore he can put what he wants in the introduction. No. NPOV is required everywhere in the article, but introductions are necessarily brief, and NPOV, with complex issues, requires substantial text, which ordinarily rules it out in the introduction. What is in the introduction should be *unquestioned*. Note the history of this issue with Robert's Rules in this article. When I pointed out, long ago, that Robert's Rules was *not* "recommending" "IRV," this was considered, by the IRV advocates who have long attempted to dominate this article, preposterous. However, Wikipedia policy is clear: a source may not be paraphrased or interpreted, unless this is not controversial in the least, and this particular paraphrase is not only controversial, it's wrong, and that has been explained in detail, over and over. Now, in an edit replacing the RR mention in the introduction, Miles Again summarizes: "... RFC respondents are split on the question of RONR summary (actually one no was for an earlier version)) "


 * Actually, RFC solicits comment from new editors, those not already participating, and only one appeared, who voted No on the actual question posed by MilesAgain. The analysis above shows that the two Yes responses are coming only from MilesAgain, an acknowledged sock puppet, and a COI editor. Both of these should not, by Wikipedia policy, be engaging in controversial edits (and Tbouricius seems to be staying away from the worst edits, often accepting reasonable compromises, MilesAgain is the one seriously pushing POV.)


 * We *can* have the RR information in the introduction. But it must be balanced. We dealt with the problem in the description of this in the middle of the article by going to an almost exact quote of Robert's Rules, which does describe a method which can certainly be given this name "Instant Runoff Voting." But that is a name covering a whole host of methods, depending on details, and IRV as it is normally being implemented in the United States is *not* what Robert's Rules actually recommends. Again, this interpretation was ridiculed by Tbouricius when first advanced; yet, in fact, the legislation he introduced in Vermont was actually consistent with the Robert's Rules form: if no true majority appeared, the election failed (in Vermont, the constitution requires a majority vote for governor; if none is obtained, the legislature picks from among the top three. That, of course, is not Robert's Rules.... but the election failure is, unless bylaws specifically allow election by plurality, which Robert's Rules discourages.)


 * The section in Robert's Rules on "preferential voting," after describing the particular form of sequential elimination used by IRV, proceeds to note a major problem with it. If we are going to note the description of IRV in the introduction, we must also note, there, the problem. Otherwise we have extracted one side of a story and have preferentially presented it. Which if these facts is more important: that Robert's Rules describes IRV, or that it criticizes it? Could it be that the answer to this question depends on one's POV. To an advocate, the "recommendation" is important. To a critic, the "problem" is important. In the middle of the article, we can use sufficient text to have both.


 * Robert's Rules, introducing its preference for repeated balloting, says "Although this type of preferential ballot is preferable to an election by plurality..." This has not been understood by the IRV advocates -- or if they do understand it, they choose not to act on the understanding. Robert's Rules dislikes any election by plurality *including one that takes place with IRV.* IRV may have offered the voters more opportunity to express a majority preference, as would other forms of preferential voting or Approval, and this is the meaning of the phrase. From that context, it's clear that Robert's Rules considers IRV as problematic, but what the POV-pushers here have done is to extract this out of context, making it into a pure affirmation, as we have it now as I write this:


 * "Robert's Rules "describes IRV as an example of preferential voting, which it states is "more representative" than plurality elections."


 * I agree that statement is biased, and I have replaced it with the "suggests PV...giving IRV as the example" version which I believe is not biased. Incidentally, the narrative in the four preceding paragraph is incorrect, but I don't think the mistakes are so substantial that they need to be corrected.  I just want to be on record as saying there are inaccuracies in Abd's text above, in cause it is ever repeated as truth. MilesAgain (talk) 09:09, 8 January 2008 (UTC)


 * Is that what it says? What it says gives a "more representative" result is "preferential voting," not "IRV." *Then* it describes one example "by way of illustration." This is essentially IRV; but the background is that the last round does not automatically produce a majority; the counting method refers to "more than half of the ballots," not of "unexhausted ballots." Then it criticizes *this specific form of preferential voting,* with a criticism that does not apply to all forms. "Preferential voting" refers to the ballot form, not to the counting method. As to other forms, there is Borda count, Bucklin voting, any Condorcet criterion-satisfying method such as Schulze method, and, indeed, Range voting could be considered a form of preferential voting because it does allow ranking of preferences -- it is little more than Borda count with equal ranking and blank ranks allowed.


 * Why did not Robert's Rules give a better method? Well, Robert's Rules is *descriptive* of parliamentary practice, and there are too few examples of other forms of voting in use. Robert's Rules cannot lead by recommending some method not in reasonably common use.


 * What to do? It is not acceptable to have a biased "sound bite" from Robert's Rules in the introduction. So, pending some other resolution here, I will move the whole RR description to the introduction. I'd prefer to have it later, it is, in my view, too much detail for the introduction. However, the presence of firm opposition from MilesAgain, supported by Tbouricius, makes this an approach that does not involve constantly reverting the edits of MilesAgain. --Abd (talk) 03:57, 6 January 2008 (UTC)


 * I agree that the entire paragraph is completely wrong for the introduction. How would you summarize that paragraph? MilesAgain (talk) 20:45, 6 January 2008 (UTC)


 * Well, we came, quite some time ago, to the placement of the entire paragraph later in the article. Now, MilesAgain, would you do what you have not yet done: tell why this should be in the *introduction*? I know of only promotional reasons. As it happens, this particular claim -- in its unexpurgated form -- is found all over the internet, as a result of the diligent work of IRV supporters and it's long-standing placement in this article. Cities (or at least one city) quote it in their pages explaining Instant runoff voting: "Robert's Rules recommends IRV." FairVote certainly was effective. I do not know how to "summarize" the paragraph without removing something important. It's already dense writing, not fluff. An anonymous editor took it all out, suggesting it go back later in the article, below the TOC. I think this editor was right. But you put it back later *and* in the introduction as the "sound bite" again. Why? *Why must this be in the introduction*? I reverted it back, it's in the introduction because you and Bouricius insisted that it be there. *Why must a "summary" of this paragraph be in the introduction, and are you aware that summarizing a source can violate WP:RS and WP:NPOV if it imbalances the reference? I do think we might successfully summarize, but I don't think we will get it down to the point that it could go in the intro. --Abd (talk) 04:24, 8 January 2008 (UTC)


 * It should be in the introduction because there has for at least a year been summary paragraph in the introduction listing all of the major places that IRV is used, and Robert's Rules represents the policy of all sizes of organizations throughout the globe. I think you need to read WP:LEAD. Tell me, what is inaccurate or biased about, "And Roberts Rules of Order, Newly Revised suggests using preferential voting when an exhaustive ballot election is infeasible, such as for elections by mail, giving IRV as its example."? MilesAgain (talk) 08:58, 8 January 2008 (UTC)


 * Length of time in intro is irrelevant. Article was controlled by a cabal of sock puppets and the Executive Director of FairVote as an IP editor (self-acknowledged in protesting the resulting block). The text quoted above is incorrect (it implies RR would approve of exhaustive ballot, when the opposite is true), but if we correct that to "repeated balloting with no automatic candidate eliminations," there is still the problem that the same passage criticizes the specific sequential elimination method used in IRV, which implies that there might be other methods of preferential voting without that problem, and, indeed, there are. The language the sock MilesAgain took out today was written by Tbouricius, thus, I had hoped, resolving the edit war MilesAgain had started by his insistence on having a mention of Robert's Rules in the introduction, since my objection was not to the mention of Robert's Rules, per se, but to an imbalanced mention creating the appearance of a recommendation when the truth is not so simple. I have reverted MilesAgain's latest deletion, and the next step, should this edit warring continue, would be pursuing administrative notice regarding disruptive editing by a sock puppet. --Abd (talk) 04:00, 12 January 2008 (UTC)

Okay, let me make this as simple as possible. This is the language that Tbouricius wants: "Roberts Rules of Order, Newly Revised describes IRV as an example of preferential voting, which it states is 'more representative' than plurality elections in cases where multiple rounds of voting are impractical, such as in elections by mail." That is completely correct, but "more representative" is unquestionably biased in favor of IRV, and therefore unbalanced.

This is the language that Abd wants: "Roberts Rules of Order, Newly Revised suggests using preferential voting when repeated balloting to find a majority winner is infeasible, such as for elections by mail, giving fully-ranked IRV as an example; however it also notes that sequential elimination under IRV may eliminate a compromise candidate, due to insufficient first preferences, who could be supported by a majority over the winner IRV finds." That is also completely correct, but "may eliminate a compromise candidate" is unquestionably biased against IRV, and therefore unbalanced.

This is the language that I want: "Roberts Rules of Order, Newly Revised suggests using preferential voting when repeated balloting to find a majority winner is infeasible, such as for elections by mail, giving fully-ranked IRV as an example." That is also completely correct, and contains no biased language whatsoever. Moreover, it SUMMARIZES, which is what the introduction is supposed to do, unlike the other two excerpts above which include incidental details that are best, and are in fact, described in more detail in the body of the article. MilesAgain (talk) 00:28, 15 January 2008 (UTC)


 * The problem with MilesAgain's interpretation above is that "Roberts Rules suggests IRV" is an interpretation of the source to make it into a promotional "fact," just as much as the acknowledged inappropriate language "recommends" or "advises," which were also tried. The language Tbouricius used is quite close to the source, almost an exact quote, plus he allowed the criticism of IRV found in that source to remain in brief summary; thus his compromise was acceptable. MilesAgain's text is not a compromise, it is actually an insistence on keeping a promotional "fact" extracted from the source while excluding the caveat which the source provides.


 * Facts are not "biased," and the normal response to imbalance created by citing sourced facts, that may have some imbalanced implication alone, is not to remove them, but to balance them. What MilesAgain wants is actually incorrect. Robert's Rules describes IRV, indeed, as an example, but it then turns around and warns about that very example. Is this a "recommendation"? Actually, no. RR is pointing out that IRV is a possible solution to the mail ballot problem, one that the parliamentarians who were writing the text clearly dislike (they don't like voting by mail and they don't like runoffs with candidate eliminations), but one which is in actual use. They were probably aware of other methods (after all, they've been known for centuries), but they were constrained, they cannot recommend a method that is not in reasonably common use.


 * MilesAgain has been persistent on this, reverting his favorite language back in every few days. One of the problem with socks is that they don't risk as much by edit warring as an open editor. If he's been careful, and has not used his other account for more than a month (and that's what he claims), it could be very difficult to identify that account; if he gets blocked for edit warring, he could merely wait a little while and then resume activity with his old account. However, I'm reverting him again, since he has removed consensus language, with an utterly spurious argument, language that was found to be acceptable by an expert associated with FairVote -- he wrote it! -- as well as myself, a critic, language that appeared to have ended conflict on this issue. He is removing *sourced fact*, necessary for NPOV balance; without at least the level of detail that was agreed upon, the mention of Robert's Rules in the introduction is POV, creating the appearance of pure recommendation. There was all along another non-controversial position for this information, deeper in the article. There is no necessity that the Roberts Rules mention be in the introduction. But I agreed that it's significant enough to be there *if* it is balanced. A pro-IRV activist, though, of the kind who manipulate public perception, might not be content with that; such a person knows that many readers will just look at the introduction for a quick explanation, and would want that idea {Hey! Robert's Rules suggests IRV! Must be great!) implanted through it. And this is how MilesAgain has conducted himself, again and again and again. Tbouricius, on the other hand, was a legislator, and understands compromise and how to work together toward common goals, in this case, improving the article. Both of us, he and I, want to see this article be correctly informative, interesting, and balanced. We certainly have differed and will probably continue to differ on many points, but the debate has generally been fruitful. --Abd (talk) 06:28, 15 January 2008 (UTC)


 * Your own version includes "Roberts Rules suggests IRV". Your idea of "balance" is to include facts showing a disadvantage, which serve to negatively bias the statement.  Why not include Tbouricius's facts, which point out an advantage?  Neither are appropriate.  You acknowledge that "recommends" and "advises" have been tried and rejected, and at the same time imply that I am not working towards compromise.  And you stoop to explicit personal attacks, again, after several warnings.  You leave me no choice but to start another RFC. MilesAgain (talk) 09:27, 15 January 2008 (UTC)

Summary: explanation of why "instant runoff" in Intro
No longer contested? MilesAgain (talk) 09:02, 8 January 2008 (UTC)

Summary: IRV is a "single vote" in Intro

 * Yes: MilesAgain, sock, unknown master.
 * Yes: Tbouricius, COI editor: paid consultant to FairVote, legislative proponent of IRV.
 * No: Abd known critic of IRV, no affiliation.
 * Yes/No: meamemg, presumed neutral. Application of Yes/No to this issue not explicit, however explained that it "doesn't make a lot of sense," and "I question the need for that statement to be in the introduction."
 * WTF - User:Tomruen, known unpaid advocate of common sense. This consensus question is incomprehensible to me. (i.e. 100% support single vote as vital fact worthy of introduction.) [Clearly 'Yes' per below. MilesAgain (talk) 08:59, 8 January 2008 (UTC)]

I see no consensus for the usage of the term "single vote" in the introduction. The only argument actually presented, as far as I noticed, was Bouricius arguing that "this is the FUNDAMENTAL difference between IRV and other preferential voting methods," yet the method is fully explained without using the term. Is there a source for this analytical claim? Again, *why must this be in the introduction*? I pointed out that the alleged "fact" that IRV involves only a single vote is legally controversial, possibly to be decided in Minnesota. I pointed out that the name of the method "Single Transferable Vote" refers to the "single vote" preserved and cut up in multiwinner elections, that is, normally most voters actually end up having created a winner with their vote, or winners with fractions of votes spread out. I have no objection to calling IRV, "single-winner Single Transferable Vote," but that is not what the article had, it had single vote, thus *interpreting* the name. Unless someone comes up with a good reason it has to be there, I'm continuing to oppose it and intend to remove it if it is replaced, or a consensus of *neutral* editors appears. --Abd (talk) 04:46, 8 January 2008 (UTC)

User:Tomruen has now made a comment on this (his comment above was actually an edit conflict as I was fixing the subhead). This "vital fact" is an interpretation. There are other interpretations that consider IRV to represent more than one vote (it's an argument actually being advanced against IRV), and this has been sourced here.--Abd (talk) 05:07, 8 January 2008 (UTC)


 * If STV is single vote, then IRV is single vote. If two round system and exhaustive ballot are single vote, then IRV is single vote. You can argue people may have LESS votes (if they fail to rank), but never more than one! Tom Ruen (talk) 05:57, 8 January 2008 (UTC)

The source of Minnesota controversy is a citizen's group under the name Minnesota Voters Alliance. I don't think a lawsuit implies anything unless it is ruled in the favor of one side or other, but quoting from their website.  1. Proponents claim IRV is constitutional even though the 1915 Minnesota Supreme Court ruled in Brown v. Smallwood that preferential voting is “contrary to the intent of the Constitution.”  They claim the ruling was “...not because it was preferential voting per se, but because it involved the Bucklin method of counting which had the effect of giving some voters more than one vote.”   (see FairVote letter)  Our response: Proponents argue that IRV does not share this “fatal flaw” because, in IRV, only one of each voter's rankings ends up counting - the one applied to the highest preferred candidate eligible to receive it. Their argument fails to address the real issue, which is NOT how votes are counted, transferred or otherwise manipulated, but rather how many votes each voter is allowed to cast!  The state's highest court emphasized that the Constitution, by implication, forbids any elector from casting more than a “single expression of opinion or choice.” Since IRV, like the Bucklin method, allows voters to cast multiple votes (choices), it too shares this fatal flaw!  IRV has a 2nd fatal flaw; all ballots are not counted equally! While the Bucklin method counts all secondary votes when necessary, IRV only tallies the second choice votes on the ballots cast for defeated candidates.  This means some voters' have their first AND subsequent choices counted while others have only their first choice votes counted. This is highly undemocratic! 

A citizen's group can sue that red stop signs are actually green - should we withhold the scientific consensus on the color of stop signs until the lawsuit fails?! Tom Ruen (talk) 07:29, 8 January 2008 (UTC)

Tom, there is no "scientific consensus" that a voter on an IRV ballot casts a "single vote." The voter, quite clearly, may cast more than one, and depending on which ones the voter casts, somewhere from none to more than one end up affecting the outcome. The lawsuit is based on clear precedent in Minnesota -- commonly misrepresented by FairVote attorneys --, which conflicts with some precedents elsewhere. I think the MN precedent should be overturned, but .... that does not mean that an IRV voter only casts a single vote, because the voter clearly may cast more than that, with only one effective at a time. We solved this problem by avoiding the "single vote" language and simply using the link text that was concealed underneath it. I.e., "single vote", appearing as a link to Single transferable vote simply became Single transferable vote, and while I could quibble with that, believe it or not, quibbling is not what I'm doing here. The method has been known as a form of Single transferable vote for a very long time. It was not necessary in the introduction to make the "single vote" argument as such. --Abd (talk) 05:59, 15 January 2008 (UTC)

Practical implications - Tally each possible ballot combination
Practical implications

"Under IRV, unlike some other voting methods, the record of votes cast in a particular area cannot be conveniently summarized..."

I would note that only the total of each possible ballot combination needs to be sent to the central location. This is explained using the example earlier in the page.

I would say something like this. "Under IRV the record of votes cast in a particular area cannot be conveniently summarized into one simple total for transfer to a central location. Instead it is often necessary to either transport the actual ballot papers or data files that contain a total of each possible ballot combination to the central location to allow for the IRV tabulation. Using the total of each possible ballot combination is seen in the above example."

Bill Arden (talk) 21:28, 11 January 2008 (UTC)


 * The current language is easy to understand. Any replacement language should be as simple as possible so that the average reader can get it. --Ask10questions (talk) 16:24, 13 January 2008 (UTC)

(unindent)The suggested language ignores some practical realities. In Australia, what is done, I understand, is to count the first rank votes, which are transmitted to a central location; lower ranks are only counted as needed. Yes, if there are few candidates, or if the data is initially collected by computer, complete ballot data could be transmitted, but this requires much more detailed counting. With many candidates (consider San Francisco, with 17 candidates for one office) transmitting the raw ballot data could be more efficient than transmitting all possible combinations. Voting security experts consider the failure of the Summability criterion by IRV to be a problem. In particular, there are those who think that IRV increases reliance on computerized voting machines, and thus is a force toward the use of these, and they think this increases fraud risk over hand counting. Personally, I think this is merely one additional consideration, and not a crucial reason not to use IRV. As the Australian experience shows, hand counting is possible; the down side of this, though, is that complete ballot data may never become available and thus failure to elect a majority winner, even though one exists as shown by the ballots, may not be detected. --Abd (talk) 16:47, 13 January 2008 (UTC)

Cary, NC election result.
I've finally been able to find actual IRV results from Cary, it was hard for a while. This election was nonpartisan, and the real contest was between two challengers to the incumbent, Roseland, who did not make the "runoff." (I've read articles about this election in NC newspapers that don't mention the IRV but which do mention the runoff.) The Cary official results page shows Frantz, 1151, Maxwell, 1075, and Roseland 793, with 3 write-in votes, total votes, 3022. Because Frantz did not receive a majority, the IRV method was used to transfer ranked votes. This is a Ranked Choice Voting scheme, it appears, like San Francisco (which, in this case, should be adequate, even if a voter casts a write-in vote.) The official results page, which only shows first round votes is at. Then there is the IRV page, which shows the results of vote transfers, it's an Excel spreadsheet at.

There is a 1 vote discrepancy, the spreadsheet only shows 1150 first rank votes for Frantz. With Roseland eliminated, and the three write-in votes as well, Frantz received 248 second choice votes and 3 third choice votes, total 1401. Maxwell received 274 second choice votes and 4 third choice votes, total 1353. I've seen this with the San Francisco votes: the IRV distribution tends to not favor one candidate over another, which is why plurality winners are going on to win the election. This was a fairly close election, though. Now, Maxwell won with only 46.36% of the votes; it looks like a good number of the 793 Roseland voters did not rank Frantz or Maxwell; at most 66% did.

The third rank votes are a bit mysterious. There are a total of 7 that were assigned to Maxwell or Frantz, there may have been more assigned to Roseland. There were only 3 write-in votes in first rank (which is what the official results page shows). One possible explanation is that there were 4 write-in candidates in second rank; then the voters voted for Maxwell or Frantz in third rank.

This is another contest which is just the same as all the San Francisco contests: uncontested, won in the first round with a majority, or, as in this case, the plurality winner went on to win, but not with a majority vote (except one obtained by disregarding exhausted ballots, and, note, take any election, discard ballots for all but two candidates, it will either be a tie or one will win with a "majority." That is not what Robert's Rules considers a majority, which would always be a majority of ballots containing a valid vote. Whether or not the 3 or 7 or ? write-ins were valid is not important here, the ballots with Roseland votes were valid.

So, sure, as FairVote trumpets for Cary, an "expensive runoff" was avoided, but at what cost? Not only was counting this election problematic (it was counted by hand, and "mistakes were made"), but the reason for the runoff in the first place would be to ensure an actual choice by a majority of voters voting. That happens in a runoff, though the set of voters may be different. (Whether or not a real runoff is superior to an "instant" one is debatable, to be sure; I'll merely note that real runoffs are a kind of ad hoc Range Voting. If the election is important to you, you vote, if not, you don't. But from an expense point of view, Plurality is usually producing the same result, victories due to vote-splitting don't seem to be that common, in spite of the fact that IRV -- as with any reformed voting method on the table -- would encourage more candidates and thus, theoretically, more apparent vote-splitting in the first round. My guess is that Approval voting-- just Count All the Votes, i.e., allow voters to vote for more than one, which is really just another form of alternative vote, or Bucklin voting, with plurality sufficient for election (as is really the case with IRV as all these examples show), would accomplish the same salutary effect as IRV -- runoff avoidance -- at much lower cost; with both methods, just count the votes and add them up. --Abd (talk) 03:17, 15 January 2008 (UTC)

By the way, turnout was quite low in the Wake County election. 56,032 ballots were cast with the number of registered voters being 512,515. For Cary, the election was "top two" runoff. To be elected in this way was the office of Mayor, plus three council seats. There were only two candidates on the ballot for Mayor, it was won with 58.31% of the ballots cast. Council District 1 had four candidates, it was won in the first round with 69.10% of the votes cast, District B had 3 candidates, and is the one described above, and District D was won in the first round with 55.00% of the votes cast.

To find these figures, I looked at the "Canvass" file at

Now, just to show, once again, how effective FairVote propaganda has been, I wrote above that a runoff was avoided. FairVote says "expensive runoff," and this claim has been repeated all over the internet. Wait a minute, I just realized. This election was a *primary*, as with San Francisco, before IRV was implemented, the primary was held, then the top-two faced off, unless someone got a majority. Unlike San Francisco, however, this "primary" was not held at the same time as the regular November general elections, which explains the poor turnout. It was held on October 9, 2007. If Cary were to hold their "primary" two months before the general election in even-numbered years, they would have almost no expense for runoffs.

What was the turnout for this particular election? To find out, I have to identify the precincts in the vote totals and then determine the number of registered voters in those precincts. (The same data is available that I used for vote totals, it is part of the Canvass file.). I just looked at the Mayoral election, so far: Total turnout across all Cary precincts was 20.11%. The overall turnout in Wake County for the November 2006 general election was 40.53%. I haven't looked at the Cary specific turnout figures, at least not yet. It's late....

So Cary runs their municipal elections in odd years. Let me say I'm not surprised to find this. General elections have much higher turnout, typically; the low turnout normally suits the status quo, it keeps the riff-raff out. It can backfire, though, if some constituency that does not normally vote in off-years comes out in droves.... This is why top-two runoff can be interesting, for it will sometimes motivate voters who'd normally skip an odd election. As I wrote above, it's a kind of ad-hoc Range voting, weighting votes according to how much people care about the result. --Abd (talk) 05:22, 15 January 2008 (UTC)


 * Cary also had a contest for "At Large" City Council member. Two at large candidates with drew from the contest because they were afraid they would dillute the vote and throw the contest to the opposition.  One of the candidates who with drew still had her name on the ballot and got a significant number of votes.
 * The raw vote data showing how voters ranked their choices has not been provided. NC has no intention to count these votes nor to account for these votes since there was no need for a "runoff".


 * In the District B, Nels Roseland told his supporters to rank him all three times, and he lost.
 * Don Frantz won in the "runoff". Maybe in part because he spent alot of time correctly explaining IRV to his supporters. --Ask10questions (talk) 08:41, 15 January 2008 (UTC)


 * Of course a first round held the same day as a major election will get more turnout. The problem is that both the first round and the runoff can not both be held on the same day as a major election.  What improvements to the article are being contemplated in the discussion of this section? MilesAgain (talk) 15:35, 16 January 2008 (UTC)


 * Miles again asks the exact same question in two sections, one right after another.... Same answer: Election detail in the section on implementation in the U.S., plus *possible* facts for the Practical implementations section, as well as *possible* comparison with runoff voting in that section.

As to the argument, MilesAgain has stated the obvious, indeed, it's what I said, except for the stunningly obvious "not the same day," and even if it was the same day, what use would it be to require the voters to hang around or come back later? There are other issues with runoffs, but it would be new analysis, and we all know about WP:OR, don't we? We *can* bring up OR here, though, especially because then someone may be able to find secondary sources that review the same or similar information. Surely there have been scholarly articles written about the situation with runoff voting.
 * --Abd (talk) 19:21, 16 January 2008 (UTC)

Astonishing IRV statistics
Since FairVote began their campaign in the 1990s, and particularly in recent years, quite a few jurisdictions have passed IRV legislation. However, as the article shows, by far, most of them have not implemented it. So far, there have been, as I count it, 25 IRV elections in three cities in the U.S. in recent years (i.e., after Ann Arbor IRV went down in flames). Every single one of these elections was won by the plurality winner from the first round, IRV vote transfers made no difference, except in avoiding mandated runoffs for majority failure, yet, ironically, out of the nine elections that went to runoff, none of them reached a true majority vote as a result of the vote transfers. Where runoffs were avoided, it could be argued in a few cases that the public was deprived of the right to make an informed choice. This is particularly poignant in the Cary, North Carolina, election, which was very close; the incumbent did poorly, and the two "insurgent" candidates split the vote fairly closely. It looks like about one-third of the incumbent's supporters voted second rank for one of the insurgents and one third for the other, and one-third did not add a second rank vote (which could mean that they cared not at all for either of the insurgents; but with an actual runoff, they might have decided there was a difference). Turnout was poor (20% of registered voters), because Cary holds their municipal elections at an odd time, in October of odd numbered years, not in connection with the General election, which seems to have double the turnout (i.e., 40% of registered voters). This was an election crying for an actual runoff, there is no reason to expect that turnout would have been worse; it would only have been held in the precincts for that Council seat. --Abd (talk) 05:50, 15 January 2008 (UTC)


 * Is this a suggestion for improving the article, or grandstanding? Wikipedia is not a WP:SOAPBOX, and talk pages are notaforum for general discussion of the topic which does not aim to improve the article.  Having said that, your statement, "there is no reason to expect that turnout would have been worse; it would only have been held in the precincts for that Council seat," shows a profound lack of understan ding of actual turnouts for runoff elections. MilesAgain (talk) 09:42, 15 January 2008 (UTC)


 * It's background. Note that MilesAgain seems to think this discussion inappropriate, in which case he's trolling in arguing against it, using a personal attack ("profound lack of understanding") -- albeit a mild one by Wikipedia standards.... Ordinary runoff elections, such as those in San Francisco, are held when the main election, typically held as part of a general election, fail to reach a majority. Because they are not part of a general election, the runoffs have very low turnout compared to the main election. Now, I could look at the Cary statistics for runoffs, they exist, I think. I was just commenting that there is no particular reason to expect low turnout, these are already motivated voters. MilesAgain's comment took what is actually two comments of mine (turnout and expense) and makes them appear as if they were one. So I made an additional *not connected* comment about the number of voters involved, in a very few precincts. That's not about turnout, that's about election expense. Cary *already* conducts relatively expensive elections by scheduling them in off-years. But this is dicta, and for MilesAgain to object to dicta as part of a substantial discussion is, quite simply, offensive. It's background for an understanding of how the facts which might go into the article have significance, why they might be considered important. That is, indeed, what Talk is about. In this case, it's about proposed material for the article, and is now about actual material in the article. Any question about the facts? Hey, I must have made *some* mistake!--Abd (talk) 20:48, 15 January 2008 (UTC)


 * Saying that a comment about the content and not the contributor is a personal attack shows a profound lack of understanding of the WP:NPA policy. Runoffs almost always get less turnout than their first round. An exception to that general fact was not uncommon in San Francisco, where IRV was adopted anyway. MilesAgain (talk) 15:30, 16 January 2008 (UTC)


 * Okay, so it's not a personal attack for me to note that, if I assume good faith, MilesAgain has a profound lack of understanding? I wrote what I wrote previously about a specific election, in Cary, NC. I was quite aware that runoffs in other situations commonly get much lower turnout. In Cary, in fact, turnout for runoffs was sometimes *higher* than for the primary election, and I explained an obvious reason. Is this relevant to the article? Could be. That's the point of talking about it. Lower turnout for runoffs is a common argument for IRV; but when primaries are scheduled prior to the general election, with runoffs, if needed, at the general election, this is turned on its head.


 * I have not researched turnout in San Francisco, but prior reports in the article -- I haven't looked lately -- claimed that the winner in a runoff sometimes got less votes than a loser in the primary. That certainly implies low turnout, and my point was that turnout was related to the timing of the election, and that is a fairly obvious point and I don't know why, except for general disagreement, MilesAgain would attack it.


 * (What I wrote about "your statement shows a profound lack of understanding" -- MilesAgain's language -- is that it is a personal attack, though a mild one. It is trolling, and offensive because it immediately followed a claim that the discussion did not belong here, but a comment like that invites response. And it was, quite simply, wrong. But, hey, so what! This is Wikipedia. Understanding is not a prerequisite for editing. --Abd (talk) 18:36, 16 January 2008 (UTC)
 * --Abd (talk) 18:36, 16 January 2008 (UTC)


 * No, please read WP:NPA. "MilesAgain has a profound lack of understanding," is a derogatory comment about a contributor and therefore a personal attack. "Your question shows a profound lack of understanding," is a comment about the content, not the contributor. MilesAgain (talk) 23:12, 16 January 2008 (UTC)

"Every word he has written here is a lie." Not an attack on him, eh? A distinction without a difference. If I thought it acceptable, I would say that MilesAgain's contributions here are nothing but disruptive, his arguments are specious and deceptive, and his objections evince a contemptuous disregard for courtesy and collaboration, but he, himself, is a nice guy, can't say anything bad about him, I'm sure he means well, but as far as his writing shows, he is apparently unable to accomplish his good intentions, "there are those unfortunates." But I wouldn't say that, all of it would be uncivil.

There was an arbitration case recently where a sysop lost his bit for blocking, based on NPA, another admin who had called his arguments a "steaming pile of crap." The admin who had made that statement was reprimanded for incivility, but it was also considered, correctly, as a comment on the argument. People with profound understanding can still sometimes make a bad argument, even a very bad one. (The sysop bit was lost because the admin was involved in the dispute. If he believed he had been attacked, and absent some emergency, he should have referred it to another -- who would doubtlessly have encouraged him to blow it off.) In this case, though, "your argument shows a profound lack of understanding" is actually a reference to the person, i.e., to how *the person appears,* since, in particular, arguments can't understand anything, they are only a collection of symbols, with meaning supplied by the reader, whether or not that was intended by the writer. "Your argument is a steaming pile of crap" doesn't actually indicate anything about the person who made the argument, except possibly that the person is capable of defecation -- hopefully, we all are -- but it was still uncivil. As I said, MilesAgain's "personal attack" was mild by Wikipedia standards. Nobody would block him for it, nor, indeed, would I be blocked for the allegedly more serious "personal attack" that I wrote. However, if we keep discussing each other rather than the project, we might indeed get slapped with a wikitrout. I mention from time to time that Miles is a sock *because it is relevant to editing guidelines.* That's all. I apologize for any incivility, but I was correct in what I'd written about runoffs, and the lack of understanding, if any, was not mine. Now, back to building the project. --Abd (talk) 02:55, 17 January 2008 (UTC)


 * If you truly apologize, does that mean that you will no refer to me as sockpuppet in a derogatory manner, and no longer take actions against me "stronger for socks than for identified editors," as you write above? I have abandoned my original real-name account and have told you why.  I have not edited with it for months now; therefore I am no longer a sockpuppet. This is my primary account now, and it is the only account I will be using unless a message to which I need to respond appears on my old account's talk page.  If you are unable to assume enough good faith, without evidence, then you need to explain why, in detail.  You have no evidence that I am doing anything malicious.


 * As for your statement, "there is no reason to expect that turnout would have been worse," it is false. I said it shows a profound lack of understanding, because you are editing a detailed article which is intimately involved with the subject of runoff turnout, and have been for months, but you apparently have not yet studied the literature on the subject of runoff turnouts, which I think is profoundly odd.


 * If you did study the literature, this is the sort of thing you would find: "turnout declined in almost 77% of all Democratic gubernatorial, senatorial, and congressional runoffs held from 1956 to 1984" If that is not a reason to expect that turnout would decline in a runoff, then what is? MilesAgain (talk) 03:30, 17 January 2008 (UTC)


 * (para 1) incorporated accusation, "will you no longer do XXX." As to actions, depends. Sock is a sock, MilesAgain began as an obvious sock, when I said so, he acknowledged it. There is a procedure for changing accounts, Miles has not followed it. I'm obligated to assume good faith, to a point. I'm not obligated to be stupid.


 * (para 2)MilesAgain, typically, is completely neglecting context. I was not writing about runoffs in general, I was writing about a very specific situation, an unusual one, where, indeed, runoffs don't have poor turnout, and he now knows why, so it is beyond me why he beats a dead horse, citing congressional elections where runoffs are special and the only issue on the ballot. Indeed, I expected to see lower turnout, until I actually looked. Apparently actually looking at evidence is a strange thing to some editors. This is what I wrote:


 * Where runoffs were avoided, it could be argued in a few cases that the public was deprived of the right to make an informed choice. This is particularly poignant in the Cary, North Carolina, election, which was very close; the incumbent did poorly, and the two "insurgent" candidates split the vote fairly closely. It looks like about one-third of the incumbent's supporters voted second rank for one of the insurgents and one third for the other, and one-third did not add a second rank vote (which could mean that they cared not at all for either of the insurgents; but with an actual runoff, they might have decided there was a difference). Turnout was poor (20% of registered voters), because Cary holds their municipal elections at an odd time, in October of odd numbered years, not in connection with the General election, which seems to have double the turnout (i.e., 40% of registered voters). This was an election crying for an actual runoff, there is no reason to expect that turnout would have been worse; it would only have been held in the precincts for that Council seat.


 * Why is this different than the normal case for runoffs? *The runoff was, by design, held as part of the general election.* Now, it is not a *major* general election, but actual statistics for runoffs from Cary showed no significant decline in turnout for the runoffs they had. (It was slight decline in two and slight increase in one). Turnout, *period*, was poor, probably because these are off-year elections. I'll stand with what I wrote, it's supported by the evidence I gave, which, it appears, MilesAgain is so concerned with trying to stop it from being presented in talk that he hasn't actually read and considered it. The comment about "only ... in the precincts" was related to expense, not to turnout, per se. Let me repeat this: the common wisdom that runoffs have lower turnout did not apply to Cary, and it's obvious why.
 * --Abd (talk) 04:38, 17 January 2008 (UTC)

RFC: How to summarize the mention in Robert's Rules?
What is the most appropriate way to summarize the mention of IRV in Robert's Rules of Order? The language about it in the introduction has changed since the previous RFC on this topic. Then, the words "recommends" and "advises" were used, instead of "suggests" or "describes." However, a compromise still has not been reached. Here are the most recent versions added to the article by different editors:

Which is the most appropriate for the article's introduction, or would other wording be more appropriate?
 * 1) Roberts Rules of Order, Newly Revised describes IRV as an example of preferential voting, which it states is "more representative" than plurality elections in cases where multiple rounds of voting are impractical, such as in elections by mail.
 * 2) Roberts Rules of Order, Newly Revised suggests using preferential voting when repeated balloting to find a majority winner is infeasible, such as for elections by mail, giving fully-ranked IRV as an example; however it also notes that sequential elimination under IRV may eliminate a compromise candidate, due to insufficient first preferences, who could be supported by a majority over the winner IRV finds.
 * 3) Roberts Rules of Order, Newly Revised suggests using preferential voting when repeated balloting to find a majority winner is infeasible, such as for elections by mail, giving fully-ranked IRV as an example.
 * 4) Roberts Rules of Order, Newly Revised describes IRV as an example of preferential voting, which it states is "more representative" than plurality elections in cases where multiple rounds of voting are impractical, such as in elections by mail. However, it also notes that IRV may eliminate a compromise candidate who could be supported by a majority over the IRV winner.
 * 5) Roberts Rules of Order, Newly Revised states that preferential voting is "more representative" than plurality elections in cases where multiple rounds of voting are impractical, such as in elections by mail, and it describes IRV as an example. However, it also notes that IRV may eliminate a compromise candidate who could be supported by a majority over the IRV winner.'' [This language is currently in article as this is written, was not when MilesAgain and Tbouricius first commented below, and was not listed here by MilesAgain.]


 * #3 is most appropriate. It is straight and to the point, and contains no incidental facts which serve to bias the statement one way or the other.  The incidental facts in #1 and #2 are already in the body of the article. #4, my second choice, is roughly balanced, but contains both of details inappropriate for the lead. MilesAgain (talk) 10:38, 15 January 2008 (UTC)


 * #3 or #1 are appropriate for the introduction. The overly-long quote from RONR in the body of the article is still too long, but even a shortened version of that could easily cover the details appropriate there. Tbouricius (talk) 15:11, 15 January 2008 (UTC)


 * #2 or #4 or #5 or other language that is balanced (with both the positive and negative implications), or not in the introduction at all, but only deeper in the article where it can be explained in a balanced way. I'm a little surprised that Tbouricius only approved 1 or 3, given that he wrote, I think, number 4. --Abd (talk) 18:12, 15 January 2008 (UTC)


 * It is not clear that RONR needs to be mentioned. I understand that many orgs are required to use RONR, but the text quoted above shows that neither preferential voting, generally, nor IRV, specifically, is required or recommended by RONR. It merely offers up PV as an option and IRV as an example of it. Among the five, #4 and #5 are best. -Rrius (talk) 06:07, 23 January 2008 (UTC)

Comment on RFC
There was a previous RFC, see Talk above; however, the discussion was long, and the matter confused because the RFC covered three distinct topics. The problem here is that what is actually in Robert's Rules is complex, and details matter. Robert's Rules prefers repeated balloting, no eliminations, until a true majority is obtained, and it does not recommend electing officers by plurality vote. However, some organizations have decided that their situation necessitates election by a single mail ballot. In that situation, Robert's Rules suggests, "preferential voting," which refers to the ballot used for a class of election methods that have been known for 200 years. This allows voters to list candidates in order of preference, and then the method processes these preferences to find, presumably, a better winner, more widely supported, than might be the case in an ordinary election by plurality, only one vote allowed. Hence Robert's Rules notes that preferential voting would be more fair than an election by plurality. Now, there are many preferential voting methods, but only one is currently in wide enough use that it could be described in Robert's Rules, which is not a book prescribing new practice, but describing existing practice, and that method is a full-ranking form recently named "Instant-runoff voting." So, as an example of "preferential voting," RR describes a particular form of IRV (not all forms of IRV described by the article would be considered as good by Robert's Rules). However, because instant-runoff voting has a well-known problem due to its candidate elimination process, which can actually eliminate a candidate preferred by up to two-thirds of the voters in a 3-candidate election, over the IRV winner, under some situations, RR then explains the problem, that the specific method it has described, IRV, can eliminate a compromise winner.

In short, say we have a tad more than 1/3 of the voters who rank A > B > C, 1/3 who prefer C > B > A, and 1/6 rank B > A > C and a tad less than 1/6 rank B > C > A. B is eliminated in the first round, the B votes go to A and C, and A wins. However, two-thirds of the voters preferred B to A. This is a classic "center-squeeze" problem. IRV works reasonably well in a two-party context, where the only concern is with spoiler candidates, but can perform badly in a three-way contest, or where there are many candidates. This problem is not shared by all preferential voting methods, so what is correct is that:

Roberts Rules, where a single ballot by mail is needed, suggests preferential voting, describes IRV as an example, and then describes a serious shortcoming of that particular example. We had compromise language in the introduction that reflected this, placed there by Tbouricius (a pro-IRV editor who actually introduced IRV legislation in Vermont, and is a COI editor by virtue of his relationship with FairVote, and explicitly accepted by me as a critic of IRV, accurate as a summary, and sourced. MilesAgain, an acknowledged sock, clearly representing a pro-IRV intention in his edits, raised the issue anew, edit warring to remove it.

WP:LEAD is trumped by WP:NPOV. Any language I've been able to imagine which only mentions that Robert's Rules "recommends" (the original language) IRV, or "advises" or "suggests" it, or even "mentions" it or "describes" it, without placing that in context (i.e., the same source, in the same passage, provides a warning), introduces a POV slant, in the introduction, and it is clear why this is so important that pro-IRV editors have long been willing to edit war over it, removing sourced text as "too much detail," even when it was in the middle of the article. Several editors who participated in these edit wars have been blocked, User:Tbouricius was himself blocked for a time as an SPA involved in this. MilesAgain is a sock puppet, acknowledged (but who claims a "legitimate reason" for being a sock -- in spite of the fact that WP:SOCK strongly discourages socks from engaging in contentious editing).

As to the text in the middle of the article, it's over-long because the pro-IRV editors removed every attempt to summarize the critical part, so ultimately it became exact quotation, which does indeed take more words. It seemed we had a decent and acceptable summary in the introduction for a few days, written by Tbouricius, but .... MilesAgain Again. Nice name for a sock. Reminds me of User:BenB4, but if MilesAgain has the same master as User:BenB4, he's learned to avoid the particular edit patterns that made User:Acct4 obvious. --Abd (talk) 18:40, 15 January 2008 (UTC)

It's a moving target, User:Tbouricius again changed the wording. I don't mind it, but it is not correct, that is, Roberts Rules does not say what he says. It specifically refers to the method it has described, i.e., IRV. What Bouricius says is true, this is a problem with all bottom-elimination methods. But it isn't from that source. I'd appreciate it if he takes it out. We are here reporting what Roberts Rules says. (The following is what Bouricius put in.)


 * 'Roberts Rules of Order, Newly Revised'' states that preferential voting is "more representative" than plurality elections in cases where multiple rounds of voting are impractical, such as in elections by mail, and it describes IRV as an example. However, it also notes that runoff procedures that drop bottom candidates may eliminate a compromise candidate who could be supported by a majority over a runoff winner.

It's true, so I'm not terribly exercised by the technical error.... --Abd (talk) 04:32, 16 January 2008 (UTC)


 * Abd is mistaken when he states that the language I offerred is a "technical error" miss-stating what RONR states....On pages 426-427 RONR explains why removing bottom candidates is not ideal, and this point is NOT limited to preferential voting. Perhaps he missed this because it is not in the preferential voting section and is not specific to preferential voting, but applies to it and any other election. However, I still agree with MilesAgain, that this level of detail is not appropriate to the introduction (although I drafted it). Since probably over 99% of the article readers will want to compare IRV to plurality elections or two-round runoffs, contrasting IRV to a rarely used method (open ended new balloting where a dark horse can emerge) is not appropriate for the introduction. The point is that RONR proposes IRV as a superior  method than plurality (though not perfect) when repeated voting is impractical. If one were to insert MORE detail than this in the introduction, it should be what RONR says comparing IRV to the shortcomings of "pluraltiy,"  not what it says comparing it to a method that hardly any readers are interested in, familar with, or would consider.
 * Tbouricius (talk) 16:39, 16 January 2008 (UTC)


 * Yes, RONR explains what Bouricius describes. And if Bouricius wants that to be in the article, fine. *However,* we are talking about the *introduction,* and, specifically about Robert's Rules and whether or not this fact regarding bottom-elimination runoffs is so intimately connected with the implied suggestion for the use of IRV that to mention one without the other introduces POV imbalance It is. It is immediately after the description of IRV, and what is said is about the *specific* form of preferential voting described. It is not a general characteristic of preferential voting, it is something that Robert's Rules, which does not dilate at length on topics, chooses to mention as a caution, a negative associated with the positive of making an improvement over plurality. The rest of what Bouricius raises is, essentially, smoke-screen. It is as if Roberts Rules says "IRV is better than sliced bread, except that, under some conditions, it might poison you." What Bouricius wants to quote is just the "better than sliced bread" part, but not the poison part, since, after all, what conditions? Plurality can poison you too, so where's the beef? I've explained why I think RR did not go ahead and describe better options, though we, of course, already know one, which is repeated balloting until someone gets a majority. RR isn't making recommendations for public elections, which is why it is possibly deceptive to promote IRV based on what's in IRV. I'm still trying to figure out what the Cary City Council is referring to when it claims that IRV is "approved by the U.S. Justice Department." Approved for what? Anyone know? --Abd (talk) 19:45, 16 January 2008 (UTC)


 * Approved by the U.S. Justice Dept - why did the Cary City Council say that? Because North Carolina is a "Jim Crow" state any election legislation passed in our state must be pre-cleared through the DOJ.  Since the IRV experiment in Cary was a Pilot Program and not permanent legislation, more leeway is given - i.e for Cary to do IRV again would require new legislation, this was an experiment.  The NC State Board of Elections also created a "work-around" for the touch screens in Hendersonville that circumvented current election integrity laws.  This "work-around" wasn't used, since Hendersonville didn't have a runoff. The SBOE only could do this in the case of a "Pilot Program".  --Ask10questions (talk) 02:28, 23 January 2008 (UTC)

about election detail in the United States section
I have been collecting information about actual use of IRV in the U.S., placing what I find, sourced, here in Talk. Then, in the article, I've continued what I did previously, and which was allowed to remain: I described briefly each example of adoption. There are a number of categories in the political adoptions:

(1) An adoption formally took place, but implementation was postponed, typically indefinitely. It's questionable whether or not these should even be mentioned, but FairVote definitely likes to promote the idea of a wave of successes, I'm sure it helps their bottom line and maintains the impression of "momentum" (which is the argument that they make when discussing possible cooperation with supporters of other election reforms: those other reforms are "impractical" -- even though one of them is zero-cost and widely respected among experts -- and IRV "has momentum.")

(2) It's actually been used for one or more elections, but the elections were with only one or two candidates on the ballot, so the IRV ranking was almost useless. (A write-in voter, however, could find it useful).

(3) There were more than two candidates, but one obtained a majority in the first round and so the runoff provisions were not exercised. Note that the existence of a top-two runoff or IRV makes it more likely, almost certainly, for additional candidates to run, because they can do so without spoiling an election (as long as they don't get upwards of one-third of the vote, where center-squeeze becomes a problem); so it becomes harder for a candidate to gain a majority in the first round, or at all.

(4) There were more than two candidates, no candidate obtained a majority, so the runoff provisions were used, but did not change the result (plurality winner) from the first round, however, the candidate did gain a majority of votes through the runoff process. *This is an empty category.*

(5) There were more than two candidates, runoff provisions were exercised, but the leader did not change from the plurality winner in the first round, and no majority was obtained. All IRV elections I found fall in this category, so far, and I was quite surprised at this, it is not at all what I expected. As far as I know, indeed, this has largely escaped notice, but it isn't "original research," because it can be directly verified by any reader, and no interpretation is necessary.

(6) The election went to runoff, and the IRV winner was not the winner of the first round, and a majority was obtained for the IRV winner. This is an empty category.

(7) The election went to runoff, the IRV winner was not the first round plurality winner, and no majority was obtained. This is an empty category.

Then, in the data on the Talk page, above, there is another interesting fact which I did not put into the article. For the San Francisco elections -- I did not check out the Burlington and Cary runoffs -- the *second place* candidate did not change from the Plurality results in the first round.

Now, what do these facts suggest? *I can't put a suggestion in the article, that would be original research, unless I find reliable source for the conclusion.* But the facts on which such a suggestion might be based *can* be in the article, and they are relevant. They are about how IRV is actually working. Now, all this could be, instead, in the History article, which was created as a fork by a probably sock puppet, but that article is currently in sad shape. My position here is that a list of "successes" is out of place unless the actual performance is examined. What's the *history*?

Note that IRV was used in various places in the U.S. in the past. That's not in this section on the United States. Why? Well, those who were maintaining the article quite clearly wanted to present a series of successes, a wave of adoptions, and mentioning Ann Arbor, well, that could be seen as a failure. Note that a common FairVote argument against Bucklin voting is that it was tried here and allegedly rejected -- though the story of how Bucklin came to be rescinded has never been fully told, and in the most obvious case, Duluth, it was corrupt legal judgement that did it, Bucklin was very popular and was working. The whole history should be there, not just the very recent work of FairVote.

So, I see two acceptable paths here: (1) take the list of recent adoptions out of this article, move them to a History article -- they are *not* necessary for this article --, or (2) put the balancing information in this article. It's quite similar to the situation with Robert's Rules, though I wouldn't dream of trying to put this vote detail in the introduction.

As to reliable source for the conclusions that might be drawn, eventually it will exist. San Francisco is spending millions of dollars to "avoid runoffs" that could be avoided much more simply, in any of various ways, all due to a desire to have majority winners -- noble intention -- which is *failing* to happen due to IRV being substituted for top-two runoff. One winner only received 33% of the vote. One way of interpreting that is that two-thirds of the voters opposed the election of this person; however, the limited-rank implementation in San Francisco makes that unclear. It's a mess, folks.

I replaced the sourced, accurate, material which had been improperly removed by User:Tbouricius. As I wrote before when attempts were made to remove similar information about the other elections, if new information has been cherry-picked to create imbalance, and the information is correct, sourced, and relevant, the remedy is to add additional balancing information, not to delete it. It is *always* expected that POV editors will think inconvenient information "irrelevant."

The suggestion made that the information presented was biased or cherry-picked is offensive. I would certainly have reported any runoffs that reached a majority, for example, and I'm also quite sure that if any had, another editor would have noticed and fixed my error of omission. I saw the same WP:ABF charge when I started to add election detail, even though I simply went down the list. --Abd (talk) 20:29, 15 January 2008 (UTC)


 * The reason Abd does not recognize that he is cherry-picking information, is because the facts he selects fit neatly with his POV goal of making IRV seem less deisrable. He didn't happen to seek out or pick information supporting the view that there are substantial cost savings of avoiding runoffs, or that IRV gives voters a greater willingness to vote their true preferences, or that effective turnout in the SF elections (even with exhausted ballots) was FAR greater than in most of the traditional runoffs, etc., etc. Yes, of course, plurality leaders in the first round are far more likely to be the ultimate winner (in Australia I think over 80% of the time). A handful of U.S. elections show little about IRV in broad terms (probably says more about San Francisco than IRV). Contrary to Abd's assertion, the appropriate response is not to insert "pro-IRV" facts into the article to "balance" the biased facts that Abd chooses to cram into every nook and cranny. The appropriate response is to fashion a balanced article without lengthy and excessively detailed tangents.
 * Tbouricius (talk) 21:19, 15 January 2008 (UTC)


 * Contrary to the assertion of Tbouricius, my goal is a balanced and informative article, one that will be eligible for Featured Article status, not "making IRV seem less desirable." However, the context is that the article was carefully crafted over a long period of time by a cabal of editors, and I won't repeat the whole saga; but bottom line, the article was full of POV content, selected to -- usually -- be true *and* misleading, or, at least, to promote certain interpretations. So, for example, there was a section on IRV adoptions, and the number of adoptions was mentioned in the introduction. This creates a picture of expanding use, which then makes it easier to get more use, after all, if San Francisco did it, must be a reasonable thing to do. Every adoption, no matter how small or trivial or, indeed, useless, was placed in the article. It was mostly true. (Hendersonville NC, which I just took out, wasn't actually IRV.) Many people, reading the article, could sense that it was a promo piece, but just threw up their hands. After all, the "facts" were true. But a selected group of facts can be highly POV, and, clearly, Tbouricius recognizes that, since he is accusing me of selecting facts. Am I selecting facts? *Of course I am.* I notice what I notice; however, I'm not suppressing anything that I notice which is relevant to the article, even if it makes IRV look *good*. It just happens that, so many times, when I have tracked down the references, gotten past all the references to FairVote, and looked at what was actually there, the picture in the article turned out to be distorted, incomplete, spin. The immediate situation shows this. With a entire list of "adoptions," before I began working on the section, would people have had any clue how few actual IRV elections there have been? Would they have had any clue that most of these elections did not use the runoff features? That every single runoff election that *did* use the runoff feature did *not* result in a majority vote. After all, one of the *big* arguments for IRV, repeated constantly, is "guarantee majority winners." *Of course* that was hype, no method can guarantee that a majority support a candidate, unless voting is made obligatory, as in Australia, which is faint support indeed! However, what about the more substantial claim, that IRV makes it easier to find a majority? Surely IRV was doing that, if there were more than a few runoffs. No, it hasn't done it *once*, as far as I could find. I'm sure that some IRV supporters are a bit upset, to find their pretty picture of IRV success showing some wear. The fact is that I didn't expect this, at all. I thought IRV would be working better than it was. Was it saving runoffs, at least? Sure. *If* you insist on having a majority winner, so you have runoff elections, then IRV saves runoffs. But, wait a minute! Why did we insist on having a majority winner in the first place? I am *not* putting this argument in the article, and it might not even be appropriate -- yet -- for the Controversies article. It simply explains why the facts described in the article are important and of interest. Is this POV pushing? Well, I should ask in response, was it POV pushing to put the adoption of IRV by Takoma Park, Maryland, into the article? Was it POV pushing to put the San Francisco election performance in 2000 in the article, where almost every election went to runoff? (I haven't checked on this, but from other elections I've seen, that may have been an extraordinary year, and the real problem in San Francisco was a combination of nonpartisan elections -- which gets rid of the filtering effect of party endorsements -- with top-two runoff, in a major city with lots of active interest groups. The classic American solution would be partisan elections and plurality voting. Now what I find interesting is that it's quite likely that IRV has not changed one election outcome in San Francisco. It might some day.


 * What I wrote about adding balance instead of removing allegedly biased sourced fact comes from Wikipedia policy, I didn't make it up. Facts aren't biased, only the selection of facts can be biased. POV editors, when they aren't actually inserting nonsense, speculation, etc., into articles, are typically selecting facts, it is a very common Wikipedia problem. And there is a well-developed solution: don't delete someone else's sourced fact unless you can make a good case it is irrelevant, instead balance it with other sourced fact. It is even-handed. As to the article becoming too complicated, too long, there are other solutions for that, in particular, article forks. If the Controversies section becomes too large, shove most of it off to a Controversies article. Note that forks are *not* allowed in order to create safe havens for some POV. But as articles grow in complexity, the article fissions into a main article and sub-articles, dealing with details. The main article should, however, keep enough of what is *important* about the sub-articles, it is not allowed to select facts from the subtopic in order to create or maintain some desired impression on readers. Hence there will always be *some* controversy mentioned in the main article, some election detail, etc. But not a long list of implementations with details on each.


 * There was some agreement in the past that we might create a template for election methods, a format that such articles would follow, in an attempt to remove bias introduced by how each method is presented. I'm not sure we have the editorial labor available to do it, I'm just noting that this idea came up.


 * Tbouricius mentions costs. I have not detailed costs in the article because I have no access to the information. IRV has possibly saved some cost in San Francisco, but compared to what? What was the implementation and operational cost, and how many runoffs were saved? What would it have cost to run them? *I don't know, and I don't have ready access to the information.* I did find that the Cary election saved one runoff, an unusual election situation, with municipal elections held as *special elections*, not with other elections. I can only speculate that they do this to suppress turnout, it would be a common device. But maybe they have other reasons. FairVote claims that they saved an "expensive runoff." Did they? What's the basis for that claim? How much would it have cost, to gather votes from less than two thousand voters, with two candidates? And how much did they spend to implement IRV, and to count it, they had extensive trouble. And they got the same result as they would have gotten with Plurality. Plus, in this case, because of the circumstances, a real runoff would have made quite a bit of sense, democratically. Look at the votes, Mr. Bouricius! This election made no clear choice, and due to what might be considered candidate misbehavior (pretty weird), the *public* was deprived of the ability to make that choice, had they been faced with a reality of the top two. That's what top-two runoff is about, and the *purpose* of it was defeated by this trial of IRV. Would I put cost information in if I had it? *Of course I would*. But where would I get it? Is it published? I can't travel to North Carolina and demand to see the records!


 * Let me tell you, though, my suspicion is that, as with other aspects of IRV, the benefit of "avoiding runoffs" has been oversold. I rather strongly suspect that San Francisco is going to see a net negative, financially, from its implementation of IRV. I'm not so sure of that, if the comparison is only between the crazy system they had and IRV, but that is a limited comparison. And there is more to be discovered about it, I'm sure. Notice one thing, though. Lots of jurisdictions have "adopted" IRV, but can't find the money to do it. If it was expected to save money, why wouldn't they invest in it? Are all these jurisdictions financially stupid? --Abd (talk) 00:23, 16 January 2008 (UTC)


 * The failure to launch: jurisdictions adopt IRV but can't impliment it because they don't have voting machines that can accommodate it. Yes I know that possibly it can be counted by hand, but the fact is jurisdictions are waiting for compatible voting machines. Just look to Scotland who abandoned hand counted paper ballots for STV.
 * Minnesota's SOS did a feasilibity review and stated that it would be several years before they could try to impliment IRV for a statewide contest because of need to purchase new machines that can accommodate it.
 * Maine turned it down because of voting machines. Another interesting point - Maine has alot of hand counted paper ballot jurisdictions who would need to buy central count optical scanners, and all ballots would need to be in optical scan format. The first choices could still be counted by hand, but second and third choices would be counted at county offices or central locations.
 * A key theme with IRV is that the actual tabulating of 2nd and 3rd choices must be done at a central office. Even in San Francisco, the voting machine cartridges were trucked in to the elections office so that 2nd and 3rd choices could be tabulated there. The fact that jurisdictions adopt IRV but don't implement it is tied to the cost issue. A lobbyist has advised a colleague of mine that "cost" is the key selling point to North Carolina lawmakers. They believe(d) the talking points because no one (had) provided them with other information.  --Ask10questions (talk) 06:58, 16 January 2008 (UTC)

What improvements to the article are being contemplated in the discussion of this section? MilesAgain (talk) 15:33, 16 January 2008 (UTC)


 * Election detail in the section on implementation in the U.S., plus possible facts for the Practical implementations section, as well as comparison with runoff voting. --Abd (talk) 18:59, 16 January 2008 (UTC)

... I moved old content to Archive 4, generally older than Jan 6-8. Now my browser scrollbar almost allows me to read these pages. It's more ideal if the talk page can be moved ENTIRELY to an archive rather than piecewise, but with ongoing content, it seems impossible to set any clear boundary except to try to extract inactive sections by hand. Tom Ruen (talk) 06:35, 23 January 2008 (UTC)