Wikipedia talk:Manual of Style (biographies)/Survey on Style-Prefixed Honorary Titles/Vote Corralling

Vote Corralling

Jtdirl is attempting to corral votes in this survey. The following text has been posted to the Talk pages of those voters who have expressed a preference for Alternative 1 and voted "First and Only choice":

Re your vote on styles. I understand and agree. But only casting one vote is effectively a vote against Alternative 1 because it means that less opposition is recorded against its nearest rival. Ireland uses an electoral system called Proportional Representation using a Single Transferable Vote. It works on the same principle as the one being used (only less complicated! I never thought I would find a system more complicated than PR.STV!) What you do is give your bottom preference to the people you want to defeat, and spread your vote in a way that boosts the rivals of the alternative you do not want. So if for example, you find Alternative 3 the one you least like, give it your bottom vote so that opposition to it is recorded. And spread the other votes to ensure the weakest get votes ahead of it. If for example in Ireland I want to ensure candidate 'x' of Fianna Fáil is elected, and ensure candidate 'y' of Sinn Féin is defeated, and there are 15 candidates, I give my number 1 to 'x', my number '15' to 'y' and spread my other votes to ensure that all other candidates beat 'y'.

Sinn Féin and Fianna Fáil voters famously used to practice a 'first and only choice' vote by just voting for their own preferred candidate and then stopping. They eventually realised that they were wasting their vote because they weren't using it to block those they were most opposed to, or to build up the rivals to the candidate they were opposed to. To stop Alternative 3 winning, if that is what you want, give it your fifth choice and give your second, third and fourth choices to the weakest options.

Just be careful though not to copy everyone else doing it. If everyone gives the same other alternatives the same order of votes they may win. So if option 4 gets a lot of 2s, give it a 4. Doing a full vote right down the line will have the effect of strengthening Alternative 1 vis-a-vis 3 or whatever. Just voting for 1 and stopping actually weakens it against its rivals if everyone else votes down the line, because while their opposition to different alternatives is recorded, by stopping at 1 your's isn't. That is why though very popular Alternative 1 is being beaten. Remember the winner won't be decided by who has more votes for, but which faces the least opposition. By voting first and only choice you haven't recorded your opposition to the other options. Slán FearÉIREANN

^(talk)

I believe this is an inappropriate attempt to manipulate the outcome of a survey by encouraging people not to vote their true preferences but instead to practice a "voting strategy" which defeats the objective of determining the alternative which is honestly preferred by the voters. This strategy will not work to secure victory for a particular alternative in a Condorcet survey, but it may very well block a consensus because it is in effect encouraging people to vote in a more or less random fashion, which prevents useful conclusions as to the "true preferences" of voters from being known.

I ask what Wikipedia guidelines exist for dealing with this kind of intentional tampering. I do not wish to discard votes which are honestly made, nor do I wish to cancel the current survey, but I believe that some kind of censure is appropriate for Jtdirl and if the consequence of his actions is purely to cause a consensus to be blocked, further discussion ought to be considered with respect to what conclusions we may draw or appropriate actions to take.

The list of users who have so far changed their votes in conformity with Jtdirl's instructions are:

It is also important to note that several of these users explicitly commented while changing their votes that they were "playing the system" or "trying to game the voting system".

Should we disqualify any or all of these votes? What action if any should we take with respect to Jtdirl's conduct?

Comments, please. Whig 13:59, 9 May 2005 (UTC)[reply]

I did not "change my vote in conformity with Jtdirl's instructions." Among other things, I had the sense that he was wrong, and that the system couldn't be gamed like that. I added extra votes because I decided that I would prefer option 2 or (by a very slight margin) option 5 to options 3 and 4. At any rate, I don't see how you can censure someone simply for trying to do a vote as votes are meant to be done. If he was right, and the system can be gamed, that is the fault of the system, not the people strategically voting. If he is wrong, then it doesn't matter anyway - the system cannot be gamed. I'd add that almost everybody who changed their vote did not change their votes in a random fashion. Almost all picked option 2 second, option 5 third, and then options 3 and 4 last (although which one was fourth and which fifth differed). This would be how one would guess that people who prefer option 1 would vote, don't you think? john k 14:22, 9 May 2005 (UTC)[reply]

As much as it frustrates me that Jtdirl and others fail to understand what NPOV means (and instead make cute, but silly, acronyms like LPOV), I concur with John Kenney that this encouragement by Jtdirl is exactly appropriate. If some voters had failed to express preferences they actually hold, the voting is less accurate than it should be. It is not unreasonable for Jtdirl to guess that the listed voters might rank non-1 options in a particular manner--suggesting that they complete their rankings does not compel them to do so, nor to do so in a manner Jtdirl wants. Lulu of the Lotus-Eaters 17:48, 2005 May 9 (UTC)

However, Jtdirl apparently misunderstands why Condorcet is different from IRV/STV. Both involve ranked preferences, but the tallying is a whole different algorithm. Neither is hard to understand (nor is Borda); the confusion around this mystifies me. Lulu of the Lotus-Eaters

My objection is absolutely not to the idea that Jtdirl should encourage people to vote a full set of preferences, and indeed I encouraged everyone (no matter which alternative they favored) to express all preferences. Rather, the fact that he encouraged people to vote deceptively and purely for the purpose of trying to spoil the vote. We may not get consensus out of this process, but we should want to get useful information about what people's true preferences are. Dishonest voting will not help change the outcome of the survey, but it may make it more difficult for the post-survey analysis to draw meaningful conclusions. Despite the attempt, I believe we should consider all ballots as having been voted honestly. Whig 06:38, 10 May 2005 (UTC)[reply]

If these people are changing their votes because they want the option they voted for to win, I'm not sure that their actions are inappropriate: one casts a vote for precisely this reason. The point of voting is to make the choice you agree with win and the point you disagree with lose. Concordat voting is not always intuitive, and if it takes people some time to see how to best cast their vote, so be it. Though people have unfortunately been phrasing it as "gaming the system", I'd say that the result is that the voting counts now more accurately reflect the desires of these votes: their favorite option as #1 and their least favorite as #5.

To put it another way, what would have been a better course of action, had you voted for a choice as "First an only" and then discovered that your vote might not have been having the effect you had wanted? Not to act would have been pointless. The only reasonable action would have been to follow Jtdirl advice. Some users then chose to make it clear that they were only voting in this manner because of the strange way that Concordat treats single votes. I don't think that we can punish these users for this, and I don't think we can remove votes that listed an option as #5 when that was indeed the user's last choice. The only reason that this looks like an instance of abuse of the system is because it was done at another user's instigation. Their own votes, however, are their own business, and unless you can prove that someone who listed an option as choice #5 was actually lying in their choice of ranking, I can't see how you could possibly dispute their right to vote.

— Asbestos | Talk 14:28, 9 May 2005 (UTC)[reply]

Asbestos, I view you as an honest broker here, so let me express my concern. I don't want to deny anyone their honest vote. I feel it was inappropriate for a user to instigate other users as a bloc to vote in a particular way so as to scuttle consensus, and to do so as deceptively as possible, viz., "Just be careful though not to copy everyone else doing it. If everyone gives the same other alternatives the same order of votes they may win." While not every user who changed their votes stated explicitly that they were attempting to "game" the system, several did say so. Should we consider those "honest" votes? And no, the Condorcet result is not altered. Alternative 3 is still the present Condorcet/CSSD winner. Not Alternative 5. In fact, that alternative is now the Condorcet loser (whereas Alternative 2 was previously least preferred). Whig 14:43, 9 May 2005 (UTC)[reply]

"Should we consider those "honest" votes?" Of course! Their choice for which options to vote for is still their own choice. There were no preconditions on voting that specified that a voter may not examine the current votes and decide a manner of ranking which would affect the results in the most desirable way. Along similar lines, if I were to tell voting officials that, while I'm voting for Nader, I actually swapped my vote for a Kerry one and want Kerry to win, they would still not invalidate my vote, even if I told them I was doing it to "game the system." So long as I was not payed or otherwise pressured into voting, I'll vote exactly how I choose to vote. I think you'll have a hard time showing that Jtdirl unduely pressured anyone into voting, he merely explained a better way of making their votes count.

This is all pointless, of course. While I know you are doing what you think is best for the integrity of the poll, were you actually to exclude their votes you'd have an extremely slim chance of convincing anyone at all that the poll was conducted legitimately.

— Asbestos | Talk

No votes will be discarded. Whig 17:05, 9 May 2005 (UTC)[reply]

Jtdirl has never placed a message on my talk page on any subject whatsoever. I was unaware of any messages he might have placed on other talk pages until reading Whig's message above. As I have not checked the other talk pages, I do not know whether or not Whig's accusations are true with regard to other users. His accusations are most certainly false as regards attempts to influence my vote.

I did not change my vote. I voted first choice for Alternative 1, and wrote a short comment. (I did not write first and only choice.) I then returned to work on an Open University assignment, occasionally looking at Wikipedia as a break, and perhaps making other contributions which I considered more urgent, since I understood that the vote would remain open until 14 May. I knew quite clearly that I intended to vote second choice for Alternative 2, but intended to write quite a lengthy comment, so waited until I had a little more time. At the time that I voted for my first choice, I had not entirely decided on my third, fourth, and fifth choices. When I finally got round to making a fairly long contribution to Alternative 2, and making a simple vote on the last three alternatives, I was not in any way influenced by Jtdirl. Apart from public discussion on the talk pages for Pope Benedict XVI and Marcial Maciel, (in which we disagreed with each other), neither of us has ever made any communication with each other. I would therefore request that Whig withdraw the statement that I "changed" my vote "in conformity with Jtdirl's instructions". Ann Heneghan 15:03, 9 May 2005 (UTC)[reply]

I am sorry for including you in the list if you were not one of the recipients of Jtdirl's message. I have removed your name from the above list, and I withdraw my statement that you were in any way influenced by his instructions. Whig 15:15, 9 May 2005 (UTC)[reply]

Thanks, Whig. Apology accepted. Ann Heneghan 20:30, 10 May 2005 (UTC)[reply]

I must admit, I actually laughed out loud at the suggestion that you might be capable of somehow "disqualifying" my vote. Nice try, though. Proteus (Talk) 14:38, 9 May 2005 (UTC)[reply]

This is foolish. One can't game the system. As Rspeer pointed out above, voting on every option is how the thing is meant to be done. Jtdirl's advice is, indeed, to be deceptive, and I disapprove, but you have no right to invalidate anyone's vote on those grounds. Votes from established users of en are to be counted. Furthermore, it's pretty clear that nobody has followed his advice to not vote the same way - just about everyone picked option 2 as 2nd choice, for instance. john k 14:53, 9 May 2005 (UTC)[reply]

No votes are being disqualified. CSSD cannot be gamed. The only thing that this has established is a weakness in Debian's modification of the method, which we are not bound by and was not declared as the method for counting. Whig 15:20, 9 May 2005 (UTC)[reply]

You yourself suggested that votes should be disqualified. If you have changed your mind, you should say that you have changed your mind, and not act as though those of us saying you shouldn't disqualify votes are crazy for thinking that is what you were saying when, in fact, that is what you were saying. john k 15:48, 9 May 2005 (UTC)[reply]

I had not made up my mind in the first place in order for it to have to have been changed. I asked for opinions, I considered the matter, and determined that the Debian modification of CSSD was subject to tactical voting and would not be used, we will use standard CSSD and all votes will be counted. Please see below. Whig 16:57, 9 May 2005 (UTC)[reply]

I've got to laugh my head off at Lulu's dillusions. I have attempted to corral nobody. I simply pointed out to some users who in effect voted along the principles of First Past the Post that their one vote by failing to express other opinions failed to register their scale of oppostion to other options. So all they were doing was expressing an opinion for one option, not all of them. They had the absolute freedom to express preferences nor not to express preferences. Some did and some didn't. That is 100% their right, and every user has the right to point out to users that they had failed to take on board the full complexity of the system being used. And alternative votings are based not just on positive but negative preferences; to pick a preferred choice and pick an option you want to block. If Lulu can't cope with people using their vote in a way that does not suit him, that's his problem. FearÉIREANN

^(talk) 18:14, 9 May 2005 (UTC)[reply]

Speaking of delusions, if you think anyone who might find your behavior inappropriate must be Lulu....? Whig 18:21, 9 May 2005 (UTC)[reply]

To paraphrase a Barbi doll: "Reading is hard..." Lulu of the Lotus-Eaters

Whig, I find it rather offensive that you would suggest that my votes be discounted, based on your (entirely incorrect) assumptions that (a) I was aware that James has said anything about the topic, (b) that this amounted to a recommendation on how the system works and so can be used to vote, or (c) that I would have listened to him. Unfamiliar with the voting system, I at first thought that by saying "first and only vote", I would emphasise how very strongly against I am any of the other options (something which, it turns out, this voting method fails to achieve, and something that I think speaks rather poorly to it). On further exploration, I ammended my votes so as to use the system in the way that it has been designed to be used. I request that your withdraw your iniquitous comment, please. James F. (talk) 19:28, 9 May 2005 (UTC)[reply]

Your votes will not be discounted, as discussed both above and below. Per your request, and although your change comments suggested you were explicitly "playing the system, then," after voting "First -> first and only" I will withdraw your name from the list of users above. My main beef is not with you or the others who changed their votes, it is with Jtdirl's attempt to corral votes in a dishonest fashion. Whig 23:49, 9 May 2005 (UTC)[reply]

I demand you withdraw that comment. I did not attempt to corral votes in a dishonest fashion. I pointed out in an OPEN manner ON USER PAGES (not secretly in emails) that some people has not fully cast their vote in a public vote. How exactly is publicly reminding people to cast their full votes in a public vote in any way dishonest? To claim that I have in any way been part of dishonesty is defamation and will be treated as such if not withdrawn. The only sham thing here is this ridiculous vote, with a voting system that users on this page are confused by, has no transparency whatsoever because of its complexity, and has failed to produce even the slightest semblance of consensus. In my years here I have never come across a more ridiculous vote.FearÉIREANN^(talk) 18:38, 10 May 2005 (UTC)[reply]

I will not withdraw my comment but I will clarify for the purpose of making myself crystal clear. I have not claimed that you acted secretly (although I would not know if you have or have not) to corral votes. You encouraged people to vote dishonestly in order to make your favored alternative look more attractive. It does not work when plain CSSD is being used, although it did serve as an effective demonstration of tactical disruption in Debian's particular modification of the system. Your objections to the vote are noted, but frankly seem to me like sour grapes from someone whose favored position seems not to be currently prevailing (even if by fraudulent means). Perhaps you would prefer a voting method you could tactically manipulate to your desired result. I'm sorry that I cannot oblige you. Whig 20:46, 10 May 2005 (UTC)[reply]

Yet another complete lie and another defamation. I did not encourage people to vote dishonestly. I drew attention to the fact that they could vote tactically, which is what billions of people do worldwide in elections, or do you think the millions of people who voted tactically in the British general election were also voting dishonestly? I pointed out that others through voting for all options had been able to strengthen or weaken the case of individual options and pointed out that by not doing so people who voted first and only choice had failed to register their scaled opposition to other options. There is absolutely nothing underhand in pointing out that if a person voting for one option wants to strengthen its chances they can do so by voting tactically. That is elementary voting technique, practiced worldwide. If your only answer is to tell lies about what happened, then you are just showing both the bankrupcy of your analysis and the patiently absurd nature of this vote. At this stage this vote is worthless and valueness. About its only use is to show that this ridiculous electoral methodology should never ever be used on wikipedia again. But then it barely qualifies to be called an electoral methodology anyway. Any system that is not instantly comprehendable to users deserves to be binned. FearÉIREANN^(talk) 22:44, 10 May 2005 (UTC)[reply]

A true statement cannot be defamatory. Wikipedia:No Personal Attacks Whig 13:05, 11 May 2005 (UTC)[reply]

A lie twisting facts to justify your initial misunderstanding of what I wrote, is defamatory. Your inability to admit you didn't understand what I wrote and your continued determination to twist it to suggest something it does not, is revealing. But then it does give you a chance to distract attention from the worthless farce your voting system has produced. FearÉIREANN^(talk) 20:58, 11 May 2005 (UTC)[reply]

I have presented the facts, and I understood what you wrote perfectly well. Presumably others who might read this correspondence can review the facts and competently judge whether I have misrepresented your statements. I have endeavored to carry out a fair survey and I do not appreciate the righteousness of your attempt to spoil it. Be that as it may, you accuse me of defamation, and you accuse me of lying, these are personal attacks unsupported by fact. I have not accused you of lying, nor of secrecy, nor of anything except making the statements that you do not dispute having made: I have characterized those statements as encouragement to deception, which they were. Whig 04:18, 13 May 2005 (UTC)[reply]

Prior Consensus

Before this vote corralling began, Alternative 3 was not only the Condorcet winner, it was preferred by exactly 3:1 to the default (Alternative 5: None of the Above). In the current results posted at the time, I did not declare a consensus then existing because it was shy precisely 1 vote from exceeding the 3:1 ratio. It was, nevertheless, preferred by a 75% supermajority to the default. Whig 14:17, 9 May 2005 (UTC)[reply]

But that was only because people many people had not voted for option 5. Now that they have, it's a moot point. — Asbestos | Talk 14:32, 9 May 2005 (UTC)[reply]

Yeah, but they also voted for Alternative 2, which means that Alternative 5 is now the least preferred option. Whig 14:47, 9 May 2005 (UTC)[reply]

Surely if the vote is to close on 14 May, there was never any possibility of declaring a consensus at some earlier stage. In a general election, if the voting is to close at 9 p.m., it closes at 9 p.m. You don't declare a particular party the winner just because most of the morning and early afternoon voters voted for that party. Ann Heneghan 15:13, 9 May 2005 (UTC)[reply]

Of course. If you have been watching the Current Results section, these are just current results, not final, and intended only to be informative about the course of the voting so far. Any current consensus that might exist at some time may be erased before the voting concludes. The final results will not be counted until after the close of May 14 UTC. Whig 15:41, 9 May 2005 (UTC)[reply]

This suggests that this system is working rather poorly, does it not? Since we clearly don't have a consensus. john k 14:53, 9 May 2005 (UTC)[reply]

Not really. If people vote dishonestly weird results might crop up as the rankings of all the also-rans might get pretty scrambled, but the ultimate outcome will not be altered. Whig 15:09, 9 May 2005 (UTC)[reply]

Bratsche's Response: I object to having my own vote being discussed without my knowledge. No one discussing the legitimacy of the votes actually contacted those first contacted by Jtdirl, and asked them about the vote or vote changes. In my view, Jtdirl merely informed me about this type of voting system, and the effects of my vote or vote changes. This system is needlessly confusing, and I thank Jtdirl for taking the time to explain it to me and the other users.

I don't think that some of you are familiar with the voting processes here on the Wikipedia. People explain and influence others on vote related pages all the time, particularly on WP:VFD. Many times, a page up for deletion is rewritten by another user to a point where it is an acceptable article. Usually, users who have voted Delete for that article are contacted, to show that their vote might be something that could be changed in light of the article rewrite. I have done this several times, and also been contacted by others doing the same thing. There's nothing wrong with this. It could even be call "vote coralling." It's completely fine. Remember the Gdansk/Danzig vote? People were educating others all the time on these votes.

Jtdirl is a seasoned Wikipedian with many, many edits under his belt. I trust his judgement on these issues. I would look gladly to any user who would care to educate me on a vote or voting process in the future. Stop making personal attacks on users who care to speak their mind and help others. If anyone has questions, feel free to contact me through email or my talk page. Thank you. Bratsche^{talk random} 22:14, May 10, 2005 (UTC)

If Jtdirl had tried to persuade people on their Talk pages that a particular position was correct, or that they should give a second place vote to some particular option for certain reasons, etc., I wouldn't object. My problem is that he encouraged people to vote dishonestly in order to tactically manipulate the survey. I stand by my objection; it is not a personal attack, as it is not directed at the person but at his actual and documented behavior. Whig 22:37, 10 May 2005 (UTC)[reply]

I do agree with your point that convincing others to vote dishonestly is wrong. But I fail to see how Jtdirl's behavior is dishonest. The message that I recieved seems just to explain how the choices (first choice, second choice, etc.) work out. His advice boils down to ranking your favorite option 1, and your least favorite 5. Then you use your remaining votes to choose their respective rankings. I had thought that my first vote, a first and only choice vote, had the same effect. Apparently, the CSSD method works a bit differently than I am used to. Until now, I have only encountered simple up-or-down votes, or the like. What points in Jtdirl's message do you find dishonest? Bratsche^{talk random} 22:53, May 10, 2005 (UTC)

Voting a full slate of preferences is desirable, and not only would Jtdirl not have been wrong if that were what he had advised, but as you presumably know I have also been encouraging everyone to vote all preferences all along.

What Jtdirl wrote that I take special objection to is, "Just be careful though not to copy everyone else doing it. If everyone gives the same other alternatives the same order of votes they may win. So if option 4 gets a lot of 2s, give it a 4." As you say above, "convincing others to vote dishonestly is wrong." That is what Jtdirl attempted to do. Whig 23:00, 10 May 2005 (UTC)[reply]

That was not "convincing others to vote dishonestly". What is dishonest is pretending it is. All it is is highlighting to users is that in voting tactically there is a danger that it can backfire. In other words if everyone voting tactically replicates the same tactical voting, they can create a result they don't want. It is exactly the same warning as Blair gave to Labour supporters - if you tactically vote for The Lib Dems to express your opposition to current New Labour policies, you may inadvertently let the Tories win in a seat. My advice to people looking to vote tactically was self-evident: tactical voting is a double-edged sword. If you choose to vote tactically (a democratic right throughout the world) think carefully about what you are doing and how you do it, or you may cause an option you don't want to win to win. But then this would not be a problem in a credible, transparent poll. This ridiculous poll is neither transparent nor credible and is clearly worthless as a consultation exercise. Maybe when this worthless nonsense if over we can have a proper, professional, transparent exercise. FearÉIREANN^(talk) 23:15, 10 May 2005 (UTC)[reply]

You were not discouraging people from voting tactically. You were explicitly encouraging them to do so. Calling this "a democratic right throughout the world" to justify yourself is pretty weak, it is done where it works because most voting systems are vulnerable to deceptive voting. Fortunately CSSD is not. Whig 23:59, 10 May 2005 (UTC)[reply]

So obviously you believe your system (which is used in how many countries exactly?) is honest and the rest of the world's systems corrupt. What a strange understanding of democracy you possess. BTW I neither discouraging nor encouraged people from voting tactically. I was pointing out that the way they had previously voted represented the principles of FPtP but that this system did not operate that way. I advised them to use their full votes. I pointed out that they could vote tactically. I pointed out that tactical voting has pluses and minuses. That is all. I left it up to them to decide what to do. FearÉIREANN

^(talk) 20:58, 11 May 2005 (UTC)[reply]

So encouraging others to vote tactically is dishonest? This is what you have just stated, unless I have misunderstood you. Please clarify. Bratsche^{talk random} 02:02, May 11, 2005 (UTC)

Bratsche's words above were: "Convincing others to vote dishonestly is wrong." I think the record speaks for itself. Whig 03:16, 11 May 2005 (UTC)[reply]

But I don't know how what is dishonest?!?!?!?! So let me ask you outright: is encouraging others to vote tactically dishonest? Bratsche^{talk random} 20:41, May 11, 2005 (UTC)

Too broad a question. A compromise can be considered tactical, but it isn't dishonest. Whig 20:57, 11 May 2005 (UTC)[reply]

Interesting Conclusion

Debian's implementation of CSSD which tries to provide for supermajority consensus is subject to strategic voting.

The method we are using for this survey was declared before the survey began. We are not bound to use Debian's implementation, as has been discussed before. Go ahead and vote as strategically as you want. You are only hurting your own position if you do not vote your true preferences, because your vote won't count the way you want it to.

The vote will be counted according to standard Cloneproof Schwartz Sequential Dropping.

Every vote will be counted. Whig 15:06, 9 May 2005 (UTC)[reply]