Talk:Ranked voting/Archive 1

AU Senate election ballot.

Would Tomruen care to try making a sample AUstralian Senate ballot paper. Looks a bit like this:

'''couldnt someone just tell me what the is preferential voting?'''
--------------------------------------------------------------------
*A[_]Green_ B[_]NLP__ C[_]Dem__ D[_]Libtn___ E[_]Rep___ UnGrouped 
*XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
*A[_]Nader_ B[_]Smith C[_]Kerry D[_]deSoto__ E[_]Bush__ UG[_]Rubble
*A[_]Brown_ B[_]Jones C[_]Derry D[_]Friedman E[_]Cheney UG[_]Flintsone
*__________ B[_]Smyth C[_]Perry D[_]Milton__ E[_]Laura
*__________ _________ _________ D[_]Forbes
*Instructions, either
* write the number "1" in a box above the line; or
* write the numbers "1", "2", "3" ......."16", "17" in the boxes below the line.
---------------------------------------------------------------------

— Preceding unsigned comment added by Syd1435 (talk • contribs) 13:12, 3 October 2004 (UTC)

Mystery section

I have deleted this section:

A potential problem with preferential votes is that they can be used to undermine a secret ballot, and thus enable corruption by vote buying. If there are enough candidates then the number of possible voting patterns may be much larger than the number of voters, and it then becomes possible to use early preferences to vote for the desired candidates and then to use later preferences to identify the voter to the person who has purchased the vote and looks at the ballot papers.

As an example, in the Irish general election, 2002, the electronic votes were published for the Dublin North constituency. There were 17 candidates allowing more than 966 million million possible patterns of preferences, but there were fewer than 44,000 votes cast. The most common pattern (for three of the candidates in a particular order) was chosen by 800 voters, and more than 16,000 patterns were chosen by just one voter each.

One way to avoid this possibility for buying a vote and confirming it has been cast as specified is to prevent partisan observers from systematically viewing each voter's preferences.

I have no idea what this is supposed to mean. I live in a country which has used preferential voting for 80 years and I have never heard if anything remotely like this (inasmuch as I understand what "this" actually is). If someone can explain it in plain English perhaps it can be reinstated. Adam 23:54, 22 September 2005 (UTC)

I'm glad you cut it. I'd at least want a reputable source for this claim. KVenzke 04:45, 24 September 2005 (UTC)

It is very simple. A wants to buy or bully a vote from B, but wants to know that the vote has actually been cast. A tells B how to cast the first few preferences to suit A's political purposes and how to cast the remaining preferences to identify B individually. A then looks at the individual votesas cast. If there a vote like this then A pays B, if not then kneecaps may be at risk. It does not work where voting is a single mark.

Here is an article about it [1]. Here are two quotes from the Irish government's Commission on Electronic Voting report[2] though they have got the number too low (you do not have to vote a full preference list in Ireland):

"The claim here is that someone could bribe or intimidate a voter to give a first preference vote in a certain way by requiring that this voter register a distinctive and unique sequence of preferences for lower-ranking candidates. This is in principle feasible, given the huge number of different ways in which an STV ballot can be completed. In the 2002 Dáil election trial in Meath, for example, there were 14 candidates – giving 14x13x12x11x … x1 (= c.87,178,291,200) different ways to complete the ballot. This gives plenty of opportunity for every voter in the constituency to be given a distinctive signature sequence for lower order preferences."

"Publication of the ballot results in full but in random order, as happened after the 2002 pilots, is a very valuable aid in ensuring the accuracy of the results, since anyone is free to recount these for themselves. Nonetheless it has been submitted that this can in theory reveal deliberate and distinctive voter “signatures” of low-preference votes (highly improbable rankings of the candidates ranked 11 to 15 in a 15-candidate contest, for example), which could allow voters to identify themselves in a context of corruption or intimidation. This may be deemed improbable, but it remains a possibility."

If you want to check the votes yourself, they can be found at [3] --Henrygb 13:12, 3 October 2005 (UTC)

That is the uttermost nonsense. How does "A" inspect all the ballots? In a Dail constituency there are 45-60,000 of them. In an Australian House of Reps seat there are 80-90,000. Certainly in Australia it would be totally impossible. If "A" wants to influence an election there are a dozen easier ways to do it. This is entirely hypothetical, extremely improbable, and not worth including. To say that preferential voting is more susceptible to fraud than any other method of voting is completely false. Adam 13:23, 3 October 2005 (UTC)

"How does A inspect all the ballots?" Well if A can, and they could in the electronic cases in Ireland (follow the links) as well as US elections (where newspapers looked at each hanging chad, then it is easy. And if A cannot, then elections become less transparent and easy to fix. Preferential voting provides a method for fraud which others do not: denial as "nonsense" is simple POV. --Henrygb 15:33, 3 October 2005 (UTC)

And just to prove it is possible, here is somebody's actual unique vote in Dublin North in 2002. It is not evidence of corruption (just of peculiar cross party voting, though in the event under Irish STV it went to Jim Glennon and stayed there and so helped contribute to Fine Gael's collapse and to Fianna Fail's victory), but it is evidence that corruption can easily work this way with preferential voting. I can easily look for any pattern of voting in this seat you might want to suggest.

1 Glennon,Jim,F.F.; 2 Boland,Cathal,F.G.; 3 Owen,Nora,F.G.; 4 Sargent,Trevor,G.P.; 5 Ryan,Seán,Lab; 6 Goulding,Ciaran,Non-P; 7 Daly,Clare,S.P.; 8 Kennedy,Michael,F.F.; 9 Wright,G.V.,F.F.; 10 Walshe,David Henry,C.C. CSP; 11 Quinn,Eamonn,Non-P; 12 Davis,Mick,S.F.

You cannot say "impossible" just because it is is not known in Australia. And it is specific to preferential voting. --Henrygb 23:33, 3 October 2005 (UTC)

Fot this to be included in an article, it has to be shown that it has actually occurred, not just it is a remote theoretical possibility. Voters can be bribed or intimidated under any system. Adam 00:22, 4 October 2005 (UTC)

Indeed they can, but in most systems not using preferential voting a secret ballot makes it harder. The issue has been identified as a possibility by an official source in one of the few countries using preferential voting for national elections, making it encyclopedic. --Henrygb 08:53, 4 October 2005 (UTC)

Both Australia and Ireland have the secret ballot, so it is impossible for anyone to know how any individual voted, even by looking at all the ballot papers, even if this were allowed, which it certainly isn't here and shouldn't be anywhere. In the example you give above, are you telling me that this was the only voter out of 43,000 in Dublin North who voted in that way? I don't believe it. How could anyone possibly tell? They would not only have to inspect 43,000 ballot papers, but record the voting order on each of them and then compare them. And even if it were the only one with that voting pattern, what does it prove? That someone paid the voter to do it? Why would someone who wanted to influence an election do it in such a laborious way, and in a way that makes it possible to detect what they had done? Why not just pay voters to vote for Glennon, if that was what they wanted to do? The whole notion is ridiculous. Australia has been using preferential voting for federal, state and local government elections since 1918, and such an idea has never even been suggested, let alone demonstrated, and believe me Australia has a long history of electoral fraud of various kinds. If someone in Ireland has suggested it is as a possibility, that just shows they have too much time on their hands. Adam 10:09, 4 October 2005 (UTC)

It appears that the key issue here isn't preferential voting, but the combination of preferential voting with the publication of individual votes in electronic, searchable format, which doesn't happen in Australia. The problem seems to be that easy access to these votes, together with the large number of unlikely "natural" votes, makes it possible for a voter to "prove" they have voted as desired, something that would otherwise be impossible. JPD 10:31, 4 October 2005 (UTC)

You may not believe the example vote was unique, but it is easy to check - follow the link. One other voter had the identical five early preferences but then stopped expressing any more preferences. Another five voters had the same first four preferences, but each of these had distinct overall patterns. Another twenty-five voters had the same first three preferences (a few identical patterns here), another 135 had the same two first preferences, and another 5,726 had the same first preference, out of the 43,942 total voters. --Henrygb 21:59, 4 October 2005 (UTC)

OK, I give in - I really didn't believe a national election authority could be so stupid as to allow full electronic scrutiny of its ballot papers, which seems to open the door to all kinds of mischief. As JPD says, the problem, if it is one in Ireland (and I'm still not persuaded that this actually happens), does not lie with preferential voting. It lies with post-election public access to ballot papers, which should never be allowed. In this country we trust our election authorities to count the ballots (under scrutiny of course), and the ballots are then destroyed, so the problem cannot arise. So if there is to be a paragraph written about this, it belongs in Irish election system rather than in this article. Adam 00:09, 5 October 2005 (UTC)

In this country … the ballots are then destroyed, In federal and most standard territory elections, though not in elections that use countback to fill seats which come up before general elections, rather than holding by-elections. Just being pedantic.

Still, I'm not convinced that it's not a problem of preferential voting, and as such I think it should be mentioned here. The bulk of the detail might be more appropriate in an article on the Irish election system, ’tis true, so it might just be a sentence directing those concerned to that article.

—Felix the Cassowary (ɑe hɪː jɐ) 03:49, 5 October 2005 (UTC)

Even in Ireland, it's not a "problem" at all: it is a hypothetical situation. I have conceded that it is a theoretical possibility, but so is the election being hijacked by aliens. I am still opposed to disproportionate mention of this completely hypothetical problem in an article of this length.

On Australian elections, are the actual ballot papers kept and recounted in Tasmania etc, or are the results simply recalculated from the initial count? In either case, the ballots are not available for public scrutiny, and even if kept they would be destroyed at the expiry of the Parliament when there can be no further countbacks. Adam 03:58, 5 October 2005 (UTC)

This section about Australian voting system needs clarification;

Voters preferences are now data-entered into computer systems, which then process the recorded votes to determine the results of the election.

At what point does this occur? I have scrutineered many vote counts. The "voters preferences" are sorted and counted at the voting booth immediately the polls close. They are recorded, in an agggregate, and when the number of ballots cast is reconsiled to the number of ballots given out it is all phoned through by the election official to some other official at the tally room. (as the scrutineer I am also phoning the results through to the candidate of the party). Of course at that point it will be entered into a computer - but the votes are in aggregates, as first preference votes and 2 party preferred (2PP), e.g. First Preference: J.Brown 1250, B.Green 350, R.White 1301, Informal 35. 2PP: J.Brown 1550, R.White 1351. There is a little bit more to it than that, but that's the gist of it.

BTW where you have say 6 candidates on the ballot the preference distribution is always; which is ranked higher, Liberal/National or Labor? They never do the full preference distro through all the candidates, or at least I'v never seen it done like that. — Preceding unsigned comment added by 203.206.5.177 (talk) 12:53, 24 September 2006 (UTC)

Uniqueness of Votes

This section is fairly coherent and then drops way off the radar scope

To wit: "If there are large number of candidates, more common in STV elections, . . . .patterns were chosen by just one voter each."

This reads fine. I had a few people look over my shoulder and they got the point--a bit arcane but they got the point.

Here it gets murky: "The number of possible complete rankings with no ties is the factorial of the number of candidates, but with ties it's equal to the corresponding ordered Bell number and is asymptotic to .[1]"

I understand what it is trying to say but I think there are going to be more than a few that need a little more information. Are there any possible links to be made here?

Malangthon 13, Oct. 2006 12:35 EPT

I would drop the second paragraph altogether. --Henrygb 00:44, 13 October 2006 (UTC)

RONR's multiple-member preferential voting

See Voting methods in Robert's Rules of Order#Single transferable vote. What is the name of the voting system that uses preferential ballots, but instead of transferring surplus votes, continues dropping the candidates with the lowest vote counts until the number of candidates remaining equals the number of seats to be filled? Captain Zyrain 00:40, 17 September 2007 (UTC)

I remember now! It's preferential bloc voting. Captain Zyrain 04:18, 27 September 2007 (UTC)

The preferential bloc voting method is what the City of Hopkins, Minnesota is considering, but I've never heard a name for it before. Does RROO use this term? (Also it is different from STV since multiple ranked votes are counted at once.) Tom Ruen 04:50, 27 September 2007 (UTC)

Ballot images added

I added 4 types of ballot images to this page. I just used MSPaint to make them, but I tried to make them pretty. I think they all cover the range of possible formats. If anyone would like to improve the ballot images or descriptive text above, feel free. --Tomruen 06:00, 8 May 2004 (UTC)

It is all very well to have examples of different ballot papers BUT there is not one word on this page about how preferential votes are tallied and candidates allocated as winners of seats which makes the information rather incpmplete to say the least. I gather instant-run off is one method of tallying preferenece votes but there is no link the page of that topic. Are there other methods? Really needs a section on how preferential votes are tallied even if that section is mostly cross references. —Preceding unsigned comment added by 220.253.4.133 (talk) 08:00, 22 August 2010 (UTC)

Countries using Preferential Voting systems?

I can't help thinking that this article would be improved by including a list of nations using preferential voting systems - if nothing else, it would provide some worked examples to help counteract the somewhat theoretical sound of this entry. Lokicarbis (talk) 03:28, 5 September 2010 (UTC)

See history and use of the Single Transferable Vote, list of US cities that have used STV, history and use of instant-runoff voting, and instant-runoff voting in the United States. Markus Schulze 20:53, 5 September 2010 (UTC)

By UN law a voting card should not be heavier than an onion?

This seems like vandalism not verifiable fact... — Preceding unsigned comment added by 141.212.196.110 (talk) 19:45, 23 November 2010 (UTC)

STV - NOT used on many European countries

STV is only used in Europe for national elections in the Republic of Ireland and Malta, local elections in northern Ireland, Republic of Ireland and Scotland, European elections in Northern Ireland and Republic of Ireland. (Coachtripfan (talk) 17:34, 14 August 2013 (UTC))

Neutral viewpoint needed for comparison to "FPTP"

The section about the comparison to FPTP needs to be reworked with a neutral viewpoint. Especially phrases like "Superiority v. FPTP" and "preferential voting achieves outcomes better than those produced by a first-past-the-post system" should be replaced by facts. Also, the whole first paragraph just expresses the opinion of a politician that supports AV. After reading the section, the reader knows that some wikipedia author(s) think(s), that preferential voting is better than FPTP, but the reader does not know the facts behind this "preference". This should be the other way round. Best regards, --Arno Nymus (talk) 16:55, 30 July 2012 (UTC)

It may sound biased, but I don't think you'll find any source that supports FPTP as being "better" than preferential voting. I think more academics support cardinal voting systems than preferential voting, but I don't think anyone thinks it's worse than FPTP. Of course, if sources can be found they should be added to the page. Fleetham (talk) 23:40, 30 July 2012 (UTC)

I don't contradict the opinion. I just say, that it has to be written in a neutral viewpoint with facts (as "With preferential voting the ballots contain more information than FPTP ballots") instead of opinions (as "preferential ballots are better than FPTP ballots"). The facts should be described - best without opinion words like "better". Then, the reader can judge on his own.

• E.g. "preferential voting achieves outcomes better than those produced by a first-past-the-post system" What is a "better outcome"?

The cites are more or less meaningless:

• "first-past-the-post leads to a whole host of problems... it means MPs can go about their business without ever having to appeal to a majority of their own constituents..." <- that is also true for any preferential voting system; that is easy to see for a constituency, in which no candidate is liked by more than 49 %.
• »He also stated that a preferential voting method like instant runoff voting would result in politicians "work[ing] harder to appeal to more people than before« Ok, that is the opinion of Nick Clegg. But, by which hints or facts did he come to this opinion? What hints or facts contradict with that opinion? Why do we have this opinion without any words about its validity?

Additionally to the fact/opinion problem, the following parts lack a comparison. Just say a bad thing about FPTP without comparing it to the compliance by preferential voting is deceptive:

• Academic experts have also rejected first-past-the-post; at a 2011 LSE workshop attended by 22 voting theory specialists, none endorsed first-past-the-post as the "best voting procedure". <- Sure, but why? The readers want to know why it is bad (or not the best). Facts are needed, not a popularity contest. This is also worsened by the fact, that no word is said how the results for preferential voting methods were. Also, the first sentence ("rejected") is a very biased description of the result of this "popularity contest".
• It's true that choosing between options using a first-past-the-post system will sometimes select a winner that would have lost out to every other option in a two-way contest, the so-called Condorcet loser. <- OK, FPTP fails the Condorcet loser criterion - why are in this "comparison" no words about preferential voting systems compliances to that criterion (e.g. Bucklin, Contingent vote, Minimax fail; Borda, Schulze, Ranked pairs comply)
• What is more often the case is that the choice selected by a first-past-the-post vote is not the one wanted by the largest number of people. <- This is only correct for some definitions of "wanted". In fact, by the easiest definition, it is not. However, for most preferential voting systems it is not satisfied, too.
• etc. Please don't understand this criticism as an attack. I'm surely not a fan of FPTP. All I say is that we could and should have a clearly more informative section to give the reader the chance to make his own decision. --Arno Nymus (talk) 03:11, 1 August 2012 (UTC)

Well, I think the point is basically that "FPTP sucks". Why is preferential vote a better system than FPTP? Because FPTP sucks. I think the "better outcomes" that occur under most any preferential voting system are evidenced by the failings of FPTP--failings that preferential voting systems don't share. Again, if you any source can be found praising FPTP, it should be included. I didn't even bother to look for any because I doubt there are any. Fleetham (talk) 03:19, 1 August 2012 (UTC)

OK, I think my text is too lengthy. So, only a short question to see if I could make my point clear: Do you understand what I mean with the difference between fact and opinion? --Arno Nymus (talk) 18:16, 1 August 2012 (UTC)

Apparently not as I don't think any of that section is opinion :) Fleetham (talk) 18:35, 1 August 2012 (UTC)

The only portions that can be construed as opinion are the Nick Clegg quotes and the "academic experts" part. I don't know why we should assume Nick Clegg is lying or misinformed (shouldn't he be quite knowledgeable on the topic?), and I think the statements of academic experts, like those of Nick Clegg, reflect the amount of knowledge they have amassed about this area. As these sources are undoubtedly more informed on the matter than you or I are, I'm not sure it's quite right to call their statements an opinion. If it was, how would we know? I'd be happy to agree to removal of those bits if evidence can be found that contradicts them. Fleetham (talk) 19:09, 1 August 2012 (UTC)

• OK, so at least the problem is revealed: I consider statements like "A is good", "B is bad", "C is better than D" opinions, you consider them facts, as long as no source disagrees (more or less). Maybe it is more clearly, if you think of the word "opinion" (as I use it) in the meaning of "view" or "judgement" and instead of "fact" you maybe should think of as "something denotable by data information"
• Now, an opinion does not become a fact, only because the facts support it. For example, both "Stan Smith is a better wide receiver than Jack Johnson." and "Jack Johnson is is a better wide receiver than Stan Smith", are opinions. In contrast, "Stan Smith has achieved more touchdowns than Jack Johnson." is a fact. Now, one can come to the conclusion, that therefore "Stan Smith is a better wide receiver than Jack Johnson.", because the fact supports that opinion. Nevertheless, it still is just an opinion.
• What does this mean for our article?

1. As can be seen, if an opinion is "supported by facts", instead of writing the opinion into the article, you should just state the facts: "Stan Smith has achieved more touchdowns than Jack Johnson." The reader reads the facts and can come to a conclusion on his own. This is how an article in an encylopedia should be: it presents the facts and the reader makes it opinion out of the facts.
2. If we can't write down facts that support an opinion, this opinion is maybe not very reasonable, and so we should not write it into an article.

So, in either case the opinion does not have a place in the article, but the fact has and is absolutely necessary and sufficient.

• Only for the record: I don't have a problem with Nick Clegg and especially don't say that he is misinformed or even lying. However, I would not call him an academic expert, but a politician - but that's not an offense, I just don't know any scientific work of Clegg in the voting theory, whereas I obviously heard of his politic work. However, I also don't agree with "As these sources are undoubtedly more informed on the matter than you or I are" - but if they are, facts(/data) exist that justify their opinions(/views). ;) --Arno Nymus (talk) 23:24, 1 August 2012 (UTC)

In reply:

• The statement that "better outcomes are to be had with preferential voting" is predicated on the idea that a voting system should select the Condorcet winner if there is one and reflect majority rule. It's a fact that preferential voting systems meet these goals better than FPTP. Perhaps it should be explicitly stated that by "better" we really mean "more likely to provide outcomes consistent with mainstream ideals for electoral systems"? Fleetham (talk) 01:12, 2 August 2012 (UTC)

Yes, a formulation like "more likely to choose a Condorcet winner" (and thus, looking about compliance with "ideals"/criterions) is a step in the right direction of "facts not opinions(/views)". Now, with the regard to the content: There are many different preferential voting methods with very different results in this respect: Ranked pairs, Minimax and the other Condorcet methods obviously satisfy it, because they choose the Condorcet winner every time it exists. However, IRV, Borda and Coombs sometimes don't choose Condorcet winners, just like FPTP. For each of these systems, it is easy to construct cases where they choose a Condorcet winner and FPTP does not - or the other way round. And obviously, systems can be constructed that are "preferential voting systems", that clearly are much worse than FPTP in this respect.

The thing about "preferential voting methods" is, that it is NOT defined over the output of the election (the winner), but over the input (the votes). Because of that, it will be difficult to find (non-trivial and non-abstract) facts about the output that hold for all preferential voting methods. One of the abstract facts about preferential voting systems is Arrows theorem. On the other hand, we can easily make conclusions about the inputs: e.g. "Preferential voting systems give the voter the possibility to specify more information on the ballot than FPTP" and so on. Thus, maybe also the comparison section should concentrate on the input. --Arno Nymus (talk) 17:04, 2 August 2012 (UTC)

I agree that it's difficult to find abstract facts that hold true about all preferential voting methods. I think this section should focus on "majority rule" and "Condorcet winner" and make broad statements such as the ones in the quotes, and then break down the differences between the different systems. I think the Borda count is deprecated and the Coombs rule little used, so the only source we would really need would be something that speaks to the Condorcent efficiency of AV/IRV. Fleetham (talk) 18:05, 2 August 2012 (UTC)

(We don't have to show anything other than the fact that most preferential voting systems likely to be used best FPTP in those two categories to make broad statements like "preferential voting systems are better than FPTP") Fleetham (talk) 18:44, 2 August 2012 (UTC)

It has been two years and nothing happened as a result of this discussion; Arno Nymus is completely correct. Wikipedia may not pass value judgments, even against things that are obviously stupid (like FPTP). Consider this lede from the current version of Murder:

"Murder is the unlawful killing, with malice aforethought, of another human, and generally this premeditated state of mind distinguishes murder from other forms of unlawful homicide (such as manslaughter). As the loss of a human being inflicts enormous grief upon the individuals close to the victim, and the commission of a murder is highly detrimental to the good order within society, most societies both present and in antiquity have considered it a most serious crime worthy of the harshest of punishment. In most countries, a person convicted of murder is typically given a long prison sentence, possibly a life sentence where permitted, and in some countries, the death penalty may be imposed for such an act – though this practice is becoming less common."

You will notice that the article does not make a value judgment. Murder is wrong and everyone knows it, but the closest the article comes to condemning it is calling it "highly detrimental to the good order within society" - and even that isn't sufficiently neutral! The article communicates the general unacceptability of murder by presenting objective facts about it, rather than calling it "bad." Similarly, it is not neutral to say AV leads to "better outcomes" than FPTP; we should say what those outcomes are. For instance, instead of "preferential voting achieves outcomes better than those produced by a first-past-the-post system", we should say something like, "unlike FPTP, AV prevents most cases of strategic voting, will never elect a condorcet loser, and has greater pareto efficiency." have attached a sticky note to my monitor and if no one objects I will rewrite this paragraph when I get around to it. - 66.189.105.217 (talk) 04:40, 3 April 2014 (UTC)

Over the past few weeks I have made a couple of attempts to come up with a better rewrite for this section; I repeatedly found that what I was trying to write was a comparison of IRV with FPTP, and lost the generality of "ranked voting". I've concluded that trying to compare "ranked voting" with FPTP is not really something one should do - the Borda Count ends up with very different properties than Instant Runoff, which is itself different from STV. I don't think this kind of comparison belongs here, and I have struck it. A decent alternative might be to add a section like "comparison with plurality voting", which may be more appropriate. 66.189.105.217 (talk) 23:12, 16 May 2014 (UTC)

First section should probably say what Ranked voting is.

Because this Wiki page was so broken, I just elected to reference a non-Wikipedia page on Ranked Voting, rather than this page.

Suggest you all say only what ranked voting is at the top.

In other words, please move the controversy down towards the end if this article, where most Wiki pages have it. — Preceding unsigned comment added by 29Flavors (talk • contribs) 04:56, 17 February 2018 (UTC)

Proportional representation

How different is this to proportional representation? Chaosdruid (talk) 12:28, 7 November 2018 (UTC)

Lead section too technical, too broad

Just count all the (lazy?) hyperlinks. Compare to the clean, focused writing at https://ballotpedia.org/Ranked-choice_voting_(RCV) and others.

Here's an idea. Any 12-year old knows how "standard" voting works, is the lead section of RCV the best place to give it technical names and define it. ...and many other kinds too? It's obvious the writers here are are very enthusiastic. Suggest putting that later, and focus. Who is the intended audience? (Joe Blow who likely followed a link and only wants a sketchy working definition.) and what is the purpose of the Lede section? see MOS:LEDE — Preceding unsigned comment added by 2602:306:CFCE:1EE0:2830:2E3D:F911:8E1E (talk) 02:07, 1 December 2019 (UTC)

Is there a difference between instant runoff and ranked voting?

There is a separate article for ranked voting, I wonder if ranked voting and instant run off are the same thing? This is an issue because Rep Jared Golden was elected under a system USA Today called Ranked Choice Voting.

Should these 2 articles be one? Geo8rge (talk) 14:29, 18 November 2018 (UTC)

Yes, there is a difference. "Ranked voting" happens when each voter ranks the candidates, puts them in order of preference. "Ranked voting" as such does not tell us what to do with those ranks, how to find out who won. And there are different ways to deal with it. One of them is Instant-runoff voting, another one is Single transferable vote, there are many more. The article actually does list some of them.

Thus no, a merge would not be a good idea. --Martynas Patasius (talk) 16:05, 18 November 2018 (UTC)

Well, it looks like editors have made the merge happen over the past year. This article has become useless to describe anything other than Ranked-choice voting in the United States. Now this article fails to properly describe alternatives to Instant-runoff voting. Is it time for someone to suggest a merge between Ranked-choice voting in the United States and this article? -- RobLa (talk) 02:46, 2 December 2019 (UTC)

No, this article should not be merged. And it is NOT useless! Ranked ballots is what this article is about, and "ranked choice" is just one of many ways in which ranked ballots can be counted. Instead of allowing this important article to disappear, I suggest that this article be renamed to "Ranked ballots". If that is not acceptable then this article can be improved to better clarify what "ranked voting" means. VoteFair (talk) 16:37, 2 December 2019 (UTC)

You're right, VoteFair, and my apologies for the overreaction (based on my skim). My reaction was based on seeing "ranked-choice voting" bolded in the introduction:

Ranked voting, also called ranked-choice voting, is an election voting system in which voters rank choices in a hierarchy on the ordinal scale (ordinal voting systems): 1st, 2nd, 3rd, etc. In some areas, ranked-choice voting is called preferential voting, but in other places this term has various more-specialized meanings. The other major branch of voting systems is cardinal voting, where candidates are independently rated rather than ranked relative to each other.

Doesn't that wording imply that "Ranked voting" (a generic term for ballots used by IRV, Condorcet methods, and the Borda system) and "Ranked-choice voting" (FairVote's branding for IRV) are equivalent? -- RobLa (talk) 00:10, 8 December 2019 (UTC)

The current wording is not correct. To correct it, here is the wording I suggest for the first three paragraphs:

Ranked voting is an election voting system in which voters use a preferential (or ranked) ballot to rank choices in a sequence on the ordinal scale: 1st, 2nd, 3rd, etc. The rankings can be counted in multiple ways to determine which candidate (or candidates) is (or are) elected. The other major branch of voting systems is cardinal voting, where candidates are independently rated, rather than ranked, relative to each other.

The similar term Ranked Choice voting refers to using ranked ballots and instant-runoff voting, which is a specific counting method. In some locations the term preferential voting is used to refer to this combination of ballot type and counting method, but in other locations this term has various more-specialized meanings.

A ranked voting system collects more information from voters compared to the single-mark ballots currently used in most governmental elections, many of which use First-Past-The-Post and Mixed-Member Proportional voting systems.

I am attempting to offer a neutral wording that avoids the conflicts that often arise between fans of IRV and fans of Condorcet methods. Any suggestions for improvement would be appreciated. VoteFair (talk) 18:56, 9 December 2019 (UTC)

Thanks for taking this up! One first suggestion: how about changing "Ranked voting is an election voting system..." to "Ranked voting is any election voting system..." (i.e. changing "an" to "any") -- RobLa (talk) 00:18, 10 December 2019 (UTC)

"Ranked choice voting" listed at Redirects for discussion

A discussion is taking place to address the redirect Ranked choice voting. The discussion will occur at Wikipedia:Redirects for discussion/Log/2020 December 1#Ranked choice voting until a consensus is reached, and readers of this page are welcome to contribute to the discussion. Mdewman6 (talk) 01:55, 1 December 2020 (UTC)

Proposal to delete OR template and one sentence of text

I can’t see the justification for the OR template in the section Ranked voting#Uniqueness of votes. Everything in it is supported by a reference except for trivial mathematical statements and the claim that the number of ways of filling in a ballot (allowing ties) is the ‘ordered Bell number’, and this has a wikilink showing that the property in question is the definition of the ordered Bell number.

I also plan to delete the ‘con’ which begins “Because RCV is a new concept... ”. The references say nothing about ranked voting being a new concept, and if ranked voting is understood in the sense of the first sentence of the article, then it dates back to 1299. Colin.champion (talk) 17:04, 10 January 2021 (UTC)

Done. Colin.champion (talk) 14:40, 12 January 2021 (UTC)

Descriptions of systems

Should this really have descriptions of three specific systems in it when they have their own articles? Why those three? It would probably be better to have a nested list of various types, maybe a description of the various classifications, etc. — Omegatron (talk) 01:22, 21 June 2019 (UTC)

I have briefly expanded on the Condorcet method area, including noting that non-Condorcet methods such as IRV can be made Condorcet or Smith-efficient by restricting to the Smith set before full tabulation. Explanations here should be kept brief, but they do give readers a handle from which to explore those on which they are curious and interested. John Moser (talk) 21:51, 2 June 2021 (UTC)

This article confuses preferential voting with STV/IRV

It starts well but then starts describing the mechanism by which ranking are used to arrive at a winner, with rounds and elimination. That approach is not used by all preferential voting methods. Notably, Condorcet methods, which have been around longer, do not. This page presents a very STV/IRV-centric view of what preferential voting is. In fact, I would say it is not viewpoint-neutral. 24.58.57.189 (talk) 13:25, 6 February 2012 (UTC)

I agree. Should we remove the sections that talk about how the ballots are counted? Perhaps we can rename the section that lists different ways to count them (and let the article end with that section). VoteFair (talk) 19:34, 6 February 2012 (UTC)

I have replaced these with the Tennessee example template and an example of IRV and ranked pairs, rather than an example of IRV and IRV. John Moser (talk) 22:48, 2 June 2021 (UTC)

Requested move 2 June 2021

The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this section.

---

Ranked voting → Ranked choice voting – Requesting a swap of these and the redirect from ranked voting to ranked choice voting, with a note at the top that if the user was looking for instant runoff voting they should go there.

FairVote's 2009 Form 990 discusses "instant runoff voting" beginning on page 43, as do prior Form 990 filings. On page 37 of their 2010 Form 990, they instead state "ranked choice voting (RCV, or 'instant runoff voting')" and continue with that name from there.

The term "ranked choice voting" refers to all systems of ranked choice voting, as is currently indicated on the Wikipedia page. Public choice and social choice academic literature references the system as "instant runoff voting" or "alternative vote," and not as "ranked choice voting."

These are uncontroversial supporting arguments.

What is controversial is whether Wikipedia should list pages based on the current common usage in all cases. The change to "ranked choice voting" by FairVote has had visible and predictable effects: defining a generic concept as a specific thing, or a specific thing as a generic concept, is a common and powerful propaganda technique. The latter is visible in the United States with every political discussion of universal healthcare being referenced as "Medicare for all," while whenever a politician proposes a universal healthcare system that is not the specific policy called "Medicare for all" they are sharply criticized by advocates of M4A. Wikipedia correctly redirects Medicare for all to a subsection of single-payer healthcare, while in the United States it is now commonly used to mean universal healthcare.

When discussing ranked choice voting with naive Americans—those who are aware of instant runoff, but not well-read and not greatly familiar with the "ranked choice voting" term—they immediately understand various systems, including Condorcet systems, and the differences between these and IRV. When discussing with those who are familiar with "ranked choice voting," comprehension is lacking in the extreme. This is not unexpected: it is a peripheral technical concept which most people have only heard of in a certain context, so redefining the basic concept of tabulating ranked ballots becomes mentally taxing. This is a topic several of us discussed while at the Public Choice Society annual member's meeting (I was surprised somebody else actually brought it up).

It is not the purpose and mission of Wikipedia to cause this kind of distortion; and it is inoffensive to the purpose and mission of Wikipedia to correctly name those subjects which it discusses. Directing users away from ranked choice voting and to a specific form of ranked choice voting as if they are synonymous is damaging to the clear comprehension of a highly complex and not-well-known subject.

I see two logical propositions here:

• The propaganda argument is correct, and Wikipedia should not further a propaganda campaign; or
• The propaganda argument is incorrect or irrelevant, and Wikipedia should ensure articles discuss the topic regardless, so ranked choice voting should discuss RCV and not be presented as meaning IRV

In any case, ranked choice voting should not redirect to instant-runoff voting, but should list at the top that visitors may be looking for IRV instead. John Moser (talk) 19:48, 2 June 2021 (UTC)

Note: the page Ranked choice voting is a redirect and so is ineligible as a current title in a requested move. This proposal has been altered to reflect that fact. P.I. Ellsworth  ed. ^put'r there 01:46, 3 June 2021 (UTC)

User:Bluefoxicy (John Moser) wrote: "When discussing ranked choice voting with naive Americans [..] they immediately understand various systems, including Condorcet systems, and the differences between these and IRV." This rationale is confusing to me. My experience discussing this area of electoral reform with "naive Americans" has never resulted in them "immediately understanding" anything. We should assume that all readers of English Wikipedia are fairly naive with respect to the subject of electoral systems. Clarifying the distinction between ranked voting methods (like Instant-runoff voting and Condorcet methods) would be useful, but I'm not sure swapping ranked-choice voting and ranked voting helps readers understand the distinction. It seems that the current ranked voting article needs to be modified to stop incorrectly conflating "ranked voting" and "ranked choice voting" (as it currently does in the first sentence, which states "Ranked voting, also known as ranked-choice voting or preferential voting, is..."). Much of the introduction of the current ranked voting article needs to be moved into new sections in the article body. It may also make sense to create a "ranked choice voting" section to describe the difference between "ranked voting" (the generic concept) and "ranked choice voting" (the brand being established by FairVote to replace the "instant runoff voting" brand also established by FairVote circa 1997 or so). It seems that many people who are knowledgeable in electoral systems tend to conflate the ballot given to voters (for example: ranked or rated) and the method used to tally/count the ballots (for example: Condorcet versus Borda versus Hare), so that may also be worth clarifying in this article. -- RobLa (talk) 22:38, 2 June 2021 (UTC)

You make by point for me, in suggesting "ranked voting" is being conflated with "ranked choice voting." It is "ranked choice voting" that is being conflated with "instant runoff voting." I refer you to page 41 of FairVote's 2005 IRS Form 990, which states: "In contrast, IRV elects candidates with majority support [...] this ranked choice voting method saves taxpayers money," and so forth. FairVote rebranded IRV as RCV in 2010, and is referring here in their 2005 IRS filing to Instant Runoff Voting as an example of many Ranked Choice Voting methods. "Ranked voting" means "ranked choice voting" (do you see anyone calling anything "ranked voting"?), and both mean any method by which a voter ranks candidates ordinally on a ballot and a method of tabulation is used accounting for the ordered preferences expressed by the voter on that ballot. You have been confused by sloppy (or not so sloppy…) misuse of terminology.

I was invited to the 2020 Public Choice Society meeting as a speaker because I routinely explain IRV, the Condorcet concept, the Smith set, and the ranked pairs system to people I meet on the streets (I canvass door to door) in three minutes. My experience has been that people who are already excited about "ranked choice voting" will repeatedly recognize the things I'm saying, but then respond by conflating the mechanics of instant runoff as the "fix" after being told instant runoff is broken. People who are not that engaged with the concept become interested, then excited, and tend to stop me along the way and reason on what I'm saying in an (almost always successful) attempt to verify they're understanding what I'm saying. They don't just parrot it back; they absorb it, understand it, and use it to reason on the concepts being explained. (Five minutes in I've already explained Meek's method, what it fixes versus the naive approach to STV, and why STV inherits the problems in IRV but contains them so they're less of a problem. This is not hard.) I live in a high-poverty community and these people are underserved by the education system—this turns out to have no relationship whatsoever to their actual intelligence.

This has all lead me to conclude the confusion in terminology is not harmless; other evidence leads me to suspect it is deliberate. Redirecting "ranked choice voting" to "instant runoff voting" seems harmful, and I'd like to neutralize the article so it does not endorse any particular system, and does not imply RCV has a technical meaning regarding one specific system, with the note at the top that the term is often used as synonymous to IRV in the United States (because it is) and that the user may have been looking for that. As to the issue of many IRV advocates not believing various manipulations are possible? That's for me to deal with in other venues; Wikipedia is not the place for having that debate, only for presenting information in a clear, correct, and neutral manner. "Ranked choice = instant runoff" is not neutral (nor would it be neutral to bias the presentation to lead readers to find Condorcet systems superior), and that it was good marketing and clever propaganda does not change that. My argument is only that the redefinition of terminology has been harmful to the spread of information by teaching people that a specific thing is the general concept, and landing on "Ranked choice voting" should broaden readers's understanding, not corral them into yours or mine or anyone else's idea of which form of RCV we should be using. John Moser (talk) 04:07, 3 June 2021 (UTC)

In addition to FairVote's 2005 Form 990 referencing IRV as a form of RCV, rather than as RCV, I should also point out FairVote didn't use the term "ranked choice" until into the mid 2000s, and did not name IRV "RCV" until 2010. Karl Sims uses the term "Ranked Choice Voting" to describe both IRV and Condorcet methods, introducing the latter with the phrase, "Another method for determining a winner using ranked choice voting was originally proposed by the French mathematician Marquis de Condorcet in the 1700s." This is a reference describing both instant-runoff and Condorcet methods as "ranked choice voting." It is perhaps better that I use a concrete example other than FairVote's evolution of their own terminology to indicate that the term has a different meaning than just IRV. John Moser (talk) 04:20, 3 June 2021 (UTC)

As well, here (his explanation about how IRV fails is incorrect: under the conditions he gave, it would elect the progressive, not the conservative), at Cornell (biased to Condorcet), Equal.Vote (biased toward Minimax).John Moser (talk) 04:34, 3 June 2021 (UTC)

“Redirecting "ranked choice voting" to "instant runoff voting" seems harmful...” I couldn’t agree more. I don’t know enough to comment on the detail of John’s proposal, but fixing this would be a great improvement. Colin.champion (talk) 08:45, 3 June 2021 (UTC)

• Oppose the current redirect reflects the common usage of the term "ranked choice voting" (see e.g. NYT WSJ CNN). All of these describe the form of IRV used in NYC as "ranked choice voting", not as a type of ranked choice voting system. Iffy★Chat -- 13:38, 4 June 2021 (UTC)

  Isn’t this backwards reasoning? The common use of the expression ‘The Worlds’s Favourite Airline’ is to refer to BA, and Google searches will probably confirm that this is its most frequent use. The question is whether (a) BA is most commonly referred to as ‘the world’s favourite airline’, and (b) if so, whether this dominance is sufficient to set aside concerns of neutrality. In the present case, it is relevant that RCV as a name for IRV appears to be limited to the US (as indeed is ‘IRV’), and it looks wrong from the UK to use a misleading Americanism in preference to a neutral term. Colin.champion (talk) 14:26, 4 June 2021 (UTC)

  Suppose there was a Wikipedia article ‘The World’s Favourite Airline (1983 film)’, but all current mentions of ‘the world’s favourite airline’ were to BA. Then it seems to me that ‘The World’s Favourite Airline’ should not redirect to BA in spite of the prevalence of use. Non-neutral redirects should be admitted only in extreme cases. Colin.champion (talk) 14:49, 4 June 2021 (UTC)

  No, Wikipedia's job as an encyclopedia is to summarize what is reported in academic journals, books, news reports etc while using terms commonly used in them so that our readers can understand them, not just those with specialist knowledge in the subject. In your example, then yes, the promotional tagline should redirect to the airline as a WP:PRIMARYREDIRECT with a hatnote to the film. Non-neutral redirects are explicitly allowed per WP:RNEUTRAL. "Ranked choice voting" redirects to "Instant runoff voting" for the benefit of US readers just as "Alternative vote" redirects there for the benefit of UK readers, "Instant runoff voting" is used as the article title because neither the name used in the UK or the US are used outside of their countries. Iffy★Chat -- 16:39, 4 June 2021 (UTC)

  "Alternative vote" is actually a common term used in scientific papers, along with "hare" and "instant runoff voting," to describe a particular tabulation algorithm. These aren't marketing terms designed to reduce people's ability to understand competing concepts. I fail to see how it is "for the benefit of US readers" to support a narrative which on the one hand is technically incorrect, and on the other makes it more difficult for them to place ranked choice voting in the context of ranked choice voting algorithms that aren't instant runoff voting. Imagine if Microsoft had popularized the term "Computer Operating System" to specifically mean just "Windows," and you talk to people like "yeah I need to pick a new Linux operating system" and after explaining Linux to them they're like "damn, can I get Linux Windows from the Microsoft Store?" Because that's what's happening: the naming is actually making people not just less informed, but harder to later inform. John Moser (talk) 13:11, 6 June 2021 (UTC)

• Oppose I think the current set up adequately explains the terminology. Reywas92^Talk 18:47, 6 June 2021 (UTC)

• It "adequately explains the terminology" in the same way that redirecting United States Democratic party to Communist Party of the United States would adequately explain the names of US political parties. My assertion is that the terminology "Ranked Choice Voting" does not mean "Instant Runoff Voting," and the misuse has been popularized to further the political gains of specific interested parties. It has of course become common usage: the propaganda campaign was successful. If an American media company routinely referred to the US Democratic Party as a branch of the Communist Party, or the US Republican Party as the rebirth of the American Nazi Party, so much so that the common usage of "Nazi Party" or "Communist Party" was to mean these parties, would Wikipedia then reinforce that by redirecting Nazi Party to United States Republican Party with a small-font note at the top that maybe the user meant the German WW2 regime? John Moser (talk) 02:22, 10 June 2021 (UTC)

  This hypothetical comparison is childish and irrelevant. Reywas92^Talk 03:01, 10 June 2021 (UTC)

• Oppose per common use of "ranked choice voting" to really mean "instant runoff voting" (see: any news article, or just do a google search for "NYC mayoral election ranked choice"). It would be misleading for that to redirect to ranked voting in general. SnowFire (talk) 21:02, 6 June 2021 (UTC)

• That's how "Ranked Choice voting" got to be the term in general: it was a general term for…ranked choice voting…and an organization decided their opposition to other forms of ranked choice voting would be better served by using the term to mean specifically instant runoff voting, i.e. that it would be best to use the term in a misleading manner until everyone else started using it in a misleading manner. That it is now hard to explain to Americans who have become sufficiently aware of this campaign the basic principle of voting on a ranked ballot and counting the votes in any way that is not equivalent to instant runoff voting, it appears to me they have been successful. Your argument is that the success of a misleading campaign legitimizes propaganda and requires Wikipedia to act to further that propaganda. Again, I refer you to FairVote's 2005 form 990 and the statement, "In contrast, IRV elects candidates with majority support [...] this ranked choice voting method saves taxpayers money," and so forth. It was FairVote who, in 2010, popularized "ranked choice voting" to mean IRV. They aggressively campaign against Condorcet methods in particular, examples here (still calling it IRV), here (differentiating RCV from Condorcet), here (calling Condorcet an alternative to RCV, but still using the IRV term in the text), subtle renaming approach here (directly stating RCV is synonymous with IRV and STV). The very source that told America RCV means IRV first told the IRS that IRV was a type of RCV, then told everyone RCV was IRV and nothing else. If your argument is that Wikipedia should support and aid successful propaganda that reduces clarity of understanding in favor of a specific goal of some group's political campaign, then say so. John Moser (talk) 02:12, 10 June 2021 (UTC)

* It’s worth adding that the meaning of terminology in a technical field (in this case voting theory) is determined by the relevant technical literature, not by news reports. There are well known biases in taking Google listings as evidence. In fact even in academic writing IRV seems to be the primary meaning of RCV, but only through a sort of implicit quotation. There are lots of studies of current affairs topics (such as the ‘Maine RCV experiment’): obviously these use the terminology of the public discussion, and would do so even if it was acknowledged as incorrect. In proportion as you focus on reliable sources using the term ‘ranked choice voting’ as a technical concept the generic meaning becomes commoner, and in my impression ultimately predominant. I can give several examples. But it shouldn’t need hours of discussion to establish the fact that ‘ranked choice voting’ can mean ‘voting by ranked choices’. Colin.champion (talk) 08:10, 10 June 2021 (UTC)

• Very oppose Realistically, the term 'Ranked-Choice voting' is only used in The United States politics since the 2000s, not anywhere in your planet, as either referring to Instant-runoff voting or Single-Transferable Vote, but it used only the former as interchangeable word, more so, than both of them equally. Chad The Goatman (talk) 00:15, 7 June 2021 (UTC)

  Here’s an example of the generic use of the expression. Colin.champion (talk) 07:11, 7 June 2021 (UTC)

• Qualified support. I haven’t formally voted yet. I think the current redirect is an indefensible violation of neutrality; I’m not certain what is the best fix. As a non-American I was completely unaware that the term RCV had been hijacked for promotional purposes; as an occasional Wikipedia editor who understands the words ranked, choice and voting I could easily have used the term without realising that Wikipedia itself would consider me to be referring to IRV. In fact I have used the exact synonym ‘ranked preference voting’ generically. A Google search shows that serious authors use ‘ranked preference voting’ generically and newspapers misunderstand it as a synonym for IRV. It would only need a concerted promotional campaign for that expression too to be hijacked under iffy’s principles. Colin.champion (talk) 09:33, 7 June 2021 (UTC)
• Comment: While there is some good stuff there, our coverage of voting systems and their attempts to avoid problems such as the Condorcet paradox is appalling. And it is a difficult subject! We all have our favourites. (And whenever someone appeals to Condorcet as the reason they are losing, you can be quite sure that they know that they have lost fair and square. We even had an example in Wikipedia.) Serious work needed. Andrewa (talk) 06:36, 10 June 2021 (UTC)

References

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Disambiguating 'Ranked choice voting'

bluefoxicy recently proposed that ‘Ranked choice voting’ should be the name of this page rather than of IRV, with ‘ranked voting’ redirecting to it. The proposal was rejected leaving open the possibility of other changes. For my part I found the discussion unsatisfying: for one thing, there wasn’t much input from editors familiar with the voting theory literature. It was also rather polarised with no compromise options being considered.

My preferred meaning for ‘ranked choice voting’ is the generic one (ie. ‘ranked voting’), and this is for two reasons. Firstly, I believe it to be the commoner meaning in the technical literature (which is what Wikipedia should be guided by); and secondly it is an honest use of language, interpreting ‘ranked choice voting’ as meaning ‘voting by ranked choices’. At the same time the alternative meaning as IRV is entrenched in North American political discourse, and Wikipedia does not sweep such uses under the carpet: to do so would be as much a breach of neutrality as is the deliberate conflation of two concepts. For this reason it seems to me that the redirect should be replaced by a disambiguation page. Wikipedia guidance explicitly says that ‘a name that could refer to several different articles may require disambiguation’.

There are advantages to this approach. It cannot be suspected of trying to hoodwink the reader. Redirecting to IRV may look like trying to reinforce the untruthful implication that ranked voting is coextensive with IRV; but bluefoxicy’s proposal may look like Wikipedia is trying to divert readers interested in American IRV proposals to the broader topic.

Moreover there will be a sizeable class of readers who do not attach a precise meaning to ‘ranked choice voting’ or who have not clearly distinguished the specific from the generic sense. These are the readers who will be materially affected by the decision taken, and they are positively misled by either of the polarised options.

A recent reference to ‘ranked choice voting’ is in a paper by Steven Heilman (marked ‘to appear, Notices of the AMS’). My impression is that the phrase barely counts as a technical term: authors such as Heilman simply put the words ranked, choice and voting together in their normal meanings without necesarily being guided by other authors’ usage and often without feeling any need to explain the term (or capitalise it). For this reason it doesn’t stick in one’s mind; and Google is pretty useless because it swamps the reader with politicised articles. But when I sample the literature, it is the generic sense which I most often encounter. Colin.champion (talk) 10:28, 17 June 2021 (UTC)

I have done this. Colin.champion (talk) 07:36, 24 June 2021 (UTC)

Moved {{Electoral systems}} to the end

I just moved {{Electoral systems}} to the end.

How many readers want to wade through a page of material like that before they get to the first paragraph of the article?

They may only want the first paragraph of the article, and having templates like that at the top, I think, makes Wikipedia articles less inviting and harder to use. People who want more context expect, I think, to look for that kind of information at or near the end. Maybe before the reference but not before the lede, I think. Comments? DavidMCEddy (talk) 19:02, 1 July 2021 (UTC)

But these templates are always at the top. For most people they don’t require “wading through a page of material” – are you using a device with a small screen? Maybe there’s something wrong with the way Wikipedia renders on your device. But I don’t see what these templates add either. Colin.champion (talk) 08:03, 5 July 2021 (UTC)

I'm using Firefox and Safari under macOS 11.4.

Did anything change for you in the display when I moved {{Electoral systems}} to the end?

I started noticing it recently in several different articles. It really gets in the way of using Wikipedia in a way that seems sensible to me. DavidMCEddy (talk) 09:46, 5 July 2021 (UTC)

Yes, but it’s just a small panel in a wide window and it didn’t get in the way. Colin.champion (talk) 09:51, 5 July 2021 (UTC)

Two more questions:

1. Is it a problem for you that I moved the {{Electoral systems}} template to the end?
2. Do you have a suggestion about where I should complain? I've seen it in several Wikipedia articles, and I find it a problem. It's like the front doors to your car were welded shut, and you had to climb in through the trunk to get to the driver's seat ;-) DavidMCEddy (talk) 13:22, 5 July 2021 (UTC)

[image deleted] David – I’ve uploaded a screenshot from my old and mediocre laptop (I’ll delete it in a few days). There’s enough space for text where the template occurs, and the lines might be so long without it that the eye would get lost scanning them. So I’m not sure why you’re worried. Personally I don’t care, but it’s a bit inconsistent modifying individual pages to fix what you consider to be a fault in the whole of Wikipedia. If you want to raise the matter somewhere official you could start from the Village pump, but I’ve never posted there myself and I have no idea how helpful it is. Colin.champion (talk) 19:56, 5 July 2021 (UTC)

Rewrite of list of RV methods

I rewrote the summary of the main ranked voting methods. I found the emphasis in the previous version lop-sided, with a lot of space given to ranked pairs and IRV, very little to minimax, and none to Copeland’s or Coombs’ methods. The list now gives methods a level of attention consistent with their support in Laslier’s workshop write-up.

STV receives less attention than it merits. I have no objection if anyone wants to amplify it, but the list is mostly comparative and STV is the only multiwinner system in it, so it doesn’t belong to the comparison. The article on multiwinner voting systems is the right place for an introductory mention of it.

Another thing I found odd was the grouping of methods according to whether they satisfied a particular criterion (Condorcet).

I cut out some of the lead para, which was still a little over-long and contained material that had very little to do with its subject matter. Please comment if I’ve done anything you don’t like. There are some other changes which I have in mind, but which might be more controversial and which I won’t put into effect without proposing them here first. Colin.champion (talk) 15:59, 29 July 2021 (UTC)