Talk:Schulze STV/Archive 1

Archive 1

Advertisement?

I don't agree to Tdh5609's opinion that this article was "written like an advertisement". The problem rather is that there are no established criteria for STV methods. Therefore, it is not possible to discuss the "merits" of the different STV methods. The only thing, that we could do, is to describe the underlying heuristic of an STV method; however, an election method is always the best possible election method according to its own underlying heuristic. Markus Schulze 23:32, 11 November 2008 (UTC)

When I wrote the original version of the article, my main objective was to provide info about the method. Many people were unclear about how it was supposed to operate. I feel that giving a description of the comparison method used and an example showing the resistance to vote management (which is one of the main features of the method) is reasonable. This resistance fails if needed to meet the Droop criteron, and I guess an additional example could be added showing this (but I felt that they were complex enough already).

In effect, the article says "This is how the method works" and "This is what the method is supposed to do".

The practical issues associated with its complexity are covered in the article.

I based the article on the CPO-STV one, and I don't think they are much different.

--Ivnryn (talk) 11:20, 18 November 2008 (UTC)

I removed the advert tag. This article didn't seem like it could meet any of the five possible forms of soapboxing at Wikipedia:NOT#SOAPBOX. Ivnryn is not MarkusSchulze, so it is not self-promotion, and there really is no case for this being advocacy. The other three obviously don't apply. The article itself is mostly a description of the method, and the discussion covers both the positive aspects (resistance to Vote Management) and drawbacks (complexity). If someone feels this was wrong, put the tag back but please give reasons why you believe it qualifies. -- Wingedsubmariner (talk) 05:18, 19 November 2008 (UTC)

Voting system criteria

Other than the Weak Invulnerability to Hylland Free Riding criterion, Does Schulze STV pass all the same voting system criteria as traditional STV? --Explodicle (T/C) 18:15, 23 December 2008 (UTC)

Well, the only established criterion for STV methods is Dummett's proportionality criterion for solid coalitions. And yes, the Schulze STV method satisfies this criterion. Markus Schulze 19:45, 23 December 2008 (UTC)

Hylland Free Riding

Given the importance placed on defeating this kind of free riding in justification of the method, I think there should be a directly linked explanation of HFR and it should be explained how SSTV defeats it. Tesspub (talk) 10:52, 4 May 2011 (UTC)

Deletion

ThurnerRupert recommends that the Schulze STV article should be deleted. He writes: "It is not a bad article, but it smells somehow bad. article from schulze, about schulze, with references to schulze. in general it seems that schulze enjoys to put his own name all over the place in wikipedia. he does this in multiple language versions, in german as well. but that is not forbidden, isn't it?" Actually, this article has been created by Ivnryn. Anyway, a conflict of interest is not a reason for deletion. Markus Schulze 06:36, 2 July 2011 (UTC)

multiple policies seem to be violated, WP:NN, Wikipedia:OR, WP:SELFCITE, WP:SELFPROMOTE just to name a few. the general editing pattern is "any person besides markus schulze is not tolerated around something markus schulze is involved in", and "anything markus schulze is involved in has to bear his name". but this one is not black and white, at least imo, as we can see in a small part of markus behaviour in general:

• markus helps with tideman
• admitting "somehow" that others work is used and important, e.g. also tideman's clone proof criteria
• at the same time not tolerating tideman's names as influencing the schulze method
• and even get into an edit war about his own name, or does not tolerate schulze method called with another name
• removes deletion discussion before finished

--ThurnerRupert (talk) 10:42, 2 July 2011 (UTC)

This article should not be deleted. I agree that there are issues about the name of the method/article, and Schulze has edited other articles in biased ways that attempt to make "his" methods seem better. Those issues are worth considering. However, this article should not be deleted. I'll add that Schulze is an expert about voting methods and has been a very helpful contributor to this and other voting-method articles. Any resolution should take his helpfulness into account. VoteFair (talk) 18:16, 6 July 2011 (UTC)

is there any evidence about schulze beeing helpful with articles? is there evidence that this article is notable and/or relevant? --ThurnerRupert (talk) 20:56, 14 July 2011 (UTC)

I do not like the Schulze STV method (it's unnecessarily complex), and it competes with an earlier and better method (VoteFair representation ranking, which I created), yet I am defending this article from being removed because the method has become sufficiently "notable" and relevant. Specifically the method has been discussed elsewhere among voting-method experts, and it has been implemented in open-source software, so its removal would leave us without a clear description of the method.

Of course Wikipedia requires references that prove the method to be notable and/or relevant, and that is easy for Schulze to manufacture because (as you point out) he has connections with Tideman, and Tideman is the editor of an academic publication about voting methods. Schulze is (I believe) an academic professor, so he can create more such publications if the ones listed are not sufficient.

Of course, just relying on academic publications alone is not enough. The most humorous example of this issue (that I've seen) occurred in a Wikipedia article about the American Idol TV show when a Wikipedia editor requested a reference as evidence of a statement that the host made during the show, even though recordings of the show (and the memories of the people who watched it) provided proof that the host made the statement. (Alas, I couldn't find the Wikipedia edit because the American Idol articles are edited so frequently.) Less humorously, but more meaningfully, I regularly see Wikipedia editors asking for references that support facts about some notable software applications/projects and programming languages, even though Wikipedia is the only place such facts are documented beyond the documentation supplied by the software maker (which Wikipedia editors regard as a biased source). Now that computers are available to handle the calculations required by fairer voting methods, many Wikipedia articles about voting methods would be outdated and irrelevant if academic publications were the only source of information about them. In fact, the Wikimedia Foundation now officially uses a voting method that was created just a few years ago (the "Schulze method"), and the article about that method appeared in Wikipedia before the method became indisputably notable.

This brings me to an important point. Detailed peer review takes place on Wikipedia itself. Previously the only place peer reviewing was done was in printed academic publications (hence their long-standing importance). Now, early discussions of voting methods appear in forums such as the "Election Methods" mailing list. Flawed voting methods do not survive those discussions. Voting methods that do survive such peer review are allowed to be added to Wikipedia (provided no one acts as a dictator in creating the article), and I think this article has survived that peer-review process.

This brings me to the next important point, which is that experts about voting methods are monitoring the voting-method articles (if they are listed as such), and the Schulze STV method has been discussed (elsewhere) and it would have been heavily criticized if it had significant flaws in fairness or rigorousness. (Its flaw of being unnecessarily complex is not a major impediment from the perspective of election-method experts, and is just an issue when it comes to understanding the method.)

Yes, when this article was first created, the method was not sufficiently "notable," but that has changed. Yes, Schulze often aggressively edits voting-method articles without sufficient discussion on the discussion page, and he has previously edited articles in a dictatorial way, but those are different issues. Yes, when Schulze names things after himself it disrespects the honor that is given when something is named after someone else, but that too is a separate issue. In terms of deleting this article, I am strongly in favor of keeping it. The method itself will survive or not, just as software applications, medical treatments, and TV shows that are popular at one time may not be popular at a later time, but even then information about them is useful for historical perspective. (I doubt that "plain" STV will still be in use two decades from now, yet it certainly deserves a Wikipedia article about it.) VoteFair (talk) 18:57, 16 July 2011 (UTC)

what link proves that it survived the peer review on electorama, and the article does not only contain markus own links? with a trustworthy source, i'd be fine to keep the article if somebody else knowledgeable in the topic but schulze favors it. --ThurnerRupert (talk) 09:34, 19 July 2011 (UTC)

To Markus Schulze: You can find the requested links much faster than I can. If you post them here then I or someone else can verify that they meet the requested criteria. (And the next time you get the urge to edit any Wikipedia article in a way that is dictatorial, please consider that the underlying purpose of any good voting method is to increase collaboration.) VoteFair (talk) 18:54, 20 July 2011 (UTC)

The "Schulze STV" article has been created by Raph Frank [1]. My paper Free Riding has been quoted by at least 4 other papers ([2]; [3]; [4]; [5]). Markus Schulze 20:09, 25 July 2011 (UTC)

I can verify that Raph Frank (who created this Wikipedia article) is a frequent and well-respected contributor to the Election Methods discussions, and his opinions do not suggest any clear bias in favor of Markus Schulze. The publication written (in German) by Martin Wilke refers to Schulze STV without apparent criticism, and that article was peer-reviewed prior to publication. These two supplied references, plus other references to Schulze STV that I've seen in the Election Methods discussions, without significant disapprovals of the method, plus the lack of any election-method expert expressing disapproval of this article or method here on this discussion page, provide evidence that the Schulze STV method is notable enough for inclusion in Wikipedia. (The other three other-author references provided by Schulze, and his article that he refers to here, do not refer to the Schulze STV method, so they are not relevant in this discussion about deletion.)

Note that this discussion about not deleting this Wikipedia article is not related to the above-listed concerns about the self-promotional naming of the method, and does not address accusations that Markus Schulze has participated in edit wars and has violated Wikipedia rules (such his removal of the discussion about deletion before the discussion was finished). Those are separate issues. My goal is to verify that the Schulze STV method (although it is unnecessarily complex) is notable among election-method experts. VoteFair (talk) 18:01, 29 July 2011 (UTC)

There seems to be some misunderstanding. ThurnerRupert nominated this article for speedy deletion by adding the {{delete}} template. I challenged this nomination by removing this template. This template says: "If this article does not meet the criteria for speedy deletion, please remove this notice." If ThurnerRupert wants to nominate this article for deletion, he has to add the {{subst:afd1}} template. See: Wikipedia:Articles for deletion. Markus Schulze 19:52, 29 July 2011 (UTC)

Suggest adding History section

Based on the editor's comment the removed sentence (about who invented this method) probably would be acceptable to them if it were moved to a "History" section. The editor's comment implies that it just isn't appropriate in the "Summary" section. And I agree; people don't want to read historical information before the topic itself has been described. VoteFair (talk) 18:25, 4 October 2011 (UTC)

Algorithm description

The description of the algorithm is different from the description that Schulze gives in his paper schulze2.pdf, there he talks about possible winning sets and the strongest path through a graph, and here it's described as moving votes, for example "51 prefer Both to Brad (assign 19.5 to Andrea and 31.5 to Carter)". It might be that this are just two different ways to describe the same algorithm? But it's not very clear to me that this is the case. Capitol (talk) 21:27, 24 June 2013 (UTC)

I was also confused about this apparent discrepancy, but looking at Schulze's paper, I believe the two are referring to the same thing. I'm no expert on anything vaguely relevant to this discussion, but p. 37 appears to describe a calculation where voter's preferences are distributed to maximise the number of votes assigned to the candidate with the fewest votes:

$N[{a_{1},...,a_{M}},b]$ ... is the largest value such that there is a $t$ ... such that ... $\sum _{i=1}^{N_{W}}t_{ij}\geq N[{a_{1},...,a_{M}},b]$

where $\sum _{i=1}^{N_{W}}t_{ij}$ is the number of votes assigned to each candidate. Obviously, this can be achieved by (if possible) assigning the same number of votes to all candidates being considered, as the article states. --RunasSudo (talk) 23:28, 3 September 2015 (UTC)