User talk:MarkusSchulze

From Wikipedia, the free encyclopedia

User:MarkusSchulze

Though I don't think much of IRV as a voting system I agree that IRV is full of POV. I despair of fixing it at the moment but I certainly encourage you to be bold in updating the page to fix the problems with it if the mood takes you. ciphergoth 00:42, 2004 Nov 25 (UTC)

Markus, Donald Saari says hello and apologizes for not being able to lecture at your university while he was in Germany. He stated that he was very pressed for time but would like to in the future.--Fahrenheit451 20:38, 6 Jun 2005 (UTC)

Contents 1 Inclusion of criteria 2 Wikipedia_talk:How to hold a consensus vote 3 Secret voting system page 4 CMU GSA voting method 5 Your conflict with Dissident 6 request for comment - not in wikispeak sense of the term :) 7 Extentions to Arrow's theorem 8 Table on K-Y

[edit] Inclusion of criteria

I haven't been able to judge the notability of criterion articles like [the favorite betrayal criterion, weak/strong defensive strategy, and (generalized) strategy free. Do you think they are not notable enough for Wikipedia? I always suspected that Favorite Betrayal was a criterion designed specifically to promote Approval, but at least I've seen it in multiple sources. I also didn't want to go on a VfD rampage across the voting systems project, because I'd rather improve the articles than delete them.

Are there any criteria (besides the consistency criterion, which I think clearly needs to be added) that don't have articles, but you think should?

I still believe that Majority Choice Approval shouldn't be a Wikipedia article. I can understand why this is a gray area - election methods are an important topic, and that topic has lots of subtopics which are important to understand it, and each of those subtopics has important subtopics of their own - but I think once you get to MCA it can't be anything but original research.

RSpeer 14:29, Apr 14, 2005 (UTC)

In my opinion, Weak defensive strategy criterion, Strong defensive strategy criterion, Strategy-Free criterion, Generalized Strategy-Free criterion, and Favorite Betrayal criterion should be deleted because of the following reasons:

• Original research, few Google hits. These terms are only used by Mike Ossipoff.

• Misleading. For example, if I have understood SFC and GSFC correctly, then they say:

Strategy-Free criterion:

Suppose (1) d[A,B] > V/2 and (2) the partial individual rankings can be completed in such a manner that candidate A is a Condorcet candidate. Then candidate B must be elected with zero probability.

Generalized Strategy-Free criterion:

Suppose (1) d[A,B] > V/2 and (2) the partial individual rankings can be completed in such a manner that candidate A is in the Schwartz set set and candidate B is not in the Schwartz set. Then candidate B must be elected with zero probability.

In other words, SFC says that the Condorcet candidate is protected from truncation and GSFC says that the Schwartz set is protected from truncation. But then these criteria should be called "invulnerability to truncation" but not "strategy-free".

• Undefined. Several people have asked whether these terms are well defined. See e.g. here. Also I doubt that these terms are well defined. For example: (1) I have asked Mike Ossipoff several times to define his criteria in terms of cast preferences; but he always refused to do so. (2) I have asked Mike Ossipoff several times to demonstrate whether Cloneproof Schwartz Sequential Dropping or Ranked Pairs satisfy the Favorite Betrayal criterion; but he always refused to do so.

Markus Schulze

They're Mike's fault? That explains a lot. He did a lot of damage to these articles.

I agree that we should VfD them. I'll start by removing them from the voting system article.

RSpeer 01:36, Apr 16, 2005 (UTC)

How do you feel about later-no-harm? It currently doesn't have an article, but has been popping its head up in a few of the method articles (such as Schulze method). I think that if it's not original research, maybe we should include it. - McCart42 (talk) 05:28, 16 July 2005 (UTC)

[edit] Wikipedia_talk:How to hold a consensus vote

Would appreciate any insight/corrections you might have in respect to the discussion being had here principally between RSpeer and myself. Whig 12:05, 25 May 2005 (UTC)

Dear Whig, in 2003 the Debian project adopted Cloneproof Schwartz Sequential Dropping. The discussion took more than 3 years (2000--2003), because the Debian project wanted to use the same method for all internal decisions (elections, referendums, decision-making in committees). Therefore, I suggest that we should discuss, in connection with the upcoming elections to the Arbitration Committee of the Wikipedia project, which method is the most suitable. And then this method should be used for all internal decisions of the Wikipedia project. In my opinion, everything else would be arbitrary. MarkusSchulze 00:28, 27 May 2005 (UTC)

Markus, in a recent survey conducted here using CSSD, it was briefly considered to use the Debian modification of your approach, which compares each option to the default ("None of the Above" or "Further Discussion") and requiring a N:1 ratio for passage. It was determined in the course of this survey that this method is subject to tactical consensus blocking through deceptive voting preferences. I strongly advocate that Debian's modification not be used. Whig 05:52, 27 May 2005 (UTC)

Clarifying my above statement. I do not advocate against having a default option. I think that N must be set as 1, however, i.e., that no multiplier should be given to the default option as is done in Debian for votes other than for Debian Project Leader elections. Leaving N=1 and having a default option allows for truly unacceptable options to be removed from consideration altogether, and this is desirable. Given the Wikipedia project's desire to obtain better-than-majority support for decisions, CSSD might best be considered as a method for "pruning" the acceptable options and then subjecting the winner(s) (more than one only in case of tied result) to consensus Approval. Whig 06:03, 27 May 2005 (UTC)

[edit] Secret voting system page

Markus,

If you're interested, RSpeer set up a deliberation page at Talk:Voting_system/Included_methods_and_criteria. So far only he, Fahrenheit, myself, and Hermitage seem aware of it. KVenzke 21:02, Jun 21, 2005 (UTC)

I don't see why I should have to convince RSpeer when I want to edit the "voting system" article. Markus Schulze 22 Jun 2005

Well, he started it after Fahrenheit decided to try to delete clone independence from Wikipedia. So he was trying to reduce time spent reverting. I don't mean to imply that you should pay attention to what others say on this page. KVenzke 13:57, Jun 22, 2005 (UTC)

[edit] CMU GSA voting method

Thanks for the info. I'll try to talk to some of the people in GSA about changing the vote-counting method to a better form of Condorcet, or at least about why they chose MinMax. - McCart42 (talk) 14:29:58, 2005-07-28 (UTC)

[edit] Your conflict with Dissident

I think you may be escalating your conflict with Dissident by referring to his comments on your talk page as "vandalism". Of course you are free to ignore him, but "vandalism" is a serious charge that doesn't apply to what he's doing. Please remember to assume good faith. rspeer / ɹəədsɹ 00:23, 2 March 2006 (UTC)

Dear RSpeer, I consider Dissident's recent edits to be pure vandalism. What I criticize is:

1. He spams Wikipedia with original research. He starts claiming that this method satisfies this criterion and that that method violates this criterion. And when someone asks for explanations, Dissident claims that it was obvious that he was correct and that all those who don't see that were hypocritical, dishonest, and Mike Ossipoff devotees.
2. Even when he admits an error, he doesn't correct this error and he keeps on insulting those people who pointed him to this error. For example: At talk:Schulze method he admitted that there is no connection between the defeat strength measure on the one side and compliance with the sincere expectation criterion on the other side. Nevertheless, he corrected this error neither at the Condorcet method article (where I corrected this meanwhile) nor at the sincere expectation criterion article.
3. He misuses Wikipedia as a mailing list. Many of his edits contain only opinions and insults. But Wikipedia is an encyclopedia and not a mailing list. When he wants to share his political opinions with other people, then he should do this outside Wikipedia.
4. To circumvent the probable deletion of the sincere expectation criterion article, he posts comments that belong to talk:sincere expectation criterion to talk:Schulze method.
5. When he nominates an article for deletion, he removes links from other articles to this article to make the readers mistakenly believe that there was no connection between the topic of this article and those of other articles. For example: To make the summability criterion look less wide spread, he deleted 10 links from other articles to the summability criterion article.

Markus Schulze 09:44, 2 March 2006 (UTC)

You've described a content dispute with occasional incivility, not vandalism. The sincere expectation criterion is going away, solving many of the problems you name. Removing links to an article on AfD is actually rather common, and should be done if it's likely to be concluded that the linked article is original research. It may be controversial in this case; you should mention on the AfD for Summability that you feel it is important to keep.

Dissident's argument style wouldn't affect you so much if you didn't bother arguing with him. If he makes a personal attack on you, you can respond with the {{npa}} template and leave it at that. He makes convoluted arguments because he knows you're listening.

You should try, at every step possible, to de-escalate the conflict. Calling a content dispute "vandalism" is not at all the way to do that. Try not responding to Dissident's arguments, and calmly using Wikipedia processes like WP:AFD if he puts something in article space that you believe is inappropriate. I doubt he will continue to argue with you if you don't hold up your end of the argument, and you will come out looking much better. rspeer / ɹəədsɹ 15:12, 2 March 2006 (UTC)

Dear RSpeer, thank you for your suggestions. However, how can I make sure that, when the sincere expectation criterion article has been deleted, Dissident won't misuse talk:Schulze method to continue talk:sincere expectation criterion? Do you think that, when the sincere expectation criterion article has been deleted, it will be appropriate simply to delete that section from talk:Schulze method? Markus Schulze 18:51, 2 March 2006 (UTC)

[edit] request for comment - not in wikispeak sense of the term :)

Hi! I wanted to ask you if you would be willing to take a look and possibly comment on a comment I wrote on the Talk:Sainte-Laguë_method. It would argue using two examples that Hare + Sainte-Lague would give better proportionality. (And I guess by implication, would make QPQ-variants the best prefferential ballot no-party list proportional method, adding to it computational cheapness in comparison with Meek or Warren STV). --Aryah 10:56, 19 July 2006 (UTC)

[edit] Extentions to Arrow's theorem

Hi. I know you've been involved in many of the voting/social choice theory articles (heck, I even read your A New Monotonic and Clone-Independent Single-Winner Election Method), so I though I'd ask your opinion on my mockup of a possible section on Arrow's Theorem. It's barebones at the moment (just quick desciptions and references), but it gets the idea across. In particular, I'm curious if you know of any other results or an earlier source for my last point (Taylor wasn't the first one to come up with these, I'm sure). Also, do you think this should be a section when it's done, an article, or something else?

I'd appreciate any feedback on this -- here, on my talk page, or on that page's talk. Thanks. CRGreathouse (talk | contribs) 16:00, 24 August 2006 (UTC)

[edit] Table on K-Y

If you want to revert the table to its original form, I have no particular objections. I edited it only because I figure whatever form it ends up in, I'd like to have polished.

Overall I lean toward putting the table in a form closer to the original to keep it in line with other articles. If a reasonable number of the criteria can be true or false without "circular ambiguities" (properly defined) for arbitrary Condorcet methods, then perhaps the best solution would be to instead make other articles like this one. Thoughts?

CRGreathouse (t | c) 23:13, 14 February 2007 (UTC)