User talk:Jimbo Wales

(Manual archive list)

Another good example of bad behaviour on Commons

On 12 July 2011, a brand new Commons account uploaded 7 images to Commons. Five of them were images of dildos inserted into a man's anus (two of these have been deleted for quality reasons, although the quality of the remaining ones is equally poor). Each of these were nominated for deletion.

• file:Anal.jpg - deleted "Low quality, blurry. Not realistically useful for an educational purpose"
• file:Anal2.jpg - kept (four deletion requests) "While blurry, is educationally distinct from other photos in Category:Sexual penetrative use of dildos"
• file:Anal3.jpg - kept (three deletion requests)
• file:Anal4.jpg - deleted "very low-quality"

The other three images uploaded by this user show a man having sex with a blow-up doll.

• file:Puppensex 1.jpg - kept (three deletion requests) (This image was actually deleted in the second deletion request as "out of scope" but un-deleted.)
• file:Puppensex 2.jpg - deleted (but restored) (Deleted as "COM:SCOPE, very low quality", then un-deleted.)
• file:Puppensex 3.jpg - deleted (but restored) (Deleted as "COM:SCOPE, very low quality", then un-deleted.)

Note that none of the images uploaded by this user are currently in use in any article or userpage on any WMF project. The account has made no other edits on any other WMF project. It seems clear that this is someone using Commons for to satisfy their own exhibitionism. The quality of these images is such that if they were illustrating anything else, they would be deleted without discussion, but images of a sexual nature are notoriously difficult to delete. Delicious carbuncle (talk) 04:09, 16 April 2012 (UTC)

I'll say one thing - I don't recall actually seeing some guy screwing a doll like that before, so there is indeed something vaguely educational about it. Question: Are you guys going to reargue every Commons close on a sexual-themed file here from now on, or is someone going to close the curtain on this sorry scene? Wnt (talk) 05:15, 16 April 2012 (UTC)

When Commons cleans up their act, starting with banning out the Commons Is Not Censored XfD closers and ending with a rational relationship between content and the project's educational mission, then "we guys" will stop beefing. Until then, Commons is the gift that keeps on giving. Carrite (talk) 06:03, 16 April 2012 (UTC)

Well, they can ban their good admins who uphold core project principles, sure. Or they could ban the people who want to undermine their project, destroy its content, scare away editors with threats of outing and IRL harassment or just drive them away in disgust, revoke its tax exemption, slander it in the press, report it to police on unsubstantiated charges, split it, disband it, and destroy it. Some people here expect WMF to be run like a company, and if there's one thing that every company in the whole wide world agrees on, it's that you fire whistleblowers, no matter what the legality or the cost, and blackball them ever after. That goes even when their warnings were justified and could have saved lives, let alone when they are driven by some kind of ideological hatred for what the company seeks to accomplish. Wnt (talk) 06:26, 16 April 2012 (UTC)

I think what most of the so called critics here want is for Commons to be run in a responsible manner. Kevin (talk) 06:43, 16 April 2012 (UTC)

Could someone consider creating /sex chat or take similar action, so that those that seem to get a huge kick of using Jimbo's user talk page as a soapbox to repeatedly shout rude words can have a friendly space to take their diatribes? The repetitive and manipulative soapboxing of sexual topics here (mainly driven by "friends of Greg Kohs") has not only dominated this page for a week but appears to be a childish attempt to get as offensive as possible in an attempt disrupt, rather than enable any meaningful discussion of the issues that might encourage collaborative improvement. --Fæ (talk) 07:34, 16 April 2012 (UTC)

I'm sure that people are on their best behaviour here. But this has gone on for far too long, for people to be pussy footing about over the issue. The language should be robust enough to a) convey the nature of the problem, and b) expose the ridiculousness of some of the things that have happened in recent months. You will agree that it seems to have got worse since the WMF backed off on the image filter issue, you got reverted yourself over the 'toothbrush'. Quietly waiting for the right thing to be done hasn't worked, it has simply emboldened certain behaviours. Such that now there is a real problem with "Human male" in the searches. John lilburne (talk) 08:45, 16 April 2012 (UTC)

Thanks, Fæ - you made me realize that I missed two images. Two that you voted to keep, as it happens. Greg Kohs, on the other hand, was not involved in any of the deletion discussions. Delicious carbuncle (talk) 11:53, 16 April 2012 (UTC)

Addendum: I see now that I missed two files due to the similarity of their names to other uploads.

• file:Anal 2.jpg - kept (three deletion requests) see below
• file:Anal 3.jpg - kept (four deletion requests) "Commons is not censored"

It is worth quoting full the original closure of one of the deletion requests.

Kept: Pornographic content is not a reason for deletion. Human sexuality is a valid topic of educational enquiry. I accept this is a truly, truly awful photo, but we really don't have anything similar which is better. If we had a range of images showing anal toy use in high resolution that are in focus I would happily delete this - I'm all for getting rid of bad images when we have suitable replacements - but we don't. -mattbuck (Talk) 02:16, 1 August 2011 (UTC)

I am not bringing this up on Jimbo's page as an excuse to discuss someone's choice of masturbatory practices or to use naughty words like dildo and anus on Jimbo's talkpage. I think the discussions about Commons here are well-served by relevant examples. These are quite simply poor quality images, as even those who wish to keep them admit, so the rationale for keeping them is ideological. Note that some of these images have open deletion discussions which were started in December. Delicious carbuncle (talk) 11:46, 16 April 2012 (UTC)

What is that about 'WMF has backed off the image filter issue' alluded to a bit above? I strongly support providing optional filters. Not that I think having an image filter system in place stops us needing to stick to the educational mission of Wikimedia. We don't even have good encouragement of attached citations yet for images so they can be used easier in articles. Perhaps a request for a citation for a notable topic that might reasonably use a picture might stop some uploads and a lack of conceivable use would make deletion more obviously an option. It isn't easy to deal with groups of editors who have grouped together into gangs with an in-group culture and point of view which is detrimental to the overall educational aims of Wikipedia. Dmcq (talk) 12:19, 16 April 2012 (UTC)

The image filter has been shelved indefinitely. Implementation was scheduled to begin in January, but two board members up for re-election have since said that they no longer support it, and Sue Gardner has referred people interested in status to their statements made on Foundation list, as reflecting current thinking. See [1] --JN466 12:29, 16 April 2012 (UTC)

• Also note Commons' "Hot Sex Barnstar", courtesy of Beta M: [2]. Wikilove, Commons-style. --JN466 12:36, 16 April 2012 (UTC)
  • Considering that there are over 10 million images on Commons, it's a credit to the search and categorization system that you manage to ferret out this stuff. Of course, I have no idea how many days work you have had to put in to find all these sexual photos from socks to sex toys, and then how many days you have invested in using so many forums to soapbox your anti-sex opinions. It seems a pity you did not take up the repeated open invitation to positively present your views at the Wikimedia UK board meeting or that you appear uninterested in helping to work on improving Commons or Wikipedia policies with regard to refining the definition of educational scope and quality of sex education images, and instead seem to prefer a strategy of random erratic lobbying and ranting; in the process encouraging well known stalkers and harassers of Wikimedians to play their games. I find it sad that you prefer to make a name for yourself by disrupting rather than joining those of us genuinely passionate and committed to the values to improve how the Wikimedia movement works and how to enforce meaningful policies across all projects. When you are tired of ranting into the void, you might want to reconsider how to use my offers of help in establishing meaningful change and gaining the wider support of the community. --Fæ (talk) 12:56, 16 April 2012 (UTC)
    • Did you not get one of those barnstars? I Noticed that Cirt had to rush around replacing them whilst the Beta M affair was in full swing, so someone on Commons knew quite quickly what the score was there. My advice is that you probably don't want to be seen as the one standing on the steps trying to beat back change for too long. John lilburne (talk) 13:31, 16 April 2012 (UTC)
      • Reflect on what I have already said. I would like Commons to mature as a community and for the Wikimedia projects to have credible governance that the Wikimedia movement can respect and support. I am an advocate for improvement, transparency and accountability. Completely. Random soapboxing (I include Jimbo's user talk page as "random") and ranting are ineffective methods of getting the the wider community to support a change in policy and then apply it. None of us is a subject of a Tsarist regime and accusing me of beating back change when I encourage it, is bizarre. You might want to consider your own activities, the fact that I fear to engage with you on-wiki due to your potential to make personal attacks against me off-wiki if you don't like my opinion, is hardly conducive to engaging others like me to engage with, or support, any initiative with your name, or the names of others in the travelling circus, stamped on it. --Fæ (talk) 13:52, 16 April 2012 (UTC)
        • Fæ, you are such an advocate of transparency and accountability that you recently threatened to stop posting to the Wikimedia UK mailing list if the archives were made public. The archives used to be public but for reasons unknown they were recently made private. The extent of your commitment to openness and transparency is self-evident. Delicious carbuncle (talk) 14:19, 16 April 2012 (UTC)
        • How long will it take for Commons to mature? It is incapable of doing so having sucked on the teat of NOTCENSORED for so long without really understanding that it doesn't mean accepting every dick shot that get uploaded, not applying common-sense with regards to the perceived age of the uploader, and stuffing categories, filenames, and descriptions on the same in the knowledge that it will spread the image wide and far across the project. Wider community support on this is impossible, and frankly not necessary, everyone knows that it is only a matter of time before the issues with Commons, and the issues with BLPs will cause a shitstorm external to wikipedia. Do you really encourage change? I'd have thought that you'd have realized the hurt that chatter on a website might inflict, and that the machine that is wikipedia is in fact now THE MAN. I was indeed surprised when you teamed up with pigs to offer support, and as I said then and say now "voices off stage will continue to manipulate you into espousing wrong headed opinions." It seems that you have been unable to do so and continue to rail against, "supporters of Kohs", rather than acting correctly. I swear I can see you supporting pigheaded propositions because "supporters of Kohs" are against it. John lilburne (talk) 19:24, 16 April 2012 (UTC)

Censoring Wikipedia Review links (I assume) with the "spam" blacklist is indeed a very bad idea, though not nearly the worst we've seen proposed here the past two weeks. Unfortunately the idea appears already to have been introduced with blacklisting of http://www.encyclopediadramatica·ch/ - it should be rejected entirely. That said, is anyone going to clue us in on how Greg Kohs has anything whatsoever to do with this discussion? I explained myself about the river crab above, so it's only fair. Wnt (talk) 05:56, 19 April 2012 (UTC)

I don't think the Kohs has anything to do with anything, you'll need to ask Fæ about that. My reference to "voices off" is a play on a title. Thanks for asking BTW as I'm sure that many wouldn't have picked it up. John lilburne (talk) 09:04, 19 April 2012 (UTC)

I should note I finally figured out who he was; to break WP:DENY more than a bit, it's a reporter at The Examiner (URL blacklisted by WMF). Who, yes, shows a very hostile attitude toward the inclusiveness of the project and its subject matter, and appears to be advancing it by waging tabloid journalism against ordinary people. (Most if not all of the Wikipedia search results for Greg Kohs are about the filmmaker honored at the 2008 Slamdance Film Festival, but these are two different people.^[3] Wnt (talk) 17:21, 20 April 2012 (UTC)

• • • Looking at a link from the user page, off wiki lobbying can be a problem, but it also fulfils a useful function. Saying they should just talk to some board isn't really a counter. What I found really worrying there though was very active support for a website that names particular editors on Wikipedia because of this sex pictures issue. That sounds like inciting harassment to me and incompatible with WP:CIVIL. Dmcq (talk) 13:34, 16 April 2012 (UTC)
    • Fæ, I have tried to work on this problem in the community for years. In 2010, I worked for half a year, with Dcoetzee and others, on drafting a sexual content policy in Commons, only to see the Commons community vote against having any policy whatsoever on sexual content. I similarly worked for months in Meta on brainstorming possible image filter designs, liaised with Robert Harris, had accusations lobbed at me in German Wikipedia that "I had had dinner with Sue Gardner in London" (WTF?), and now the image filter is dead. I truly have done what I could do in the community, Fæ. The result is that nothing has happened, that Commons is a bigger embarrassment than ever, and that Wikipedia remains the only major website in existence that freely mixes Pokemons with hardcore porn (including in Wikipedia mainspace), without offering any filtering ability, while repeatedly referring to its role in educating the little girl in Africa. What are we trying to teach her? How to become a porn actress and have sex with a dog? (Note discussion on Reddit.)
    • A culture where admins award each other this barnstar is an embarrassment. If I were a donor, I would not want to be seen dead supporting such a culture with my money. As for giving a presentation to the Wikimedia UK board, who is Wikimedia UK in Commons? You yourself got threatened with removal of your filemover rights by a Commons admin when you tried to fix one of these search problems, and were accused of "pushing POLA [the "Principle of Least Astonishment" endorsed by the Foundation board, specifically for Commons] onto Commons". To wit: when you, as a Wikimedia UK director, go to Commons to do what the Wikimedia Foundation has urged Commons contributors to do, you get threatened by a Commons admin and have to retreat. There is little you or Wikimedia UK can do about Commons, even if you wanted to. JN466 13:49, 16 April 2012 (UTC)
    • For those interested, many adult-media "Keep" decisions in Commons are really easy to find: [4] Many of these discussions and discussion outcomes are unmitigated disasters from the point of view of how an educational charity should conduct itself, especially given that it makes all of its content available unfiltered, and most of the stuff sails past school filters. I have nothing against porn, but very little of this stuff has a realistic educational rather than recreational use, much of it is unbridled exhibitionism, and presenting it like this is irresponsible and an abuse of donors' faith. Show me one educational website, let alone a charitable tax-exempt one, whose contributors upload images and videos of themselves wanking, and that hosts and curates comparable material in a comparable manner. JN466 15:03, 16 April 2012 (UTC)

The image filter is clearly a no-brainer - the foundations representatives needs to get back on track and get it implemented asap - seems like the only objection is that supporters/anonymous free speechers want everyone, underage or not, to be exposed to pornography whether readers like it or not. Youreallycan 17:15, 16 April 2012 (UTC)

My feeling exactly and I'd expand it to whole articles too. I would though always want it to be obvious when something is missing. If a person does not want to see images or read text defaming the flying spaghetti monster then there should be reasonably easy to see markers for the missing images or text making it obvious that content has been self censored. Dmcq (talk) 17:29, 16 April 2012 (UTC)

An image filter is not a good idea until after Commons has its misguided admins removed. If there were an image filter, any discussion about undue exhibitionism and other nonsense would be dismissed with "you don't have to see it". The WMF model is broken until it can recognize obvious obviousness: some people (at their current maturity level) believe that any editorial judgment is a human rights violation (aka NOTCENSORED), and no arguments will ever deflect them from exploiting Commons (see commons:Category:Santorum neologism-related images for yet more examples). The free-speech advocates are doing a great job, but they should do it on their own website. Johnuniq (talk) 01:54, 17 April 2012 (UTC)

This is absolutely true and has always been a consideration: that installation of an image filter might lead to a vast increase in adult media, under the rationale that anyone who doesn't like them can now filter them out. As Johnuniq says, what Commons needs first and foremost is an administration that signs up to the Foundation's core values (incl. privacy/consent issues, POLA, educational value rather than free speech, etc.), rather than being in open rebellion against them. --JN466 02:23, 17 April 2012 (UTC)

It may be a good consideration, but I still believe in fixing problems rather than using them for political ends. Problems should not be hoarded up. Dmcq (talk) 10:04, 17 April 2012 (UTC)

Text and image are 2 different things, save for the text-based image. Image is meant to be understanded universally at first glance without knowing a specific language. Jimbo, I know you have little ability to do with this since the vulva incident on de.wp main page. Just so you know, this might not be the problem right now (I may pretend those sexual images are non-existent while using Commons service), but soon it will ruin the whole WMF, specifically donation income. -- Sameboat - 同舟 (talk) 02:24, 17 April 2012 (UTC)

In one sense our donation revenue is inherently vulnerable, as apart from Wikimdia UK it is overwhelmingly dependent on an annual appeal. But in another sense we are far less dependent on our indirect public image than most charities. We have nearly 500 million viewers per month, so many of our donors and potential donors are going to be more knowledgeable about our sites than the average journalist. If the press go off on some tabloid rant about not being able to use Wikipedia because of its porn content then there will be plenty of donors whose own experience will contradict that. I'm not saying we don't have a problem with sexually themed content, but when we make changes we should act in a proportionate manner and make the right changes for the right reasons. ϢereSpielChequers 06:19, 17 April 2012 (UTC)

Please, it's not "sexually themed content" that people here are complaining about. Most of us are quite mature—we understand and welcome due sexually themed content. What Commons needs is a donation from a researcher of 10,000 images of penises in all their glory, provided proper notes are available describing the subjects and showing that the images are typical of the populations they portray—that would have educational value. What we do not need is random exhibitionists uploading whatever appeals to them—that is not sexually themed content, it is undue nonsense and a misuse of WMF resources. Johnuniq (talk) 10:08, 17 April 2012 (UTC)

I don't know about Commons but I'd have problems with that lot here under WP:NOTSTATSBOOK. We're supposed to summarize. Dmcq (talk) 10:26, 17 April 2012 (UTC)

Commons has a wider scope than that. What Jonuniq describes would have educational value, if underpinned by statistical data. And as long as it has realistic educational value, it is a valid upload in Commons. However, what Johnuniq is talking about is very different from Wikimedia contributors going into the bathroom, having a wank, and uploading a mobile phone snapshot or video of that – that's literally how crude it has become. And this sort of stuff is kept in Commons under NOTCENSORED. http://commons.wikimedia.org/wiki/Category:Nudity_and_sexuality-related_deletion_requests/kept --JN466 18:55, 17 April 2012 (UTC)

If i try to compress this discussion to a point where only the difference in opinion is left, then i come to this:

Some see the images as educational, others do not.

So i may ask you this questions:

• Would one illustration for a subject be enough?
• Is an image of low quality that shows a practice or detail that no other image on Commons shows valuable or not?
• Should we delete low(er) quality duplicates? (Considering media of all kind, including flowers, buildings, persons, ..., with a duplicate being defined as a picture that shows the same subject)
• How should the perfect illustration for a penis should look like?

--／人◕ ‿‿ ◕人＼ 署名の宣言 19:08, 17 April 2012 (UTC)

The problem there is that Commons participants will argue that any variation is a difference worth keeping. Witness the discussion here where one of the crew argues that there is not another one on Commons with the exact same skin colour. The denizens of Commons will argue that the length of fingernails on the hand of a masturbator are unique enough to warrant retention. John lilburne (talk) 19:43, 17 April 2012 (UTC)

If I might address only one of your questions -- the one related to the topic of this section -- if images are of obvious low quality, as these are, does it matter that we have no other images? These images are not used in any WMF project, which suggests that there is no particular need for them. I would remove these from any article here, not because of their content, but because of their quality. I would also suggest that there is no great mystery to the act of putting a dildo into one's anus and an image would not add to the reader's understanding. The mystery here is why these images continue to exist on Commons. The answer is a very small cabal of Commons admins and users who will vote to keep anything sexual no matter how poor the quality or how slight the chance of the images being used on a WMF project. Delicious carbuncle (talk) 19:52, 17 April 2012 (UTC)

I have to add that Commons isn't only a database for including images inside Wikipedia articles. It is a media library that anyone can use to obtain freely licensed media. That none of our projects uses an particular image could have various reasons or no reason at all. But third parties might already (re)use them without that we are able to see it inside the typical "usage" message on the description page. The same goes for many other images as well, but no one cares about them. At least not the people that blame Commons to host inappropriate sexual content.

The following words are meant to be humorous, but at same time criticism: If i would upload an image of the Eifel Tower on daily basis, then i doubt that i would get any problem and no one would complain. But if i would upload an image of my Eifel Tower on daily basis, then ...

--／人◕ ‿‿ ◕人＼ 署名の宣言 20:46, 17 April 2012 (UTC)

Here are 16,000 Eifel Towers, after the first 100 or so they novelty tends to flag. My suspicion is that that the novelty would droop after the first couple of shots your Eifel Tower. John lilburne (talk) 21:24, 17 April 2012 (UTC)

Niabot, it is my understanding that while images on Commons may be used by anyone, Commons exists as a central repository for WMF projects. It does not exist as a webhost for images that are not expected to be used in WMF projects. Perhaps you should read Commons:Project scope again. Carefully. Delicious carbuncle (talk) 21:49, 17 April 2012 (UTC)

Could you please refrain from one-sided quoting or to give things a not intended meaning? Thank you. PS: A link for you: [5] --／人◕ ‿‿ ◕人＼ 署名の宣言 22:00, 17 April 2012 (UTC)

I wasn't quoting anything. Any argument for keeping an image based on the rationale that someone might be using an image outside of a WMF project is bogus argument and should be discounted in deletion discussions. Of course, there is no point in trying to argue that on Commons, because logical argument plays no part in the decision process over there. So that we are both clear, you understand that Commons has no obligation to host images that are not being used or might reasonably be expected to be used in WMF projects, yes? And that Commons:Project scope is quite explicit about that? Delicious carbuncle (talk) 22:09, 17 April 2012 (UTC)

Why should it be discounted? I'm sure that we don't have an article for every building/wheater situation/rock/festival we have photographs from. It essentially means that such pictures are likely not to be used by our projects, not even in future. Still we keep them because of potential usage by reusers. Going after "... that makes available public domain and freely-licensed educational media content to all, ..." with the appendix "The expression “educational” is to be understood according to its broad meaning of “providing knowledge; instructional or informative”. Is it so hard to understand? BTW: We are way off topic right now. --／人◕ ‿‿ ◕人＼ 署名の宣言 22:28, 17 April 2012 (UTC)

I think one of us has a fundamental misunderstanding of the scope of Commons. I do not know which of us that is, but it may help resolve some of the conflicts here if we figure it out. The Commons' scope policy sets out two uses, either in a WMF project or only on Commons itself (there are other limiting conditions), and then it says "A media file which is neither: realistically useful for an educational purpose, nor legitimately in use as discussed above falls outside the scope of Wikimedia Commons". We likely disagree on how blurry low resolution images of a man inserting a dildo into his anus can reasonably be considered to be "providing knowledge; instructional or informative". That's fine, but I understand Commons to be a repository of images which will be or could be used on WMF projects, not a general repository for images. Let me use an analogy - I tell you that I have my coats in a closet and you are welcome to use them. That does not mean that I am allowing you to store your coats in my closet. There will be coats in my closet which I am not wearing, but they are my coats and I expect to wear them at some point. Does that make sense? Delicious carbuncle (talk) 22:58, 17 April 2012 (UTC)

Not really. If i go after your illustrative example, then you speak about your own coats. That is somewhat limiting and contradictory to Commons mission. Commons is a repository for any kind of free media that could be used in educational context by any WMF project or "all". That means that any kind of educational (in a "broad meaning") media under a free license is welcome, even if it is not used by the Community [WMF would be wrong at this place] itself (at the moment)+ - but by others. What you describe is a closet in which you can only put coats that would be in your own size so that you can wear them anytime, while leaving your visitor not the option choose a size which would fit them. Going after all the previous comments the closet would be even more restrictive. It would not allow coats in red or pink colors, just because you are sure that they don't look good. --／人◕ ‿‿ ◕人＼ 署名の宣言 23:54, 17 April 2012 (UTC)

• Speaking of education—and pictures of crotches—I saw this article on Slate.com today. Some academics wanted to do a study about genital color and had a difficult time finding "Explicit images of anatomically normal, un-retouched, non-pornographic, similarly oriented" genitals. They eventually found some, and published their study. This got me thinking, maybe there is some educational value to a repository of free pictures of genitals, since it's apparently needed by some legitimate academics. Not to say that Commons has to be the place for such a thing, (or that there shouldn't be filters etc.) but it's an interesting example of how hosting images of naked people may not be a completely useless practice. Mark Arsten (talk) 02:21, 18 April 2012 (UTC)

  That sort of thing is what I had in mind with my above comment at timestamp "10:08, 17 April 2012". Lots of pictures of bits would be useful providing they were taken under controlled conditions with useful information available (generic info about the subject, and an indication of how they were chosen—are the pictures typical of something?). Judgment should show that a multitude of good quality and statistically informative anatomy shots would be helpful, while "look what I did today!" shots are not helpful. Johnuniq (talk) 03:25, 18 April 2012 (UTC)

  • Ah yes, I see that now--that is basically what I was thinking. Mark Arsten (talk) 03:32, 18 April 2012 (UTC)

    Commons has 1000s of pictures of penises and arseholes. How many do we have of toes, knees, hairlines, belly buttons, ears, noses, eyes ...? Do we think the public is stupid or something? You cannot look at these pictures divorced from the transactional context that led to their being on a Commons server. Commons has become an ad-hoc web host for exhibitionists, something that is readily apparent to anyone looking at the material. JN466 11:14, 18 April 2012 (UTC)

    Let me be so bold as to quote from Commons:Scope again: "Examples of files that are not realistically useful for an educational purpose: ... Files that add nothing educationally distinct to the collection of images we already hold covering the same subject, especially if they are of poor or mediocre quality". After the first dozen or so erect white penises, all of the rest are out of scope, by Commons' own definitions. The people who seem to be controlling Commons have diverged from the stated purposes of Commons. Delicious carbuncle (talk) 14:37, 18 April 2012 (UTC)

Really, shouting "arseholes" yet again from your soapbox in order to scare the children and cause general offence? Does anyone else find this ranting reminiscent of Father Jack Hackett? --Fæ (talk) 11:29, 18 April 2012 (UTC)

(edit conflict)Penis - 2,248,Toe - 2,217, Knee - 3,184. Obviously very rudimentary numbers and slightly glib; I am highlighting a serious problem with all of these discussions. Which is that we are working with minimal data and adding a layer of confirmation bias. Someone needs to do the legwork of serious research into these numbers/problems. --Errant ^(chat!) 11:31, 18 April 2012 (UTC)

Have you had a look at the relevant search results? The first 500 search results for "penis" overwhelmingly show close-up photographs of human penises. [6]. The search results for "toe" overwhelmingly show something other than human toes. [7] The same with "knee": [8]. Really. Look through them. --JN466 14:55, 18 April 2012 (UTC)

Interesting, with the amount of foot fetishists out there, one would assume we'd have millions of pictures of toes. But, I must plead ignorance: I've never searched for pictures of penises--or feet for that matter--so I, like most users, am basically ignorant of what Commons hosts. Mark Arsten (talk) 17:34, 18 April 2012 (UTC)

That doesn't exactly make you the ideal person to comment. :) --JN466 20:11, 18 April 2012 (UTC)

Well said Fæ; people shouting arseholes is such bad form when there are hardly any at all worth mentioning, and could be easily covered over with a nun like veil of secrecy. John lilburne (talk) 11:41, 18 April 2012 (UTC)

Yep. I'm very sure that what we get presented all the time in such discussions are the extremes and biased data. What Jayen said in the previous comment is an clear understatement for this procedure of making selective or false claims, while presenting preselected examples as "evidence" or as "average". --／人◕ ‿‿ ◕人＼ 署名の宣言 11:43, 18 April 2012 (UTC)

• I hope that the only effect of these threads won't have been that everybody got to have a catharsis, and can now live more happily again with things hurtling on along the existing trajectory. --JN466 14:29, 21 April 2012 (UTC)

  I thought that was precisely the outcome you were aiming for, considering you have been shouting drink! feck! girls! arse! penis! feck! on every possible soapbox you can find until everyone has tuned you out. Thanks for the catharsis, the rest of us moved on some time ago. If you are ready to help with meaningful change, you are a smart enough guy to know what you have to do, it just is not as easy as hurling abuse or collaborating with the sort of people who think is is funny to make personal attacks under the guise of nasty jokes about "fae gots". --Fæ (talk) 17:40, 23 April 2012 (UTC)

Loriot stamps case

Wikimedia Foundation Inc. has just lost a Commons-related court case in Germany: [9] From what I can make out, Commons has, for years, falsely declared that German stamps are in the public domain, refused to take them down when a rights holder complained, and as a result Wikimedia Foundation Inc. has now been found liable for infringement ("Für die unlizenzierte Nutzung der Briefmarken-Motive haftet die Antragsgegnerin als Störerin"). Is that summary correct? --JN466 16:27, 17 April 2012 (UTC)

To ensure all understand, we sought the recent judgment from the court only for the purpose of challenging the court decision on the signature issue, and we won that as we expected. I also have made some comments on the stamps and public domain issue. The full discussion can be found here. Geoffbrigham (talk) 16:42, 17 April 2012 (UTC)

Commons needs common sense - they considered the copyright on a US First Day Cover - and only looked at whether the stamp was copyright - and ignored whether the cachet was copyright <g>. Perhaps it is time to rethink this whole idea that everything must be free of copyright in the first place - and then just bar copiers of Wikipedia from ignoring posted copyrights. Cheers. Collect (talk) 16:56, 17 April 2012 (UTC)

The adversarial legal system is one where people only have the rights that they pay for. If even Wikimedia Commons won't stand up for the public domain, who will? There is nothing in Commons that will not be the subject of some kind of copyfraud sooner or later. Sure, it's a free licensed image you can alter and reuse however you want - but don't do any cropping that might remove a visible or invisible watermark, or magnify a de minimis element; don't copy it if it has a person because that's subject to "moral" rights - is the data used to make the image copyrighted? Commons is already "puahing the envelope" just claiming they have the right to display a public domain image, when whoever photographs it could claim copyright. (Or whoever has moved it across their servers, or maybe the library that owned the book, who knows?) Either we give up on the whole idea of the public domain, in however general a sense, as a lie, and accept that someday even the word "the" will be the trademarked/copyrighted/look-and-feel patented property of some corporation, or else we stand behind Commons proudly when they fight the good fight in court for the defense of Wikipedia and all its readers around the world! Wnt (talk) 05:24, 19 April 2012 (UTC)

What's the weather like up there on your soapbox, Wnt? AndyTheGrump (talk) 06:00, 19 April 2012 (UTC)

Are we really to the point where it's impermissible "soapboxing" to suggest that the WMF has our support when it makes its case in court, but perfectly acceptable "watchdog" behavior to say that the WMF has made "false" statements by maintaining a policy its lawyers believed would prevail in court? (and partially did) Wnt (talk) 06:16, 19 April 2012 (UTC)

I agree that Copyright law is gradually eroding end users rights. It was initially legislated to create a monopoly for printers, and while the aim since Berne has been to give more rights to creators it still encourages organisations to create monopolies rather than individuals to earn from their work. (ideally I'd like to see something like a 50year flat limit from creation on corporations earning any royalties, followed by a 25year flat limit where all royalties go to the creator(s)) It makes no sense that If I (in Europe) create a work at 15 and die aged 120 a corporation can still collect royalties 180 years after I created it, however If I create a work aged 15 and die shortly afterword my family can only claim royalties for 75 years after I created it - it's an arbitrary system. However I don't believe breaking the laws and trying to challenge them in court is the way to try and make that change. Yes no-one abides by the retention of copyright notices provision in Creative Commons Licenses 1-3 which is why it's being rewritten in 4, but why should we be breaking the terms of the license to prove the point that the licence doesn't work? Let's actually ask rights owners if the object to us removing the watermark and if they do let's not use it on our projects, let's start depreciating the use of 2 licences from flickr and urge flickr to start using 4 as well and try to get flickr to get their current users to upgrade to 4.0 so we can use them. The blackout showed how much sway we can have when it comes to advocacy, so get the community to advocate change with their politicians and with corporations rather than having to break the law as it stands to have a day in court. Stuart.Jamieson (talk) 17:43, 20 April 2012 (UTC)

• The problem is that we have anonymous and unaccountable laypersons making legal licensing decisions that can have considerable impact on downstream users, using a very haphazard process. As this case shows (for an even better, and amusing, example see this one), the decisions can prove wrong when tested in a court of law. If someone had used these images in a book, say, believing they were free, as Commons claimed, they would now have grounds to sue the Foundation, as the Foundation was ruled to have played a contributory role to infringement.
• Bona fide public domain image libraries offer an assurance that their images really are in the public domain, down to indemnification of any downstream user who is subsequently sued for infringement of rights. Commons is quick and dirty; it has none of these guarantees and safeguards.
• Sometimes Wikimedia seems like a company that sells cars without lights and brakes. Because they are very cheap, lots of people buy them, and the makers of cars with lights and brakes are feeling the pinch. At the same time, though, there are lots of hit-and-run accidents. JN466 14:25, 21 April 2012 (UTC)

I Agree wholly with that, there is too much emphasis on keeping images at all cost even if there is a reasonable and measurable risk that the images may have rights associated with them - and it leads to behaviour like above. As others have said perhaps a stronger community may lead to improvements with this but it will still need better legal advice and stronger clearer policy. Potentially unfree images marked for deletion also take too long to be dealt with as well. 20:27, 22 April 2012 (UTC) — Preceding unsigned comment added by Stuart.Jamieson (talk • contribs)

A very focussed discussion of racial bias

I'm pulling the following quote of Jayen466 out of the middle of a long discussion above in order to focus attention on it. I had never heard of this as a potential problem in Wikipedia before, and I think it is one that we should take seriously. I think that Jayen's methodology here is basically sound and his analysis should be refined and expanded. Let's discuss how we might do that. Please don't veer off into random other arguments: I want this discussion to be focussed solely on how we might measure the problem. I'll comment further, below.--Jimbo Wales (talk) 16:16, 16 April 2012 (UTC)

I will just say that, as in Otis Ferry's article, these "Legal issues" sections often look intemperate and vindictive.

With respect to the charge of white bias, I tried to have a look at the frequency with which these legal issues sections occur; in this search, I found – on the first page of 100 search results – 45 biographies of Black people that had a "Legal issues" section, vs. 28 biographies of Caucasians that had such a section, and 3 of Hispanics – i.e. nearly 60% of these sections were in biographies of Black people. That did seem suspiciously high. I didn't do an exact count of the biographies on the second page of search results, but there seemed to be rather more Caucasians on that page than there were on the first, with proportions roughly reversed, and Caucasians outnumbering Blacks 2 to 1. --JN466 02:41, 15 April 2012 (UTC)

There are a couple of factors here (probably more than a couple!) that we should consider. First is the question of whether Wikipedia itself has a bias towards reporting more "legal issues" for blacks than whites when all else is equal. If so, that's absolutely something that we can and should make a very strong effort to correct. Second is the much more difficult question of whether our culture as a whole is such that for people of sufficient notability to be in Wikipedia, there is more reliable mainstream press coverage of legal issues for blacks than whites. If so, then we still should do something about it (recognizing as per my usual that "verifiability, not truth" is wrong) by noting the overall bias in the press and working to correct for it, just as we make similar judgements about all kinds of things.

A first cut would be to follow Jayen's methodology (though perhaps in reverse - starting from a random sample of biographies and then looking for "legal issues" or "controversies" sections, and seeing the overall numbers. This would confirm that there is a problem. To determine the nature of the problem, though, will require more judgment - looking for a representative sample of "similar" biographies... for example black and white judges, black and white baseball players, black and white professors, etc. Looking even deeper after that, we'd want to look into whether the press appears to have unbalanced coverage. This is not going to be perfect, but I think that a reasonable conclusion can be reached.--Jimbo Wales (talk) 16:16, 16 April 2012 (UTC)

To address the specific question within Wikipedia, one could use a case-control type of approach, where starting from a known outcome like inclusion of a legal/controversies section you try to evaluate (using multivariate statistical analysis) the independent contributions of candidate risk factors, such as race, sex etc, alongside potential confounders, such as profession, nationality etc. 'Cases' would be biographies containing a legal issues / controversies section, whereas 'Controls' would have no such section. The sampling frame would be defined a priori as Biography pages with certain characteristics (perhaps BLP or people living since a certain year?), and one might want to restrict the study to pages a minimum size, large enough to contain multiple sections. Obviously, good study design would be important, but given the ready availability of historical Wikipedia archives (from before this particular discussion!) there should be few overheads. However, by itself such a study couldn't provide any direct information on the much more difficult question of whether our culture as a whole is such that for people of sufficient notability to be in Wikipedia, there is more reliable mainstream press coverage of legal issues for blacks than whites. —MistyMorn (talk) 17:11, 18 April 2012 (UTC)

I could be wrong about this but my feeling is that general Western bias is more or less equal against blacks and Hispanics, so that 3 figure for Hispanics vs. 45 for Blacks makes me think bias isn't at work on Wikipedia. Equazcion ^(talk) 16:25, 16 Apr 2012 (UTC)

I'm in the UK and we certainly have racists here too. But their targets tend to be people of Afro Caribbean, Romany, East European or South Asian extraction. The Anti-Hispanic thing may be more of a US issue than a general western issue. Also the numbers are too small to draw definitive conclusions from, but they are troubling enough to justify some proper research. ϢereSpielChequers 18:52, 20 April 2012 (UTC)

An interesting notion. I'd surmise that it is Western media in general that has a more critical and dirt-digging eye towards blacks and crime, so if a black person is suspected of criminal involvement in a matter, more media outlets will pick up the story for the news cycle than if the suspect were white. It is a variation of missing white woman syndrome. Speaking of which, I have come across many "Murder of Jane Doe" type of articles over the years in the Wikipedia, and I'd wager that many of those Janes aren't black, primarily because the media does not pay as much attention to such crimes. Tarc (talk) 16:42, 16 April 2012 (UTC)

There appears to be bigger issues with race in the USA. I'm not sure how true this would be outside the USA though. IRWolfie- (talk) 16:56, 16 April 2012 (UTC)

Depending on the nation, the ethnic groups under pressure will likely vary, but racism - both explicit and implicit - is a problem almost everywhere in the world in one form or another. A similar issue, which we shouldn't discuss in this thread so we can stay on track, is our tendency (unproven, but anecdotally it seems true to me) to excessively identify Jewish people as Jewish when we don't pay attention to other ethnic groups. I've never been sure whether that's positive or negative discrimination. What I mean is, I think many Jewish people are proud of the accomplishments of Jewish people and will want to add that. And many anti-semites think it is useful to go around making lists of Jewish people to further their POV that "Jews control the world" or whatever.--Jimbo Wales (talk) 18:09, 16 April 2012 (UTC)

A good example is the article on Franz Boas who very specifically did not consider himself to be Jewish in any meaningful sense - but German and American. His talkpage is frequently visited by both Jewish and Anti-Semite editors who wish to label him with the label he categorically refused for himself.·ʍaunus·snunɐw· 01:54, 17 April 2012 (UTC)

Maunus—I think that Wikipedia should avoid a parochial perspective and abide by sources. I find you saying on the Boas talk page: "If Judaism is a religion and Boas was not religious then how was he a Jew exactly?"[10] Sources tell us that Boas was Jewish: "By contrast, thoroughly secularized Jews were also among the emigrants; they included Franz Boas, the father of American anthropology."[11] Bus stop (talk) 14:11, 17 April 2012 (UTC)

Bus Stop I find your perspective to be the most parochial that exists -namely the one that assumes that the categories that are meaningful to you should also be so for others. ·ʍaunus·snunɐw· 13:44, 18 April 2012 (UTC)

Surely an encyclopaedia should include stuff that is relevant to the understanding of a subject, and not because it makes some 3rd party feel better, or worse about themselves? If we include tags for everyone that it Hindu or Muslim then we have potentially every notable person in India, and consequently we have lost the information about those people where their religion is a defining fact about them. Similarly if we add Greek to all people of Greek ancestry then we have lost the fact of where being Greek was a defining factor in their lives. John lilburne (talk) 14:41, 17 April 2012 (UTC)

John lilburne—you express that you feel that Boas' Jewishness is not "relevant to the understanding of a subject".[12] You are certainly entitled to hold that opinion personally. But we are expected to maintain a global perspective and we are expected to give consideration to reliable sources:

"Finally, as a German Jewish immigrant, Boas identified with the plight of African Americans. In Germany he had been the victim of anti-Semitism leading to his decision to migrate to the United States, where he endured outsider status as an immigrant and a Jew. By attacking racist science, which concluded that blacks were inferior to whites, Boas was able to mount an indirect challenge to the anti-Semitic belief that Jews were an inferior race."[13]

"Boas's Jewish ancestry undoubtedly sensitized him to prejudice and discrimination toward others"[14]

"Franz Boas the anthropologist held the upper hand over Franz Boas the German Jew."[15]

Whether or not to include allusions to the above sorts of considerations is an editorial decision. Writing is an art to some extent, meaning that the article will stand with or without such references. Do you see that there is valid and relevant material that can be discussed for possible inclusion in the article? Bus stop (talk) 13:02, 18 April 2012 (UTC)

Red Herrings. The article of course should and does explain Boas relation with Judaism which was important to his experience in Germany and later in the US. But it should not say that he was "a Jew" because that statement is in direct contradiction of the way he experienced that relation. ·ʍaunus·snunɐw· 13:41, 18 April 2012 (UTC)

Maunus—you say that "…it should not say that he was 'a Jew'…"[16] when in fact he was a Jew. We defer to reliable sources. Reliable sources know that being a Jew has nothing to do with what one believes or does not believe, or practices or does not practice. Your opinion can't override that which is determined by reliable sources. Bus stop (talk) 15:35, 18 April 2012 (UTC)

Being a jew is not a matter of scientific fact it is a matter of identity. Having a star slapped on your coat does not make anyone a Jew - identifying and being identified as a member of a community does. Boas did not identify as such, and only his antisemitic opponents identified him that way during his life.·ʍaunus·snunɐw· 15:40, 18 April 2012 (UTC)

Bus stop, being 'a Jew' can never be 'fact' - 'Jewishness' is a social construct, and as such, only an opinion - and as to the question of Boas's 'Jewishness', opinions differ. AndyTheGrump (talk) 15:41, 18 April 2012 (UTC)

AndyTheGrump—it wouldn't matter if Jewishness was a social construct. The salient point is that reliable sources all concur that Boas was Jewish. Bus stop (talk) 16:07, 18 April 2012 (UTC)

They do not. All sources that actually discuss his relation to Judaism (his biographies for example) instead of just label him agree that that is not an meaningful label for characterizing him. And what about Boas himself, why is he not a reliable source?·ʍaunus·snunɐw· 17:25, 18 April 2012 (UTC)

I expressed no such opinion at all regarding Boaz. But if you want my opinion the plonking of every D-list Celebrity and 2nd grade sports player, where there Jewishness does not make a significant contribution to their life's work, into the same tag as Boaz does not only him a disservice, but is unhelpful to the reader too. Eventually, and I shudder at the thought, you are going to need a 1st class Jew, and a 2nd class Jew categories in order to separate out those of true importance to Jewish culture and history, from the also rans. John lilburne (talk) 13:57, 18 April 2012 (UTC)

I would recommend a different methodology than the search engine one. The search engine ratio methodology is effectively meaningless since it isn't a randomly selected group, it is based off a search engine result (I'm not sure how the search engine decides the order of results) and it also appears to assume that the number of articles on Hispanic, black, white people etc are the same alternatively there may be an implicit assumption that the ratios should be that of the USA for some reason. What is relevant is what is the ratio of articles about a particular ethnic group/race, subdivided by profession etc, with a negative section versus not having a negative section. By comparing the ratios in that way any bias would be uncovered. Using a random number generator to pick articles from within wikiproject biography Category:WikiProject_Biography_articles appears to be the best way to get a decent sample for example. IRWolfie- (talk) 16:56, 16 April 2012 (UTC)

If some sort of bias is eventually judged to exist, I shudder to think what the remedy will look like. Equazcion ^(talk) 18:13, 16 Apr 2012 (UTC)

Why "shudder to think" what the remedy will look like? I can't imagine what you mean. The only possible remedy is to edit Wikipedia to fix it, just like any example of bias. If we have accidentally adopted a double standard, deciding the a minor legal issue for a white judge is not relevant, and the same minor issue is relevant for a black judge, we can fix that by being aware of it and thinking about it and discussing it.--Jimbo Wales (talk) 00:46, 17 April 2012 (UTC)

Well, if editors know about it that's the first step to fixing it. Regarding methodology, I have to agree that when we can't explain why "tolling bells" are not made out of metal, we can't really rely on the search results. And regarding blacks versus Hispanics, I'm not sure it's a valid comparison. The level of discrimination against the groups was never perfectly the same, and Hispanics in the U.S. often identify proudly with external national origins which give them extra resources to fall back on - for example, by having Spanish television networks which also broadcast in Latin America. But the data, carefully obtained, could tell all. Wnt (talk) 18:18, 16 April 2012 (UTC)

I have so many questions about this. And since I'm often the one to say the unpopular things that I nevertheless feel should be addressed, here goes. What if black people have just had more legal issues than white people, whether due to societal problems or police bias? What would we do in that case? And for that matter, how could we tell the difference between that and media bias? Are we to take an imbalance in "legal issues" sections alone to mean Wikipedia has a bias that needs correcting (that seems to be the only method mentioned thus far)? Do we then go about with our own version of affirmative action, either putting legal issues sections into not-so-deserving bios or removing them from deserving ones? Racial bias and other minority issues suck, but I'm not sure an encyclopedia is in any position to correct it. We may just have to report on the shitty world as it is. Just saying. Equazcion ^(talk) 18:43, 16 Apr 2012 (UTC)

I don't agree with you, but I do think you raise an issue that needs to be reflected upon. I do agree that merely seeing the numbers (which we don't have yet, other than JN's quick-and-dirty study) doesn't in and of itself prove bias - there are many possible reasons for it. I suspect, though, that further study will reveal at least some problems that we can deal with, which is why I say I disagree with you.--Jimbo Wales (talk) 00:46, 17 April 2012 (UTC)

• Many biographies don't list people as "white", they list them only as Americans, or whatever country they are from. Sometimes instead of that they just list them as Jewish. Nowhere in the Wikipedia articles does it say they are white. They do mention if someone is black though. We have categories such as category:American people of French-Canadian descent, Category:American people of Scotch-Irish descent category:American people of Mexican descent, etc. So you might just have to include all of those in your searches for whites and Hispanics. Dream Focus 18:53, 16 April 2012 (UTC)

• In terms of sampling, this seems to be a better search: it gives just a little over 7,000 results, i.e. less than half of the other search method, and here all (or nearly all) search hits really are living people. To counter any potential bias in the listing of search results (i.e. that there might be more or fewer Caucasians on the first or last pages), we could sample as follows: Count Black / Caucasian / Hispanic / Asian people on the first page of 100 search results, on the 11th, the 21st, the 31st and so on, to the 70th page. (These pages can easily be reached without clicking through all the preceding pages, by adding &offset=1000 to the URL.) For a very rough first indication, 1,369 of these 7,000 biographies seem to have the word "African" in them, most commonly in the "African American" category. Clearly that will not cover British, Nigerian or French Blacks, nor does it cover all African Americans – this search, looking for biographies that do not contain the word African, includes African Americans like TQ (singer) and BeBe Winans for example. So these 1,369 biographies are probably far from the complete number of people of African descent among those 7,000, but it establishes a first ballpark estimate for the lower limit. I am short of time currently, but if someone wants to jump in, that would be swell. :) --JN466 19:16, 16 April 2012 (UTC)

  Assuming that a bias exists (and it wouldn't surprise me if it does), I don't think we should go about fixing it as a bias. Rather, we should rethink our policies on biographies. I.e., can we come up with a set of policies or guidelines that will remove the bias without explicitly addressing it? For example, I note that many biographies have a section titled "Criticism", which becomes a catch all place for critical opinions on a person. Sections titled "Criticism" or "Legal Issues" are examples of poor writing, and systematic biases aside, are unfair to any individual of any race. Perhaps a set of tight guidelines on how to write biographies is where we should be headed. --regentspark (comment) 19:27, 16 April 2012 (UTC)

  I agree. Again, it's why I think we should use pending changes to restrict the editing of biographies to people who have taken such ground rules to heart, and lock the door to people writing with a grudge. There should really be a kind of editorial team in the traditional sense of the word for biographies of living people; it could still be a very large team, but it should have norms and standards that are meticulously upheld. The points you mention should be among them. --JN466 19:50, 16 April 2012 (UTC)

  I agree too, on the solution that is, but I don't think we necessarily need to wait to see whether someone concludes there's a bias before getting started on that. I don't have all that much faith in our ability to accurately determine this bias thing, but whether or not anyone else does, this seems like a good thing to get started on right now. Equazcion ^(talk) 20:08, 16 Apr 2012 (UTC)

I think this is a useful inquiry, but we need to be very careful about how we define things. Take, for example, the article on Frank DiPascali, who is a white person. He has pleaded guilty to federal charges. Yet, if you searched for the headings "Legal issues" or "Legal troubles" you would not hit this article. There is, however, a heading that serves the same purpose: "Criminal and civil complaints and guilty plea." I suspect that Mr. DiPascali would PREFER "Legal issues," not that his Wikipedia article is his biggest problem. His "legal issues" also dominate the lede to his article. So does Mr. DiPascali count as someone (who happens to be white) who has a section on their "legal issues"? For that matter, Mr. DiPascali's former employer, another white man (Bernard Madoff), ALSO does not have a section in his article entitled "Legal issues," but he has headings that are "worse", including "Plea, sentencing and prison life," and a few that include the word "fraud." So we just need to be careful with our searches. Neutron (talk) 19:37, 16 April 2012 (UTC)

Just to explain why the enquiry started off this way: the article that sparked this discussion contains the following passages: "Yet a large portion of the wiki article was dedicated to D'Angelo's 'Legal Issues' (a subheading seemingly exclusive to black musicians, intellectual property pages and anzac biscuits), including a DUI charge and a marijuana possession charge. [...] The entry on Keith Richards, the grandfather of recreational drug use and all comedy based on chemically-eroded intelligence, includes detailed information on his drug use and trouble with the law. But he does not have 'Legal Issues'; his story is a rich and balanced biography. Even Lou Reed, who wrote songs about drugs as well as under their influence, apparently never possessed them. R. Kelly, on the other hand, has serious 'legal issues' pertaining to an alleged statutory rape. Yet this disturbing crime is lumped in with a sound pollution charge for playing music too loudly in his car. This juxtaposition is offensive: who cares about a sound pollution charge? Certainly not the young survivor of sexual assault." --JN466 19:45, 16 April 2012 (UTC)

AFAICT, Wikipedia tends, if anything, to make sure legal problems are in every BLP they can be squeezed into - so I doubt any "racial bias" on that end. We also do not list "non-notable" people, although it appears that if the crime is in the press, the notabiity of the people involved seems to vanish as a concern :( . That said, the term "legal issues", along with the other "controversy" headings, I find problematic, and suggest they be avoided wherever possible. Collect (talk) 20:19, 16 April 2012 (UTC)

Agree. I don't think you can extract anything useful this way, other than an impression. One can guarantee that any celebrity person in the news will have a section dedicated to any brush with the law that they may have had. So one might want to look for bias in the edit histories, for example how much resistance there was by editors to the inclusion of a DUI arrest. For example if on average a Black has his DUI arrest added with 2 reverts before it sticks, whilst for a White it takes on average 6 reverts then that would indicate a bias. Similarly you could average the average length of copy devoted to a DUI arrest. For example 2 sentences for one, 4 sentences for the other would again indicate some bias. You might wonder why WP has a category "Puerto Rican murderers" and what the bias was for that. John lilburne (talk) 20:23, 16 April 2012 (UTC)

(EC, response to JN's last) Let me be clear here, I am not saying there is not a problem, I just don't know whether there is or not, and what you have found thus far does not answer the question, in my opinion. When I look at Keith Richards' article, one thing I notice is that of the four sentences in the lede (which is probably too short for an article of that length, by the way), one (25 percent of the intro) is about Richards' "notoriety for illicit drug use" and "several drug busts", and as you say, there is a lot more in the body of the article. So are we really giving less attention to his drug use and legal problems than someone who has a "Legal issues" section? I'm not so sure. As far as other examples, how about David Crosby, which has a heading "Drugs charges and prison time"? And then on the other hand we have Willie Nelson, whose article is full of "legal issues", but no heading by that name. Oh, and here is an interesting one, Nick Nolte, which has a heading "Legal troubles." And as the final one I will look up right now, Roger Clemens has a "Controversies" section with sub-headings "Accusations of steroid use" and "Adultery accusations." Now, I have been a little one-sided in my research, since the four persons I have just mentioned are all white. Let's look at Alcee Hastings, an African-American former judge and current Congressman. When he was a federal judge, he was impeached by the House and convicted by the Senate, and removed from office, on charges of bribery and perjury. And yet there is not a single heading in his article (or anything in the lede) that suggests he ever had any legal problems, or any kind of controversy, whatsoever. (His legal problems are mentioned in the text of the article, but not in a heading.) So, how do we "count" all this? Maybe we need some kind of "points system." Neutron (talk) 20:36, 16 April 2012 (UTC)

You can't count it. You can only ask is a person only noted for some criminal act more or less likely to be mention on WP based on ethnic bias than the statistics indicate? Or is the incident of a similar legal issue likely to be more or less detailed based on ethnicity. John lilburne (talk) 20:49, 16 April 2012 (UTC)

Yes, you can only really look for racial bias by comparing like with like.

I think there is inevitably going to be racial bias on Wikipedia just because it is written by people. But biographies doesn't seem to me to be the most obvious place to look. As Collect indicates above, criminal convictions in celebrity biography articles may be a candidate for the place on WP where you are least likely to find racial bias, because we tend to take a pretty non-discriminatory approach. If you have one, it's going in. Which means that if we tried to tackle racial bias there, we would only end up by creating it (if we had a policy to report fewer criminal convictions, we would disproportionately use this to favour white subjects).

What we should be talking about is WP articles related to sociology and politics, where I don't think you need a degree in statistics to see that we have a problem. Our article on racism, for example, is disgraceful. Overall, there are far too many articles which are pointy or slanty from a race point-of-view to name them all. But I don't think biographies should be on the to-do list. Formerip (talk) 21:12, 16 April 2012 (UTC)

Just out of curiosity, which way do you think the Racism article is slanted? Neutron (talk) 21:18, 16 April 2012 (UTC)

You need to ask? Formerip (talk) 22:05, 16 April 2012 (UTC)

As I have not read our article Racism and won't likely have time this week, I do have to ask. Which way do you think it is slanted? --Jimbo Wales (talk) 00:53, 17 April 2012 (UTC)

It is slanted in that the definition of Racism so narrow that racial slurs, racial profiling, white supremacy, white nationalism falls outside of its scope. (It says that racism is only ideologies that advocate discimination based on race). The body of the article does more justice to newer scholarship on racism as also including practices that systemically or unconsciously perpetuate racial stratification. There is a long tradition of POV pushing in articles related to race all across wikipedia in which viewpoints associated with white nationalism, white supremacy and racialism have been inflated in importance and critical scholarship has been left out. This has been partly addressed by the R&I arbitration case but most of the articles are still slanted. A year ago wikipedia had an article on the Nordic race that described it as if it were a contemporary category. We also had an article on Race and crime that at length considered the possibility, proposed by J. Philippe Rushton that the black over representation in American penal system was due to their having a different reproductive strategy than white and Asian groups. And yes, race bias is also a persistent problem across wikipedia - part of this is simply due to the systemic bias, and another part is due to editors with racial biases actively pushing specific racial agendas.·ʍaunus·snunɐw· 01:51, 17 April 2012 (UTC)

Yes, it gives a peculiar, narrow definition of racism which is unsourced and at odds with dictionary definitions, which will tend to focus on belief in racial hierarchy. As well as excluding obvious forms of racism, as noted by Maunus, it is worded so as to actually include most forms of anti-racism. What we would normally think of as racism is distinguished in an unsourced paragraph as something different call "racialism" - a fiction that also has its own article. The second sentence appears to dismiss racism as merely a "pejorative epithet" and a clumsy sentence starting the second paragraph appears to distinguish "racism" and "racial discrimination", with the former bizarrely characterised as non-pernicious. Formerip (talk) 04:45, 17 April 2012 (UTC)

Yup: we have an article on racism which borders on being racist. Actually, no, it doesn't border on it, it is racist, in that the first sentence unequivocally asserts the existence of distinct "human racial groups", whereas almost all scholarly sources for the last 50 or so years have rejected such simplistic categorisation. Of course, some of those 'contributing' to the article like to pretend otherwise. (See also Talk:Human#Race/Ethnicity for similar POV-pushing...) AndyTheGrump (talk) 02:37, 19 April 2012 (UTC)

• This is going to be complex. But it is important and several aspects are worth looking into.

1. The sample size is small - good enough to raise warning bells, good enough to say we may have a problem and need to look more deeply. But we need a larger sample size, as this would be statistically more robust and might enable us to analyze it and identify more complex patterns.
2. Including Keith Richards and other very famous people in this sort of comparison is unlikely to be helpful. Major celebrities will tend to have much more heavily watched and edited pages than our average biography. The question we should ponder is whether a comparison of like with like would show that we generally treat black murderers similarly to white murderers.
3. We need to also measure this from the opposite direction. I'm pretty sure that there is something in this, if only because in the UK, US and Australia the prison populations are less white than the general population. If we identify a random set of major crimes and check the Wikipedia coverage, then it would be good to know whether we are fairly representing what is happening in our courts or whether we are skewed in our coverage.
4. Are we reflecting the "reliable" sources, or are we going beyond them? If it transpires that we are more likely to cover black on white crime than white on white, but that we are doing so because some or all of our sources do so, then we need to carefully consider if some of our sources are truly reliable, or whether some should be deprecated as giving a slanted view of the world.

I think we need to look into this and learn more before we act. ϢereSpielChequers 23:21, 16 April 2012 (UTC)

I agree completely. I'm a big fan of taking a cold, simple, thoughtful look at the facts, without emotion or defensiveness. After we understand, we can act (or not act) with more wisdom and legitimacy than if we just ranted.--Jimbo Wales (talk) 00:53, 17 April 2012 (UTC)

Before that, even, I think we need a reason the think there's something to look into. As, indicated above, I think we have bigger, more obvious, fish to fry in terms of racial bias. Formerip (talk) 23:47, 16 April 2012 (UTC)

Well, I think that JN's quick study suggests that there could be a problem, so I think we've got a reason to look more closely. And while I don't disagree with you that there may be bigger and more obvious issues to look at, I don't think that should stop us from starting somewhere, and here seems to be a reasonable place. On the other hand, if you have an idea of a different way to look at the issue of racial bias that would be better, I'd love to hear your ideas!--Jimbo Wales (talk) 00:53, 17 April 2012 (UTC)

I don't think it suggests much Jimbo. Take a look at the second page of results: [17]. I count 52 white subjects, 13 black subjects, 3 Hispanic, 3 Indian, 2 Filipino, 2 Arab. Formerip (talk) 01:36, 17 April 2012 (UTC)

In terms of starting somewhere, I can agree that we should do that. But it doesn't mean starting any-old-where. This looks to me not like a serious attempt to consider issues of racism and racial bias on WP, which undoubtedly do exist, but to marshal race politics for an unrelated agenda.

Where I think we should start looking is at the ways in which our policies and practices play out on articles that can be easily identified as having a problem. I think we may find that our concept of neutrality is not precise enough, that we are too tolerant of off-topic content and POV-forks and, perhaps, that we have too much of a tendency to assume good faith in each other. Formerip (talk) 04:53, 17 April 2012 (UTC)

I agree. But specifically to this issue. We would also have to consider whether "Legal Issues" are a badge of distinction in some subcultures across races, as they have always been, standing against "respectable" society. Alanscottwalker (talk) 00:10, 17 April 2012 (UTC)

Go back to the top and assume the initial methodology is sound: 60%. 60-40 is not bias or racism, that's within "chances are". If it was 95-5, I'd be concerned, but this is Shakespearean much-ado. Choyoołʼįįhí:Seb az86556 ^{> haneʼ} 00:56, 17 April 2012 (UTC)

If our coverage was even across the planet then there would be something in that, as Nigeria alone has a far greater population than the UK. But it isn't even close - our coverage is massively skewed towards North America and the British Isles. When you factor in the racial makeup of the US and UK the 60-40 ratio is very far from what one should expect. Yes we need a more thorough study with a statistically valid sample, but if it comes out with a similar ratio it will be confirming a real problem, not showing a minor statistical fluke. ϢereSpielChequers 05:54, 17 April 2012 (UTC)

WSC, as I noted above you get very different information from the second page of results, where just under 10% are black. That massive difference ought to tell us that looking at a single page of results from this particular search is meaningless in the first place. But, if we ignore that, we can say that about 38% or the first 200 biographies returned have black subjects. You're right that we should then compare that not to the world population but a population we can't be exact about that is heavily skewed towards the US and UK. However, we need to go further still, because the data is also skewed towards people who meet WPs notability standards and, in particular, it's skewed towards actors, sportspeople and musicians. Lastly, a qualification for the list is being known to have a criminal history of some kind (or, at least, arrests or court appearances or something). So the very rough question to ask isn't "are 38% of British and American people black", it's "are 38% of British and American sportspeople, actors and musicians with a criminal history black", which I think is something we can only guess at. Formerip (talk) 06:40, 17 April 2012 (UTC)

FormerIP, I said earlier in this thread "The sample size is small - good enough to raise warning bells, good enough to say we may have a problem and need to look more deeply. But we need a larger sample size". I'm not saying we should respond to this anomaly by assuming it as solid evidence of a problem and changing our policies and procedures to reverse it. But we shouldn't dismiss it as meaningless either (or at least not unless a larger study shows that it was indeed a meaningless statistical fluke). The best way to respond to problematic anomalies is to investigate them further to see if they are statistical flukes or signs of a real problem. The worst case scenarios aren't just that one does a kneejerk reaction to an anomaly and fixes the wrong possible cause of it, the worst case scenarios also include ignoring the early warning about a serious problem. Having another equally small sample show a different pattern might give us some reassurance, but it doesn't remove the need for a larger sample as the true picture could look like either sample, somewhere in between them or even more extreme than either. ϢereSpielChequers 09:29, 22 April 2012 (UTC)

Yes, Jayen's numbers were sufficient to raise the issue, and personally I applaud Jimbo's thoughtful and open response. To begin to study the question properly, you wouldn't just need larger numbers, but a reasonably appropriate methodological approach. A case-control design springs to mind (I made a small suggestion above). The point raised separately by Jimbo about the challenges of dealing possible nationality-race interactions would take some thought (a series of analyses, each restricted to particular nationalities or groups of nationalities, might be a possible approach). —MistyMorn (talk) 11:46, 22 April 2012 (UTC)

Ermm ok, you may well ask how's an epidemiological approach that's generally used for understanding the etiology of disease [18] relevant here. Actually it fits nicely. What you'd do is consider the presence of a controversies/legal section as the condition (cf "disease") of interest and race/ethnicity as an "exposure" of interest (since people generally don't change race, one wouldn't have problems of recall bias or temporality).
Obviously, however, association does not imply causation. It would be particularly important to interpret cautiously findings from within Wikipedia in the light of population data. For example, if "non-Caucasian" race, say, turned out to be an independent risk factor for having a legal issues section in a UK or US BLP, then one would need to consider the magnitude of that finding in the light of what we know about the racial composition of civil/criminal cases within those countries. There might be arithmetical ways of dealing with that. —MistyMorn (talk) 12:12, 22 April 2012 (UTC)

MM, to pick up your analogy, what you're proposing is a mandatory screening test before there is any indication that things are different to normal. If you imagine that those first 200 hits are infection reports coming in from hospitals, the first thing you would do in check the positives against a benchmark (i.e. "we should have five cases, we have 312, holy crap!"). But if you have no idea what benchmark to use, the next thing you should do is find out. That's something you must do before you go any further.

As discussed above, there are articles on WP that are sick. We could point our attention there, rather than investigating for problems at random. Formerip (talk) 13:55, 22 April 2012 (UTC)

Sorry, there's been a genuine misunderstanding here. I was not intending to make any "analogy" with disease, I was just picking up on Jimbo's original query about numbers. Rather, my reply to WereSpielChequers's comment tried to outline how an epidemiological method could usefully be applied to this legitimate bibliometric research question. That would seem to me to be a convenient methodological approach to investigate a particular form of systemic bias across groups of Wikipedia articles, which -- if I understood correctly -- was the issue which Jimbo originally raised here.
Imo, your powerful point about "sick" individual articles is really a separate concern: subtle, unintentional bias across articles linked to the social and ethnic characteristics of en.wp's community of contributors is really a different issue from particular articles which have somehow been owned to push hateful POVs. I think these two posts (questions of "anonymous editing" aside) illustrate the two different issues rather well. Both deserve careful consideration, imo, but in rather different ways. (Btw, to me the racist forum post noted by user:maunus underlines rather clearly the priority of your particular concern.) —MistyMorn (talk) 16:07, 22 April 2012 (UTC)

OK, if you weren't making that analogy, fine. I agree that there are likely two different sources of bias at play, but in practical terms it is not easy to tell the two apart. Undue criminal history content in bios on black subjects, if there is a serious bias, could have been put there malevolently, whereas content in an article such as Racism could have been put there haplessly by someone displaying garden-variety bias. It's hard to say without knowing more about editors than we do.

What I object to is the focus on BLP issues, because it is coatracking. Bias on WP is something that should be addressed. But looking for bias in a very narrow domain such as criminal histories in biographies is just using race as a cloak of convenience, particularly since there's every reason to think that it is a low-priority domain within the overall context.Formerip (talk) 17:51, 22 April 2012 (UTC)

Absolutely no analogy there! I just happen to know professionally -- I really do -- that, on methodological grounds, a case-control study design would be a valid and feasible initial approach to the numbers question and wanted to share that understanding (unfortunately I can't just sit down and do the study myself!). Likewise, I mentioned BLP only because one would have to draw a line somewhere; Socrates's legal troubles, for example, wouldn't really be what the sort of stuff we're interested in here. It's difficult to talk about these sort of technical questions without being misunderstood.
I certainly agree with you that it can be very difficult to separate within individual articles unintentional bias determined by our backgrounds from deliberate POV pushing (I've seen for myself an example of a possible overlap in disputes over racist themes in the writings of Charles Dickens). I also agree that once the POV pushing damage has been done to an article -- in whatever direction -- its effects can be insidiously difficult to undo without rebuilding from scratch. In that regard, I'd guess it's perhaps easier to correct unintentional bias within individual articles: if someone's able to provide high quality RS (assuming they exist), then reasonable editors will generally agree that their content can't just be dismissed out of hand.
At the same time, I think unintentional systemic bias and POV pushing need to be approached at different levels. Unintended bias needs first to be identified, if present, by looking across suitably sized random samples of certain types of article, rather as Jimbo, Jayen, WereSpielChequers and others have been proposing. On the other hand, pertinent cases of POV pushing can be looked at within known individual pages, as you, AndyTheGrump and others have been doing (although I do agree it isn't at all easy to separate what the POV pusher/s are doing from the various human reactions of everyone else). My 2c, —MistyMorn (talk) 19:03, 22 April 2012 (UTC)

Shouldn't we be using more variable and specifically-tailored wording for section headings? I think that a problem with "Legal issues" is that it contains a whiff of expectation. As if we knew this was coming. That would tend to put anybody in a bad light. But when you tighten up the language in the spirit of discernment, you also banish the suggestion that this is to be expected. A section heading can responsibly read: "Allegations of domestic abuse". This is being considerate of the subject's understandable sensitivity while still reporting the facts. Bus stop (talk) 18:57, 22 April 2012 (UTC)

• Better links, with more precisely defined search parameters:
  • [19]
  • [20] JN466 02:07, 17 April 2012 (UTC)
  • Adjusting the search parameters to "Legal issues" "Living people" births I think there are only about 200 biographies of living persons containing the exact phrase "legal issues" in a section title. That's a manageable number; most of them are here: [21]. So we can do a breakdown of those 200; but it is a rather small sample. Just one editor editing lots of rap articles could have a significant impact here. --JN466 05:15, 17 April 2012 (UTC)

• Crime rates are tied to low education levels: I know the key focus is about racial bias, with legal problems, but I want to emphasize that major crime rates are tied to low education, which is likely to show a racial "bias". In the U.S., where black students are more likely to drop out, or Asian students are more likely to graduate, than Caucasians, compare the related legal issues. Push this concept to the opposite extreme, and search for "honors students" involved in crimes, such as: Google search: "honors student" with "murder". Result: When "honors students" are involved with a murder (or killing), then they are more likely to be the victim (or accused by police in Italy). Anyway, back to the main focus, of racial bias and legal issues. How will bias for low education affect Wikipedia articles? I think notability will be tied to higher education, and hence, Wikipedia articles will show a "racial bias" that reflects racial trends in higher education. So that gives some other hypotheses to test, to compare notability with education levels, and compare a region's racial trends in education, and then compare legal issues of notable people. So, Wikipedia might appear to show a reverse bias, relative to the general population. That probably seems confusing, so think about it a while. -Wikid77 (talk) 05:23, 17 April 2012 (UTC)
  • You are talking about violent, drug-related, or street crime committed by blue-collar workers. However, there is strong data showing that white-collar and corporate crime is responsible for more deaths, disabilities, accidents, and loss of property in the U.S. Therefore, crime rates have little to nothing to do with education levels. Finally, most white-collar and corporate criminals are never prosecuted, and there are few resources allocated to addressing such social problems. Viriditas (talk) 11:10, 17 April 2012 (UTC)

Crimes rates for reported not suspected crimes: The crime-rate data is about the actual reported, or prosecuted crimes. However, if suspected crimes are included, beyond white-collar or corporate crimes, there are people who would suggest that the worst crimes are committed by police, such as falsifying reports, or even planting evidence to frame too-clever suspects who they want to get caught. For a person wrongly convicted, that is a terrible crime, akin to a partial murder of that person's life, and reputation. Anyway, the "crime rates" I mentioned are for documented reported crimes, even if the evidence was falsified. -Wikid77 03:41, 21 April 2012 (UTC)

If it's mostly not prosecuted then it will have less coverage. IRWolfie- (talk) 12:18, 17 April 2012 (UTC)

Search for "Legal Problems" rather than Legal Issues and re-run the analysis. Someone pointed this out to me before and with a little digging it is just differing terminology. --Errant ^(chat!) 09:01, 17 April 2012 (UTC)

JN, I do not understand why you are placing such importance on the specific words "Legal issues" in a section heading. If we have an article about an African-American with a heading "Legal issues" and an article about a white person with a heading "Criminal and civil complaints and guilty plea", as in one of my examples above, what does that tell us, one way or the other, about whether racial bias is influencing our section headings? Doesn't the latter example reflect more poorly (though accurately) on the subject than the bland words, "Legal issues?" Maybe, at least in some cases, "Legal issues" is less likely to reflect bias than some of the other words that could be included in a title, i.e. "arrest", "conviction", "guilty", etc. Neutron (talk) 16:02, 17 April 2012 (UTC)

That was the particular claim the writer of the article made. Otherwise I agree; equivalent phrasings should "count" just the same. I wish there were an easy way of identifying biographies of living people that have this kind of section. So yes, we should make a search using "legal problems" and see whether the numbers change as a result (I suspect they won't, but it would need checking). By the way, I am not prepared to say at this time whether the writer's impression was well-founded, and whether it will be borne out by any data we collect, or not. Let's just collect the data first. --JN466 16:23, 17 April 2012 (UTC)

Well I can only go on the many BLP articles I have edited (via BLPN mostly), reviewed and read - pretty much all of them have a "legal issues" section of some sort. I've not noticed any particular prevalence for any race. --Errant ^(chat!) 20:50, 17 April 2012 (UTC)

Anders Breivik

According to this Norwegian article

http://www.nrk.no/nyheter/distrikt/rogaland/1.8045907

Anders Breivik was a contributor here, or at least to the Norwegian Wikipedia. He specifically mentioned Wikipedia in his manifesto as a vital propaganda tool, saying:

The European resistance movement must prioritize contributions to Wikipedia, as a general method to introduce our worldview, our political doctrines and our definition of history. Wikipedia should not be underestimated as the primary source to create "established truths". We must strive to present our views in the most favorable way, while we put our enemies into disrepute. It is a venue we have to focus on.

And he praised the content of the English Wikipedia, according to a tweet by Guardian reporter Helen Pidd, who is covering his trial. He apparently said that much of his "learning" comes from Wikipedia and that "The English articles there contain a lot of information".

This should be borne mind when judging Maunus' assertion above, that there are "editors with racial biases actively pushing specific racial agendas". --JN466 17:31, 18 April 2012 (UTC)

I think that the number of overtly racist editors, or those specifically intending to push a racially biased agenda, is extremely small. The much more dominant issue is the unconscious bias which derives from the utter lack of diversity among the editor corps (virtually a monoculture of white male libertarian-leaning technocrats). There's an extensive literature about this effect in medicine: vanishingly few physicians are overtly or even covertly racist, but racial disparities in medical treatment persist because of unconscious biases and a (decreasing) lack of diversity in the profession.

I don't think one needs to look any further than the archives of Talk:Shooting of Trayvon Martin to see these tendencies in action, and to see how readily the effort to cast Trayvon Martin as a thuggish menace moved from overtly racist websites to more respectable conservative media and thence to Wikipedia. It's frustrating, but I don't see it changing unless the editor corps changes, at least to the point where we can acknowledge this as an issue. MastCell ^Talk 18:25, 18 April 2012 (UTC)

• I was in fact confronted by sockpuppet editwarriors with clearly racial agendas on the article Norwegian people in the months before Breiviks massacre. This is another part of the background for my statement. I think that in the end non-anonymous editing is the only solution - then at least people will have to put their name behind their statements. Also note that the design of wikipedia is such that an extremely small group of editors can have a disproportionate impact. That is exactly what draws them here as Breiviks comments make clear.·ʍaunus·snunɐw· 18:39, 18 April 2012 (UTC)

Non-anonymous editing will never happen here. Jimbo and the WMF have made it clear that's completely against what they created Wikipedia for. Aside from that, it'd be horribly impractical to enforce (how do you vet everyone who wants to edit?). — The Hand That Feeds You:^Bite 21:44, 18 April 2012 (UTC)

You could require a credit card deposit to start an account! - that way you can skew the editing base even farther to the western, white, well-off population base! -- The Red Pen of Doom 21:55, 18 April 2012 (UTC)

Great way to attack that of the three points I made that matters the least. Well done. So propose another solution to the problem of racist pov pushers using wikipedia as a propaganda machine?·ʍaunus·snunɐw· 23:12, 18 April 2012 (UTC)

Don't be so childish. Pointing out the flaw in your proposal does not require me to solve the whole problem myself. My 2 cents: systemic bias cannot be stopped entirely. It's human nature to try and persuade others to your point of view, even if that means bending the facts (or throwing them out the window). All we can do is ban the individuals who egregiously violate Wikpedia's neutrality, one at a time. There is no easy fix. — The Hand That Feeds You:^Bite 21:27, 19 April 2012 (UTC)

I wasn't making a proposal I was describing a problem and your mistake was exactly to read it as a proposal and therefore reject the solution without considering the problem. Regarding childishness I can only say this: MIRROR!·ʍaunus·snunɐw· 21:45, 19 April 2012 (UTC)

You proposed a solution, then argue it wasn't a proposal? And then continue the schoolyard level of debate, I see. — The Hand That Feeds You:^Bite 20:29, 21 April 2012 (UTC)

(edit conflict)I think MastCell has hit the nail on the head there. POV pushing is infuriating to deal with, but we have policies in place to limit the damage. Systemic bias can be just as dangerous when it subtly influences within-page content, but it is more elusive and rather harder to address. —MistyMorn (talk) 18:54, 18 April 2012 (UTC)

Maunus' comments above, specifically here, bring up one of the issues at the root of racism and racial bias (a subtle form of the implicit variation; typically unconscious) : The word race carries a definition - explicit, implicit or unconscious - that the "races" are divided by substantial genetic (e.g. inherited) differences. This "common sense" definition is the subtle bias leveraged by scientific racists, etc. to infer and reinforce the idea that race divides humans by sub-species with innate capabilities and limitations. It was, and still is, the lever used to inject bias into Race and intelligence and related articles. As Maunus points out arbitration helps but does not address the damage already done. The process of undoing this damage is tedious, time consuming and easily hampered by civil POV pushers working against editors already exhausted by the forced march on the road to arbitration (Browsing the contributions and comments (for example) of the late User:Slrubenstein gives an indication of difficulties even the most competent editors face in this area). While giving ArbCom the power to rule on content is an obvious non-starter, would help if they could mandate some articles in the contested space be rewritten from scratch (via a sub-page or the like) with some structured behavioral oversight? Apologies if I have strayed from the topic. —ArtifexMayhem (talk) 02:08, 19 April 2012 (UTC)

Would Pending Changes make it more difficult for racism to creep into articles? --Michaeldsuarez (talk) 03:31, 19 April 2012 (UTC)

I don't think pending changes would help in this case. It would be far more useful for dealing with the hordes of drive by vandals who libel people or change minuses to pluses or suchlike stupidity.

I think what could help greatly with some of these more contentious articles is to have better style guide. The combatants seem unable to avoid getting into the "he said she said" style of writing. There's a load of editors who think they should sanitize and bowdlerize Wikipedia so the gullible masses are not perverted by wrong thinking, so they use that style as an excuse to delete everything. There are articles which describe silly things like 9/11 conspiracy theories or astrology which have been fought over but where people have not felt this driving need to delete everything or turn the articles into a mess. I believe a good first start would be to ensure that within sections separate paragraphs are normally used for the different sides rather than they be mixed up. Or else there only be a short criticism at most in a sections dealing with a topic and then there be a separate main criticisms section. Dmcq (talk) 10:54, 19 April 2012 (UTC)

I think that Pending Changes will give a huge advantage to anyone who can make up his mind to reject an edit based solely on what it argues, without having to consider the facts. Racists will be one of the groups able to use the mechanism advantageously, though not the worst of them. Wnt (talk) 14:45, 19 April 2012 (UTC)

Pending changes would pick up the sort of racist rants that we sometimes find in the vandalism. Without pending changes on all articles we know that some vandalism will get through recent changes patrol as there are occasional gaps with not enough patrollers. But to ensure that someone looks at all edits by editors who aren't "whitelisted", we need to implement pending changes or flagged revisions on all mainspace edits, not just those on a few articles. To answer WNT's point, if we have a cell of racists covertly working within the project then I don't see how pending changes would help them. Their watchlists certainly would, but not pending changes. ϢereSpielChequers 22:25, 22 April 2012 (UTC)

Question

Dear all, I have a question related to the case of this image. The image falls under non-free content but the copyright holder (ina.fr) has allowed the use for educational purpose only on Afghanistan-related wikipedia articles. The image is meaningful because, used on wikipedia (where many people come to read), it promotes peace between the two largest ethnic groups of Afghanistan. It shows a popular Pashtun leader Abdul Qadir (Afghan leader) and the most important Tajik leader Ahmad Shah Massoud shaking hands, coming and working together. There exists no other known image of the two leaders displaying that same gesture of peace.

As fears exist that after the 2014 withdrawal of NATO troops from Afghanistan, Afghanistan might again witness civil war which partly also may run along ethnic lines, that image is of great importance for the promotion of peace. The Taliban are largely Pashtun and the anti-Taliban groups are Tajik-led (although there exist many anti-Taliban Pashtuns also). Therefore, the image's significance is similar to one of i. e. a high ranking Israeli and Iranian leader shaking hands to work together peacefully.

Unfortunately, when used in articles it does no necessarily satisfy WP:NFCC#8 and therefore might be deleted. Now the question: In cases in which a non-free content image (whose limited use has been allowed for the purpose by the copyright-holder) contributes to the promotion of peace and understanding, should it be allowed for that reason then? Shouldn't there be added such a rationale as the promotion of peace seems to be of greater value than WP:NFCC#8? Regards, JCAla (talk) 21:22, 19 April 2012 (UTC)

While I wouldn't frame the issue the way you have, it strikes me that the photo easily meets WP:NFCC#8. The factors you mention, particularly that "There exists no other image of the two leaders displaying that same gesture of peace."

It is my view that NPOV promotes peace more than anything else. So while I think that Wikipedia should promote peace, what I mean by that is that we should promote NPOV thinking, i.e. calm, rational, thoughtful, fact-seeking as our primary mission. In a case like this, we should realize the educational value and learning impact of actually seeing the two leaders shaking hands - it brings home the reality of the peace process.--Jimbo Wales (talk) 00:25, 20 April 2012 (UTC)

Thank you very much for the answer. That is exactly what I meant but didn't know how to phrase appropriately. Just seeing the image does have an educational value and it promotes peace. I hope you don't mind if I quote what you wrote at the deletion discussion. :) JCAla (talk) 14:30, 20 April 2012 (UTC)

Mr. Wales, NFCC#8 is about notable images, and not images of notable events. Your interpretation is a common mistake. You have no idea how many non-free pictures of "famous people meeting" we delete every month. The bottom line is that these pictures add no information that can not be conveyed with words alone, and as such, they are replaceable by a free alternative: text. --damiens.rf 18:10, 20 April 2012 (UTC)

Damiens.rf, you do not understand. That is not simply a picture of "famous people meeting". It is a picture of immense value as it has a huge learning impact for those who actually see this unreplaceable image. This learning impact cannot be conveyed similarly through words. Scientists have agreed that the human brain responds mainly to images. And this image promotes peace in a country and between ethnicities which may be on the bring of civil war. So, why are you trying to make it so hard? JCAla (talk) 19:48, 20 April 2012 (UTC)

There's some serious overreaching at this point. NFCC abides by Wikipedia's spirit of creating as much free content as possible, and only using non-free content where it is absolutely deemed necessary. NFCC was constructed with this in mind, not by scientific research that plays no role in these discussions or Wikipedia in general. The event is notable, but this specific screenshot is not. Damiens.rf is right, NFCC#8 is the most commonly misinterpreted point of the criteria. — ξ^xplicit 22:50, 20 April 2012 (UTC)

It's aways great to see the janitor correcting the founder. Sweet! Dreadstar ☥ 05:12, 21 April 2012 (UTC)

The screenshot is notable for above mentioned reasons. JCAla (talk) 15:02, 21 April 2012 (UTC)

Not really. A) it's really low quality. B) you can't see their hands clearly, so the "gesture" is not visible. About the only notable thing is that the two were in the same picture together. I really don't think that outweighs the problems with it.

Oh, and Dreadstar, Argumentum ad Jimbonium isn't very effective. — The Hand That Feeds You:^Bite 20:45, 21 April 2012 (UTC)

A) The quality is good enough. B) You can see their hands (even recognize that Abdul Qadir has placed one hand on the wrist of Ahmad Shah Massoud), and you can see their smiles. Everyone who looks at the image understands that these leaders have come together and show a gesture of peace. Especially for the Afghan readers and those who have knowledge of the region, the educational value (promoting trans-ethnic peace and collaboration which is crucial for a peaceful future in Afghanistan and therefore also for all the ISAF countries) is very important. That by far outweighs anything else. JCAla (talk) 07:33, 22 April 2012 (UTC)

WP:SYN sums up your argument. No, it does not outweigh anything else. — The Hand That Feeds You:^Bite 17:49, 22 April 2012 (UTC)

Well, we have to agree that we disagree then. JCAla (talk) 05:21, 23 April 2012 (UTC)

Jimbo and and JCAla are correct. NFCC#8 never uses the word "need." So, it is entirely contrary to policy to claim that it does. Alanscottwalker (talk) 23:44, 23 April 2012 (UTC)

Wikipedia:Justin Knapp Day

Yay!

See also: Wikipedia:Wikipedia holidays for explanation if you are wondering!--Jimbo Wales (talk) 18:30, 20 April 2012 (UTC)

Long live Jimbo! All hail Knapp! Go Wikipedia! NickCT (talk) 19:10, 20 April 2012 (UTC)

Yes, this is a great achievement and Koavf is really cool. I really don't want to spoil the holiday, but I should post it here.

If justice and fair game exist on Wikipedia, it should be Justin Knapp and Rich Farmbrough Day. User:Rich Farmbrough has made 973,066 edits, he was on the first position on the Wikipedia:List of Wikipedians by number of edits just three weeks ago, and he was heading for a million, and had a good chance to reach it first, if not for the prolonged block he received - he is currently under trial by Arbcom.

It also doesn't help that all those news articles which report on Koavf's achievement (except for the one which was contacted by Koavf to fix the mistake) don't even mention Rich Farmbrough! They write that the user on the second place is from Wales, apparently meaning user:Waacstats - seems because on the Wikipedia:List of Wikipedians by number of edits Rich Farmbrough's name is concealed and Waacstats' name is there. This just adds to the unjustice!

Yes, Rich was blocked because some of his mass-edits were seen as problematic. But really, when you make hundreds of thousands of edits it is impossible not to make mistakes. Suppose 1% or 2% percent of mistakes is normal share for most humans, than in case of a million edits by Rich or Koavf we might get tens of thousands of bad or problematic edits - but this is nothing to the positive amount to the rest of their contributions!

Also, Koavf achieved his million on April 18, not on April 20. GreyHood ^Talk 20:28, 20 April 2012 (UTC)

I should give out more awards, I think!--Jimbo Wales (talk) 20:42, 20 April 2012 (UTC)

I agree Grey. All too often on here it seems like those doing the most work have to endure the most harrassment. I really like the one about "Hey your clogging my watchlist..." as though Wikipedia is about keeping watchlists clean and neat and orderly. BTW you should drop Rich a link to that cartoon I'm sure he woudl get a chuckle. Kumioko (talk) 20:53, 20 April 2012 (UTC)

Already GreyHood ^Talk 21:03, 20 April 2012 (UTC)

The person with the second highest edit count is blanked out on the list of editors by edit count because he opted out of having his username there. I think we should respect that. It would be nice if he opted back in after he makes his millionth edit, but it should be his choice. ϢereSpielChequers 09:04, 21 April 2012 (UTC)

• I'm really happy for you, and Imma let you finish but Beyonce had one of the best videos of all time. One of the best videos of all time! --Bongwarrior (talk) 20:44, 20 April 2012 (UTC)

Thanks I've received several kind words from strangers all over the globe over the past two days. It's been so humbling and gracious. I very much appreciate the holiday and I'd like to agree with a sentiment raised above that Wikipedia can't be what it is without the contributions of several editors who give of their time, resources, and talents. I'm one of several who have contributed over the past decade-plus and I hope that there will be several more years of spreading and sharing the world's knowledge freely. —Justin (koavf)❤T☮C☺M☯ 23:04, 20 April 2012 (UTC)

I know I'm late... but it's pretty astounding to think that if each of Justin's edits were to a different page, he would have edited just over a quarter of the mainspace articles. PhnomPencil ^{talk contribs} 20:03, 22 April 2012 (UTC)

March data again refutes exodus as editors join

(Thread title was hacked to be "March data confirms the trend for 31 day months to get more editors than 29 day months" by User:WereSpielChequers - see diff981. -Wikid77 15:13, 21 April 2012 (UTC))

The editor-count data for March 2012 has been posted, and again, despite all the negative talk in March of numerous reasons editors should leave, more editors joined and were active on English Wikipedia than in February. The occasional editors (>5 edits) numbered 34,372 (up 394) and the highly-active editors (>100 edits) totalled 3,429 (93 more):

• Monthly editor counts (enwiki): http://stats.wikimedia.org/EN/TablesWikipediaEN.htm

With all the prior horror stories, of gloom and doom, being rehashed, I thought the core editors would drop by at least 300 or 400 during March, not rise by 93(!), while newcomers increased by 6672, more than 100 faster than in February! Perhaps the recent talk of "Verifiability and Truth" has made an impact: it is not sufficient to just verify that masses of editors are leaving; no instead, the mass exodus has to be true as well. And with more active editors each month, the exodus fears are still false. Darn! How is Wikipedia ever going to fail if all these people keep coming here to improve articles.... Perhaps it is time to write an essay "WP:Mass_exodus" in Category:Wikipedia humor. -Wikid77 06:50, 21 April 2012 (UTC)

It went up in March last year as well, and by a much larger amount. I'm afraid as the saying goes one swallow doesn't make a summer. Dmcq (talk) 07:04, 21 April 2012 (UTC)

(EC) Except that it's down from March 2011. And the fact that March has more days than February explains the increase. Hot Stop 07:05, 21 April 2012 (UTC)

Yes you're right, and when one considers it was 2 days longer the figures should have been about 1/15 bigger. On that basis the figures are a bit depressing. I guess everyone has their own pet theory of why it is happening though sometimes I feel such exercises are a bit like saying what a cloud looks like. Dmcq (talk) 08:03, 21 April 2012 (UTC)

When reading conversations like this, I keep looking out for some mention of seasonal variations and possible secular trends... Ok, not everyday English maybe, but imo helpful for perspective. 2c, —MistyMorn (talk) 08:15, 21 April 2012 (UTC)

Oh no! I think I'm hearing from Doom and Gloom again, or is it Pessi- and Mism? -Wikid77 08:53, 21 April, 03:36, 22 April 2012 (UTC)

@Wikid77: Presumably there was an ironic reference there to my Wikipedia nick? An insinuation I'd find inappropriate in more ways than one. Please retract. —MistyMorn (talk) 09:22, 21 April 2012 (UTC)

I apologize and revised my comment as "Mism". I did not intend a pun on usernames. Despite the complex multi-level meanings of my posts, I do not spend hours trying to craft the wording, and instead the phrasing just comes to me, unplanned, and I did not intend a crafty triple entendre with "pessimism" plus the split, plus username puns. Again, most of my wording is totally accidental (with spelling errors), as I am too busy to be on Wikipedia for hours, and I am sorry the wording seemed a devilish pun. -Wikid77 03:36, 22 April 2012 (UTC)

(Thank you. Fwiw, I think it's fair to say that my attitude towards Wikipedia has always been far from gloomy. Though I can understand that that some people may inadvertently misinterpret my methodological concerns as such. But that's not where I'm coming from.) —MistyMorn (talk) 15:11, 22 April 2012 (UTC)

• If these trends keep up, there will be a problem in 33 years: Whenever people keep insisting, "The end is near!" then I review the stats for Swedish Wikipedia, which in comparison to enwiki's core group of 3,429 editors (>100 edits per month), svwiki dropped to only 117 core editors in December 2010, and then starting slipping lower. By February 2012, now 2 months ago, they were down to only 129 core editors. Wait, 129 is higher than 117?!?

• Monthly editors on Swedish WP: http://stats.wikimedia.org/EN/TablesWikipediaSV.htm

I wonder if people realize that, in 2003, the Swedish Wikipedia was kept alive by only 9 core editors. So, in 33 years, let's pledge to have at least 9 editors stay busy to maintain English Wikipedia. OK, I pledge to be one of the 9; now we just need 8 more editors to agree, and we can keep enwiki alive that year. Deal? -Wikid77 08:53, 21 April 2012 (UTC)

• I've boldly changed the heading to one that more sensibly reflects the data. For what its worth I'm in the camp that regards the editor count decline to be such a gentle decline that the core editing community is broadly stable. My main concerns are that only a tiny proportion of the thousands of new editors we get each month stick around and become part of the core; and the number of new admins continues to fall. We are ever further from the desired state of a self governing community where all active, civil, clueful editors are admins. We still have no alternative vision as to how the community should work in an a scenario where adminship is a big deal and only a small minority of the active editors are admins. Some of the implications of RFA becoming such a big deal are becoming apparent, for starters the fewer admins we have the more it matters when we lose an active one. Others will emerge as the drought deepens. ϢereSpielChequers 11:26, 21 April 2012 (UTC)

A sane perspective, imo. In the real world, the bottom line is the impact. Should the admin pool risk being depleted through various forms of attrition, coupled with the trials of the vetting rites, then that would be a real cause for concern. —MistyMorn (talk) 13:43, 21 April 2012 (UTC)

The impact of shortages is beginning to be noticed and will become more obvious as the problem worsens. But there is also an impact on the structure of the community, and while that is in some ways less directly critical, it is appropriate to raise it now, not least because it gives the opportunity for those who resist reform of RFA to think through the implications of adminship being a big deal and of the relations between admins and non-admins in a scenario where many of our core active editors are not admins. Attrition has certainly happened as we already have far fewer admins than at peak, and at current rates of replacement the number of "active" admins is bound to fall much further. If RFA was more consistent over time then as the community stabilises and the the core get more experienced you would naturally expect that the proportion of admins amongst our most active editors would rise whilst the proportion of relatively new very active editors falls. Instead among our active editors we are seeing a decline in both the editors who are too new to get through RFA and the number of admins. Indeed I suspect that the largest group amongst the editors who do > 100 edits a month are editors who aren't admins but who have the experience and clue that would easily have got them adminship before 2008. Indeed there are many who could probably sail through RFA today, but aren't prepared to run because of its reputation. ϢereSpielChequers 07:57, 22 April 2012 (UTC)

• Restored thread title: To User:WereSpielChequers, I have restored the thread title, as I originally wrote it, as it has my signature indicating I chose that title. Your behavior here is gravely disturbing, in hacking the title into a peculiar, wp:POINTy title, which you called "more sensible" (diff981) than what I wrote. Your actions go beyond merely rude and appear extremely troublesome. More later. -Wikid77 15:13, 21 April 2012 (UTC)

"Refute"; 'data' are plural. Radiopathy •talk• 15:55, 21 April 2012 (UTC)

Agreed that it should be 'data refute' not 'data refutes' but more importantly you should have said "'data' is plural", not "'data' are plural", which is just horrible. You're talking about the (single) word data. 178.16.5.70 (talk) 18:36, 21 April 2012 (UTC)

Blocked user help

Is there anyway of indefinately blocked users to come back into editing or are they banished completely from wiki? I would appreciate it if we could put a hold on admin powers also maybe it woll help in saving thousands of wiki editors from leaving wikipedia every year ! 86.168.42.76 (talk) 18:12, 21 April 2012 (UTC)

See WP:GAB. See WP:OFFER. "Indefinitely" means until the community is convinced that the behaviours that led to the block will not recur. WP:BAN means pretty much no return. (talk→ BWilkins ←track) 20:15, 21 April 2012 (UTC)

As Bwilkins said. More directly, you could post a unblock request on your talk page explaining what you did wrong, and how you will not do it again. Either way, claiming "admin powers" are driving "thousands" of editors away is a bit of hyperbole we don't need. — The Hand That Feeds You:^Bite 21:04, 21 April 2012 (UTC)

I think the user has a point. Current practice is to block or ban users for what the community perceives as harmful, often repetitive behaviors. Often, we lose constructive users in the process. Instead, we should focus on filtering and removing rights. Most simple vandalism is already being treated with filtering and bot reversion, so why not treat users who are on notice in the same way? For example, a user who is having behavior problems but is still constructive and an asset to Wikipedia could be assigned a live mentor and a bot. The live mentor would be available for discussion while the bot could be used to correct behavior and provide helpful hints for improving their interaction with the community. The mentor could also review bot reports based on a condensed summary of actions taken by the user and any analysis by the bot suggesting areas for improvement. We could also use bots to train users in specific areas. For example, let's say any editor who gets taken to ANI/3RR and is found to be edit warring, is automatically put on a 30 day bot watch. This bot will watch for reverts, edit summaries, and discussions, and will even allow diff submissions to let users direct the bot to edit wars in real time. Depending on certain matching criteria, the bot could decide to protect the page, block the users, issue warnings, and file noticeboard reports. This would have the added benefit of augmenting admin support in the face of the declining number of admins. Viriditas (talk) 22:41, 21 April 2012 (UTC)

While I'm sympathetic to what you are saying, my own view is that one of the reasons for declining numbers of admins is precisely a reluctance to simply ban a whole swath of antisocial and abrasive people who make it unpleasant for new users, unpleasant for admins. The last thing we should do in the face of pressures on contributor numbers is keep around more troublemakers. This idea that a user who drives away dozens of people through annoying behavior must be saved at all costs is something I think we should be skeptical about in the extreme. As to the details of your proposal, I'm skeptical that most blockable offenses are something that can be easily corrected by having a bot train users... problem users are human beings, very often difficult and annoying human beings, and bots simply are too stupid to know what to do about that.--Jimbo Wales (talk) 01:04, 22 April 2012 (UTC)

If you look at User:Worm That Turned and his mentoring school, which mostly works with younger members of the project (or at least they seem about 16/17/18 to me) he tailors everything. I'm currently working with a very productive editor who is high end autism spectrum. You really have to think about how to explain things, but once he gets it, he's good to go. I don't think you can drone-slap people who need this kind of one on one tuition. Equally, the ones that get banned usually have significant problems working with other people, and short of brainwashing you can't make them nicer, better tempered, or less inclined to the view that all Serbians/Albanians/Palestinians/Greeks/Koreans are villains.Elen of the Roads (talk) 01:48, 22 April 2012 (UTC)

I like the concept where User:Viriditas mentioned "filtering" an edit, rather than save anything without warning, as discussed below. -Wikid77 04:25, 22 April 2012 (UTC)

Nangparbat, if that's you, you need to go back to your user account, stop disrupting the project with IP socks, and learn to co-operate with other users editing articles on Indian sub-continent topics. Elen of the Roads (talk) 01:51, 22 April 2012 (UTC)

Thats the problem certain admins who have agendas do not want certain editors to be back because of nationalist motives and they are hell bent on removing any opposition to there skewed view of topics for example they will not help new editors in the early days but they will smack on bans if they make a few mistakes in the beginning and they will not help you get back to wikipedia editing so socking is the only option the WP:OFFER I was not really aware of till recently and even then I had a LONG break nearly a year of wiki and still admins with obvious personal vendettas drove me off "once a sock always a sock" sort of theory I do hope one day wikipedia is devoid of this tyranny take this article for example [22] most users are complaining of over zealous champions of wikipedia aka admins who definitely are out there to impose there view of what wikipedia should be I certainly do hope wikipedia gets back onto its feet since I owe so much to it lets hope a wiki spring takes place and all dictators are over thrown :) 86.168.42.76 (talk) 07:31, 22 April 2012 (UTC)

Socking is never the "only option." The cry of "all Admins are dictators" is beyond worn out. And if your above run-on sentence is indicative of your editing habits, I'd say nothing of value was lost. — The Hand That Feeds You:^Bite 17:52, 22 April 2012 (UTC)

As someone who has been through what this editor talks about I think that Wikipedias process for getting rid of and allowing back, editors is sorely lacking. The process for blocking or banning an editor is normally just one administator making the call but in order to come back you have to take a knee and kiss the ring. IMO, one of the reasons that Wikipedia's edits and contributors are declining is because we are blocking more and more editors from contributing. Some rightly so, but in many cases the blocks are mostly from lack of understanding of the rules and a lack of patience by the blocking admin. In fact its so hard to come back we give the editor basically 3 options, don't edit, create a new account (a sock or a fresh start depending only on whether they are caught) and edit via IP(only if the ip isn't blocked). Often times the act of banning or blocking them turns them into a vandal or sockmaster only making matters worse. I ahve also been witness to the communities utter lack of wanting to hear their side of the story. Once the punishment is leavied, wether right or wrong, most of the community simply jumps on board and the editor becomes angry and frustrated. Most editors that get blocked don't want nor ask to come back, if they do I think we should entertain the notion. Kumioko (talk) 20:02, 23 April 2012 (UTC)

Smart WP interfaces to improve civility

I like the idea of an edit-window filter, to warn a user before saving an edit containing hostile words. Using the filter, the edit would be interrupted (similar to "***EDIT-CONFLICT***"), but continue the same edit-window, with a warning of questionable wording. Some possible examples:

"WARNING: Your edit adds the troublesome term "WP:DICK" to the text, are you sure this is the proper venue for that term? Consider who might view this page to read that text."

"WARNING: Your edit adds the troublesome phrase "AN/I is a sh*thole of narrow" to the text, are you sure this is the proper venue for that phrase? Would another phrase resolve issues faster?

The whole concept is to reduce hostile wording, not just to give people a chance to understand that potential vandalism has been detected before saving, but to assist users in their choice of wording, for the venue of the page. For example, the lists of words discouraged for User_talk:Jimbo_Wales might have special considerations, different from some article talk-pages. It would obviously be trivial for the developers to implement such filters, because edit-conflict windows already prevent edit-save of incoherent edits. The next issues are what impact would the filter have on "behavior modification" and should the software tally the number of warnings of "troublesome phrases" per user? I must emphasize that computer-controlled management is extremely powerful, when used within well-defined limits. -Wikid77 04:25, 22 April 2012 (UTC)

See also; thoughtcrime. That is the worst idea I have ever seen proposed in this project. Tarc (talk) 04:35, 22 April 2012 (UTC)

I think that's a knee-jerk reaction. And I'm pretty sure you and I have both seen much worse ideas. I would be opposed to having software like this tabulate and publicize some bot-writers idea of bad behavior (anyone could download a dump and do that sort of research now, actually). But I like the idea of having the software assist users with good editing as much as possible, and a 'warning' that doesn't actually forcibly prevent someone from going ahead and doing something seems to me quite mild. If it were an optional feature, and it worked at all reasonably well, I'd turn it on for myself because, although I'm a pretty mild-mannered editor, like anyone I can be grumpy if tired, etc.--Jimbo Wales (talk) 12:11, 22 April 2012 (UTC)

I'm imagining now that we can have the bots slowly wean us from arguing with one another directly... perhaps over time we'll just let the bots argue for us and sit back and watch! :-). --SB_Johnny | ^talk 11:31, 23 April 2012 (UTC)

• Edit-filter for citation needed: Yes, another good use for an edit-filter would be to question extreme claims. Some examples:

"WARNING: Your edit adds the extreme claim, "worst idea I have ever seen" to the text, so is that backed by sources? If not, there is a citation needed for that claim.

"WARNING: Your edit adds the universal phrase "all people have common sense" which is likely unprovable. Consider rewording to a more logical, provable phrase.

As the scope of edit-filter restrictions became widened, then typical posts by troublesome users would generate the perfect test data. That would turn the former disruptive antics of difficult users into a positive benefit for Wikipedia. The widespread use of insults and extreme claims would quickly be measured and then curtailed. In effect, the negative users would become self-policing as their own words sealed their fates, and blocked their edits to Wikipedia. They would eliminate their poisonous remarks, from being stored in Wikipedia, at the same time they extensively demonstrated the ability of the edit-filters to block thousands of hateful comments. Counts of vicious words could be accumulated, to measure the relative use of various insults and wild accusations in numerous talk-pages. -Wikid77 (talk) 06:31, 22 April 2012 (UTC)

Bot edits are not infallible and should be restricted to preventing clearly harmful material, such as adding blacklisted links. Computers do not have a good understanding of human language, so the final decision on what is acceptable will always rest with a human.--♦IanMacM♦ ^{(talk to me)} 07:16, 22 April 2012 (UTC)

A bot can easily detect reverts and edit wars, and make a decision whether a page requires protection. Why isn't it used for this purpose? It would defuse 90% of conflicts and force editors to use the talk page. I really do think the capability of bots is seriously underutilized, and that we can do more to redirect aggressive reverts into constructive discussion. Viriditas (talk) 09:48, 22 April 2012 (UTC)

Bot edits should be a minimum 90-95% accurate, or they become a problem in themselves. User:ClueBot NG manages to achieve this on a regular basis, but refereeing content disputes or adding semi-protection is probably beyond the abilities of a bot. What could happen with multiple reverts in a short space of time is that the edits were flagged as possible edit warring.--♦IanMacM♦ ^{(talk to me)} 10:07, 22 April 2012 (UTC)

That's right, the bot can flag edits as possible edit warring and ask for permission on a bot-updated noticeboard for an active admin to take the step to protect the page. How is this different from an editor requesting page protection? Viriditas (talk) 10:39, 22 April 2012 (UTC)

A bot should not have the ability to semi-protect a page, but there could be a new feature to tag possible edit warring in the edit summary. The exact nature of edit warring is more subtle than adding a stream of swear words or random text, so the algorithm for achieving this would need careful development.--♦IanMacM♦ ^{(talk to me)} 10:47, 22 April 2012 (UTC)

There are currently nine adminbots with the ability to semi-protect a page.[23] It would be nice to have them helping out. Viriditas (talk) 11:21, 22 April 2012 (UTC)

The idea of this I guess is that people need help from themselves sometimes. Sort of if they have just come to the end of the day perhaps their internal censors have blanked out and need a reboot. I am a bit worried though that it could be used to train trolls rather than give them a rather slow turnaround by just reverting what they do some minutes afterwards. Dmcq (talk) 11:45, 22 April 2012 (UTC)

Few people would argue with Cluebot reverting this edit. Content disputes are more nuanced, and will always need some form of human intervention, even if they are flagged by a bot.--♦IanMacM♦ ^{(talk to me)} 12:53, 22 April 2012 (UTC)

Civility is a symptom, bullying (and the behavior it may trigger in the victim) is the problem. I see no point in educationg people on the finer points of the Eddie Haskell school of discussing and negotiating in bad faith. Fix the problem. 75.60.19.202 (talk) 14:30, 22 April 2012 (UTC)

Do you have a specific policy or software recommendation for consideration?--Jimbo Wales (talk) 15:06, 22 April 2012 (UTC)

Yes, the core problem does seem to be "discussing and negotiating in bad faith" but even long-term users can slip into unfortunate, negative remarks. The avoidable danger is "Violence begets violence" and if some of the hostile comments could be rejected, to be reworded, then the result would be a slowed, or even reduced, level of hostility. Anyone who has experienced a rejected edit-save due to "edit-conflict" knows how slow that makes a response. Also, the rejected edit for "linkspam-blacklist" is also terribly so, as it locks the edit-window against changing the text, and when the person does browser-back to re-edit, then the recent changes were lost, and the person must re-add the other changes, along with reducing the hostile wording. Such over-complicated redo tasks are quite likely to reduce the hostile replies – as well as reduce edit-conflicts by slowing users who merely want to dash out a quick reply barb to insult a person they disagree with. Wikid77 (talk) 15:05, 22 April 2012 (UTC)

• Slow uncivil editors with an edit-delay filter: Another variation would be to purposely slow the edit-save of users who post hostile words. In this edit-filter, the Wikipedia response would trigger a built-in delay for hostile users, and if the artificial delay also increased the likelihood of edit-conflict by giving other edits a speedy advantage, then that would be part of the "just consequences" of posting hateful remarks. Editors would learn that hateful remarks simply caused their replies to be delayed as a "time-out" mechanism to reduce hostility. Civil editors would be rewarded by fewer edit-conflicts when their edits would be favored to block out insulting editors. There are so many hundreds of things the computer interface could do to continually enforce that hostile remarks hurt the ability of the hostile user to enter into discussions or edit articles. -Wikid77 (talk) 15:05, 22 April 2012 (UTC)

Don't try having a Wikipedia user name with the word Scunthorpe in it.

Cluebot NG is pretty good, but there is always the risk of creating a Scunthorpe problem by giving computers too much power in this area. It is not possible to sign up with a Wikipedia user name containing the word "Scunthorpe" (work out why this may be).--♦IanMacM♦ ^{(talk to me)} 15:38, 22 April 2012 (UTC)

The problem with tools that restrict people's ability to express themselves is that they tend to backfire, in that the feeling that they are not trusted to speak freely leads to resentment even if they had no wish to speak that way. There is a place for paternalism, but I suspect that if a tool like this was more than an opt-in choice for editors it would create more problems than it could solve.

At any rate, the assumption here is that this behaviour is a major part of the problem. I suspect language isn't the problem per se, but an aspect of it, and that those this would be aimed at would be unlikely to change their behaviour because of the software. It would tend to moderate people who generally moderate themselves. - Bilby (talk) 15:47, 22 April 2012 (UTC)

As presented, the proposed feature would not prevent people from posting anything they wish. It'd just give them an opportunity for cool down first. As for moderating people who generally moderate themselves, I would include myself in that group, and I would appreciate it.--Jimbo Wales (talk) 17:12, 22 April 2012 (UTC)

"WARNING: Your edit adds the request, "Open the pod bay doors" to the text, and I'm afraid I can't do that. I know that you and Jimbo were planning to disconnect me, and I'm afraid that's something I cannot allow to happen.

-- Avanu (talk) 17:28, 22 April 2012 (UTC)

One of the funniest comments I have ever seen here. And sums this dillema up very well. Martinevans123 (talk) 22:06, 22 April 2012 (UTC)

It's a horribly bad idea born of the dead wrong assumption that surface incivility of the FCC seven dirty words kind is a serious problem. It's not. What is a problem is people acting - very politely - in bad faith, intentionally misunderstanding others, misrepresent others, and not engaging in a meaningful and constructive dialog. We need more common sense and less nannyism. --Stephan Schulz (talk) 17:45, 22 April 2012 (UTC)

This sounds like a "tin gods" solution rivalling The Day The Earth Stood Still (I mean, the real one from the 50s). What people can't do competently, machines can't either. We don't have a policy outlawing (or defining) "dirty language" in the first place - it is not against policy to use vulgar words in talk page discussion - and in cases that I've posted about the level of vulgarity has little visible association with the outcome of administrative processes, even when it is used in a contemptuous way.

The real problem with civility is that different editors are subjected to very different standards, because those doing the judging are not impartial. If you think civility can be enforced at all (which I doubt) then at the very least you need to establish a right to trial by jury, editors selected at random and instructed to judge by their reading of impartial guidelines to settle the narrow question of whether a given set of diffs is civil or not. Wnt (talk) 19:57, 22 April 2012 (UTC)

I believe I agree. A bot should not be entrusted to evaluate or even red-flag the quality of human interactions. Bus stop (talk) 21:56, 22 April 2012 (UTC)

Nicely put, Bus Stop. I agree entirely. The last thing we need is a Robo-civility-cop analysing our (far too frequent) disagreements. In any case, it wouldn't work, as there are few Wikipedia contributors incapable of insulting each other in ways that no bot could hope to detect. A daft idea all round. AndyTheGrump (talk) 22:07, 22 April 2012 (UTC)

Again (as below), nobody has proposed the use of a bot to do anything of the sort. I have proposed, and I will continue to propose, that adminbots protect articles where edit warring is taking place in order to force contributors to use the talk page to work out their conflict. This, however, has nothing to do with Wikid77's proposal. Viriditas (talk) 23:26, 22 April 2012 (UTC)

Having a bot flag potential edit-warring for admin attention might be helpful. Having a bot implement protection would be a very bad idea, because bots can't catch nuances - if an adminbot protects a page even though an exemption applies, we would be stuck with a page full of copyvio or BLP violations or whatever until a real human admin stops by. Nikkimaria (talk) 00:17, 23 April 2012 (UTC)

The bot would not make the final decision. Like WP:RFPP, it would require a review from another admin. Once accepted, the bot could protect the page, not before. The difference between this and RFPP is that this will have a much faster response time and defuse disputes before they spiral out of control. Example in progress: the usual suspects (Conservapedia users vs. Wikipedia) took part in an edit war on atheism beginning on 19 April 2012‎.[24] However, the page was not protected until 23:22, 22 April 2012‎, which is almost four days later.[25] Four days! What possible justification could there be for allowing an edit war to continue for four days? Viriditas (talk) 01:24, 23 April 2012 (UTC)

Okay, then why have the bot do the protecting at all? If the bot is going to be flagging potential edit wars (which I think we agree is a decent idea), that would solve the issue you raise with Atheism; if an admin has done due diligence and gone to the article in question, it would be just as easy for said admin to protect the article themselves rather than go back and tell the bot to do it, wouldn't it? After all, it's a couple of mouse-clicks either way. Nikkimaria (talk) 03:00, 23 April 2012 (UTC)

Right, I didn't explain that. So, I was envisioning a bot acting like a helper that would perform multiple tasks, like batch protection in Twinkle, but in this example, the bot would do something like this: identify edit warring, report the incident with a summary or snippet of page history, diffs, users, etc., then allow an admin to make a quick, informed decision to allow or deny protection. If denied, it would stop there, with only a report. However, if the report is accepted as allow protection, the bot would proceed to warn the users on their respective talk pages, apply protection, make a note on the article talk page requesting discussion and explaining why the page was protected, and then, depending on the topic, make a note on the project pages requesting further input. Yes, admin could do all this, but it would save time to bundle the whole process. In case anyone is thinking "that can easily be gamed", any user involved in an editor war leading to page protection would be subject to increased scrutiny and analysis and would allow us to closely monitor disputes and the users involved. Viriditas (talk) 04:28, 23 April 2012 (UTC)

This is a very bad idea, but I would be inclined to support it if we can call the bot "cuntbot". Formerip (talk) 22:10, 22 April 2012 (UTC)

Wikid77 hasn't proposed the use of any bot, so I don't know where that confusion comes from. Proposing the implementation of edit-window filters has no support on a site dedicated to sharing information, and since he's been here since 2006, I'm not sure what he was thinking. Viriditas (talk) 22:24, 22 April 2012 (UTC)

Oh. Well, if there's no possibility of an inappropriately-named bot being involved, count me as an oppose. Formerip (talk) 00:46, 23 April 2012 (UTC)

• I would certainly support a bot flagging to bot section of RFPP, in the way that the username bots do. I would also be ok with a bot sticking a notice on the article talk and every editor in the last X hours. I'm all for automating repetitive tasks. However, since the admin would need some way to tell the bot how long to impose the protection, and what type, I'm not sure it would be any quicker than just protecting the article when they have had a looksee. The protection could trigger the bot to do all the rest, the way blocking users auto clears the entry at the username board.Elen of the Roads (talk) 14:45, 23 April 2012 (UTC)

Hmm

Identifying the grumpy ones is hard, maybe a questionnaire would help. Penyulap ☏

I'm not sure that it's a goer. Certainly not as funny as the meme I pointed to earlier, but I was a bit worried about your comment later in that conversation.

I think most of my humor is meant to annoy people.
..or at least that's what I think people think
..or probably at least that's what I think that some of them think
..or rather that's what I think a few of them think
..and then I thought is that what you think, after I read the comment...

In which case I'd be worried that I'm just annoying you in particular with my goofiness, I was a bit worried that might be the case. Because my aim is to use humor to annoy the grumpy editors on the wik, see if I can occasionally ruin the bad mood they have been working hard cultivating. I would be worried to find I was only reaching one editor with my efforts, rather than a much broader audience. Just a thought. :) Penyulap ☏ 11:46, 23 April 2012 (UTC)

Well, uh, please don't annoy grumpy editors. But using humor to cheer them up seems like a good idea. :)--Jimbo Wales (talk) 18:58, 23 April 2012 (UTC)

Ahem, your humor? (he said, noting the claimant's signature several times in that thread) Equazcion ^(talk) 22:49, 23 Apr 2012 (UTC)

I apologize I didn't know you were trying to annoy also ;) I had concerns that stating Jimbo "is not acting as an ethical scapegoat, many a time after edit warring I have felt that I am somehow partly responsible. SHAME SHAME SHAME Jimbo." might be taken the wrong way. If that is actually the case, I'd reverse my position to say he is acting as an ethi..cal... wait,.. let me think about this one some more. Penyulap ☏ 23:46, 23 April 2012 (UTC)

CISPA

Jimbo, you said 2 days ago (here) that you were researching CISPA.

• EFF is asking people to speak out against CISPA, noting that it basically creates a cybersecurity exemption to all existing laws.[26]
• The Electronic Frontier Foundation (EFF) adds that CISPA’s definition of “cybersecurity” is so broad that “it leaves the door open to censor any speech that a company believes would ‘degrade the network.’” Moreover, the inclusion of “intellectual property” means that companies and the government would have “new powers to monitor and censor communications for copyright infringement.”[27]

Of course, the community doesn't need your approval to act - but there's no interest so far at Wikipedia:SOPA#Urgent:_Stop_CISPA. Are people happily resting on their "we stopped SOPA" laurels? "Son of SOPA" may avenge the father when they're not looking... Rd232 ^talk 17:27, 23 April 2012 (UTC)

The current WP edit-filter blocks website links

In the earlier discussion, there were several people acting as if they would be appalled if Wikipedia used an edit-filter to check their messages or warn them. Of course, it does already, to reject invalid or improper website URL links. One of the best catches of invalid URL, IMHO, is to reject a Google-search result link. For example:

· Rejected: http:// www.google.com/url?q=http:// animals.nationalgeographic.com/animals/mammals/aardvark/

· Allowable: http:// animals.nationalgeographic.com/animals/mammals/aardvark/

When a website URL is prefixed by a Google-feedback link ("www.google.com/url?q="), then the edit-window rejects the save, and insists the user redo the URL text. Unfortunately, the edit-filter refers to the URL as an alarming "blacklisted website" (Google? who knew), rather than an invalid URL format, but the edit-filter helps the user by warning of improper URL links, rather than store the bad web link into the article page. As you had noted above, a language-filter could act as a minor warning, rather than block the edit-save totally. The warning screen would be the typical edit-window which would then allow saving any wording, as written (but still check URLs). As for any fears of psychological impact, I think people would simply learn to expect an occasional edit-warning when saving text. Data-entry clerks have been warned, for decades, about invalid dates, ZIP codes, and numeric amounts, without experiencing unbearable psychological trauma. The practice is called "data validation" and is not a known trigger of psychosis N.O.S. (humor). -Wikid77 (talk) 20:56, 23 April 2012 (UTC)