User talk:Jimbo Wales/Archive 125

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 120 Archive 123 Archive 124 Archive 125 Archive 126 Archive 127 Archive 130


A new translation project

Jimmy, I was wondering if the idea of forming an agreement with google and setting up a new sister project dedicated to translating articles between wikipedias would appeal to you. I think translation is one of the most important assets in getting information to spread and to reduce systematic bias. I want all of the other wikipedias to have assistance like the Indian and Arabic wikipedias in translating articles from English and also us translating articles from other wikipedias. I'd like to scale Wp:Intertranswiki which I started a while back into a shared big sister project which would file all of the articles on each of the wikipedias and give wikipedians access to the translated versions in many languages or something and increase the level of interwiki collaboration over translating. Any thoughts anybody?♦ Dr. ☠ Blofeld 20:45, 26 January 2013 (UTC)

Interesting concept, but article content is so dynamic that maintaining a stockpile 100 or whatever constantly changing translations of the millions of articles would be nearly impossible, one would think. Maintaining links to the various language versions with some sort of easy auto-translate function would seem a better approach, in my opinion. Carrite (talk) 22:49, 26 January 2013 (UTC)
Yes, I'm very interested. It's important to take note of existing work in this area (including by Google). Carrite, while I agree that it's interesting to think about how to make articles in different languages more similar through enhanced translation, but I think that's very far down the line, and like you, I'm a bit skeptical. However, that doesn't negate the fact that it would be very nice if users who wish to translate from English to Arabic had some enhanced tools to help them do it. It seems likely to me that if we could accelerate the growth of articles in Arabic, then natural SEO would bring in more users to help maintain and further improve those articles.
Here's an example of what I would like to see, and I think this is pretty lightweight. Here is a sample redlink in Hungarian Wikipedia. It would be nice if that page automatically loaded a machine translation from Google but in a visual edit window with a simple message at the top saying "Hungarian Wikipedia does not have this topic. Here's a translation from English, which is probably pretty bad. Please read over it, fix any grammar errors, and hit save at the bottom. Then other editors can help to begin improving it. Hungarian Wikipedia thanks you for your help!"
That translation would look something like this: Google translate.--Jimbo Wales (talk) 10:54, 27 January 2013 (UTC)
Jimbo, you wrote It would be nice if that page automatically loaded a machine translation from Google--with all due respect, it would not be nice but disastrous and it would be the fastest way to bring down the quality and the credibility of Wikipedia in all of the languages GT would set foot in. On behalf of Translators without Borders (TWB) I oversee the translation efforts into 19 languages (so far) and I can assure you this is a wonderful, efficient system, with qualified and dedicated human translators. I agree with Doc James that the process needs improvement at the integration stage; one of the solutions we've been trying to implement is to recruit Wikipedians in each target language. If you take a look at our progress over the past few months, you will see the number of translated articles awaiting integration (orange). I envision Wikipedia initiating a worldwide recruitment program to raise awareness of this need for integrators. We of TWB have been doing our part, promoting this worthy project; the same would bring results on the Wikipedia side if concerted efforts were made. Again, I strongly advise against using GT. Ildiko Santana (talk) 21:56, 27 January 2013 (UTC)
However if it is clearly marked that this is a machine translation and thus is not going to be very good this might encourage more of our readers to translated or write articles. A pilot would be a good idea to see if this does increase editor numbers and if people like it. I would not see us adding these translation directly to Wikipedia themselves. They would need to go to a different area or simply be loaded on an as you need basis. If someone hits the edit button on one of these pages up would pop an explanation on how to translate the article from English / another language or write from the ground up. These two projects (this one and the human translation one) could potentially mesh together. Doc James (talk · contribs · email) (if I write on your page reply on mine) 13:39, 28 January 2013 (UTC)
Google's so-called "translations" are at least 50% gibberish, especially if the target language is something other than English. Usually it is harder (and less fun) to produce a good text starting from the Google nonsense than to just translate from scratch. If you haven't done any translations yourself, please do not recommend something as nonsensical as using Google's services instead of translations. —Kusma (t·c) 11:34, 27 January 2013 (UTC)
It depends on the language pairs. Of course I agree with you. Each language should be able to use the feature or not as they wish, and also to choose the most useful language to use in their context.--Jimbo Wales (talk) 11:39, 27 January 2013 (UTC)
Yes, agreed, the visual window feature to be able to access a missing article in any language I think would be tremendously productive if translation could be perfected. I think that would be the first priority is to introduce the technology that anybody searching for an article on any of the wikipedias, like the Lily Cole article on Hungarian wikipedia example, that they can access automated translated versions of missing articles from other wikipedias which supplement that wikipedia until they can start the article themselves. OK, google translation is far from perfect, but the google guys are steadily improving it and I think having a translated version which is instantly accessible, even if rather poor, would be more beneficial than no access to the information. The average visitor to many of the language wikipedias might not think to visit english wikipedia and then use google translation, so I think the feature should be instantly available to any wikipedia in any language including ourselves. For instance if a German actor is missing on here that the red link mentions the German wikipedia article. "Click here" to view in English. The biggest problem of course might be translation dumps from imperfect translation, but a warning could be given to ensure that it is proof read and copyedited.
Carrite, I didn't mean a permanent stockpile, I meant some sort of technology which feeds off the live articles which allows anybody to view a google translated version and a team of editors in each language to assist with proofing them. Sort of like the toolkit but in an organized project which is geared to helping editors translate articles and have automated translations checked. Google could use it too to gradually improve their automated translations by working together as a team across the many languages. Actually google translate for a lot of languages, like French and Spanish, is considerably improved and is often 90% fine now. German is still pretty bad but it's getting better Portuguese was diabolical previously but has improved much of late. Some of the lesser languages are still pretty awful though but anybody who remembers the old Babel fish. Systran soft days it has come on leaps and bounds. The worst I think as you say Kusma are the non English language conversions, like Japanese to Belarusian or something. I just think a large-scale translation "house" so the speak would be a great thing for freeing up information and increasing collaboration between wikipedias and that we're all working together to provide knowledge. Of course we can all think of possible flaws and problems as with any project, but I'd be interested to know what the google guys would think of such a proposal. I believe that anybody in any language, whether it is Zulu, Mandarin or Breton, has a right to a perfect translation of information in any language and to be able to access any article across all of the wikipedias in their own language and to be able to have the facility in which they can do so with translation support.♦ Dr. ☠ Blofeld 12:08, 27 January 2013 (UTC)
I think for such a project to truly reach its potential we'd need more than to work together with google. I think we'd need to involve language schools and institutions, and to involve pupils in various languages to assist in translation for practice. I know one of our editors Thelma Datter who is an English teacher in Mexico has given her pupils articles to translate into English from Spanish and it has been a win-win situation I think.♦ Dr. ☠ Blofeld 12:45, 27 January 2013 (UTC)
Certificate for translation completed in 2012

────────────────────────────────────────────────────────────────────────────────────────────────────Back in 2011 we at WP:MED began formally collaborating with an NGO called Translators Without Borders. So far more than a million words of text has been translated (a total of more than 150 translations in one of more than 30 languages). How we do it is each article is broken into parts and each of these parts is translated within the ProZ plateform followed by them being put back together and the whole article checked again by a lead translator. All the translators involved are verified as competent by a professional in the field. We are only translating article that are of top importance in medicine and that have been brought to GA/FA. A table of the efforts can be seen here [1]

What problems have we run into: In Swedish some want content built from the ground up not translated [2] In Polish they have specifically used different templates for references to make translation more difficult. The infoboxes do not work in most other languages. There are a few others which I can elaborate on for those interested.

What we need help with: I am also a little behind on getting the translated articles integrated. I really need to hire someone, I think, as I have been unable to find a volunteer to take on adding them once completed. Have just received a small grant from the Indigo foundation to support developing translation capacity in East Africa. Also need help within specific languages once the article is their for localization. And we also need people to integrate old content with new content if their is some old content already present. People get upset if someone outside their community overwrites content already present even if of poor quality. Sign up is here with language ability [3]. Finally we need some tech support. To raise funds we need to be able to give results (ie how many people are using these translations). We have the results in English ( about 10 million for the 80 articles. But MrZs tool does not work in any other language. Have asked him if he or we could improve it a few times [4] He has stated that he will get to it for about a year now. Even have a volunteer programmer ready to jump in. We got a nice write up in the WHO bulletin following going their to promote these efforts [5]. Hopefully we will see this grow.

With respect to Google they tried a pilot project to sort of do the same back in 2010 [6] but it did not pan out. Have tried to pitch what we are doing both to Microsoft and Google but they are really hard to contact. I was thinking that our translation could be useful for them to build better "translation software" and if they provided some funding we could generate content faster. Regardless we will continue on slowly slowly. Drop me a note if you wish to become involved or simply jump in :-) Doc James (talk · contribs · email) (if I write on your page reply on mine) 14:53, 27 January 2013 (UTC)

It's really easy for me to contact people at Google (or, since Marissa Mayer is now CEO, Yahoo) so if you have a specific polished proposal for them, let me know (preferably by email) and I can try to help.--Jimbo Wales (talk) 21:48, 30 January 2013 (UTC)
Highly interesting I think. But I think that like with writing wikipedia we'd need to rely on the masses who are competent amateurs in language translation, although having a team of professionals as well to overlook things would be great. I definitely think there would be high demand for a translation "house" as I put it where editors wanting to translate articles can go to have articles proof read and to assist with people in easing transferring content between wikipedias. If google understands the important of translating in languages like Arabic and Hindi and had attempted to launch something to this effect in 2010, I think they might show an interest in assisting with setting up a project which unites our resources and tries to organize a translation site to spread knowledge. I think with our people resources and google's technical resources, and other lingual experts overlooking it we could produce a successful project which would grow over time. Not sure if you are in contact with any of the google guys Jimmy, but it would be good I think if the foundation could try to contact google in light of the current work google has been doing with translation and consider scaling it to all languages, and both to and from each wikipedia. I'd like such a project to unite all wikipedias and attempt to "break down language barriers". Doc James, Jimmy and anybody else here, I'm willing to make a proposal at some point if google are interested in proposing a new WikiTranslation project or something. I think it would be worth asking around for a bit to see what the sort of response would be. The way I see it, translation is arguably the most important component to building a broad, comprehensive encyclopedia across all of the languages and the key to achieving our goals by breaking down the language barriers and trying to promote content derived from any language in different languages.♦ Dr. ☠ Blofeld 16:58, 27 January 2013 (UTC)
Given the large number of sometimes very underdeveloped wikipedias we have, I definitely think it would be a great idea to have some sort of human-reviewed translation service available, particularly for articles which might be of specific importance in those language groups, like medical concerns, and some of the really "main" articles, like, for instance, pretty much every article in Encyclopedia Britannica, particularly those in the "Macropedia" section. I think it would be a wonderful feather in the Foundation's cap if the foundation could boast that it has made at least some fairly comprehensive basic information regarding major current healthcare topics, for instance, available to anyone with access to the net in all the languages it has wikipedias in. The biggest problem will probably unfortunately be finding people to do the human-reviewed translations, particularly for some "linguistic" groupings, like maybe Simple English, which may not have the vocabularly we think of an generally required to convey some of the more complicated and sometimes abstruse information. John Carter (talk) 17:29, 27 January 2013 (UTC)
If someone wishes to put together a proposal of exactly how this would work we can continue discussion. This of course will require buy in from each language version and from the Wikimedia movement as a whole. This may be hard to achieve. Even in the project I am working on which deals with only 80 articles pertaining to medicine overcoming resistance was a bit difficult. This sort of project is especially difficult as discussion would occurring in languages we do not speak.Doc James (talk · contribs · email) (if I write on your page reply on mine) 18:14, 27 January 2013 (UTC)

Automatic text translation has two major problems: (1) Texts often contain unidentifiable (or easily misunderstood) elements such as proper names, mathematical formulas, or quotations in a third language. (2) Computers often parse the structure of a complicated sentence the wrong way. Each of these two problems exacerbates the other.

I think these problems can be solved for Wikipedia's purposes. For (1), it is enough to have compulsory markup for problematic text elements. For (2), there exist various controlled natural languages such as Basic English. (Unlike Basic English, one would probably not want to reduce the vocabulary but only the range of permissible grammatical constructions.)

The most natural approach would be to extend the functionality of the Simple English Wikipedia so that the required markup can be added, the grammar of every sentence can be checked automatically to conform with the restrictions, and every conforming article will automatically be available in all translation target languages.

In practice, an editor who has entered a new sentence would get immediate feedback if there is a parsing problem. The software's hypotheses about word forms would be made explicit, and the editor would get a convenient method for correcting them, tagging what a pronoun refers to (in cases of doubt), tagging words with the shades of meaning required for proper translation, replacing words by simpler synonyms, adding new technical terms to the dictionary, or marking the fact that something that looks like a common idiom is meant literally. Once the sentence parses successfully, the software would present one or two rephrased versions of the sentence, either in English or in the user's target language of choice, to facilitate checking that it's parsed correctly.

Editors speaking various target languages would be in charge of checking the tags for shifted or split meanings. E.g. native Spanish speakers would add or correct markup for distinguishing between the ser and estar meanings of be.

This could be implemented step by step. The first step would be the development of an appropriate version of simplified English (probably by proper linguists) and its implementation on Simple English – first informally, then through automatic checking. Then various target languages could gradually be introduced. This is probably the hard part. It could be addressed by providing the dictionaries and offering an API for linguistic researchers who want to experiment with translation engines.

Such a plan could cause some resistance from the Simple English Wikipedia, which would have to change a lot. But it would make the project much more useful and give it a lot of exposure. Writing for a fork of the English Wikipedia is one thing, writing in dozens of languages simultaneously is a completely different thing. Hans Adler 12:31, 28 January 2013 (UTC)

I doubt anyone would welcome this sort of idea into any of the current project which is why I think the proposal is for a separate project. I am not yet convinced that this would work and thus not yet convinced it would be a good idea. I have seen issues with machine translation for anything complicated. But would be happy to see a proposal for a pilot put together. Doc James (talk · contribs · email) (if I write on your page reply on mine) 13:08, 28 January 2013 (UTC)
  • As a Wikipedia editor who has been involved in machine translation since the early 1970s, I am pleasantly surprised at the amount of common sense which has been expressed in the above comments. While Google has indeed made enormous progress in the field thanks to their "statistical" approach, success is still mainly based on two factors: 1) the level of similarity between the two languages involved; and 2) the syntax of the text to be translated. In other words, languages with common origins such as the romance languages (French, Italian, Spanish, Italian, Portuguese, Catalan, Romanian) will normally provide a relatively high level of translation while those of different origins (say French into German or Japanese into Russian) will provide far less satisfactory results. But the second criterion, syntax, is equally important. Well-structured sentences (preferably not more than 20 to 25 words) produce far better results in machine translation than long rambling sentences with lots of dependent clauses. With English as a source language, it is interesting to note that plain English is ideally suited to machine translation. In the 1980s, the Xerox Corporation developed what they called "Multilingual Customized English" as a basis for translating their user manuals and technical documentation into a wide variety of target languages using machine translation. Their approach was highly successful. Dr. B's proposal is therefore likely to be most successful, whatever the language combinations, if the article to be translated is written clearly and unambiguously in straightforward prose, free of academic frills. Finally, as someone who frequently translates articles from most of the European languages into English, I must say I usually prefer to work straight from the original rather than from a machine translation which needs to be revised. (PS: Though have failed to follow my own recommendation in writing these comments, I am nevertheless happy to see Google has translated them into an understandable level of French. Here is the raw output:
"Comme un éditeur de Wikipedia qui a été impliqué dans la traduction automatique depuis le début des années 1970, je suis agréablement surpris de la quantité de bon sens qui a été exprimé dans les commentaires ci-dessus. Alors que Google a en effet réalisé d'énormes progrès sur le terrain grâce à leur approche «statistique», le succès est encore principalement basée sur deux facteurs: 1) le niveau de similitude entre les deux langues concernées, et 2) la syntaxe du texte à traduire . En d'autres termes, les langues ayant des origines communes telles que les langues romanes (français, italien, espagnol, italien, portugais, catalan, roumain) fournira normalement un niveau relativement élevé de la traduction, tandis que ceux d'origines différentes (disons français vers l'allemand ou le japonais en ) Russe donnera des résultats beaucoup moins satisfaisants. Mais le second critère, la syntaxe, est tout aussi important. Des phrases bien structurées (de préférence pas plus de 20 à 25 mots) produisent des résultats bien meilleurs que la traduction automatique de longues phrases décousues, avec beaucoup de phrases subordonnées. Avec l'anglais comme langue source, il est intéressant de noter que bon français est idéalement adapté à la traduction automatique. Dans les années 1980, Xerox Corporation a développé ce qu'ils ont appelé «Français multilingue sur mesure" en tant que base pour la traduction de leurs manuels de l'utilisateur et la documentation technique dans une grande variété de langues cibles en utilisant la traduction automatique. Leur approche a été très réussie. La proposition de M. B est donc susceptible d'être plus efficace, quelles que soient les combinaisons de langues, si l'article à traduire est rédigé de manière claire et sans ambiguïté en prose simple, sans fioritures des universitaires. Enfin, comme quelqu'un qui se traduit fréquemment des articles de la plupart des langues européennes en anglais, je dois dire que je préfère travailler directement à partir de l'original plutôt que d'une traduction automatique qui doit être révisée."
I am therefore certainly ready to do what I can to forward the proposal. --Ipigott (talk) 21:55, 28 January 2013 (UTC)
I can forsee the following problem- I (or anyone) uses this function to take an article from say Arabic Wikipedia for instance and translates into English. Some editors or admins here on En Wikipedia decide despite its existence in Arabic it is not notable... huge debate with arguments and insults abound; two groups coalesce, not exactly in complete opposition but at the same time they are not mutually compatible in the same arguement- one with the dogmatic stance of "if it exists in another Wikipedia it is notable" and the other with "if it is not notable in the ENGLISH-speaking-world, then it is not notable for ENGLISH Wikipedia" (a sentiment that has been expressed before on AfD. This leads to unneeded drama, AfDs, and even edit wars at guidelines, policies, and the 5P by editors in the various dogmatic groups to push their views into !law. Of course those balanced, look at each article on its own merits case-by-case editors... well, they are admirable, but often get yelled down by fanatics on both sides depending on how the winds blow. (talk) 22:09, 28 January 2013 (UTC)
I don't think there would be any automatic translating from Arabic to English, only the other way round. One should think of this simplified English as a programming language. There would be a bias towards the Anglosphere, but I don't think it would be half as bad as you seem to imagine, and the entire thing is only meant as a crutch to create minor articles for the languages that would normally not have them.
As I see it, this is about the long tail. It's about making an article such as 1980 Turkish coup d'état accessible to a reader who can only read Urdu. Hans Adler 11:17, 30 January 2013 (UTC)
There should be an obvious issue with Google Translate that its results are copyrighted, and displaying them in a Wikipedia frame, especially if reworked to remove Google stock text and logos, might be subject to some challenge. Developing an in-house automatic translation for Wiki would be highly desirable, not only for its use on-wiki, but as an open source software application in general. Resources such as the curated set of other language wikilinks at the bottom of each article seem like they could be helpful. Maybe people running a GPL project from Comparison of machine translation applications could be recruited? I just hope that for any two given languages there aren't a thousand patents on the very idea of translating one from the other... but it is what one expects, nowadays. Wnt (talk) 13:42, 29 January 2013 (UTC)
If the source material is CC BY-SA, then running it through Google Translate results in text that is CC BY-SA. Maybe. It's not obvious. But it's also not a real problem. In any formal partnership, obviously Google would agree to license the results under CC BY-SA.
I do agree that open-source machine translation is extremely valuable, and that concern about relying on proprietary solutions is a valid concern.--Jimbo Wales (talk) 21:48, 30 January 2013 (UTC)

Is there some kind of table/graph we can put together so we can compare which language-to-language translations are the best? Points can be given by how closely related the languages are etc. The result of this will be that if (for example) we find that when translating from French to English we get a more accurate translation than from any other language, we may be able to start mass-translating articles, while ending up with an almost correct article. Who knows, we may even find out that if you translate from Hebrew to Russian, and then from Russian to German, you end up with a much more accurate translation than from Hebrew straight to German. Whatever the actual outcome of such an endevour, I think this sort of analysis would be extremely handy.--Coin945 (talk) 11:28, 30 January 2013 (UTC)

  • Yes, a 3-language translation does work better in some cases, but that is a kludge, and good direct translations should be possible, with the help of more volunteer software developers here. -Wikid77 (talk) 14:45, 31 January 2013 (UTC)

Jimmy, thanks for the offer of contacting google. Give me a few days and I'll have something drafted for you to forward to them. I definitely think this is worth pursuing and am willing to put in the time to try to get something moving on this,♦ Dr. ☠ Blofeld 16:38, 31 January 2013 (UTC)

Find other websites like Google Translate in 2004

We should keep looking for better translation software, such as the old Google Translate of 2004. Before the numerous re-workings of Google Translate, it was possible to use it to dynamically translate text, so as a webmaster, I extended one customer's website in 2004/2005 to have other-language buttons to invoke Google Translate on 25 pages into 7 other major languages, and the results were basically flawless in every language, after careful phrasing and bypassing of proper names, but no longer. Instead, the dozens (hundreds?) of modifications to Google Translate have produced "50% gibberish" results (most obvious in German output texts which have omitted the main verb of many sentences for years, but not in 2004). Long story short, it has become almost impossible to write a simple paragraph and get it to auto-translate correctly, now, into just a few other languages. We need to find better translation systems, perhaps depending on language-pair proficiency, and use whichever software works best for a particular sets of languages. -Wikid77 (talk) 14:45, 31 January 2013 (UTC)

Bypassing translation of names

With automatic translation of text, it is often necessary to bypass (or "escape") translation of names which have meanings in other languages, such as "Smith" being a toolsmith or blacksmith and "Einstein" being "one-stone" or such. In some cases it can be as easy as adding non-breaking spaces such as "Smith " or some other minor, subtle trick. The basic concept is to invisibly alter the spelling of the name so that it is no longer a common word in an English lexicon which would translate, and typically, the word will be repeated as untranslated, appearing letter-for-letter identical, as same mixed-case format, in the output language text. This whole concept, of escaping names to enable dynamic translation, should perhaps be discussed, especially for lede sections, as similar to avoiding use of templates in lede paragraphs to allow for clean mouse-over preview of intro sentences. Of course, complete machine translation would be impossible, as proven in Conceptual Dependency Theory (scripts, plans and goals) which noted the vocabulary depends on topic context for idioms, where "run home" can indicate a house but a "homerun" in baseball. However, if Wikipedians developed a core machine-translation system, then articles could be tagged with topic-context categories such as "baseball" to alter the default definitions of words during translation. As Wikipedia's interlingua capabilities grew, then all languages would have "footballer" terms in related auto-translations. -Wikid77 (talk) 14:45, 31 January 2013 (UTC)

Information Extraction

Another option is I think information extraction. Wikipedia text already includes semi-structured data like X was born in [[Y]], studied in [[Z]], played basketball in [[Team D]]. By doing an n-gram analysis, such phrases (born in, studied in, lived in, is endemic to etc.) may be derived. Once this is done, a lot of verifiable data will be gathered and then it will be a lot easier to translate this content into another language by simply translating the phrases and defining their order and relation with objects. Regular expression may be employed for this purpose. This is very similar to what DBpedia did with our data. However we could do a lot better with community effort.--Alperen (talk) 17:15, 1 February 2013 (UTC)


Hello Jimbo,

Thought you should look at this comic here, if you have not already. Now that I look into it, it seems that Wikipedia has got a really bad reputation at making the Himalayas out of a molehill. And we see that kind of criticism and mocking everywhere around the Internet. Not that we did not earn it (Also, see Apteva's blocking situation if you haven't already. It all happened over a single hyphen)

Which led me to think why it could be even remotely possible that something of this sort. And I realised its because of our no-consensus rules. For some reason, there is an interpretation of the rules according to which no issue can be solved by going either side if there are strong arguments from both the sides. If its one-sided enough, then the issue can get solved. But once there are editors fierce enough to keep fighting tooth and nail for it, no issue ever gets resolved. Not here on Wikipedia. [I consider Danzig to be an exception].

Whats more, we can easily have just a minority of the readers hijacking the entire system by making the first move, and fighting to defend it. Most of these issues could have been solved by plain simple common sense. But then someone made the first move, and it went unnoticed. Now anyone who ever tries to revert that first move will be considered to be edit warring. Which will ultimately ensure that a small minority of the editors, if dedicated enough to defend their reverts, and blindly attack any opposing arguments, can make sure the entire issue gets boiled off as a "no consensus" and nobody does anything anywhere.

Ultimately, we all lose our precious time bickering over whats not important than doing something actually important. Not to forget the fact that everyone who tries to come into the argument is forced to read through pages and pages of circular and repetitive "discussions already there".

I suggest that with such highly controversial issues, we have you or the Arbitration Committee stepping into to solve the situation. Someone neutral appointed by either of you creates a concise summary of the main arguments that are present for either side, which never is more than a single page, in my experience, after which it is put to a community-wide vote. Any discussion will be in a separate page, while all votes need to be either "Support", "Oppose" or "Neutral". No further words or comments shall be allowed in the voting area. After a given period of 30 days have passed, we simply tally the votes, and choose whatever consensus determines.

How does this sound? I believe there will be quite a bit of opposition to this proposal, but IMHO this is what is the true meaning of consensus - To be able to take a decision even when we have two equally correct arguments on both sides of the fence.

Do tell what you think of this.... We really need to solve all those issues now. Regards and cheers, TheOriginalSoni (talk) 08:31, 30 January 2013 (UTC)

See also [7], [8] and, inevitably, [9]... Wnt (talk) 14:14, 30 January 2013 (UTC)
I don't regard the fact that we have a reputation for intense discussion and debate of the most minute details to be negative at all. We should be proud of it. Ok, well maybe less proud when the issue is trivial, but the point is: we work really hard to get things right. That's a strength, not a weakness, and having a process whereby I, or ArbCom, step in to mandate editorial matters would be a step backwards, not forwards.--Jimbo Wales (talk) 16:19, 30 January 2013 (UTC)
xkcd has its funny moments. This isn't one of them. --OnoremDil 16:32, 30 January 2013 (UTC)
I think this was less a joke and more critical commentary on our edit warring habits. WP:LAME shows that we do this a lot. That being said, the fact that people will battle over what most people consider a trivial stylistic change is not inherently a sign of us having a poor reputation or a problem. Resolute 16:35, 30 January 2013 (UTC)
It's not so much that there are editors who will fight tooth and nail over these minor issues, it's the lack of an effective way to make bring the disputes to a finite a reasonably timely manner...that gives a bad impression. (IMHO). DeCausa (talk) 17:07, 30 January 2013 (UTC)
Having been the closer of the poll on whether we should write "the Beatles" or "The Beatles," and having done my best in writing about the closing to alternate between the two forms to avoid anyone's inferring that I was expressing a personal preference, I thought the comic struck home. Newyorkbrad (talk) 17:09, 30 January 2013 (UTC)
Indeed. :-) --Jimbo Wales (talk) 18:45, 30 January 2013 (UTC)
I thought it was hilarious.--Amadscientist (talk) 20:34, 31 January 2013 (UTC)

My point here is that whenever we talk of contentious issues of widespread participation, we often seem to be dragging debates round in circles into a No-consensus every time. Could we not have a simple widespread !vote, summarizing all the important points for everyone to read, and letting them choose either of the two sides? Keeping the discussion separate from the voting in most cases will end up in giving us a clear cut consensus in the end. TheOriginalSoni (talk) 07:33, 1 February 2013 (UTC)

Wikipedia misconceptions

As part of the wiki article List of common misconceptions, I strated a discussion about the inclusion of a section on Wikipedia misconceptions, based ont he fact that my research had revealed quite a few and I deemed it worthy of inclusion int he article. Negotiations fell through, but perhaps some good may still come of this idea. I'm not sure if it's been done, or at least as extensively as I am proposing, but essentially I think that in order to know about ourselves, we have to know about how others perceive us. Many of us have been in this community for so long that we have lost sight of how the general public (our readers) view this site and its articles. Therefore, I propose that we scout Google and beyond, and compile a comprehensive list of every single misconception related to Wikipedia that our audiences have (perhaps some that we never even realised existed before, or ignored due to the belief that they were widespread enough). We could use this information to help shape the way we work here, and how we design the website. We could also use it to correct the public's perception by directly addressing their concerns and unspoken turn-offs, thereby aiding the move of Wikipedia from an unreliable joke to an important and significant contribution to free knowledge and crowdsourcing.--Coin945 (talk) 17:18, 31 January 2013 (UTC)

Here are some sites to start us off: [10], [11], here, [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23].--Coin945 (talk) 17:18, 31 January 2013 (UTC)

This sounds like a very good idea - it's a pity that it's not getting more attention. (I'm not sure where such attention would manifest itself, other than here, though.) --Demiurge1000 (talk) 11:03, 1 February 2013 (UTC)
I looked at that discussion, and I wonder if you should instead try to create "Misconceptions about Wikipedia", basing it on RS:s on that subject. It might be easier to get agreement on which sources to use that way. Gråbergs Gråa Sång (talk) 12:43, 1 February 2013 (UTC)
Nahh... I'm staying away from misconception article in mainspace. That List of common misconceptions article has been in serious shit over the past month or so. Theres a huge discussion on the talk page for pretty much every entry. It's controversial and nasty. I'd rather keep this as a behind the scenes type thing. After all, who would need this information more than us? Usually the most "true" info isn't in reliable sources at all, but in some duscission forum somewhere. I wouldn't want to restrict this information from being included.--Coin945 (talk) 14:20, 1 February 2013 (UTC)
Ah, I misread you. I thought you wanted it in mainspace, but that´s not what you wrote. Gråbergs Gråa Sång (talk) 14:31, 1 February 2013 (UTC)
Sorry. Vagueness on my part. No harm done. :)--Coin945 (talk) 14:38, 1 February 2013 (UTC)

Thank you for taking a stand against homeopathy

Consensus and off-wiki canvassing

Jimbo, It seems that a weakness of the consensus system on Wikipedia is the possibility of off-wiki canvassing, which is difficult to detect or prove. Did you consider this potential problem when starting Wikipedia? Have you had any significant discussion on the matter? Regards, --Bob K31416 (talk) 13:47, 31 January 2013 (UTC)

I don't think it's a serious issue. I don't like the term 'canvassing', even on-wiki. I think it's more often used by people who want to shut down an open dialogue than people who have a righteous cause for concern. Another word for 'canvassing' is "engaging more people in the discussion" - it's open to all sides. The idea that it's bad to go out and recruit editors when you see a problem in Wikipedia is problematic. That isn't to say that some kinds of approaches to that aren't annoying - they are - but in general, this paranoia about it is not justified.--Jimbo Wales (talk) 15:44, 31 January 2013 (UTC)
Interesting answer. It looks like you don't consider it a serious issue, and that there is more of a problem of some editors making related accusations to suppress dialogue in a discussion, or of being suspicious for no good reason. Let me ask another. What are your thoughts on the following excerpt from the lead of the behavioral guideline WP:Canvassing?
"However, canvassing — which is done with the intention of influencing the outcome of a discussion towards one side of a debate — is considered inappropriate. This is because it compromises the normal consensus decision-making process, and therefore is generally considered disruptive behavior."
--Bob K31416 (talk) 17:17, 31 January 2013 (UTC)
I don't disagree with it (much) as written, but I think people tend to overstate the likelihood or importance of it, and tend to underestimate how often the real problem is people screaming 'canvassing' to prevent people from seeking outside voices. Many things on Wikipedia would benefit from more participation, more eyeballs, and the bias against recruitment means that decisions are made in obscure corners without relevant people being properly notified. This may suit the interests of a group that has a majority in that little corner, but knows that they are in the extreme minority in the broader community or world. But it doesn't suit the interests of Wikipedia.--Jimbo Wales (talk) 20:19, 31 January 2013 (UTC)
Thanks. --Bob K31416 (talk) 22:36, 31 January 2013 (UTC)
  • Topic-ban due to 1 vote canvassing: Just in case people think the screams about "improper canvassing" are not over-the-top, I would remind people that I was topic-banned for 3 months ("92" days) about "Amanda Knox" (currently on trial with parents for slander against Perugia police) and the "Murder of Meredith Kercher" because I had notified 2 supportive editors about re-creating article "Amanda Knox" but only notified 1 neutral editor, and 1 offline opposing editor, because "15" opposing editors had already been notified in the MoMK talk-page, and if I had notified more opposing editors, then they could have claimed more opposing editors were "double-notified" as already replying on the talk-page about the AfD for article "Amanda Knox" (afterward deleted, but re-created upon wp:DRV deletion review after she was acquitted of murder). That case of being sanctioned for "wp:Votestacking" of the Support/Oppose/Neutral notifications, as counting 2/1/1, was badgered by the majority of Oppose editors showing a "consensus" to topic-ban another editor as a means to silence an opponent in POV disputes. I think a good solution to such over-the-top badgering of editors is to restrict wp:ANI to follow pre-scripted "decision scripts" for how to judge a claimed violation, with proportional punishment, if found liable for sanctions.
    A related issue is that editors are supposedly not "punished" but rather "Wikipedia is protected" by blocking or topic-ban of editors, and the loophole is that admins (or friends) could be exempted as often having no intention to truly harm "Wikipedia protection" and so some people would receive no sanctions, unless issues are backed by so-named "punishment" sanctions with equal punishment for all, not capricious ideas of pre-crime suspicions to "protect Wikipedia" from imagined thoughtcrimes of future, crystal-ball harm. Instead, block anyone for 1 day, 1 week, or 1 month, depending on violations, and if they return, then treat the first violation as a probation period (3 months?), where future violations would receive longer blocks during the period, but after the probation period expired, then short blocks (or level-1/2/3 warnings) would again accompany similar violations. Currently, WP is "gunnysacking" all minor infractions to indef-block editors much faster than needed, with no return to first-time warning levels. I think that clarifies the issues. -Wikid77 (talk) 14:58, 1 February 2013 (UTC)
The situation with canvassing is impossibly fouled up. This page and Wikipediocracy have a seemingly unlimited exemption from the policy, even when it comes to people lining up to nominate the Wikipediocracy owner, a banned editor, for free Wikipedia merchandise. Meanwhile, it is constantly raised as an objection here. I think the whole policy should be scrapped and replaced with some minimalist and easily evaluated anti-spam limit on cold calling other editors. Wnt (talk) 06:11, 2 February 2013 (UTC)

Ref. Child protection rule

Hello, Mr. Wales,

I am an editor from the Russian Wikipedia. I was thinking to propose the Wikipedia:Child protection rule as a rule for ruWiki as well. I am unclear in a few points. The text have been promoted into rules by your (diff), which could be considered as a "blessing" :-) But the current version differs rather significantly from the one "blessed into a rule". Can be the current version seen as still corresponding to your interest and/or the interest of the Foundation? If so, could you or someone else "re-bless" the current version by a minimum edit? Otherwise I am not sure (actually I will remain unsure in any case) what would be the most proper procedure of introducing a rule that says "should not be the subject of community discussion, comment or consensus". I guess a standard discussion in search of a consensus would be a rather contradictory doing...

Best Regards,
Vsevolod (Seva) --NeoLexx (talk) 15:53, 31 January 2013 (UTC)

Mr. Wales, please, for the love of God, make sure that this rule is not perverted to become a means of running interference for the villainous policies that Russia has recently been enacting against providing any kind of gay-positive information on the Internet. From [25]: "The legislation being pushed by the Kremlin and the Russian Orthodox Church would make it illegal nationwide to provide minors with information that is defined as 'propaganda of sodomy, lesbianism, bisexuality and transgenderism.' It includes a ban on holding public events that promote gay rights. St. Petersburg and a number of other Russian cities already have similar laws on their books." Wikipedia obviously has the potential to be a major hole, perhaps the most important one, in Putin's campaign of censorship and hatred. Wnt (talk) 19:10, 31 January 2013 (UTC)
Eh... In what way a policy about pedophilia is connected with "providing any kind of gay-positive information on the Internet"? In what way at all do you connect pedophilia, child protection and gay activism? Also, the policy has been accepted in enWiki back in 2010 and I would strongly believe w/o any interference of "the long hands of KGB" :-) --NeoLexx (talk) 19:28, 31 January 2013 (UTC)
The timing of this suggestion, immediately after the passage of the first reading of the bill, bothers me. The Russian laws are primarily aimed at minors, i.e. at preventing "gay propaganda" to minors, such as by holding a gay pride parade. See LGBT rights in Russia, Moscow Pride for example. I don't know the situation on the ground in ru.wikipedia, but I am worried about how "advocating improper adult-child relationships" might end up being translated into Russian, for example; or whether the wording will be further altered before the policy is enacted; or whether it will simply be deliberately enforced beyond its nominal scope. I am not saying that you can't have a rule about pedophilia, but I am very suspicious that isn't its purpose. Wnt (talk) 19:37, 31 January 2013 (UTC)
I would agree with your concerns if it would be a "Russian interpretation" of the policy and not a word-by-word community verified translation. Or if the English original would have proven samples of misuse or a hot community disagreement. I don't know the whole history, but obviously there were some reasons that enWiki existed till 2010 w/o such particular policy, and then took it and in two years made it much more strict than the first policy version. I will not argue that Russia may have some democracy problems in comparison to the US, but it is not a reason to make it a potential "pedophilia oasis". A person proudly declaring "I am a pedophile!" is not welcome to enWiki (link), is not welcome to Wikiversity (link). "So let them go to ruWiki, because there they speak Russian, Russian means Russia, Russia means Putin and democracy problems..." I have some doubts about such approach to be fully fair. --NeoLexx (talk) 19:55, 31 January 2013 (UTC)
If you provide assurances that the policy will not be used to infringe on coverage of gay issues or to harass gay editors, then I should withdraw my objection with apologies. Wnt (talk) 20:08, 31 January 2013 (UTC)
Do you see any potential to misuse Wikipedia:Child protection or are you aware of real life samples of such misuse? Do you see the word "gay" at all there? Obviously for the declaration "I am a gay and my beloved boy is 12 years old, I see nothing wrong with it" the policy still applies, as being a gay is not an indulgence for any desired action or declaration. It seems that it never was a topic of doubts in the US or EU. Here is the Russian translation I have made:

Википедия принимает всерьёз вопросы использования своих ресурсов детьми. Будут бессрочно заблокированы те участники, которые пытаются использовать Википедию для доступа или для облегчения доступа к неприемлемым отношениям между взрослым и ребёнком, которые выступают в защиту неприемлемых отношений между взрослым и ребёнком (например, высказывая мнение, что такие неприемлемые отношения не наносят ребёнку вреда), те, которые сами себя объявляют {{wikt:ru:педофилия|педофилами}}.

О таких участниках следует сообщать в Арбитражный комитет для принятия адекватных действий. Неприемлемость перечисленного выше поведения не является предметом дискуссий, комментариев или текущего консенсуса. Если участник заблокирован за перечисленное выше поведение, то заблокировавший его администратор должен прокомментировать причины блокировки в нейтральных выражениях, заблокировать также страницу обсуждения бывшего участника и его доступ к википочте. Блокирующий администратор должен известить заблокированного участника о том, что любые просьбы и комментарии должны напрямую направляться по эл.почте членам АК, а затем немедленно известить АК.

Комментарии в Википедии, указывающие на то, что их автор является педофилом, будут быстро удалены из истории правок, во избежание нарушений приватности и возможного использования для диффамации. О своих личных подозрениях вы должны сообщать только по эл.почте. Подобные вопросы и обвинения конкретному участнику на самих страницах проекта могут привести к блокировке того, кто спрашивает или обвиняет.

Если вы сами юный участник или участница, и вы чувствуете, что другой участник Википедии действует в менере, заставляющей чувствовать угрозу личной безопасности или беспокойство, пожалуйста, сообщите несущему за вас отвественность взрослому и попросите его просмотреть страницу, на которой ведётся общение. Не продолжайте общение с тем человеком, полностью его игнорируйте. Никогда никому не сообщайте личной информации о себе, даже тем, кто якобы стремится вам помочь.

Если вы озабочены поведением другого участника проекта, пожалуйста, обратитесь в АК по эл.почте, но сначала, пожалуйста, внимательно прочитайте правила подачи заявок.

The style can be adjusted, but you can conduct any language expertise to see that I didn't change any meanings (or, out of evilness of my Russian nature :-) stock "gay" somewhere between the lines). --NeoLexx (talk) 20:36, 31 January 2013 (UTC)

I'm not qualified to evaluate this. I admit I don't see any obvious way to confuse the intent of most of this. However, the sentence "несущему за вас отвественность взрослому и попросите его просмотреть страницу, на которой ведётся общение", originally "please tell a responsible adult, and ask them to look at this page", comes out on Google as "inform carrier for you LIABILITY adult and ask them to view the page that is communication." Just to check: is this a suggestion to contact authorities, and could it be interpreted as urging children to report adults adding what they view as "gay propaganda" in Russia? Wnt (talk) 21:03, 31 January 2013 (UTC)
"a responsible adult" in the English policy puzzled me a bit. I eventually understood it not as "anyone who is looking adult and being able to act responsibly" but "the adult who is currently primarily responsible for your", so a parent, a foster-parent, a teacher (at the school time) etc. So I wrote "несущему за вас отвественность взрослому" — "to the adult who is carrying responsibility for you". As the text is intended for children as well, it could be actually further explained in both versions: "to your parents, your foster-parents, your teacher (at the school time) or to the other adult who is currently responsible for you". Is that what English text wanted to say? --NeoLexx (talk) 21:19, 31 January 2013 (UTC)
...and I wrote it with a typo which I just noticed: "несущему за вас ответственность взрослому". I assure you it didn't change the meaning but simply made the word mistyped. --NeoLexx (talk) 21:25, 31 January 2013 (UTC)

Child protection should be "global policy" for all Wikimedia projects. I think the best location for this policy is Meta wiki, where it could be translated into as many languages ​​as possible. The policy should come from Wikimedia and should not be put to a vote.

That would be the best IMO.

As for Russian Wikipedia, you must follow local procedure that is required to adopt a policy there, I don't see any other way.

In my original post I explained that I didn't see an obvious way to apply a standard local procedure for a text that contains "should not be the subject of community discussion, comment or consensus" There is the General disclaimer, Five pillars and office actions. As much as I understand, anything else is a subject of community discussion, comment or consensus, and consensus may change with time.

I have great respect for the Russian Wikipedia and there is no danger that the policy will be abused in a way that Wnt suggests. Their coverage of LGBT topics is fair, IMO. They have a number of good and featured articles from that topic area (see ru:Милк, Харви, ru:Стоунволлские бунты, ru:Гей-прайд, ru:Шепард, Мэттью etc.) --В и к и T 21:55, 31 January 2013 (UTC)

Thanks! (though I personally didn't participate in writing these particular articles). Just a side note — Russian Wikipedia is not Wikipedia of Russia ;-) Among active editors there are Russian speaking people from all across the world, US residents inclusive. But yes, I am from Saint Petersburg and I am a 4 years project participant — not to boast, but to note that I am not some new Russian initiatives' excited newcomer :-) --NeoLexx (talk) 22:41, 31 January 2013 (UTC)
Thanks - I'm very glad to hear that! Wnt (talk) 22:31, 31 January 2013 (UTC)

NeoLexx, Sue Gardner provided the following statement to Fox News in June 2010 in response to a m:controversial content flare-up in 2010:

Wikipedia has a long-held, zero-tolerance policy towards pedophilia or pedophilia advocacy and child pornography. The Wikimedia community is vigilant about identifying and deleting any such material. Any allegations to the contrary are outrageous and false.

The WMF globally banned a user (as an "office action") when the Commons community seemed unwilling to do so. There is a draft policy on meta, but since there is no process for approving a policy, it will never be approved (yes, I am serious about that). Also, please ignore Wnt - sometimes they get confused. Good luck. Delicious carbuncle (talk) 04:19, 1 February 2013 (UTC)

I placed a policy proposal to the ruWiki policies forum: Новое правило ВП:ДЕТИ
I also forgot to mention that I placed a question to Wikipedia:Village pump (policy)/Archive 117#Child protection policy. Please correct there if I'm wrong or misleaded.
P.S. I feel a bit bad by using a personal talk page as a forum where the page owner himself is neither the initiator nor a participant. It makes me feel a bit a better as I see this rule violation is rather common at this particular talk page and accepted by the owner. Still sorry. --NeoLexx (talk) 12:07, 1 February 2013 (UTC)
Jimbo's talk page is still his own, but it has also become an informal sounding board of sorts for a variety of issues and concerns. Some of our hottest hot-button issues and debates have spilled to or from this page over the years. If there's a conversation Jimbo has no interest in or one that goes too far, he will step in and close it. No worries. Tarc (talk) 13:35, 1 February 2013 (UTC)
The final result of the discussion: "A practical need in this separate policy is not demonstrated. It can be proposed for a public discussion again if the nominator is able to demonstrate an existing problem of the kind in the Russian Wikipedia which is hard or impossible to solve by existing policies" (Russian: "Практической необходимости в отдельном правиле не продемонстрировано. Оно может быть вынесено на обсуждение заново, если номинатор сумеет показать существующую в русской википедии проблему такого рода, которую сложно или невозможно решить действующими правилами.") No protests declared or planned from my humble side, thanks to all for comments and the overall participation here. --NeoLexx (talk) 21:04, 1 February 2013 (UTC)

Turkish Wikipedia

Turkish Wikipedia users are now questioning everything. --This unsigned article written by: User:Aguzer 21:14, 31 January 2013 (UTC)

  • Now please take concrete steps. --This unsigned article written by: User:Aguzer 21:15, 31 January 2013 (UTC)
  • Update + --This unsigned article written by: User:Aguzer 12:10, 1 February 2013 (UTC)

tr:Türkçe Vikipedi topluluğu sağlıklı bir topluluk değil artık. Doğru çözümler üretecek oylama yapamazlar. Çünkü bu konu hakkında bilgileri olmayıpta oylamaya katılanlar çok var. Zaten diğerlerinin de çoğu bu diktatörlerin yalakası olmuş. --This unsigned article written by: User:Aguzer 16:07, 1 February 2013 (UTC)

tr: Jimbo, artık bir şeyler yapmazsanız Türkçe Vikipedi iyice bozulur. Sonraları da gazetelere, internet sitelerine olumsuz olumsuz haberler düşünce de iyi kullanılmaz olur. --This unsigned article written by: User:Aguzer 16:23, 1 February 2013 (UTC)

tr: Jimbo ? --This unsigned article written by: User:Aguzer 10:22, 2 February 2013 (UTC)

Show your TAFI support

We're still in the process of getting TAFI all ready for the main page, and have got a lot of nominations in stock atm. We need 3 supports or more for a nomination for it to be chosen for the main page, and we would really appreciate for any and all users to come along and cast their votes. If you have other great ideas related to the project, do add them to the talk page? The link you'll want is at: Wikipedia:Today's article for improvement/Nominated articles. Follow the prompts and if you get stuck, give us a buzz (there's about 6 of us constantly there). :)--Coin945 (talk) 15:04, 2 February 2013 (UTC)

from the TAFI

Today's Article For Improvement star.svg

In the past few days, Today's Article for Improvement has gone through many changes.

  • We have modified the process for adding Nominations, which now uses a table format and requires only 3 supports for an article to be selected.
  • There is now a new Holding Area, where articles are kept for discussion before being selected for a particular date.
  • The new TAFI schedule now involves adding 7 articles weekly, chosen from a variety of topics.
  • We now have an Accomplishments page where we will be highlighting our older TAFI articles which have now become quality articles on the Wikipedia.

The Project is almost ready to hit the Main Page, where it will be occupying a section just below "Did you Know" section. One article from the weekly batch of 7 will be displayed randomly at the main page, the format of which can be seen at the Main Page template. There is also an ongoing discussion at the Technical Village Pump over the final details before we can go forward with the Main Page.

If you have any ideas to discuss with everyone else, please visit the TAFI Talk Page and join in on the ongoing discussions there. You are also invited to add new nominations, and comment and suport on the current ones at the Nominations page. You can also help by helping in the discussions at the Holding Area.

Above all, please do not forget to improve our current Today's Article for Improvement - Fun.

Thank you and hoping to have some productive work from you at the Project,
TheOriginalSoni (talk) 06:42, 3 February 2013 (UTC)
(From the TAFI team)

A kitten for you!

Cute grey kitten.jpg

Thaks for Wikipedia! :D

Dennis6492 (talk) 19:17, 3 February 2013 (UTC)


This article describing wikipedians as creatures in a dark room might be relevnt to those who participated in the discussion about the comic two days ago in the hello thread. Pass a Method talk 20:48, 2 February 2013 (UTC)

I love my old Ikea lamp, myself. :) --Jimbo Wales (talk) 21:53, 2 February 2013 (UTC)
RE "What If the Great Wikipedia 'Revolution' Was Actually a Reversion?" — Please note WP:3RR. Thanks. --Bob K31416 (talk) 22:40, 2 February 2013 (UTC)
Fascinating article. Wonder what it portends for WP's future.Thelmadatter (talk) 23:19, 2 February 2013 (UTC)
Sure, it's a nice article. But I can't help but feeling it's nothing new. Just the same stuff that was probably written about encyclopaedias 300 years ago. Formerip (talk) 23:44, 2 February 2013 (UTC)

Seems like the article missed an important aspect of Wikipedia that print encyclopedias never had. Wikipedia is effectively a real-time encyclopedia that rapidly includes new information. --Bob K31416 (talk) 05:50, 3 February 2013 (UTC)

I'll bet no comparable project was ever as beset with wp:vandalism as this one. Bus stop (talk) 13:53, 4 February 2013 (UTC)

Arbcom and the community

Hi Jimbo. Seeing as you set up Arbcom in the first place, would you be able to comment on my situation ? I was recently unbanned, and decided after a few months to have a clean start. Arbcom prevented me from doing so outside any normal process, even though there are no restrictions on me. I challenged this unsucessfully. A RfC on the topic of clean starts for previously sanctioned users is overwhelming in favour of my position. In terms of the reasons Arbcom was created, do you think this was within their remit in the first place ? If so, will the passage of the RfC annul the unwritten restrictions on me in the first place ? --Simone 13:57, 3 February 2013 (UTC)

I believe that editors not under sanctions are entitled to a clean start; however, as some indicated in the decision,[26] returning to the same problematic behaviors is one thing that can break a clean start according to that policy. I don't see why you shouldn't have a chance to have one, but if you were to go back and do something like Twinkle-nominating 7 articles for deletion as on December 19, well, it wouldn't last long. I was impressed by JClemens' comment in that discussion. (I should reiterate my distaste for those accursed robots - how many cascades of negative emotions on Wikipedia can be traced back, ultimately, to the doings of those machines? Our editors shouldn't be John Henry (folklore) trying to argue with a bot that wants to make it easy to delete their work.) Wnt (talk) 16:41, 3 February 2013 (UTC)
I have an issue with the idea that there are higher expectations of my conduct than of any other editor in good standing. My deletion nominations are all reasonable, and only a minority of articles I nominate are kept. Simone 11:48, 4 February 2013 (UTC)
It seems true that, as I alluded above, you have become something in a pawn in the larger debate about the role of deletions - how often and how easily things can be nominated and what the standards are for doing so. Wnt (talk) 18:46, 4 February 2013 (UTC)
You seem to be under the misapprehension that one is either banned or in good standing, a very binary approach, and one not supported by past community, arbcom, or AE decisions. You are not in "good standing" because of your past socking and recent return to the same areas. You will never be "in good standing" if you continue to revisit areas you created problems in before. Your combination of reneging on pre-unban promises and revisiting specific articles in which you'd previously socked demonstrates that you are precisely the sort of editor on whom repeated and focused scrutiny should be placed--with scrutiny, you are incentivized to behave yourself appropriately, knowing that a re-ban awaits if you resume inappropriate behavior. Jclemens (talk) 07:02, 5 February 2013 (UTC)
I would suggest that "only a minority" is a poor test, especially where something as contentious as deletion is concerned. Somebody who is trying to operate within consensus should easily be able to get their deletion nominations to a standard where only a small minority are kept. I would suggest you have a good look at your unsuccessful deletion nominations, try to learn from the reasons why some you thought reasonable failed and try to revise your practices so that in future only a small minority of your deletion nominations are kept. ϢereSpielChequers 10:30, 5 February 2013 (UTC)
Please review the contributions of, for example, JBSupreme or TenPoundHammer, who nominate lots of articles for deletion which do not get deleted without having Arbcom kicking up this sort of fuss over their behavior. @Jclemens, I am in good standing because there are no restrictions on my editing. Your desire to scrutinize me actually has detrimental effects on my editing (such as you popping up with negative comments on every single discussion which I start) and is precisely why I do not want to edit using this account. Simone 17:10, 5 February 2013 (UTC)

A barnstar for you!

Peace Barnstar Hires.png The Barnstar of Diplomacy
Hi dear Jimbo Wales. Star looks beautiful. For this reason, I've given you. :) You have set up Wikipedia. And of course, to get your attention. I do not want to be rude. But on Wikipedia, which would have been enough attention to complaints. Maybe there was a lot of work. I trust you. Look here, please: .... The articles is not empty. You or your be assigned to one or a few people ilgilenebilirmi the following situation? Criticizing the blocked. As evidence, there is a very link. You or someone else will answer. Tens of ll give link. These are "evidence of injustice". Even if I wrote you an e-mail. Please take care ... Cano58 (talk) 11:05, 4 February 2013 (UTC)
Google Translate, EN: At events are now completely blocked Turkish Wikipedia. Turkish Wikipedias disadvantage of these problems. This situation is gradually overflowing beyond Wikipedia. --This unsigned article written by: User:Aguzer 20:05, 4 February 2013 (UTC)
tr: Artık her şey birbirine girdi ve hararetlendi.

Google Translate, en:Everything came together and was no longer move. 1, 2 and 3 --This unsigned article written by: User:Aguzer 16:42, 5 February 2013 (UTC)

Quick question

I am in a situation in which I need to read deleted files. I have no need for any administrative tools and no desire to serve as an administrator. Is there any other rights status which allows the reading of deleted files and if so how does one go about getting that authorization? Thanks, —Tim /// Carrite (talk) 05:01, 5 February 2013 (UTC)

  • Because deleted files can contain sensitive information, or personal details about specific people, several admins have advised that only people fully vetted as admins should be allowed to have read-access to those files. I do not think the access to deleted files has been separated from the full set of admin-level tools. -Wikid77 (talk) 05:46, 5 February 2013 (UTC)
  • There is the WP:Researcher user status - though this may not give you as full an access to deleted edits as you want, the developers and of course the founder flag. But they all have entry requirements of their own. ϢereSpielChequers 08:23, 5 February 2013 (UTC)
  • Tim, depending on why the files were deleted, you may be able to get an admin to provide you with copies. Anything oversighted would be unavailable, as would material that was a copyright violation, but eg things deleted through PROD/AfD, or files deleted because they weren't in use would be available. --Elen of the Roads (talk) 16:34, 5 February 2013 (UTC)

Verb "wikying"

Get it right. "wikiing" means using a wiki, not necessarily this wiki. This project's name is Wikipedia, and "Wikipediaing" is only recognized by Wiktionary as yet.


Jimbo, as you know, there is article "googling" which mentions a Google co-founder saying that word in 1998. Well, with the growing popularity and enthusiasm for Wikipedia (Italian WP exceeded 1 million articles), I am wondering if you could use the word "wikying" (from 2005?) in a reliable source, as meaning either to look up topics in Wikipedia or to edit articles. I think it would be beneficial for more people to easily say, "I was wikying for more about some subject and found..." or similar easy phrases. Any thoughts or reservations about the word "wikying" (or general verb "wiky")? -Wikid77 (talk) 13:37, 3 February 2013 (UTC)

  • I think you meant "wikify". The meaning of this word is to fit written information to encyclopedic standards. It's sometimes used in templates, dealing with the quality of articles. Galzigler (talk) 17:35, 3 February 2013 (UTC)
    • He's not referring to wikify, he's referring to the act of looking something up on Wikipedia. Ryan Vesey 17:39, 3 February 2013 (UTC)
  • I've seen the term "wikied" (as in, I wikied it) used in common usage. I don't know that I've heard anyone say wikying or wiky despite the fact that they're all the same verb. Ryan Vesey 17:39, 3 February 2013 (UTC)
    • Well obviously I can't speak for everyone, but i've personally used "wiki" as in "I'll wiki [the article", plus both wikying and wikied. I've also used wiki-hop, as in to trawl aimlessly from article to article. I've also caught the Tube for that matter (:P). It's interesting that you ask Jimbo to create a reliable sourtce so you can make your article. I feel like thats breaking some sort of rule...--Coin945 (talk) 18:42, 3 February 2013 (UTC)
      • At least 20 reliable sources. Why does everyone on this site seem to expect to be spoon-fed everything rather than do any actual work themselves? (talk) 20:53, 4 February 2013 (UTC)
        • For the same reason that they also erroneously think that a handwave in the direction of a Google search results page is magically evidence, without bothering to read what the search engines point to, I suspect. Your handwaved-at search results page contains, in my part of the world: two novels (which don't qualify as sources at all, not being factual works), a For Dummies book that tells the reader that wikiing is the use of the wiki, and six books which use the word without defining it, all six of which, from context, are meaning the use of wikis in general and not what Wikid77 wants it to mean. Uncle G (talk) 21:49, 5 February 2013 (UTC)
  • Checked on "wikying" but more on "wikiing": Okay, thanks for noting the spelling as "wikiing" (or "wiki-ing") because we were checking for the less-common spelling as "wikying" where "wiki" would be a noun and "wiky" would be the verb to search a wiki or edit a wiki. -Wikid77 (talk) 23:36, 4 February 2013 (UTC)

RfA RfC round two

Hi Jimmy and talk page watchers, the second of three rounds of the requests for adminship request for comment has begun. Please comment with your proposed solutions there! Thanks, Ed [talk] [majestic titan] 04:19, 6 February 2013 (UTC)

Request for input

Hay Jimmy, can you add your two cents at Wikipedia:Village_pump (technical)#a heads-up -- may go dark about the possibility of either a partnership or merging of the two projects? Werieth (talk) 19:43, 5 February 2013 (UTC)

It would be best for someone at Webcitation to talk to somone at the Foundation about how we might join the consortium of funders. I'm not going to have time in the next several days to do anything on this, unfortunately - but it is important.--Jimbo Wales (talk) 23:50, 5 February 2013 (UTC)

Replacing the Featured Article Director position

Hi Jimbo, I thought I would give you a heads-up about a suggestion that I've posted at Wikipedia talk:Today's featured article/requests#Time to replace Raul654. The position of Featured Article Director appears to have effectively been abandoned, as the editor who holds it, Raul654 (talk · contribs), has done virtually no scheduling since late August 2012 (apart from a week-long flurry of activity in mid-November) and has been all but inactive since the New Year. The Featured Article Director post is in any case rather anomalous - I can't think of an equivalent anywhere else on Wikipedia - and is unduly dependent on one person. I've suggested eliminating it and promoting the current three delegates to the role of co-directors. I'd be interested to know what you think. Prioryman (talk) 20:28, 6 February 2013 (UTC)

Poor old Jimbo. There are dozens of threads at WT:FAC which cover this, and every time the odd fellow (including me) thinks Raul654 should be more active, every time the community says "not needed". I'm not sure, but I suspect Jimbo isn't reading all of these threads. Either way, right now there's no community consensus to remove Raul, that's all that needs to be said. Jimbo's opinion may be interesting, but nothing more than that. The Rambling Man (talk) 20:33, 6 February 2013 (UTC)
Well, the problem has gone beyond needing a community consensus to remove Raul; he's effectively removed himself, since he's not even editing Wikipedia currently to anything other than a trivial extent (17 edits since New Year's Day), let alone engaging with the FA process. There are decisions which the TFA rules say only he can take, and right now his absence and non-engagement means that there is nobody to take them. Prioryman (talk) 20:39, 6 February 2013 (UTC)
FWIW I agree that someone who is nominated to be a guardian of the things that make Wikipedia great and prominent (e.g. articles on the main page) should be more active than passive but the community have discussed this a number of times and, for whatever reason, believe that keeping an invisible FA director is better than trying to find a visible one. Jimbo may have an opinion but it's unlikely to change the consensus with the FA community. The Rambling Man (talk) 20:58, 6 February 2013 (UTC)
I've raised the question, so we'll see what people say. It wouldn't be problematic if it was a purely honorary position but Raul's absence is causing practical problems that needs to be resolved. Prioryman (talk) 21:08, 6 February 2013 (UTC)
Speaking generically (not about any individual) I've always thought that it was both un-Wikipediian and unhealthy to give king-like powers to one single-person, and have a process that more or less equates to an appointment for life (or until they officially quit). Of course kingdoms work really well if there is a good king like Raul and they are active. North8000 (talk) 21:46, 6 February 2013 (UTC)
It's purely an historical accident dating from the early days of Wikipedia (around 2003, I think). It came about because, if I remember rightly, Raul654 volunteered to schedule TFAs, he was very good at it and his role eventually became a unique position. It only seems "un-Wikipedian" because things were done very differently back then - there was no RFA, for instance, and people became administrators simply on the basis of recommendations from other administrators. Even arbitrators were occasionally appointed by Jimbo without the need for them to be elected. If we were setting up TFA from new, right now, we certainly wouldn't create an unelected Featured Article Director with absolute power to overrule consensus and his delegates. It's a relic of how things used to be done and perhaps Raul's de facto abandonment of the position gives us an opportunity to put something better in its place. Prioryman (talk) 22:09, 6 February 2013 (UTC)
It's been a while since I hung out there, but at the time the thing that struck me as most un-wikipedian is that only about half of the articles of the TFA's came from the public process, the other half were chosen by the one person based on come completely invisible process. And this is to select what content will be given the single most prominent place in the whole Wikipedia world. But my observation was that we had a good king who made the process work well. North8000 (talk) 22:40, 6 February 2013 (UTC)
North8000's observation is correct. In addition, I gather there's also a secret list of featured articles that are banned from ever being run on the Main Page. What's on it and who decides what gets banned, I don't know - I guess it would be down to Raul654. I would hope that the process can be a bit more transparent in future. Prioryman (talk) 22:53, 6 February 2013 (UTC)
Sixteen users (including Raul himself) voted to make Raul the featured article director in 2004 (discussion archive). Based upon his comments over the years, Raul regards this as a lifetime appointment to a position of absolute authority.
Historically, this attitude wasn't a major problem (as he did a good job and usually made decisions more or less in line with consensus). I agree that it's become a major problem, given his continued ownership of a process that he largely ignores, which serves only to impede the efforts of his delegates (who actually engage the community) and other active contributors. —David Levy 22:46, 6 February 2013 (UTC)
I note that the very next section in the discussion you found is "On actively choosing single points of failure", which looks quite prescient considering the current situation. Prioryman (talk) 22:59, 6 February 2013 (UTC)
FWIW David, Raul has been reconfirmed a few times since, including a very large RFC (sorry, can't find the link atm) where he was reconfirmed by a large margin, within the last couple years. That being said, he was also more active then. Now, I don't know if he's just gotten too busy in real life, or if the people who have been campaigning for a long time to try and drive him out against that consensus have succeeded, but in either case, we may well have to take another look at the position because of the practical realities. Resolute 15:19, 7 February 2013 (UTC)
I'm aware of the RfC, in which I participated. As I mentioned at Wikipedia talk:Today's featured article/requests#Time to replace Raul654, I supported Raul's reconfirmation and defended his record against those who sought to drive him out.
As you noted above, this was before Raul became largely inactive. He did a good job when he actually did his job, but he's essentially abandoned his responsibilities and hasn't fully delegated them to others (so we're stuck waiting for Raul to make decisions in areas that he's simply ignoring). —David Levy 19:14, 7 February 2013 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I've made some proposals for a new way of managing Today's Featured Articles at Wikipedia talk:Today's featured article/requests#Proposals - comments welcomed. Prioryman (talk) 08:47, 7 February 2013 (UTC)

Banning IP editing

If you ever decide to leave Wikipedia completely (and I certainly hope you do not) perhaps you will leave us a going away present and put us all out of our misery by arbitrarily banning IP editing. Or why just ban it anyway? I know, I know, it is an old and tiresome subject but the pain it inflicts through persistent vandalism that last years, on Administrators, Rollbackers, Bots and editors is getting unbearable. It wears everyone down and the fact is, there is no reason why a person cannot register if they wish to edit on Wikipedia. If they are a very important person may be the only exception though I believe they can make some arrangement rather than to IP edit. There is no point to the endless RFC or other avenues. Make it quick and easy. I do not think any good-faith editor would object to the ban on IP editing. Mugginsx (talk) 19:38, 4 February 2013 (UTC)

I would. Just the other day, an IP corrected an error in an article I wrote [27]Ryan Vesey 19:52, 4 February 2013 (UTC)
Wow! So that makes up for IPs like User talk: who has been vandalizing for three years under the same IP or the hundreds of others, perhaps thousands who people have to clean up after every day. I am sure someone else would have helped you eventually. Mugginsx (talk) 19:58, 4 February 2013 (UTC)
The number one reason we should always continue to allow IP editors is that full time editors are often produced. Had I needed to create an account to make my first edit, I probably never would have started. Ryan Vesey 20:14, 4 February 2013 (UTC)
I cannot conceive of why it would make any difference. It does not seem logical to say that being able to edit anonymously would eventually encourage someone to register. People register for Facebook and Twitter and a thousand other things online. If they wish to discontinue they simply delete their account. I believe it is the same for Wikipedia. I mean it is not like one has to give personal financial information. Mugginsx (talk) 20:22, 4 February 2013 (UTC)
Well, there are some other popular sites that allow no-account contributions, notably 4chan. Although perhaps that actually bolsters the argument in favor of mandatory registration. Mark Arsten (talk) 20:33, 4 February 2013 (UTC)
People aren't going to create an account to correct a minor error they see. People create accounts with a) the intent of creating/modifying one specific thing and leaving and b) with the intention of editing at least somewhat actively. Often the first of these are just as disruptive as IP editors, others are productive like Adambrower who created Herbert Greene (Broadway conductor). What people don't create accounts for is to make a simple correction like a spelling fix, or a correction to a minor factual detail. If a user doesn't intend to edit actively, it's not worth the effort. Heck, for really minor things, it's sometimes not worth logging in when you do have an account. My first edit was one of those, and without that edit, I would not have considered creating an account later. On a more humorous note, your comment on financial information is slightly ironic in light of this recent discussion. Ryan Vesey 20:35, 4 February 2013 (UTC)
Count me as another good faith editor opposed to banning IP editing. I see a lot of good work by IPs. Yes, there is a lot of vandalism, but IP editors are also less likely to get involved in edit wars, battleground mentalities and all the other issues registered editors become involved with in addition to vandalism. In fact, now that I think about it, Jimbo, can you ban all registered editors for those reasons? Resolute 20:39, 4 February 2013 (UTC)
I cannot remember exactly but I think it took approximately 5 minutes to register. And yes, I had not read that before. I would have to see the statistics to believe that IPs are less likely to get involved in edit wars. The IP vandals may not only because they have already made their statement through vandalism. Yes, I can think of a few registered users with battleground mentalities. There are remedies for these editors. What about the registered user who use IPs as their alter ego? At least we could eliminate that. Mugginsx (talk) 20:45, 4 February 2013 (UTC)
Registered users who use IPs as their alter ego can get caught out, if people with checkuser privileges care to check. So they shouldn't think they are immune. (Some of you know who I'm looking at.) --Demiurge1000 (talk) 20:56, 4 February 2013 (UTC)
Incidentally, and I'm not sure where it fits in the above discussion, but on a subjective level, I am seeing a lot more IP edits that appear to be clued-up former named editors who have simply made a decision to edit constructively as an IP instead of with their account, for some reason. Some say it gives them "freedom" and they don't care about the faux "status" of being a "respected" account. However much sense that makes, these really do appear to be IP editors that are here to edit constructively, really not just former banned editors looking to provoke their former targets (as of course some clued-up IP editors are). This "freedom to edit as an IP" is a fairly interesting development in all this. --Demiurge1000 (talk) 21:01, 4 February 2013 (UTC)
I think there are ways around these deceptions. I have been here too long to believe it can be stopped without an IP ban. I wish it were true but there are too many computer-sophisticated users and they take advantage. Yes, if you have been here long enough, you can figure out who they are but cannot prove it as a user. If it were that easy to stop them, how is it that it still persists along with other kinds of IP vandalism. And to the latest edit, why wouldn't someone want to be regarded as a respected user, if only by themselves? And why would they care what other people think if they are doing the right thing and in compliance with guidelines?Mugginsx (talk) 21:03, 4 February 2013 (UTC)
Because editing an an IP - particularly a dynamic IP - is just as easy as logged-in editing when it comes to making necessary changes, but avoids all your edits being dogged by a posse of disgruntled people with whom you had an disagreement three years ago, and since then have followed your every edit just waiting for you to make a slight slip so they can report you to teacher? Needless to say, I endorse every word Demiurge1000 says above. IP editing also avoids being tied to a watchlist and talkpage - you look at and edit only those pages that happen to interest you that day, without being sucked into the timesink of "I need to reply to this/I need to revert this". (talk) 21:18, 4 February 2013 (UTC)
I think IP editors can be a breath of fresh air. The environment of Registered users can be stifling. I think registered users are more likely to be carrying emotional baggage in relation to other registered users. I think this essay, poorly named, at least obliquely refers to one of the pitfalls of the otherwise wholesome concept of "community". Bus stop (talk) 21:30, 4 February 2013 (UTC)

If wikipedia didn't allow for IP editing, most of us wouldn't be editing here. Sure, you might have a small core of dedicated editors plugging away but isn't that the model already followed by britannica? Wikipedia thrives when there are more rather than fewer editors and even vandalism plays a role here by keeping us all from dying of boredom. --regentspark (comment) 21:37, 4 February 2013 (UTC)

Since IPs edit in secret, there is absolutely no way of validating how many would or would not edit as a registered user if they had to. The reasons given are counterintuitive and do not speak to the vandalism, secrecy, and dishonestly used by many IPs. The amount of vandalizing edits in any one day would invalidate any of these reasons by virtue of their numbers. They are legion. I could also say that IPs carry more emotional baggage but there is no way of proving that either. These reasons have no basis in fact or logic. I am sorry. Mugginsx (talk) 21:48, 4 February 2013 (UTC)
Just to be clear, since I don't understand your conclusions one bit, what I meant above was that many of us would be editing at all because we started out as IP editors. Not sure what that has to do with secrecy or what an IP editor would do as a registered editor. (No need to be sorry. Discussion is a good thing.) --regentspark (comment) 22:08, 4 February 2013 (UTC)
Yes, an IP can also hold a grudge, and an IP can also unthinkingly support past allies. This is what I meant by "emotional baggage". But an IP is also free to evolve personally as an editor due to the absence of obvious ties to other identifiable editors. IP editing is the training wheels of registered editing. Bus stop (talk) 21:56, 4 February 2013 (UTC)
I have now heard that registered users are rude, leave endless messages on talk pages of other registered users, and have emotional baggage. And the registered users do not speak to this? How very odd. Unfortunately since IPs edit in SECRET all we can say is that they may or may not be vandals, may or may not be uncommitted users or may or may not be users who are, in fact, registered at the same time and up to no good. The vandalism statistics are real - the rest is conjecture. Some of the comments make it sound as if signing up as a registered user is like signing up for Armed Forces. There is no other commitment to signing up except that your edits and your comments are there for all to see and you must abide by the guidelines. If one is well-intended, what possibily could be bad about that? If you make a mistake, there are good administrators who are not bogged down with vandals who can help you. If you are not good, then you are made to be RESPONSIBLE. An IP does not have to answer to anyone or to any guideline. That is what makes it so attractive to some, methinks. Mugginsx (talk) 22:09, 4 February 2013 (UTC)

Banning IP editing is not going to happen, but there is a problem that needs serious resourcing and the community should pressure the WMF to spend serious money to reduce the stress. The community should not discuss how developers might help until invited because such discussions always derail the project. Just set up a couple of clever people with a big budget and tell them their job for the next 12 months is to reduce the strain on pissed-off editors and admins. It is beyond absurd that IP-hopping vandals spend years disrupting the community, and when someone says "let's do a big range block for 12 months", the response is always "but that might stop some productive contributions", and nothing is done, again. If the answer from the devs (after spending $100k) is "it can't be done without big range blocks", then that's what should happen. In that case, resourcing should be switched to pressuring ISPs while providing alternatives for productive people behind a blocked IP. Johnuniq (talk) 22:55, 4 February 2013 (UTC)

The nuclear option is banning reading from an entire ISP's netblocks until they agree to remove the biggest problem users from their service or from editing Wikipedia. Whether we can discuss that as a serious option or not, I don't know. --Demiurge1000 (talk) 23:17, 4 February 2013 (UTC)
Interesting idea. Here is one way that it could be done:
  • Implement the technical possibility on the server level.
  • Nuclear option means that the only way to read or write Wikipedia is with a pre-existing account. No new account creation.
  • Some people may come up with inofficial work-arounds such as providing an indef blocked account and publishing its password. That should be allowed.
  • Limit the nuclear option to one arbitrarily big provider per year.
  • The nuclear option is only applicable for very serious and very long-term abuse, and after the provider has been contacted but failed to solve the problem.
  • The nuclear option requires consensus after a full, widely advertised community discussion on whether to use it for a specific long-term vandal. (If several vandals are proposed, use normal processes to decide between them before the final vote.)
  • Once a target for the nuclear option has been chosen, the provider gets 3 months to solve the problem, e.g. by terminating the vandal's contract or by blocking Wikipedia for the vandal.
  • If, after 2 months, there are no satisfactory assurances from the provider and the vandal continues to edit from the same provider, Wikimedia gives a statement to the press to the effect that the provider will be blocked.
  • If, after 3 months, there are no satisfactory assurances from the provider or the vandal continues to edit from the same provider, the provider is blocked for a year.
That way providers will learn that ignoring Wikipedia's requests to combat serious abuse can cost them a lot of money. One possible issue is that for vandals that always use accounts, their provider is necessarily disclosed. If that's impossible for legal reasons, then the method can't be used for those vandals. However, after we have first gone nuclear with something like Verizon, it is likely that compliance among providers will be high even for such cases. Hans Adler 09:07, 5 February 2013 (UTC)

It's ironic, amusing, and slightly saddening, to see a person with an account, who is using a pseudonym and is completely unidentifiable, participating in a discussion where at least one other participant can be easily traced by anyone to the town of Ilford, and yet repeatedly claiming that it is the people without accounts who supposedly edit "in SECRET". Uncle G (talk) 22:20, 5 February 2013 (UTC)

To those who wish to ban IPs on the grounds that their editing is more vandalising and disruptive than registered users': you've expressed your feeling that IPs are so much more vandalising and disruptive than registered users, so that banning them would be very positive. Are there any statistics that could support your feeling? Garsd (talk) 12:57, 6 February 2013 (UTC)

Probably not. But here is an article by Aaron Swartz that gives a statistical analysis of Wikipedia data and finds the site is mostly written by unregistered users: "Who Writes Wikipedia?". Cheers!-- (talk) 20:41, 6 February 2013 (UTC)

Move to "trusted IP editors" and erase hack edits

You don't have to pay me for solving the problem, which involves promoting IP addresses with 100-200 acceptable edits to a status as "trusted IP editor" where many (most) major pages would be edit-protected, then if an IP editor turns sour, then revoke the trusted status on that IP (or IP range) but not totally blocked, just unable to edit the major articles. Next move to a system of erased edits, where someone could temporarily hack a page and log cute edit-summary comments, but eventually an oversighter would actually erase the edits (database-delete transaction) to completely remove the hack edits and related reverts, until all traces of the problem were gone, except perhaps in separate, hidden log pages for historical purposes. Those steps should solve the problems, so give my "big-budget" payment to your favorite charity. -Wikid77 (talk) 23:55, 4 February 2013 (UTC)

Thanks, something along those lines might be very helpful. But someone (a dev) has to do think it through and do the work, and do it well. There should be tools to apply the levels of IP protection believed appropriate, and tools to monitor progress. Wikipedia is a multi-million dollar business and the WMF should take funding from cutsey stuff like article feedback and do something to reduce the stress on good editors and admins. After a while, even a saint thinks "Why should I bother? If someone cleverer than me wants IPs to keep stuffing up the biology articles, why should I fight them?". Johnuniq (talk) 03:38, 5 February 2013 (UTC)
Wikipedia is doing all the things related to discussions and to trust rather poorly, and where it tries to address them, unusable nonsense such as liquid threads comes out. Reddit is doing these things right, so let's learn from them. Karma could be implemented both for individual users and for IP ranges. Hans Adler 09:15, 5 February 2013 (UTC)
The problem with trusting an IP is that IPs get reassigned and are often shared. There are ways we could handle IP edits more efficiently, I'm keen on moving to smart blocks where you block People with the same IT configuration rather than everyone at an IP or range of IPs, that wouldn't work for schools. but it could make range blocks less heavy handed. Perhaps we could also combine this with Hans Adler's excellent suggestion that the WMF try and learn from those sites that are doing this better. ϢereSpielChequers 09:50, 5 February 2013 (UTC)

Let's look at ip editing from a different perspective

I've been a staunch supporter of "anyone can edit", and so, by extension, of IP editing.

But while reading the above, a couple things occurred to me.

  • 1.) Is there anything (and I mean anything - though not someone's guessing, but tangible "this prevents me from making an account") preventing those who edit from an IP to create an account? Noting that accounts (afaik) are free and open to everyone.
  • 2.) This is the bigger one for me, and where I might support prohibiting IP editing, depending on the answer: Attribution of edits. As far as I know, wikipedia policy is that editors cannot share accounts. So, any single account should only be used by a single editor. And all edits from that account are attributable to that single person. This is obviously not true of IPs. And this leads me to all sorts of thoughts and concerns, like who should be credited/attributed when re-using an article? If we attribute an IP, are we attributing a particular edit to every user who has ever edited from that IP? And if not, when re-using does the time stamp of the edit need to be attributed too? (Would more than the time stamp be required?)

And from the other side of this, consider that requiring editing from an account would deal with a lot of blocking issues (though of course not all).

IANAL, so I won't claim to understand the ins and outs of licencing, or re-use, etc. But it really does sound like something worth looking into. Especially if there is no reason that people can't merely take a moment and make an account.

I sincerely would like your thoughts about all of this. - jc37 23:40, 4 February 2013 (UTC)

Regarding (1), one that immediately springs to mind is that a lot of people—particularly those with high-end userrights—are very reluctant to use public terminals. (When I had CU/OS I'd have been very reluctant to log on on my main account in a library or at work, for instance; all it takes is to accidentally click "remember me" or to forget to log off, and one's handed over reams of sensitive personal data, pedophile porn et al to whoever the next person to wander in happens to be.) By banning IP editing altogether, the WMF would effectively be ordering at the very least all with CU or OS, and probably all admins, either only to edit from a secure home terminal or to create a set of sockpuppets; while there are plenty of such socks about (including User:Iridescent 2) created for just this reason, it's a fudge that goes against the one-user-one-account ideal.
Regarding (2), the relevant parts of the license are "Original Author" means, in the case of a literary or artistic work, the individual, individuals, entity or entities who created the Work or if no individual or entity can be identified, the publisher" and "You must, unless a request has been made pursuant to Section 4(a), keep intact all copyright notices for the Work and provide, reasonable to the medium or means You are utilizing: (i) the name of the Original Author (or pseudonym, if applicable) if supplied" (my emphasis). Under CC3 there's no requirement to attribute where the author can't be identified, which is why the WMF terms of use are careful to include liberal doses of "if possible". "Everything must be attributed" is one of the Wikipedia myths that isn't strictly true, along with "anyone can edit". – iridescent 00:11, 5 February 2013 (UTC)
For #1, that falls under "legitimate sock accounts". It doesn't sound that much different than having a separate account for a bot.
For #2, eeee yy uu cc kk uh. gotta love legalese loopholing...
So what I get from this (and please correct if I am misunderstanding) is that if one edits from an IP, they are essentially making their edits public domain? If so, no wonder no one wants to push to eliminate ip editing.
Though, to me, this reinforces the idea that perhaps we should move to account-only editing. I dunno. I'd like to discuss this more I think. - jc37 01:14, 5 February 2013 (UTC)
  1. "Legitimate sock" is a very controversial fudge to get around the "sometimes I can't use my main account but I don't want to disclose my IP" issue. User:Tony1 should be able to give you all the (many) arguments against it.
  2. Blame the people who voted for this change. The old license terms ("List the five most significant authors") were a heck of a lot less confusing, but the first law of Wikipedia is that any simple policy is replaced by a complicated policy which can then never be reversed because there's never "consensus" to do so. By editing from an IP you're not making your edits PD—for practical purposes, you're transferring your intellectual property rights to the WMF (since they're the closest thing to an originator that can be identified). – iridescent 01:30, 5 February 2013 (UTC)
I'll admit right now that I didn't read the full licensing page you linked to, but as I understand it, if your re-use Wikipedia information under the GFDL, you need to name the five most significant authors. If you use the CC-BY-SA 3.0 License, you need to link to the history page. There doesn't seem to be any attribution difficulty there. Ryan Vesey 01:39, 5 February 2013 (UTC)
I've edited as an IP - if you are on an insecure public network it is much more sensible than logging in and one way to kill an hour at an airport. That said we've had so many people for so long call for a ban on IP editing that maybe one year we should test it on a few language versions of Wikipedia (but please not English). My expectation is that the loss of new recruits and good IP edits would outweigh the gain of not having to revert bad IP edits, but I wouldn't object to a sensible test. It might be possible that we have got beyond the point where we need those IP edits to build the site, and we have such high prominence as a site that we could get away with requiring account creation. ϢereSpielChequers 08:28, 5 February 2013 (UTC)

wrt (1): yes, some people edit from IPs because there is a certain clique whose sole purpose is to hunt socks, instead of improving Wikipedia. they do this so they can "win at their MMORPG. ™ (The tm looks just like a tilde on this phone, but it`s fitting, nonetheless. (talk) 01:10, 6 February 2013 (UTC)

I have not noticed that, but if it is indeed true that they look for socks and find them, good for them! If you are not a sockpuppet you have nothing to worry about! Mugginsx (talk) 14:26, 6 February 2013 (UTC)
That's not true; false positives can and do happen, as with any test. The whole "if you're doing nothing wrong, you have nothing to worry about" is a non-starter, as an argument. Writ Keeper 14:34, 6 February 2013 (UTC)

IP Banning (cont'd) Survey of IP editors

I suggest that IP editors be invited, as an exclusive group, to participate in a survey with questions such as these.

  • When did you first edit Wikipedia?
  • How frequently do you edit Wikipedia?
  • What kind(s) of edits do you usually make on Wikipedia? (In what subject areas? In what namespace[s]? Adding information? Removing information? Correcting information? Organizing information?)
  • From where do you edit Wikipedia? (From home? From an educational institution? From a public library? From a workplace? From a community center? From a mobile device?)
  • Why do you edit without using a registered username? (What are the advantages and disadvantages of doing so?)
  • What is your age?
  • What is your sex?
  • Where do you live?
  • What is your first language?
  • How well do you read and write English?
  • What additional information about your editing of Wikipedia do you wish to provide?

Wavelength (talk) 00:13, 5 February 2013 (UTC) and 00:16, 5 February 2013 (UTC)

  • We can get many answers by sampling the IP-edit histories: By looking at thousands of IP-edit histories, we know that ~90% of edits to unprotected, medium-interest articles are IP hack edits+reverts. When I say, "Wikipedia is 10% information and 90% deformation" then IP edits+reverts are one example of the deformation. Extreme fringe articles get mainly bot-edits, and almost never see vandalism. "Crowd sourcing" is called a "blog" and instead, encyclopedia writing is generally very restrictive and requires dedication to editing by checklists. Thousands of IP editors edit just a few articles, then rotate to another IP, rarely returning months/years later, possibly as someone else using that IP. If IP edits were sampled for IP location, then that could indicate activity by regions of the world. Looking at the average word-count of IP edits could be another statistic. I have spent some hours to confirm there are some IP editors who have added one paragraph of important content to each of hundreds of articles. Some major articles (celebrities: "Edith Piaf") were even started by IP editors (when they could create articles from IPs). Like many editors, I began editing WP as an IP editor for months/years, unsure of what restrictions were involved to be "allowed membership" as a username. Anyway, we could sample thousands of IP-edit histories to better understand them. -Wikid77 (talk) 01:18, 5 February 2013 (UTC)
  • Let me predict the average answers generated by any 100 randomly selected Wikipedia IP editors: Q1. 1492. Q2. 325 times a day. Q3. FUCK YOU! BRIAN IS A SHIT FACE!!! Q4. ANYBODY CAN EDIT! Q5. 321. Q6. Oral!!! Q7. In a house. Q8. Ma-ma. Q9. ¡No comprendo! Q10. See attached link for picture of my dick at Commons: <URL HERE>. Carrite (talk) 04:58, 5 February 2013 (UTC)
That is unfair and rude about IP editors as a whole.--♦IanMacM♦ (talk to me) 10:35, 5 February 2013 (UTC)
It's funny because it's true. Carrite (talk) 03:47, 7 February 2013 (UTC)
+1 to ianmacml. It'd be good form to retract that, after doing a random sample of recent ip changes. By far most is good faith, some is good content too. There is a high level of spam, though I severely doubt requiring registering an account will deter many, if any, spammers. Martijn Hoekstra (talk) 10:55, 5 February 2013 (UTC)
Oh, I believe it definitely will curtail most of the spamming. Mugginsx (talk) 12:49, 5 February 2013 (UTC)
who do you think is likelier to sign up for an account, a public relations employee who has been given an assignment to improve a companies online presence, or someone who wants to fix a typo? Who would you'll rather exclude from editing? (Hint: the answers are the pr employee is more likely to sign up, and we want to retain the typo fixer. I'm leaving the conclusion on how this relates to banning ip editing as an exercise for the reader. Martijn Hoekstra (talk) 13:51, 5 February 2013 (UTC)
It takes about four seconds to log in. Mugginsx (talk) 16:25, 5 February 2013 (UTC)
just yesterday i didn't fix some formatting errors on scholarpedia because it required me to register an account. Those errors had been there since 2011. Who knows how many others didn't fix it because they didn't want to sign up. Had I been promoting my company, I would have taken that time. Martijn Hoekstra (talk) 04:54, 6 February 2013 (UTC)
I'm glad you found an old typo. Registered users also find old typos. What does that prove? Why don't you register?
Why would he? It's just a typo. Anybody else would make the same correction and surely a registered user will be along shortly to take care of it. Of course there are other editors whose motivations aren't affected by the bystander effect. Registration is no obstacle to those wishing to promote a website that no one else would recommend or push a POV that they expect the community to reject. Kilopi (talk) 07:59, 7 February 2013 (UTC)
does it really matter why? The fact I had to register prevented me from fixing it, that project would have been better off if I fixed that, rather than have shrugged it off. The same goes for Wikipedia. Martijn Hoekstra (talk) 13:07, 8 February 2013 (UTC)

IP banning and Over-intellectualization

Every time this subject of banning IPs come up this happens. Well-meaning as the computer sophisticated editors are who would like to devise ways of dealing with IP vandals, the IP abuse is just too prevalent and pervasive to design a program to discourage them. To allow the IPs to continue to edit drains important time and resources of good-faith editors and over the years becomes discouraging and can eventually lead an editor to start to distrust other editors, registered and unregistered alike.

Once again good faith editors have strayed from the fastest and easiest solution - BANNING IPs. Anyone can still edit - all they have to do is take five minutes to register. Logging in takes about four seconds, so that should never be a problem. Can we please address the banning of IPs? So what if many editors started out as IPs - who is to say that if they had to register from the start they would not have done so eventually? I simply can not agree with that rationale. The lure of Wikipedia is too strong. Some of the other reasons given for using IPs border on the absurd. Unsure of the rules? Read them before registering! We are not children here and can certainly decide if we want to participate in Wikipedia or not. If we change our mind all we have to do is stop editing!

The reason I started this discussion off with asking Jimbo Wales to order the banning of IPs was that I know from passed experience that there are too many conflicting interests for editors to ban it on their own. It is one of those few problems that takes an executive decision! Mugginsx (talk) 11:35, 5 February 2013 (UTC)

So long as there are people willing to spend the time and effort reverting IPs and new editors, there is little incentive for banning IP editing or ensuring that new editors understand our rules and guidelines. I think you will admit that there is no shortage of editors willing to do "patrolling" and "vandal fighting". It should be noted that the stimulus-response aspect of this activity is unlike most of the rest of Wikipedia, where there is generally no reward for doing something sooner than anyone else. If you consider that many of the people doing that work are not well suited to actually creating or editing articles, some of those people may leave Wikipedia if there was no need for this function. Although banning IP editors may or may not improve the quality of the project -- I tend to believe that it would be beneficial -- any loss of active editors would be viewed as a negative by the WMF. On the other hand, banning IP editors would undoubtedly increase the number of registered users, so perhaps they could simply switch metrics. Delicious carbuncle (talk) 13:25, 5 February 2013 (UTC)
  • Banning IPs is banning unregistered voices: I can sympathize with the frustration, to reduce the amount of hacked edits by banning IP edits, but the concerns expressed in earlier threads are major, common issues, and that is why people want to know more about who are the users who edit as an IP editor. Remember IP editors also post thousands of talk-page messages, and I have noticed some of the IP editors:
  • a famous singer's assistant says the songwriter's name is "Smithy" not "Smitty"
  • an expert researcher who always has the same IP address in that lab
  • a new user wary of "country club" or "good ol' boy" groups
  • a professor who "does not have time" to login and see the massive main page
  • a person who hints at the truth, but only when extremely anonymous, or
  • a grammar expert who finally fixes errors left in pages for months.
We still have the option to semi-protect the highly visible pages, to require autoconfirmed editors, while anyone can post to the talkpages. However, moving to a system of "trusted IP users" would be possible, and thereby the many researchers in their labs, or professors on faculty networks, could easily gain the status as trusted-IP users. Meanwhile, non-trusted IP editors could still post to a talk-page about how page 57 of that magazine confirms what the celebrity told them. I guess the main point is this: some complex problems require complex solutions; blocking the motion of some pawns, knights or rooks will not make Chess easy to play. Blocking all IP editors will not fix grammar errors, nor stop registered POV warriors, nor add sources, nor clarify complex text to young readers of articles. -Wikid77 (talk) 13:47, 5 February 2013 (UTC)
Simple solution: put pending changes protection everywhere except on talk pages; create an IP "user-group" (and yes, I know that's not the correct technical term) equivalent to autoconfirmed rights for registered users. That way the constructive long-term IP editors aren't inconvenienced, but the proxies, dynamics, and other unsavories are kept out. As a side note, I doubt even Jimbo would be allowed to ban IPs. The WMF would not stand for it and someone would de-founder him (if that's technically possible). Evanh2008 (talk|contribs) 13:56, 5 February 2013 (UTC)
Important people manage to edit on Facebook and Twitter - I think they can manage on Wikipedia. With the definitive statistic being given that 90% of vandals are IPs what do we really have to lose? The arguments for keeping the policy of allowing IPs are, in my opinion, weak and non-existent. As for whether Mr. Wales can make an executive decision, I have no doubt. Wikipedia is his brainchild and his innovation. He can do whatever he feels should be done and has, in fact, done so on a few occasions. I respect many of the editors here and do not know many others. I cannot find one good reason stated anywhere for not banning IPs unless an editor has something to hide. People who wish to stay will stay and, of course there will always be some degree of vandalism, but nowhere as large as it is now. Further there will be less editors quitting in disgust and frustration. Mugginsx (talk) 15:41, 5 February 2013 (UTC)
I gotta agree with banning IP editing. The benefits outweigh the drawbacks.Thelmadatter (talk) 15:50, 5 February 2013 (UTC)
Yes, and the last time Jimbo made an executive decision this short sighted, he caused many Commons editors and admins to outright bail on the project. And I may have missed it in the walls of text above, but can you show me the source of your "definitive statistic" that 90% of vandals are IPs? But even if that were true, that does not present nearly an accurate picture. What percentage of decent edits are IPs responsible for? How many edit wars do they get into? How many engage in long-term POV pushing? When was the last time an IP was hauled in front of ArbCom? You accuse others of "over intellectualizing" the issue, yet you are guilty of gross oversimplification. You are looking for a "fast and easy", and frankly stupid, solution to an overstated problem. Resolute 15:59, 5 February 2013 (UTC)
Just because people disagree with you doesn't mean that they have a conflict of interest, looking at all the above threads the closest to a conflict of interest is that I occasionally edit as an IP whilst in Public spaces. We do have debates that get toxic with conflicts of interest and alleged conflicts, paid editing being the most obvious example. But this isn't one of them. What we have here is a conflict of ideas. and such arguments are best approached by trying to convince others that your case is stronger than theirs, not by trying to get someone in authority to rule against an argument that you find difficult to refute, or indeed by restating your case in bold. You seem to be assuming that IP vandalism is something that nothing has been done to address and which eventually will drive off goodfaith editors. I and others think that the rise of editfilters and vandal reversion bots has meant that the vast majority of IP vandalism no longer requires human intervention, and the cost in volunteer time of dealing with each thousand IP vandalisms has greatly fallen and continues to fall as edit filters and vandfighting bots improve. You think that the lure of editing is so strong that the good editors will still create accounts even if we make our editor recruitment process that bit more difficult. I look at new page patrol where we already have banned IP creation of new articles, and I'm more inclined to the theory that requiring people to create an account has dissuaded at least as many of those who would have created worthwhile articles as it dissuades vandals, and I seriously doubt it dissuades any spammers. Now if you could find a way to keep the good IP edits and lose the bad ones then we'd all be interested. But the lesson from Conservapedia and Citizendium is that making a Wiki less open to editing can be more effective at throwing out babies than it is at throwing out bathwater. ϢereSpielChequers 15:54, 5 February 2013 (UTC)
I just came from dental surgery and still I can smile at this response. Difficult to register on Wikipedia? It takes five minutes! The amount of IP vandalism is still 90%. I never said anything like the rest of that so I do not know how to dignify it with a response. I was making a plea to Jimbo. That is why I brought it to this page. Mugginsx (talk) 16:11, 5 February 2013 (UTC)
I know I haven't been terribly active lately, but the statistic of 90% IP vandalism is new to me. Would you provide a citation for this claim? As a definitive statistic, it shouldn't be too hard to provide. EVula // talk // // 16:19, 5 February 2013 (UTC)
A citation? It has been said over and over again in many discussions by many editors. It was said above. This is not an RFC, it is a request made to Jimbo Wales on this page designated for discussion with him.Mugginsx (talk) 16:21, 5 February 2013 (UTC)
Translation: you just made the figure up out of thin air and are now getting defensive that a number of people who do know the realities of how Wikipedia operates have challenged it. For the record, the only person to say it above was yourself. (Speaking of "how Wikipedia operates", why is this on Jimbo's talkpage and not WP:VPR? Jimbo has no authority to enact a change like this, let alone unilaterally enact a change like this against consensus, and if he tried it would likely put Arbcom and/or the WMF in the awkward position of having to block him.) – iridescent 16:36, 5 February 2013 (UTC)
It's a known fact that 92.5% of statistics are made up :) Elen of the Roads (talk) 16:38, 5 February 2013 (UTC)
-Wikid77 said it above. This is a fine example of why an administrator should want IPs banned. Registered users have to respond to administrators or could get blocked. An IP on the other hand, can say whatever he wants and his IP is blocked, in which case he comes back in five minutes with a new IP. Hardly seems fair, does it? I cannot understand any administrator being against this. Mugginsx (talk) 16:53, 5 February 2013 (UTC)
You are describing IP-hopping vandals, i.e. people with criminal energy. They could do the same with throw-away accounts which would then have to be blocked and would probably turn out really annoying. Prevention of IP edits would only help against casual vandalism from bored students etc. Hans Adler 17:07, 5 February 2013 (UTC)
I wasn't going to comment on this thread any further, but I now have to make sure everyone knows that the comment above is a pile of uninformed rubbish. Account-hopping would be is significantly more difficult than IP-hopping. Not only would does account-hopping involve switching IP addresses (via a proxy, or by resetting one's IP settings through the command line) for every Wikipedia account the person wants to create, but it also involves registering a new email address for every Wikipedia account the person wants to create, then going through the account confirmation process for every Wikipedia account the person wants to create. Acting like banning IP editors wouldn't slow down anyone but casual trolls is ridiculous. I have dealt extensively with IP-hopping vandals in the past, and I know for a fact that, had they been required to register accounts every time their previous one was blocked, they would have been significantly slower, I would have been required to spend a far smaller portion of my time on-site dealing with them, and I and countless other users and admins could have spent our time improving the project rather than throwing the furniture up against the door and waiting for them to get bored. Evanh2008 (talk|contribs) 09:51, 6 February 2013 (UTC)
"Pile of uninformed rubbish" is hard words with which you set up high expectations for your own contribution. Unfortunately you didn't quite meet them. You are describing the need of IP-hopping for account-hopping as something that would make account-hopping even harder. In the context of vandals who are already IP-hopping I find this not convincing at all. Of course an account-hopper must also IP-hop to get around the autoblocks. Did anyone say otherwise?
The required email address hopping is not a problem at all for reasons that I am not going to disclose per WP:BEANS. For the same reason I am not going to even hint at how a general IP ban would definitely not slow me down the least bit if I were interested in that direction. Others will find the same or similar methods once they become sufficiently necessary to them and will disclose them on 4chan or similar sites.
We can reduce casual vandalism in this way, but with the determined vandals we will always be in an arms race that neither side is going to win. Every escalation in this arms race impacts functionality for regular users. For some vandals such escalation is precisely what they are trying to achieve. Hans Adler 14:22, 6 February 2013 (UTC)
I have no doubt that a notable portion of IP hoppers would become moderately successful account hoppers as well, and I am well aware that there will always be ways for certain people to circumvent whatever security measures we come up with, but you are, in my opinion, still wrong that there would be no net benefit to banning IPs. Very few people are that technically competent, and almost as few have the time to become that technically competent. Even some trolls have social lives, and it takes significantly more time and effort to create new accounts than it does to create an IP address. I am not bothering to actually advocate for banning IPs at this point, as that issue is out of the community's hands, but I do wish that people on both sides of this debate would acknowledge that there are measures that can and should be put in place to prevent vandals being able to edit. Evanh2008 (talk|contribs) 03:36, 7 February 2013 (UTC)
"Wikid77 said it above." Yes, he did, and his statement was just as unsourced as your own. His exact statement: "By looking at thousands of IP-edit histories, we know that ~90% of edits to unprotected, medium-interest articles are IP hack edits+reverts." What were those thousands of IP-edit histories? What articles? What edits? What was the sampling size? I've edited across multiple projects (Wikipedia, Wikiquote, and Wikisource), and I'd say at least 72% of the IP edits I see are good-faithed edits. See how easy that was? It's just as well-sourced as your statistic. EVula // talk // // 21:59, 5 February 2013 (UTC)
At least you didn't make up that number. I think I know where it comes from. [28]
In 2007 a study found that 97% of all vandalism and 25% of vandalism reverts were done by anonymous users. But that's not the metric we need. We don't know how many vandals would create an account if necessary, and the key question is what is the percentage of vandalism among anonymous editors. There seems to be no data on that. Hans Adler 17:07, 5 February 2013 (UTC)
My anecdotal evidence, from my watchlist. IPs are responsible for most vandalism. But every once in a while, an IP improves an article in a significant way, or gives a clue about errors and omissions in the article. Wikipedia would be a much quieter place without IPs, and it would have a lot more mistakes and a lot less content. --Enric Naval (talk) 17:39, 5 February 2013 (UTC)
  • Other solutions: faster semi-protection, edit filters, Lua-script scanners: Perhaps we need to improve the rejection of all hacked text, and then not only will IP-based editing be guarded, but any improper text could be reduced. Some possibilities:
  • Allow faster semi-protection: Perhaps some admins would be willing to have a "vigilance squad" for faster semi-protection of articles where IP editors or new-account editors are hacking the text, such as during high-profile TV coverage of topics.
  • Improve the current edit-filters: If it seems too much hacked text is getting stored, then perhaps the edit-filters (like black-listed weblinks) could be changed for higher restrictions in some articles.
  • Insert Lua-script scanners: With the new Lua script modules allowing text scans 180,000x times faster than markup-based string searches, then an unprotected article could have Lua-based watchdog templates which warn, bleep or hide any hacked text. For example, the intro text of an article could be sent through a hypothetical "{bscan}" template to bleep off-topic text, such as someone adds the insulting phrase "is a child molester jerk" but the article would show "John Doe is a ** ** **" or similar bleeping of off-topic text.
The list of off-topic phrases could be stored in a subpage of the talk-page or in common-nonsense lists, and then the Lua-based scan templates would rapidly check those lists and either bleep the words as starred-text "**" or just hide the improper words as totally omitted during display of the page. A vandal is less likely to vandalize when the added hack text is always hidden, even though they re-edit to confirm the text is still there, but just will not display through the {bscan} or related text-scan templates inserted into targeted articles which are not yet semi-protected. Only the likely targetted articles would have text-scan templates, not every article. The strategy is to focus on ways to block more hacked text or improper claims, not only the edits made by IP editors. By widening the focus beyond IP editors, then even more problems could be avoided. -Wikid77 (talk) 17:00, 5 February 2013 (UTC)
That is, by far, the most intelligent suggestion (if possible to do) that I have heard as an alternative to total IP banning, though for the respect that I have for you Wikid77, which is considerable, I do not know why you do not agree with total IP banning, unless you think it is just not feasible. Mugginsx (talk) 17:07, 5 February 2013 (UTC)
I would add to this that the BOTS are to be congratulated and I know for a fact that they try very hard to curb the vandalism, as do the Rollbackers, but it is getting too much for all of us to handle and still be able to participate in the most enjoyable aspect of Wikipedia, editing. Mugginsx (talk) 17:15, 5 February 2013 (UTC)
As I said above the bots and edit filters are steadily improving, and I'd add that if anyone can come up with a specific improvement that would accurately identify more vandalism then go talk to the people who run those bots and filters. But I'd dispute that "it is getting too much for us all to handle" as that would imply that the amount of vandalism that needed manual reversion was increasing, or at least the amount which each vandalfighter had to contend with. My experience is that though my watchlist is several times longer now than it was in 2008 there is less vandalism which gets past our bots, edit filters and recent changes patrollers and so needs correction by watchlisters such as myself. I believe that the stats bear this out, and that over time less of our vandalism is left to manual editors to revert. If we get to the point where the community is shrinking faster than our automated anti-vandalism measures improves and the burden of vandal reversion per editor starts to increase then there are other options which we could look at. Personally I'd like to see us go down the DE wiki route and have flagged revisions on all articles, perhaps if in the future the amount of vandalism that each editor has to deal with goes up we will eventually persuade the community that we need a more efficient vandal reversion system where each IP edit is checked once, but perhaps only once as opposed to the current system where some IP edits are looked at by a host of editors and others are not manually viewed. ϢereSpielChequers 17:42, 5 February 2013 (UTC)
I agree with your observation on the magnitude of the vandalism problem. A few years ago there was a point where I felt that the vandal-fighting subculture of Wikipedia was about to take over the entire project. Nowadays it seems to be rather insignificant. I don't know if they have less work nowadays (due to bots) or if they are just more professional, but it sure feels as if the someone is being overly dramatic here.
Flagged revision on all articles is very frustrating, by the way, because it can take ages for a change to an insignificant article to go live. I think it's just as bad for editor recruitment as a general IP ban, only in a different way. But at least it's more effective against vandals. Hans Adler 00:47, 6 February 2013 (UTC)
Banning IPs would have serious, predictable consequences. What you're proposing is basically to design Versailles without toilets because toilets are nasty. For example, suppose for a minute you're not a Wikipedian and you want to add something to an article. You create your free account in "4 seconds" and add the thing you're thinking about. Within 4 days you've forgotten the password and perhaps the username as well, so you start another free account and do the same thing when you see some idiot has reverted what you did, citing some alphabet soup excuse in his edit summary, so you can add roughly the same thing again. Well, congratulations - now you're a Master Sockpuppeteer and though you may not know it, there's some goofy process afoot that you have no idea about to track you down and use the cookies on your machine and your IP address to ban you off the Wiki. You try to edit again and you're told to go through some humiliating process, mea culpa, sackcloth, ashes, crow meat, the whole pizazz. The effect of banning IPs and transferring the edits to fresh accounts, in other words, is to take the routine patrolling of IP edits and turn it into formal checkuser work - which in turn greatly increases the exposure of Wikipedia to serious privacy breaches by a larger number of highly trusted users. Wnt (talk) 17:56, 5 February 2013 (UTC)
I have forgotten my password/username and got it back within about five minutes or so. I cannot comment on the rest because, frankly I am just not well versed in some of these very technical issues now being discussed. I am an editor and a rollbacker who has never been blocked in the years I have been editing. Mugginsx (talk) 18:05, 5 February 2013 (UTC)
The reasons given here (excluding those of a technical nature that is, as they do at least try to provide an alternate solution), are no excuse NOT to register. Some of them do not even make good sense. I am told that a typical proxy server offers you from dozens to hundreds of different IPs you can use to hide your own IP address. How do we keep up with this? I would still like to see the founder's opinion, if he cares to express it.Mugginsx (talk) 12:58, 6 February 2013 (UTC)
At the very least, we should ban IP addresses identified as being a proxy server. Nobody uses a proxy server for any good purpose. With regard to public terminals not being considered safe - don't use public terminals. This is not the FBI or the DIA where people have to be available to answer questions 24x7Mugginsx (talk) 14:43, 6 February 2013 (UTC)
Wikipedia already treats WP:Open proxies in that way, though there are legitimate reasons to use them, as that page mentions, such as Wikipedia:Advice to users using Tor to bypass the Great Firewall. Note however that except for specific users with a reason, this blockage affects users who create accounts, so obviously it can't scale to affect every IP address without causing a problem.
As for "not using public terminals", that would be a serious mistake. Think about where public terminals are likely to be found ---- in a library. It's hard enough to document anything requiring offline sources without adding an extra roadblock we don't really need. They are also found on vacations to exotic locales, museums, etc. No, I don't think we should be giving those up. (And I don't even know what the DIA does. Beat Bradley Manning with a wet sock? Write biographies of Megan Rice? Wnt (talk) 16:03, 6 February 2013 (UTC)
The general internet experience is that requiring people to register discourages new users by about 50%. Given that the foundation has identified declining participation as a problem that they care about, there isn't going to be anything imposed by fiat to drive away half the new users. WilyD 16:02, 6 February 2013 (UTC)
I'm sorry but I quoted a Wiki source and got re-buffed. What is it that article which is so general in his nature and has to do with purchasing items from retail stores, have to do with Wikipedia? It proves absolutely nothing about Wikipedia's membership. Mugginsx (talk) 16:26, 6 February 2013 (UTC)
I'm not sure what "Wiki source" you're claiming to quote, but it's neither here nor there. If you want to be willfully ignorant, that's your concern, but forcing people to register with a website to use it decreases participation dramatically, even though you wouldn't naively expect that. Obviously we can't know what the exact percentage of new users that would be driven away would be (though I suspect we're more precarious than a retail website, as the benefits of registering here are far less than when I'm buying a homebrewing kit or whatever). But given that Jimbo and the Foundation believe that declining participation is bad (and aren't so concerned about vandalism, since it's a pretty minor concern), you're not going to convince them to ban IP editors. While we can't know anything without trying, we can reasonably extrapolate that banning IP editors would dramatically reduce the amount of simple vandalism, and dramatically reduce the number of new editors. Few people are going to sign up for that bargain. WilyD 16:48, 6 February 2013 (UTC)
It is said that when someone resorts to insults, they know they had made a poor statement. Is it I that is willingly ignorant? Your source is a comparison made of online retail stores with different purchasing techniques and their rationale for doing so. You are comparing that with types of preferred editing on Wikipedia? I think not. Also, one must ask what type of editors are repelled by registering when they probably would not hesitate to register on Facebook, Twitter, Google, or any of the thousands of other sites that are wise enough to demand it. Are we that desperate? Please show the statement from Wales or the Foundation that they are not concerned with vandalism. Mugginsx (talk) 16:54, 6 February 2013 (UTC)
Now, don't start going off at the deep end, and do try reading The Signpost occasionally. You'll find it very enlightening as to the views from the top. Elen of the Roads (talk) 17:16, 6 February 2013 (UTC)
Thank you Elen. I actually do read Signpost. I re-read this article because I do not remember reading anything about IPs in it. I was correct. It talks about newbies and other editors being treated badly or insulted by administrators. It questions whether Wiki is losing editors is because there is a bunker-mentality and general rudeness of administrators toward editors, (hard to imagine). It goes on but does not specifically mention IP editing that I can find. Did you mean to link to a different issue?Mugginsx (talk) 17:21, 6 February 2013 (UTC)
Yes, of course Facebook and Twitter have far less users than they would if registration is optional. Facebook's revenue model is based off knowing who you are, so they don't much want anonymous users anyhow. It's more profitable for them to force you to register, even if it reduces their userbase by half (and it's unclear how unregistered users could use facebook). Google doesn't demand you register. If Google demanded you register to use their search engine (or google maps, or books, or scholar, or whatever), they usership would drop dramatically. But they don't. Google wants maximum usership, so we can search without registering. I regularly avoid sites if I have to register just to use them. I gave up on trying to buy a £300 item from Argos because the registration was too much of a hassle. And if I may be a bit of a braggart, I've made a contribution or two around here, something I probably wouldn't have done if my first attempt to help was to a hostile demand that I register. The foundation has made turning around the declining editorship a priority (here). They haven't done that for vandalism (for an excellent reason - vandalism is a pretty minor concern, less so than in the "heyday" of 2006/07). Which isn't to say they're not concerned (something I never said), but they're less concerned. Banning IP edits would reduce vandalism, but it'd be fashioning deckplates for the Titanic out of sheetmetal from the hull. A terrible idea. WilyD 17:23, 6 February 2013 (UTC)
I have made almost 20,000 edits and all registered and at no time did I ever consider not registering, nor did I feel it was a hostile demand that I do so. Did you have to register to get email, or to do banking online? I wonder how many people do not use email because they have to register or how about the online libraries? Answer is not required. Mugginsx (talk) 17:55, 6 February 2013 (UTC)
I haven't said requiring registration would mean we'd get no new editors, but that we should expect to lose about half (and in our sample size of two, we'd have probably lost one - small number statistics, but it looks sensible). "Answer is not required" is willful ignorant. I use online libraries every day, and I've never once registered. If they ask me to register, I simply use a different library. Google books doesn't make me register. The arXiv doesn't make me register. (And if I want to cite a paper, but I have to register to get at it, and it's not on the arXiv - forget it, there's someone else who wants my citation). My bank doesn't make me register to use online banking (though it's an option, it's one I've not pursued). Not everyone acts exactly the way you do. You need to stop assuming they do. WilyD 06:41, 7 February 2013 (UTC)

There was a time, that I despised IPs. However, my opinon of them has softened in these last few years. I've actually had very few conflicts with them. GoodDay (talk) 09:35, 7 February 2013 (UTC)

I started this plea to Jimbo, his was the opinion I respecfully sought, though I wouldn't mind knowing the name of that bank that allows online banking (which is what I said) without logging in. Face-smile.svg Mugginsx (talk) 12:00, 7 February 2013 (UTC)
Logging in is slightly different from registering, but I access my online banking the same way I access a hole in the wall machine (ABM) - with a Barclaycard and a PIN. (At least, this is functionally different to me than when I was with TD Canada Trust, who required me to have an online banking account, with a password and such - I had to register to use online banking with them, then log in - now I log in without having registered). That said, I wouldn't want to endorse Barclays. Compare that IPs (which are far less anonymous than accounts anyhow) still have to specify what page they want to edit, they just don't have to register to do so. WilyD 18:04, 7 February 2013 (UTC)
Thanks, I suspect that most of us are aware of the difference between registration, log-in and pin nos but it is a good reminder.
As said below by someone else, I also predict, and I hope we am wrong, that eventually, good faith editors and volunteers are going to get tired and just give up fighting IP vandalism. Mugginsx (talk) 11:06, 8 February 2013 (UTC)

Vandal patrol - Wikipedia the game

You know, it's always been interesting to me that the first solution to any given situation is the rush to create gadgets and gizmos a plenty.

In this case, to deal with IP issues, let's create a sub-group of Wikipedians to "vandal patrol". Time they could spend doing other things instead they spend racking up points (number of vandals reverted) playing Wikipedia the video game. And to perpetuate this sub-group, let's create all sorts of gadgets and tools to make it even more like playing a computer game.

And now, I read the discussions above, and see the response is more of the same.

Why is just asking people to make a free account such a problem? If anything, it follows the wiki-way of "many hands make light work". By asking each editor to create an account themselves, they are reducing the work of others.

And those same productive IP contributors are likely to be productive contributors with an account. (Is anyone out there really contesting the addictiveness of Wikipedia? In my experience people typically leave due to burnout, not disinterest...)

There's been a lot of talk about changing the mmorpg mindset that's been growing lately at Wikipedia. This would definitely be a very simple way to do so.

And last I checked, it's fully supported in mediawiki software.

Do I think it will happen?

I dunno.

Do I think it will help make things better in many ways? Yes. I'm seriously starting to think so.

But, open-minded person I am, I welcome others' thoughts on this. - jc37 20:12, 6 February 2013 (UTC)

  • I recently took down the user box on my User page regarding the opinion I had held regarding requiring mandatory editor registration, as my views on this have changed somewhat after encountering several IP's who did not appear to be socks and who had very good contributions. Problem here being our defining, core mission statement "that anyone can edit." That precludes a requirement to register, which stops some people. But I do agree that IP vandalism and the attendant need to stop it is a distracting timesink that eats up energy which could be used elsewhere. Though I revert vandals the "old school" way these days, I can testify that the people using Huggle and other tools to revert vandals, whatever their motivation, are doing the encyclopedia a great service, and I salute them. Jusdafax 22:08, 6 February 2013 (UTC)
  • Try trusted-IP restrictions and erase nonsense edits: When I first registered my username, I was shocked that more articles were not semi-protected, and someone misinformed me, "Vandalism happens to pretty much every article" which I found was wrong because rare topics almost never get vandalized. I advise to treat IP addresses as usernames which get autoconfirmed, but with 100-200 edits, not just a few edits, and then protect more articles. Plus, enable deletion of nonsense edits, where vulgar crap is just totally erased (by database delete transactions) perhaps with a quick revert also later deleted after review and potentially logged to a hidden history log, but no more cluttering of article histories with obvious hack edits. Perhaps there has been a fear that a good edit would follow a hack edit, and that could complicate erasure, but often there is time to erase a recent hack edit (as if it never happened) before a good edit is posted. Also, we need to discuss "#If IP editors were banned" as a totally separate, reverse topic thread further below. -Wikid77 (talk) 01:02, 7 February 2013 (UTC)
    • Autoconfirmed IP addresses after 200 edits? 200 edits that could have come from between 1 and 200 people? Right. That's silly. pablo 14:29, 7 February 2013 (UTC)
      Not necessarily. If we treat each IP address as a single account, then problematical IP addresses will typically get a long block for multiple violations before they are autoconfirmed. Of course the status would have to be set back after the block. Hans Adler 18:43, 7 February 2013 (UTC)

Constitutional changes update, please

You're running a bit behind schedule on your plans to shake things up. While things have been unusually peaceful lately both "on-wiki" and in the press, the things that are broken are still broken.

You said in December:

This is the moment when I normally post a bit of a "state of the union" address to the new ArbCom and to the community, but as of two days ago I came down with an awful cold and I'm not really in a position to write up my full thoughts right now. (It's really important and I want to get it just right.)

In short, I'm planning in January to submit to the community for a full project-wide vote a new charter further transitioning my powers. Because the changes I hope to make are substantial, I will seek endorsement from the wider community. (There are powers which I theoretically hold, but can't practically use without causing a lot of drama, but it is increasingly clear to me that we need those powers to be usable, which means transitioning them into a community-based model of constitutional change. One good example of this is the ongoing admin-appointment situation... a problem which I think most people agree needs to be solved, but for which our usual processes have proven ineffective for change. Some have asked me to simply use my reserve powers to appoint a bunch of admins - but I've declined on the view that this would cause a useless fight. Much better will be for us to put my traditional powers on a community-based footing so that we, as a community, can get out of "corner solutions" that aren't working for us. More to come in January.

Would prefer not to have a random speculative fear-mongering discussion about this today. Leave the end-of-the-world doomsaying to the Mayans. (Or rather, to the nutters who willfully misinterpreted the Mayans!) There will be plenty of time for panic in January. :-)--Jimbo Wales (talk) 15:59, 21 December 2012 (UTC)

(Emphasis added.) Are you still planning to push through some reforms? --SB_Johnny | talk✌ 01:35, 7 February 2013 (UTC)

Yes, and you are right - I'm running behind. I'm about to be in a period of six weeks of comparably reduced travel and other duties, so I'm looking forward to doing a lot of on-wiki work.--Jimbo Wales (talk) 04:08, 7 February 2013 (UTC)
"How to delegate" is discussed on many web pages, including
Wavelength (talk) 17:50, 8 February 2013 (UTC)

Time to scan

This is just a heads up: neurologists are beginning to better understand where "evil" resides in the brain. If you are ever asked to complete such a scan, refuse it! Or else they might find the Real Jimmy. Oh, and you remind me of Kevin Spacey in House of Cards. --Cei Trei (talk) 15:38, 7 February 2013 (UTC)

When scientists report their findings in the Daily Mail rather than in academic publications, one should retain a measure of doubt. Looie496 (talk) 00:28, 8 February 2013 (UTC)
And what makes you think he hasn't reported his findings in "academic publications"? A scientist can use different channels for exposing his work, including the media. The article also mentions a lecture held by a different professor at a university. Besides, what is mentioned there is nothing of controversy. Similar ideas have been expressed in the past. --Cei Trei (talk) 02:04, 8 February 2013 (UTC)
The specific statements there are certainly controversial, and I know he hasn't reported his findings in academic publications because I checked -- it is easy to do an author search in PubMed or Google Scholar. Looie496 (talk) 16:57, 8 February 2013 (UTC)
Fair enough. :) --Cei Trei (talk) 02:06, 9 February 2013 (UTC)

A barnstar for you!

Brilliant Idea Barnstar Hires.png The Brilliant Idea Barnstar
I Read wikipedia from about 5 years...but last 1 year was with great respect & continuity...

Today Suddenly I think its time to do some edit or else with wikipedia...then I edited a page,i dont know more about editing....i just repleced 170 with 150 in a page...... then i realize all the time searching something in wikipedia, learning something new & reading something with a great Orderly view gave me a good support with something briefly & this is the reason that I respect wikipedia all the times.......& it could nt be done without so many editers who giving their effort to for free to help someone around the world like n't it great,I thought...but who was the first ???? then I searched with google & found u..... It would not be done without u....u gave everyone a great oppurtunity...All the time I thanked & respected wikipedia was actually u.... & I just want to thank u from bottom of my heart.....

Im just a student,studying CSE 2nd year in Bangladesh....u do nt know me & I do nt even u personally,even if u do nt read or get this mail or u just delete does nt matter,coz there are thousands around the world r8 now must giving u their appreciation & gratitude to u, in their appreciation those words are as same as mine & u will read one of the thousands...Than u sooo much again Dark Bruin (talk) 00:08, 9 February 2013 (UTC)

Turkish Wikipedia that have been blocked unfairly

Türk Süvarisi

  • Update: Barrier was removed 6 months. 1 --This unsigned article written by: User:Aguzer 11:46, 9 February 2013 (UTC)

Seksen iki yüz kırk beş
Hedda Gabler

  • Update: Barrier was removed 6 months. 2 --This unsigned article written by: User:Aguzer 11:55, 9 February 2013 (UTC)

If you do not intervene many times this number increases. --This unsigned article written by: User:Aguzer 12:59, 8 February 2013 (UTC)

I'm lazy, so I only checked the first name on the list, and I already know (from experience) that that person is contributing usefully on the English language Wikipedia instead. Turkey isn't even in the global south any more, so... send us your huddled masses et al. --Demiurge1000 (talk) 01:15, 9 February 2013 (UTC)
  • Jimbo ?, Jimbo always watching :/. --This unsigned article written by: User:Aguzer 11:31, 9 February 2013 (UTC)

Declaration of injustice at Turkish Wikipedia

---Cano58 (talk) 03:15, 9 February 2013 (UTC)

Advice on systematic and possibly government-sponsored censorship?

Sex ratio at birth in mainland China, males per 100 females, 1980-2010.

I recently found what I believe may be the longest running and most successful censorship campaign on Wikipedia. Events concerning Ping Fu's recently published autobiography which details her exile from China after her study of female infanticide, involving several hundred new and sleeper forum participants on Amazon, Twitter, and some related activity here on Wikipedia, have led me to sources such as [29] and [30] which describe the magnitude and implications of the underlying social issue, respectively. My questions concern Infanticide#China, which has been repeatedly scrubbed of any reports of the practice in the past century, contrary to numerous news and scholarly reports (e.g. [31], [32], [33], and [34]) and One child policy#Infanticide which has been frequently cited on Talk:Infanticide as the reason mention in that article would be redundant, but is presently reduced to a very short statement citing three government and NGO reports, two of which are dead links.

Given that there is no systematic censorship noticeboard or anything like it as far as I know, how is it best to deal with this issue? SPA3142 (talk) 00:48, 9 February 2013 (UTC)

I encourage a thorough investigation into this. I'd be interested in looking at evidence of sockpuppetry, etc. Is it actually true that Infanticide#China has been "repeatedly scrubbed"? I have no knowledge of this issue (yet!) and hope the community will help me take a good look at it.--Jimbo Wales (talk) 03:15, 9 February 2013 (UTC)
Government censorship is usually a matter of NPOV, so WP:NPOV/N is probably the most appropriate noticeboard in most cases. Hans Adler 08:08, 9 February 2013 (UTC)
I should note that the user who asked this question has already been indefinitely blocked on three accounts by User:AGK for making a few contributions with throw-away accounts ([35], [36], [37] and here. I hope he doesn't try to log into those accounts again lest browser cookies lead to an autoblock of his main account and thus a permanent online identification of his political positions. I should note that WP:Alternate account allows using alternate accounts for privacy, and if this user is based in China this would give him a remarkably good reason. Nonetheless, things being how they are, it would appear that trusting Wikipedia to do anything but run straight to the Chinese government with your identity records would be most unwise.
Unfortunately, the OP did not have time to identify diffs marking precisely who is censoring the article and how, which would have been most helpful. Wnt (talk) 17:25, 9 February 2013 (UTC)
Looking only at the edits with "Chin" in the summary, I'm not so sure there's any more censorship on Infanticide#China than you see on typical controversial articles. There's a little section blanking such as [38] and [39] and the occasional deleted sentence or source, but the talk page archives have more opposition to including recent reports than this obvious subset of the edit history. I could have missed edits where "China" was deleted from the edit summary before saving, so perhaps it's worth the effort to look through the full history text with a script to see more closely.
And I would like to know the rationale for checkusering that account, too. I'm posting from a new account. Does that mean I get checkusered? Are there any controls in place to prevent checkusers from being overtly or covertly manipulated by powerful political interests who are known, shall we say, for their disregard of computer privacy and security measures? Lswhwt (talk) 21:59, 9 February 2013 (UTC)
It does seem that the Ping Fu article was more obviously being edited to put the subject in a negative light. It was also obvious that the page ratings for that article were the subject of the same "virulent attack by China’s Internet vigilantes, who have slammed her account of the country’s Mao-era troubles and lampooned the book on Amazon with a flood of one-star reviews"[40] Before page ratings for our article were recently put on the blacklist, there were 60-70 ratings for the article with a very heavy weighting of "1". Not very effective, since the ratings are not about the person, but it was more than a bit suspicious. First Light (talk) 22:07, 9 February 2013 (UTC)

Wikipedia, it's in space!

Did you know...

From Wikipedia's newest spatial content:


The Wikipedians (inhabitants of Wikipedia) must be feeling proud that the world's largest encyclopedia has got its name···Vanischenu「m/Talk」 21:49, 9 February 2013 (UTC)

I'm afraid the picture won't make it through DYK - as far as I know Jimbo doesn't have a halo, although that might have changed since the last time I saw him. Jimbo, can you clear this up for us? Halo or not? Prioryman (talk) 23:16, 9 February 2013 (UTC)
OK.....never making a sacrifice joke again after seeing some of that.--Amadscientist (talk) 03:28, 10 February 2013 (UTC)

Ethical problems on the Wikipedia Signpost

Jimbo, you'll recall that a while back you banned Jayen466 (talk · contribs) from this page after his repeated questioning of you over your dealings with Kazakhstan. Suppose that Jayen466 had gone off to the Wikipedia Signpost and written up a "report" on the matter, effectively using the Signpost as a platform. I don't think you would have looked very kindly on that. Something very analogous to this may be about to happen; he has just written a lengthy story in Wikipedia:Wikipedia Signpost/2013-02-11/In the media about the recently published governance review of Wikimedia UK following the Gibraltarpedia controversy. He has not disclosed his own huge conflict of interest in the matter. He was one of the loudest voices campaigning on- and off-wiki on this issue (especially through Wikipediocracy) - you will recall that he deliberately generated controversy by feeding information to the media, including your own private emails to him. It's being discussed at Wikipedia talk:Wikipedia Signpost#Complaint about Jayen466's undisclosed conflict of interest and inappropriate authorship of a Signpost story. This episode suggests that there is a gaping hole in the way that the Signpost handles ethical issues involving its contributors; I can't find any evidence that the Signpost even has a conflict of interest policy. Unfortunately it seems that the Signpost may have left itself wide open to partisans hijacking it to propagate their views without disclosing their own involvement. Prioryman (talk) 19:48, 9 February 2013 (UTC)

So now being a critic of some shady shenanigans constitutes a "conflict of interest". Geez, way to turn the meaning of words upside down Prioryman.Volunteer Marek 06:54, 10 February 2013 (UTC)
Wikipedia:Wikipedia Signpost/Index (shortcut: WP:POST/I) lists pages related to the Signpost.
Wavelength (talk) 20:11, 9 February 2013 (UTC)
For his own benefit, I think that Jayen466 should have steered clear of the issue. Beyond that, I trust that the Signpost editors will review everything and make a solid decision about the story, based primarily on the content.--Jimbo Wales (talk) 23:08, 9 February 2013 (UTC)
I really can't disagree with that. Thanks, Jimbo. Prioryman (talk) 23:19, 9 February 2013 (UTC)
  • Hey Jimmy, I'm generally supportive of the idea that free flow of ideas are one of the things that take down unjust regimes and hope you do go to Almaty... But I was goofing around with Kazakh Wikipedia the other night, playing with some translation software on the net and I noticed that V.I. Lenin was a redlink. The joke being, of course, that he'd probably rather be a redlink than a bluelink. But still... That indicates to me that there is something fundamentally wrong with the way KK-WP or whatever it is called is structured, that what is called Wikipedia is not actually freely editable. Feel free to check me on that information, I only know a little Russian and no Kazakh at all and maybe V.I. Lenin was a redlink only with a certain case ending or something — but that's a real marker, I think, that content is being severely filtered. The redlink I saw was on the KK-WP page for the USSR. Carrite (talk) 06:12, 10 February 2013 (UTC)
Oh, never mind, I was looking for the redlink to show ya and found the page for Lenin instead. They must not have a redirect up for В.И. Ленин --> Владимир Ильич Ленин or something. It is a very weak piece, 12.62k. [41] Carrite (talk) 06:17, 10 February 2013 (UTC) Last edit: Carrite (talk) 06:22, 10 February 2013 (UTC)

Compass Partnership report on WMUK governance

Jimbo, I suspect you already know this, but the WMF-commissioned Compass Partnership report on WMUK governance was released today. WMUK have a blog post about the report. Since there is a WMUK board meeting on Saturday and trustees have expressed a desire for community input prior to that meeting, I am hoping this will prompt people to read the report and comment. Delicious carbuncle (talk) 05:09, 8 February 2013 (UTC)

Just out of curiosity, what is professional development for WMUK trustees? Wnt (talk) 13:57, 8 February 2013 (UTC)
Yes, I have been tracking the issue closely and we've discussed it extensively at the Wikimedia Foundation board level. I encourage lots and lots of community members to take a look.--Jimbo Wales (talk) 16:51, 8 February 2013 (UTC)
I was thinking to copy some sections of general relevance to all chapters to the Meta site - however, although WMUK is distributing this as a CC-SA-3.0 document, it is downloaded as a secured PDF with copying "not allowed". (The PDF has no mention of Creative Commons in it) Can someone confirm that this document is in fact free licensed, so that I can crack it and upload it to Commons? Wnt (talk) 20:20, 9 February 2013 (UTC)
It is CC-BY-SA licensed, but I'm afraid we don't have the unsecured version in our possession. Richard Symonds (WMUK) (talk) 15:14, 10 February 2013 (UTC)
Thanks. For my purposes File:Wikimedia UK gov review rpt v5.djvu (which I've just uploaded) should suffice, though it is a little murky in preview.
(What I'm thinking of is Meta:Meta:Recommendations for chapters and thematic organizations. Not sure if it's really a useful idea, but here's a start...) Wnt (talk) 19:38, 10 February 2013 (UTC)


WebCite will stop accepting new submissions end of 2013, unless we reach our fundraising goals to modernize and expand this service.

Please support our crowdfunding campaign. If you are interested in keeping this service alive, please give generously - or at least share our campaign on Facebook. Funders supporting us with $250 or more will be acknowledged by name on our redesigned website.

I think all should check out and be aware of this proposal on WikiMedia that hugely effects the English Wikipedia.  — Statυs (talk, contribs) 04:46, 10 February 2013 (UTC)

Has anyone contacted to see if they are supportive of a takeover? :-) We don't really have the ability to do a hostile takeover, so if we wanted to do something like this we'd either need their cooperation or we'd need to launch our own independent service (which is not necessarily easy to support in the long run, as our engineers need to focus on building next-generation MediaWiki. I like the alternative of them applying to the Foundation for a grant - I'd put in a good word for them, although I'm not the decision maker on things like that.
One major major concern I have - their website gives no indication of how much money they need. What is their annual budget? If we're talking about $5,000 to cover some hosting bills, I'll write a check myself personally and we'll be set for a year. If we're talking about $250,000 for multiple staff, etc., then I'm a bit puzzled and concerned. Can someone find the person in charge and have them contact me?--Jimbo Wales (talk) 15:12, 10 February 2013 (UTC)
Their current fundraising goal is $50,000 for running costs and modernisation. [42] For full (i.e. funding) members they recommend donations amounting to $1 per new citation. [43] WMF joining them as a full member and paying for a percentage of the 180,000(?) [44] existing links would probably go a long way towards solving their problem. Hans Adler 15:26, 10 February 2013 (UTC)
I think that would be Gunther Eysenbach. I will email you both in a single email. Biosthmors (talk) 20:56, 10 February 2013 (UTC)
I would have expected WebCite to be a project with like Archive-It. (talk) 21:53, 10 February 2013 (UTC)

Checkuser policy: "founders"

As the sole exemplar of the class, I thought I should perhaps inform you of meta:Talk:CheckUser policy#"Founder" CU(s), on the offchance you might want to comment directly yourself. (talk) 04:52, 11 February 2013 (UTC)


I'm sure you've answered this (a lot) but, where did you get the inspiration for Wikipedia? GingerGeek (talk) 17:28, 10 February 2013 (UTC)

  • I am not sure if Jimbo has time to answer now, but it might be because the earlier effort, Nupedia, was so slow to approve articles, and adding content by wiki-editing could generate better articles, much faster. Hence, after Nupedia, came "Wikipedia" with article approval based on wiki-collaboration, rather than a formal, 7-step approval process. -Wikid77 (talk) 19:40, 10 February 2013 (UTC)
  • Nupedia was slow because it was "written by experts and reviewed under a formal process." This created an incredible bottleneck. Meanwhile, the general public wasn't allowed to edit. The Transhumanist 01:31, 11 February 2013 (UTC)
  • That's really interesting. Thanks. GingerGeek (talk) 12:28, 11 February 2013 (UTC)

Neglect of a thoughtful PR editor

Jimmy, you have repeatedly mentioned how "simple" it is for PR editors to follow your bright-line rule against direct article editing by PR specialists. Here is "Kat", a social media consultant in your home base of London. Here is Kat properly using the Talk page of one of her clients' articles to request assistance. That was on December 20, 2012. Here is Kat on January 3, 2013, asking for an assist from one of the frequent editors of the client's article. She never received a response. On January 4, 2013, Kat asked for an "Editor review", but that request remains in a backlog, with no response. On January 23, 2013, Kat asked for help from Wikiproject:Cooperation, which is a project you have called a "complex and noisy community discussion area". After five more days of waiting, that plea finally caught the attention of someone, one of the Wikiproject's main co-founders. Here then is one of your "trusted" editors finally replying to Kat on January 28, 2013, five weeks after her initial request for guidance. In my opinion, he provided only a partially useful comment, barely addressing the main share of her editorial questions. Given the lengthy period of neglect such PR editors are routinely shown, it is no wonder that Kat has taken to editing herself the article about her client. This is just one of many examples of the Wikipedia community's neglect of PR editors who are (mostly) just trying to get Wikipedia to read factually about their clients. I believe that your "bright-line rule" is failing to do anything but add a delay in PR professionals eventually editing Wikipedia directly. I doubt that the intent of your rule was to merely delay the inevitable? Do you believe your bright-line rule is working effectively? Kat is there in London with you; you might consider reaching out to her to thoughtfully guide her with some WikiWisdom. -- SimSaladBim (talk) 16:03, 6 February 2013 (UTC)

I think the "bright-line rule" is the only ethical way forward that works for both PRs/clients and for Wikipedia. Unfortunately I don't think we give good enough guidance for such editors to be able to find help in a timely fashion. Remember that I always say that such editors are free to come here and ask me directly for help. But the main thing is that we need to make sure people like Kat don't spin their wheels doing things that aren't giving them a timely result.--Jimbo Wales (talk) 16:34, 6 February 2013 (UTC)
Could have 3 mandatory sections in a biography: The good, The bad and The ugly. (talk) 23:08, 7 February 2013 (UTC)
I agree generally. I support the Bright Line adamantly, but I have also spent months waiting on Talk, begging and pleading editors to make even non-controversial edits. I appreciate Jimbo's invitation, but would any company realistically think that's a good way to avoid controversy on Wikipedia is to solicit Jimbo's involvement? I don't think it means the Bright Line is a bad idea, but rather we need to do a better job implementing it.
For what it's worth RE "good enough guidance," there's this, which has been incubating for a few months now, to help provide guidance, but it's currently focused on using {{request edit}}, which does not have a well-monitored queue. CorporateM (Talk) 23:41, 7 February 2013 (UTC) (PR guy and frequent COI contributor)

Let's think about this

Whoa. Hold on a minute people and let's step back. Let's look at some first principles: the Wikipedia has two, and only two, stakeholders: the readers, and individual living persons (per WP:BLP). That is it. Except inasmuch as it improves the Wikipedia, we needn't care and mustn't care about other powers, interests, and entities. Not to be harsh, but to be hard, plain, and honest: we do not care one whit if User:Kt1502 is inconvenienced. We do not care one whit if Trilantic Capital Partners is happy or unhappy. User:CorporateM, I don't care if you spent months waiting on Talk, begging and pleading editors to make even non-controversial edits. And neither should any other Wikipedia editor. Only inasmuch as it affects our duty to produce the best encyclopedia does any of this matter.

Yes, of course, if a PR person wants to tweak or add or update some material, and assuming that there's no POV or other problem (a lot trickier to determine than most people seem to think), and the person can't do that quickly, or even at all, because of Bright Line, well of course that's not optimal.

But good grief, let's place this in perspective. The world's not perfect and we have to weigh tradeoffs. Bright Line is a the absolute minimum necessary for the Wikipedia to function as intended (as a trustworthy neutral-point-of-view online encyclopedia, written mostly by volunteers, and with all contributions attributable to a single individual and released for reuse by the public). Against that you have that some material, a very very tiny fraction of 1% of our material (and not really touching our key areas such as science and history etc.), might need to be updated or tweaked and this might be somewhat delayed because of Bright Line. It's just all out of proportion to complain about foregoing that tiny gain when the integrity of the Wikipedia is on the other side of the scales.

And for this? It's nice to make these corrections and in a perfect world all articles would always be up to date, but this is business-directory trivia. What about Khabarovsk#Economy? Here's a huge city and that section's essentially blank. If you were complaining about that I'd have a lot more sympathy. But there's no dough in improving Khabarovsk so screw 'em, I guess.

By the way, the example given is not "properly using the Talk page" at all, and is a perversion of Bright Line to my mind. A major rewrite of an article is extremely unwelcome in my opinion, because vetting it is very difficult and time-consuming -- at the very least, the person vetting the PR rewrite has to thoroughly acquaint themselves with the entity (especially difficult for financial entities), to ensure that nothing important is being left out or spun -- and putting this onerous task on volunteers is unfriendly and frankly a jerk move, in my view.

I think Bright Line should be used for small, specific corrections, mostly of plain facts, not for "vet my press release, kthx". The miniscule price to pay for this (in terms of some good changes abjured) is very overwhelmed by the gain. People need to stop posting press releases (that is, substantially rewritten articles) and imagine that they are following the spirit of Bright Line. Herostratus (talk) 08:06, 8 February 2013 (UTC)

I don't necessarily disagree, and your arguments are reasonable. But there is this habit (and I mean it with all due respect) of raising such reasonable criticisms against PR participation, without offering any valid solution. Do you have an alternative?
I can think of only one better way of doing it, which is to have "sponsored editors," where the company has no control over the contents. CorporateM (Talk) 14:18, 8 February 2013 (UTC)
I don't buy this argument. "We're here to produce the best encyclopedia" is not a reason to ignore people's complaints on the grounds that satisfying the complaints doesn't improve the encyclopedia. In fact, you unintentionally show the flaw yourself by adding the qualifier "and individual living persons" as if Wikipedia was otherwise obligated to harm people and if we didn't have a specific policy against it, harming people to improve the encyclopedia would actually be an obligation.
If we don't care if some person is happy or unhappy, that is a problem on our side. We certainly do not need to care about all possible sources of some person's happiness, but neither should we reject the complaint because we aren't required to care at all. What we say about companies can harm companies--and ultimately, people--in the same way that what we say about individuals can harm them. We are obliged to care, not because of Wikipedia policies, but as human beings.
Companies should be allowed to edit articles about themselves under exactly the same circumstances that individuals are. We won't allow individuals to do the wholesale rewrites you are talking about, so there is no need to allow companies to do so. But we also wouldn't totally ban individuals, and there are some edits by them that we would allow. Companies need to be allowed to make edits analogous to those. Sure they have a conflict of interest. So does the individual, but sometimes we allow him anyway. Ken Arromdee (talk) 16:29, 8 February 2013 (UTC)
Like in all things, there is a balance to be had. Wikipedia editors, unlike professional journalists, are often overly sympathetic to PR representatives. We don't serve the PR person, but if the PR person has something of value to the reader they would like to contribute, there should be reasonable pathways to do so. If a PR person provides 10 hours worth of work, it should not be unreasonable to ask for 20 minutes of impartial eyes on it. CorporateM (Talk) 19:59, 8 February 2013 (UTC) (PR guy and frequent COI contributor)
Bluntly, CM, I've yet to see one of these drastic overhauls that didn't make the article worse (i.e., more in accordance with the subject's wishes and less bluntly impartial, with extra little fluffy bits floating on top). They should be de facto forbidden, in favor of gradual and open improvement to the parts of the article that actually need improvement. The client's desires do not constitute the volunteer editors' emergency, especially when there are articles like vocational rehabilitation that need improvement so much more, and matter to so many more people. --Orange Mike | Talk 21:04, 8 February 2013 (UTC)
Incremental improvements that don't require a time investment from the community would then suggest direct-editing no? What would you suggest the company who has a terrible entry do? CorporateM (Talk) 00:21, 9 February 2013 (UTC)
Well "terrible entry" is in the eye of the beholder. If it's clearly, unambiguously, and incontrovertably terrible, well then changing it should be easy. No reasonable person will prevent the redaction of learly, unambiguously, and incontrovertably material. If "terrible" is used to mean "not presented the way I, personally, would like" then what they should "do" is suck it up. If its somewhere in between, then some reasonable engagement on the talk page is called for. There's lot of things in this life that we can't control. Our Wikipedia entry is one of them (except for individuals, and even then only somewhat).
As to "...without offering any valid solution. Do you have an alternative?" yeah I do and Ive offered it before and you've seen it: User:Herostratus/Articles about extant organizations. It's sort of WP:BLP, but for organizations. It's a reasonable idea, and if people want to talk it up, go for it. It hasn't gained much traction. That indicates that the community doesn't see a problem to be solved here. Agree or disagree, the community rules and you have to respect that -- and accept it.
@User:Ken Arromdee, I get what you are saying: corporations consist of people, who can have their individual interests (and reputations, and feelings) hurt by the contents of a Wikipedia article, and thus should be subject to BLP-like considerations. The problem is, the community doesn't agree with you. Get the community to accept User:Herostratus/Articles about extant organizations or something like that and we can revisit the issue.
Of course we shouldn't go out of our way to make organizations unhappy. But when we're writing about an organization, or town, or party, or religion, or whatever, then "is it accurate and fair?" is a hella more important consideration than "will they like the article?", don't you think?
By the way, should we be troubled if the Workers' Party of Korea, Exxon-Mobile, Gazprom, the Westboro Baptist Church and so forth are not happy with their articles? That's certainly debatable; if those entities are completely happy with their articles then we're probably doing something wrong. And picking and choosing which organizations are or are not benign enough to merit special consideration is not something we can do well, I don't think. Herostratus (talk) 02:50, 9 February 2013 (UTC)
I would strongly support BLP-like considerations for companies. It's much more practical for volunteers to clean things up then to ask paid editors to get involved when the company is not fairly represented. But thus far it hasn't gotten support from the community... unfortunately. It's a good idea though - you could make another run at it - maybe have better luck than I have. CorporateM (Talk) 14:12, 9 February 2013 (UTC)
The answer to "would we be troubled if (evil corporation) is not happy with their article?" is the same as the answer to "would we be troubled if (evil person) is not happy with their article?". You may suggest that we should ignore complaints and not allow editing from the Westboro Baptist Church, but even by your own standards, we should consider complaints and allow editing by Fred Phelps. Surely whatever horrors you're trying to stop by not allowing editing by the church could still happen if Fred Phelps himself edits his own article. Fred Phelps editing an article about himself is still subject to all our rules about editing articles, and in practice, I would not expect him to have much to add that obeys our rules. But you would not flat-out ban him sight unseen. Ken Arromdee (talk) 16:04, 9 February 2013 (UTC)
Upton Sinclair said it best "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" It is difficult enough that they are already funding much of the research, ghost writing the articles, selectively publishing the results they like, hiding the patient level data from independent analysis, and paying the journals large amount for reprints and advertising. Not to mention paying "experts" to go around and promote their products, paying for "consensus conferences" and lobbying the government. Plus in some areas directly advertising to the populous.
Wikipedia is one tiny place where PR people must not be allowed. The big issue we have is that we occasionally allow these sorts of folks to hide behind the WP:OUTing policy. If a PR person is editing nefariously they should get zero protection. It is amazing that the science has any chance at all. Doc James (talk · contribs · email) (if I write on your page reply on mine) 17:13, 9 February 2013 (UTC)
In some cases the COI - though not declared - is obvious enough to warrant action. I like this statement: "when his salary depends upon his not understanding it." But what if their job is reliant on them understanding it? There is honorable work to do helping companies be fair and honest; companies who struggle to be neutral and need help. As oppose to being hired with nefarious or corrupt intentions. CorporateM (Talk) 18:28, 9 February 2013 (UTC)
And when people are paid they are unable to tell the difference. Companies struggle to make a profit for their shareholders not to be neutral. I have not seen a neutral tv or journal ad yet. And the evidence states they are few an far between Othman, N; Vitry, A; Roughead, EE (2009 Jul 22). "Quality of pharmaceutical advertisements in medical journals: a systematic review.". PloS one. 4 (7): e6350. PMID 19623259.  Cite uses deprecated parameter |coauthors= (help); Check date values in: |date= (help). They even know about adding references to make what they write look more scientific but it does not always.Lankinen, KS; Levola, T; Marttinen, K; Puumalainen, I; Helin-Salmivaara, A (2004 Nov). "Industry guidelines, laws and regulations ignored: quality of drug advertising in medical journals.". Pharmacoepidemiology and drug safety. 13 (11): 789–95. PMID 15486957.  Cite uses deprecated parameter |coauthors= (help); Check date values in: |date= (help) Doc James (talk · contribs ·email) (if I write on your page reply on mine) 18:52, 9 February 2013 (UTC)
I would suggest that the personal stake a BLP subject has in their own article is as great as the financial stake an employee has in the company's article. (Indeed, there are plenty of cases where a BLP subject has a direct financial stake in his own article anyway.) Yet we don't prohibit BLP subjects from editing their own article. I would also suggest that a BLP subject is about as unlikely to be neutral as a company spokesman is going to be neutral. Again, this isn't a reason to prohibit BLP subjects from editing their own article. The arguments you're giving have been proffered, and rejected, when it comes to BLP subjects, and the exact same reasoning as to why they are rejected for BLP subjects also applies to companies. Ken Arromdee (talk) 06:09, 10 February 2013 (UTC)
Let's get 2 things straight:
1) The article wasn't a "terrible entry", nor was the company "not fairly represented". The only factual correction concerned the new name "Marex Spectron" instead of the old "Marex".
2) The PR people requested that the intro be changed from:
changed to new lede:
    • Trilantic Capital Partners (Trilantic) is an independent, transatlantic private equity firm seeking to make investments in fundamentally sound and growing companies in North America and Europe with the aim of achieving superior (risk-adjusted) returns. Trilantic has a particular focus on risk management, distinctive investment strategies and takes an active role in creating value post an acquisition. The firm uses flexible transaction structures and has historically partnered with family-owned businesses as well as providing growth capital to outstanding management teams. The firm was formerly known as Lehman Brothers Merchant Banking (LBMB), the buyout arm of Lehman Brothers. In 2009, Trilantic was formed by its five founding partners as part of the bankruptcy of Lehman, the LBMB business was sold to its partners and a minority interest was acquired by Reinet Investments, a listed investment trust controlled by the Rupert family and chaired by South African businessman Johann Rupert.
I'm not sure this is really an improvement of the lede. Quite a lot of nice words for a positive sentiment analysis, I guess. Conveniently pushing back the bad word "bankruptcy" to the fourth sentence. Further, the PR people asked for "adding more info on the key partners and updating the investment section." For a company of 35 employees, they wanted 10 "key people" to be listed or "boxed out" (some twice) in the article (presumably their bosses). I'm sorry, I don't see the emergency here, nor encyclopedic knowledge that is sorely missing. In my understanding the bright line doesn't mean that editors have to cater immediately to every wish of the PR people of companies. --Atlasowa (talk) 19:08, 10 February 2013 (UTC)
This example in particular would be a good time for this:
CorporateM (Talk) 21:01, 10 February 2013 (UTC)
Yes perfect example of why we do not allow COI editors to edit directly. Just a lot of fluff. We are an encyclopedia not an advertising forum. Doc James (talk · contribs · email) (if I write on your page reply on mine) 22:14, 10 February 2013 (UTC)
We do allow COI editors to edit directly. They still have to not make COI edits, of course, but it's false that we don't allow them to edit. At least when they are BLP subjects.
As for this particular case, I couldn't care less. If you're going to argue that the edits made by the PR people here are poor, I'm not going to dispute that. What I disagree with is that this justifies a general rule "never let a company edit". This is yet another of the arguments that has been made and rejected for BLPs--you can easily point to cases where BLP subjects made poor edits for their own aggrandizement, but those cases don't justify a general rule of "never let BLP subjects edit". If they don't justify such a rule for BLP subjects, why should they justify such a rule for companies? Ken Arromdee (talk) 16:38, 11 February 2013 (UTC)
In theory a COI that does not make COI edits is acceptable, but in practice COIs have a strong propensity to make COI edits, such that it is not unreasonable for us to make the assumption that they will. Why would any COI create a new article, when they can submit it to AfC and let an impartial editor review and assess it? By the same token, if the Request Edit system is adequately manned, even non-COI edits made by COIs could use a second pair of eyes. CorporateM (Talk) 20:00, 11 February 2013 (UTC)
I'm not talking about "in theory". BLP subjects have a conflict of interest and we allow them to edit. We do not use their conflict of interest as a reason to ban them from editing their article, nor do we force them to use a request edit system.
Again, your reasoning applies to BLPs--but we reject it there. Ken Arromdee (talk) 06:24, 12 February 2013 (UTC)
I agree that there is way too much focus on "forcing" PR people, using threats by the anti-PR crowd, when any good-faith PR person would do so voluntarily if they are both (a) given clear instructions and process and (b) the system works well. I think the difference is minor BLPs are not expected to know better, while professionals are; professionals are subject to legal guidelines, more likely to be humiliated in the media, etc. and we should give them advice on how to stay out of trouble. The whole "evil corporation" bit is not an argument at all and saying PRs should only use Talk pages and that we should also not respond to them is basically telling them to go stick-it. CorporateM (Talk) 14:01, 12 February 2013 (UTC)
I don't think that makes sense as a reason BLPs are treated differently. If that was the reason, we'd give BLPs a warning the first time, and ban them from editing their own article once they have been warned and can no longer be said to not know better.
The reason we don't do that is that there's no person better motivated to fix problems with the BLP than the BLP subject and the subject should not have to tolerate lies and bias even one second longer than necessary, let alone weeks or months. If the BLP subject is himself prone to bias and letting him edit means we have to deal with self-promotion, then we have to deal with it. I do think this rationale applies to companies; if anything, even more so since Wikipedia has neither the procedures nor the willingness to fix them like it does with BLPs. Ken Arromdee (talk) 16:28, 12 February 2013 (UTC)
I've reverted back to the old lead - as well as not being neutral, the new version cited FT articles for its content, but they said nothing like what they were being cited for. SmartSE (talk) 14:29, 12 February 2013 (UTC)

About Turkish Wikipedia Admins

Turkish Wikipedia admins do not do the work to find the middle way. Admins are not constructive. Availability drag them into chaos. Bureaucrats not doing their duties. Turkish Wikipedia admins not longer enough. Or are they doing on purpose. --This unsigned article written by: User:Aguzer 20:15, 11 February 2013 (UTC)

  • Turkish Wikipedia admins quiets users. Continuous blocking. -Writer: User:Aguzer 10:16, 12 February 2013 (UTC)
Jimbo ! And deal with these issues now.. ?

-Writer: User:Aguzer 10:47, 12 February 2013 (UTC)

QRpedia - going forward

After at a protracted negotiation, a deal has been reached to transfer ownership of QRpedia to WMUK. Echoing comments I have made on Meta, I strongly believe that this is a facility that should be centrally controlled and supported. For that and other reasons, I suggest that the WMF create (or maybe and the existing (as of now controlled by WMUK) be a redirect to that site. The code is available under a free license, but it would not be difficult to develop the same functionality. This is a mechanism that is likely to be used internationally and it seems misplaced to have it so strongly associated with a specific chapter, particularly under the circumstances which prompted the governance review. Delicious carbuncle (talk) 19:27, 9 February 2013 (UTC)

I think QR codes are robot barf that will fade into obscurity in due course. An unexplored part of this scandal is that people have been asked to do something stupid, which is slap QR codes all over their towns. There's a great technology that is comfortable for both humans and machines to read. It's called "text". I think WMUK and all chapters should be warned that the best technical advice around QR codes is this: run away from them as fast as you possibly can.--Jimbo Wales (talk) 23:16, 9 February 2013 (UTC)
Can you explain which text you mean which could achieve the same result? Because the important part is the automated language selection without the need for specific software.--Saehrimnir (talk) 07:34, 10 February 2013 (UTC)
As an afterthought you are right for this purpose QR Codes are overkill one could just say and achieve the same result and the half of the world population who speaks English could be relatively sure that they get to the right spot without malicious content.--Saehrimnir (talk) 10:52, 11 February 2013 (UTC)
To any monolingual audience that would be true, and not just in the English speaking world, if you were somewhere where everyone spoke Spanish or somewhere else where they all speak Japanese then the whole multilingual aspect of QR codes would be wasted. However my understanding of them is that they tend to be used in museums and heritage venues and especially in places which have or aspire to have a multilingual audience. ϢereSpielChequers 10:00, 12 February 2013 (UTC)
Frankly, I'm not sure about that. The reason QR codes exist in the first place is because URLs are too long and cumbersome for mobile device users to type in. As long as that remains the case, machine-readable shortcuts like QR codes will be necessary. Of course, QR codes aren't the only way of achieving that. The other day I saw an unusually high-tech bus shelter that had, next to an advertising poster, a QR code on an LCD display built into the surround, with an NFC device below it. Both provided machine-readable shortcuts (reprogrammable, I assume) to the online version of the advertising campaign. It may well be that NFC will replace QR codes for some applications - I think it more than likely - but the ease of use of QR codes, which you can generate and print at home, is likely to mean that they'll be around for a long time to come. Prioryman (talk) 23:31, 9 February 2013 (UTC)
What nonsense. Any smartphone with Google Goggles (or similar) installed easily has the OCR capability to translate a text sign into machine-readable text, and will go to a website just by pointing your phone's camera at the URL, and anything capable of running a QR scanner is capable of running OCR. I have, quite literally, never seen anyone actually scan a QR code. I have often seen people photographing signs. (talk) 23:36, 9 February 2013 (UTC)
One bored kid (and or "banned user") with a laser printer and some glossy adhesive paper could easily turn Monmouthpedia and Gibraltarpedia into goatsepedia. Worse, the same strategy could be used to install some malware on the way to Wikipedia. That will not be good PR for anyone involved. --SB_Johnny | talk✌ 23:50, 9 February 2013 (UTC)
That would work with laser printers that produce ceramic plaques, but most print on paper. ϢereSpielChequers 00:00, 10 February 2013 (UTC)
Plaques can presumably be covered by stickers, though as someone pointed out we may be dishing out the "WP:BEANS" by hashing out the theoretical possibilities ;-). My point is that vandalism in the real world might be harder to fix. --SB_Johnny | talk✌ 00:07, 10 February 2013 (UTC)
If that's a real risk, presumably it's a risk for QR codes generally. Is it a documented problem? Formerip (talk) 00:09, 10 February 2013 (UTC)
Yes, it has. Tarc (talk) 01:52, 10 February 2013 (UTC)
That article seems to be talking about an online thing, rather than a real world thing. Formerip (talk) 16:57, 10 February 2013 (UTC)
A lot of silly things are said about QR codes. QR codes are lot more trouble to vandalise than Wikipedia articles. As everyone knows Wikipedia should in the eyes of the uninformed be full of errors and misinformation, but this is not the case. QR codes will of course be superseded but they will make their 20th anniversary next year! Text is a fine solution - but as we know it has been superseded by hypertext. Towns are not sticking QR codes all over themselves are far as I know unless you look at every crisp packet. NFC is in some ways a better idea and QRpedia is upwardly compatible. However as Terence Eden notes you cannot put NFC on a packet of salt and you cannot get much text either. Thousands of new articles and hundreds of editors working together. Imagine yourselves in a museum in China or Korea with lots of text on display. All you have to do is type the text into your phone and Wikipedia (Or another similar site) can tell you what it is in Korean. I think we can do better. Victuallers (talk) 11:16, 10 February 2013 (UTC)
Jimbo, your opinion of QR codes and the people who love them notwithstanding, when someone in a museum scans a QR code expecting a Wikipedia article and gets a poor result for whatever reason, it is Wikipedia that will look bad, not WMUK or the museum. Given that these are already out there, I think it would be wise to mitigate that risk to the project's reputation. If it is to be taken under the WMF's control, it would be better if this were to be negotiated before WMUK expends any effort or funding on the project (I believe there are some concerns over data privacy that they intend to address in the near term). Delicious carbuncle (talk) 11:58, 10 February 2013 (UTC)
Yes, I agree that we should ensure that these codes work solidly for a long time to come, and that the safety of this is an important consideration. My personal views of QR codes (that they are a technology that will go away very soon) should not be taken to be a negative view of museum/city partnerships in general, of course! Just this one little technical aspect.--Jimbo Wales (talk) 03:27, 11 February 2013 (UTC)
An interesting aspect of the QR system is that it would allow volunteers to decide to document local attractions even if they are not judged notable enough for Wikipedia. For example, what should the local volunteers do if they arrange for a museum to buy a dozen fancy ceramic plaques, then half the articles are deleted? Well, if WMUK controls a qr domain, it might point that article at some "local project", e.g. a standalone Monmouthpedia or whatever, where these "supplemental" entries can still be read by tourists. Personally I view this as a very good thing in concept, though one can see some serious potential problems that could come up. However the system is handled, I think it would be useful to plan ahead for that situation, to try to ensure that such a supplemental site is well-run - that it can exempt itself from notability but not become terribly biased, infiltrated by spammers, etc.
I agree with Jimbo (I've said it myself) that the alphabet is a remarkable cutting-edge technology to use in situations like this. I can picture a plaque with a cute little logo like "LnK:" in an oval, followed by text, e.g. "LnK: Wikipedia/Gibraltar Manhole Cover 136" - first, however, someone would have to convince people to use such an app, which is where QR has excelled, namely in marketing. Unfortunately, across the Internet we see people and companies feeling forced to establish presences on really bad platforms - like Facebook - distinguished only by their cleverness in using non-traditional methods to do their advertising. Wnt (talk) 15:42, 10 February 2013 (UTC)
More likely you'll just be able to murmur quietly to your smartphone "show me the wikipedia article for that whatsit", and it will oblige. The problem with QR codes is that they aren't human-readable, so you don't know what you're getting into when you point your phone at them. --SB_Johnny | talk✌ 00:53, 11 February 2013 (UTC)
The future is more along the lines of location awareness in general, where specific apps use AR. As for the smartphone, it's probably going to go the way of the Dodo in favor of wearable computers, watches, glasses, and transparent computing embedded in everyday things like furniture, cars, etc. Hardware is quickly disappearing into the background as it gets smaller while data is becoming visualized as a normal overlay over everyday reality. Blissenobiarella 22:08, 12 February 2013 (UTC)
This concept image of a proposed Apple smart watch, made out of bendable smart glass and designed as a bracelet, is a pretty good example of the kind of thing people are thinking about. One could also envisage smart devices embedded in clothes. We certainly won't be interacting with them in the same way that we do with smartphones. Prioryman (talk) 22:20, 12 February 2013 (UTC)

WebCite has responded

There was a reply from Gunther Eysenbach here. I assume at this point we should leave it up to the WMF to start discussions on what the proper action on the WMF's part would be. SilverserenC 02:03, 13 February 2013 (UTC)

Poem and picture of the day

Ubud Cremation 4.jpg

Ash Wednesday

Letting Go of the Past
It is frightening to let go of the past. It is like letting go of something that is precious.
That includes the feeling for everything that has been, and also, for what was once a solid identity built on deprivation.
Letting go of the past must be done gradually and with special care; one old belief at a time
and only one fear. ...
By Poeticbent, with permission, thank you.

--Gerda Arendt (talk) 10:52, 13 February 2013 (UTC)

If IP editors were banned

(Reverse topic of "#Banning IP editing" above)

In this thread, let's imagine Wikipedia has banned all IP edits, and discuss what happens next. I wonder if it would be like, "Let's ban all children under age 4 in public until they can walk and talk better". Anyway, with IP editors banned, then every editor must create a username and login. The first result I think would be no more easy spotting of IP edits as potential vandalism; instead every username must be suspected, but perhaps the redlinked usernames could be considered more likely to post a hacked, or misguided, edit. How often would people ask about forgotten passwords? If we banned IP edits for 2 weeks, what complaints should be expected then? -Wikid77 (talk) 01:21, 7 February 2013 (UTC)

Maybe it would result in a proposal to ban editing by redlinked usernames, since it would be easily established (during the two weeks) that redlinked usernames were responsible for the vast majority of vandalism. --Demiurge1000 (talk) 12:13, 7 February 2013 (UTC)
why stop there? It has been adequately proven that 100% of vandalism occurs because of edits. If we ban editing altogether, that would completely eliminate both the need for and the possibility to fight vandals. This would also stop editwarring, pov pushing, wikilawering, trolling, and many,many other problems. Martijn Hoekstra (talk) 09:15, 10 February 2013 (UTC)
  • Support - I support Wikid77's proposal to ban IP editing on a two week trial basis, although I am somewhat surprised that there was no reference to how Lua could speed up that process. Delicious carbuncle (talk) 04:24, 7 February 2013 (UTC)
  • Support - Anything is better than the status quo. Mugginsx (talk) 12:28, 7 February 2013 (UTC)
  • Oppose. This is a perennial proposal, see Wikipedia:Perennial_proposal#Prohibit_anonymous_users_from_editing. It is also a Foundation issue and not subject to discussion lacking a change in WMF's central policies. Dcoetzee 13:17, 7 February 2013 (UTC)
  • Oppose. As I explained in the other section above, this would primarily affect casual vandals. Determined IP-hoppers would easily develop swift and convenient work-arounds (I certainly know one way of doing it). As a result we would have a few determined vandals less, very few casual vandals (who are not a big problem), and lots of inconvenience due to the irreversible step forward in the arms race.
    Two weeks might not be enough time for the negative effects to kick in, and if they are, the damage would be done at that point. There is no step back once sophisticated tools have been developed and widely distributed. At the moment there is no market for these. Let's not create one. Hans Adler 14:05, 7 February 2013 (UTC)
  • I don't think WIkid77 actually proposed banning IPs for two weeks. He simply asked what, hypothetically, we could expect. Actually implementing such a short-sighted and frankly idiotic idea is something I vehemently oppose. Resolute 15:12, 7 February 2013 (UTC)
  • Many help-pages would need to be altered: The references to "IP editing" are scattered everywhere through the various help-pages about Wikipedia, and it would be confusing if new users then came to Wikipedia and read how "anyone can edit" (without prior approval of usernames) if it were no longer allowed. Even the login/logout text would need to be updated:
  • The logout message would need to remove the phrase, "This computer may be used to browse and edit Wikipedia without a username".
  • The edit-message would need to be changed to request login, rather than saying, "You are not logged in. Your IP address will be publicly visible if you make any edits".
Hence, even a 2-week test would require changing many pages to no longer state, "edit without a username". Then, after the test, all those pages would need to be reversed to mention IP editors being accepted again. Fortunately, as noted above, the acceptance of IP editors is a WMF decision, so we do not need to discuss every problem that banning IP edits would cause. Obviously, people who lacked usernames would not even be able to report a login failure, because their IP edits would be blocked, so another dialogue would be needed to allow questions from people who could not create a username. Overall, banning of IP editors would cause quite a mess at first, as more of these issues are considered. -Wikid77 (talk) 17:06, 7 February 2013 (UTC)
  • Oppose. For one thing, we could expect a drastic drop-off in new editors. How many regular and semi-regular editors started off editing anonymously, without any issue or incident? Banning anonymous editing would make this "The Encyclopedia that authorized users can edit", and that's not what I signed up for. UltraExactZZ Said ~ Did 17:42, 7 February 2013 (UTC)
    • I hate to break it to you, but "The Encyclopedia that authorized users can edit" is what we have now (since we ban users). Nor would making users spend a few seconds creating an account qualify as requiring authorization. Rd232 talk 18:21, 7 February 2013 (UTC)
  • Since the point of the post was to prompt discussion: banning anonymous editing altogether is a difficult and dramatic step with many pros and cons that are hard to weigh. What we could do is remove privileges from anonymous editing - as we did years ago by removing the privilege of creating new articles. The obvious privilege (to me) is for edits to appear on the site live immediately to the public. Queue all anonymous edits, then tell users (again - tell them before/during/after editing) that if they want the edit to go live immediately, they need to log in or create an account - otherwise, it'll be queued for review. Rd232 talk 18:21, 7 February 2013 (UTC)
  • Support - I consider IP editors in the same manner I consider the vermin in my garden, so anything we can do to cull the numbers is a good thing. Tarc (talk) 18:32, 7 February 2013 (UTC)
That's a tad harsh. Having said that, I'm starting to think that eliminating anon editing is inevitable in the long run. I don't think it is sustainable to rely on a volunteer army of patrollers to act as playground monitors. There will always be vandalism, but monitoring and blocking accounts would be infinitely more manageable than monitoring and blocking billions of IP addresses. Philosophically speaking, it would be a huge change and I don't expect it to happen anytime soon. But creating an account is far from difficult - it takes about 20 seconds, and it doesn't even require an email address. Honestly, anyone who wants to contribute wouldn't see this as much of a barrier. In my opinion, allowing anon editing is a gimmick that has outlived its usefulness. --Bongwarrior (talk) 23:18, 7 February 2013 (UTC)
Err, this "gimmick" built Wikipedia. --MZMcBride (talk) 07:24, 8 February 2013 (UTC)
Absolutely, no argument from me. But now that it's built, is it still necessary or prudent going forward? The cost/benefit of allowing anon editing for a fledgling website is much different than it is for a top-ten website. --Bongwarrior (talk) 08:12, 8 February 2013 (UTC)
It wasn't harsh at all; "harsh" was the first draft when I directly labeled IP editors as vermin, but altered before final submission. Tarc (talk) 14:22, 8 February 2013 (UTC)
@Tarc there is not so much a difference if you say they are vermin or if you imply they are as bad a problem which has to be handled as such. @Bongwarrior if you consider Wikipedia to be finished then yes but there will be fewer new authors and the effort which now goes into fighting vandalism has to be channeled into making the useful contributions of IP editors and getting the same amount of different information with fewer searchers is always difficult.--Saehrimnir (talk) 13:48, 12 February 2013 (UTC)

Oppose a two week test isn't really sufficient to to test the important question "What fraction of new editors are we going to drive away with this kind of hostility?" It'd take a much longer test (and by the end of that, the only people left apart from Vandals will be Visigoths). Given how significant the issue of the dropoff in new editors (and correspondingly, admins) is viewed, something that severely exacerbate that process can't be done on a lark. We shouldn't burn down the encyclopaedia without first giving it significant thought. WilyD 08:37, 8 February 2013 (UTC)

  • Support a test, without a sense of how it will turn out. The obvious "this will drive constructive IPs" argument is of obvious concern, but might be overshadowed by the increased communication we would have with those editors. Helping people get past the various learning barriers to making constructive edits often requires communication, communication which is nearly precluded (how often do we really see new editors reply at by a place and a notification system for that communication. If the goal is increasing the eventual number of constructive edits and editors, the idea that this might be a win, instead of a lose, is plausible enough to deserve a test. --j⚛e deckertalk 19:09, 9 February 2013 (UTC)
  • Oppose. I consider Tarc and his like to be a bigger problem for Wikipedia. (talk) 20:11, 9 February 2013 (UTC)
  • Verifying IP edits will fix 90% of vandalism: Well, it seems clear that, in the past, over 90% of vandalism was posted by IP editors (while over 70% of IP edits were good). So, even though some hack-edits were posted years ago, we can fix most factual vandalism by easily re-verifying the old IP edits (and only those IP edits not already reverted) and look for a suspicious change in a date or other fact as made by an IP edit. By looking at mainly those edits, then we can easily correct facts from years ago, and less than 10% of hack-edits will slip past. However, if IP editors were banned, then we would need new tactics to detect most hack-edits, as "escalating the weapons of vandalism". Instead, perhaps we need to inspect only 6 edits per 50 history entries, going back 3 years, as pinpointing non-reverted IP edits, to fix such problems. That is so much easier, compared to vandals learning to create usernames so their edits would become hidden among long-term usernames, where most of each 50 history entries would then need to be checked for altered facts. I thank everyone, above, for clarifying those issues, as to how much worse it could be if IP editors were banned, and how easy it is, now, to verify just 6 of each 50 entries for non-reverted IP hack-edits posted 1-3 years ago. -Wikid77 (talk) 07:39, 10 February 2013 (UTC)
  • The level of support for this proposal is disturbing. The negative IP contributions that we all know about necessarily receive much more attention than those small improvements to the encyclopedia made by anonymous passers-by from all over the world. I don't have stats, but I find it hard to imagine that IP contributions are anywhere near a net negative. — This, that and the other (talk) 09:31, 11 February 2013 (UTC)
  • I'm not convinced that a two week trial would be long enough to test the premise that we need to allow IP editing as an easy first step to becoming a new registered editor. But a two week trial would be long enough to confirm the theory that putting up barriers is more likely to dissuade good faith editors who want to help than badfaith editors who despite the challenge want to commit vandalism. So my prediction is that if we did this we'd lose a lot of edits, but disproportionately we'd lose more of the hard to replace good edits than the easily reverted bad ones. We'd also have to be much less trusting of new accounts if creating an account became the minimum effort needed to start vandalising articles. ϢereSpielChequers 10:10, 12 February 2013 (UTC)
  • Oppose, "anyone can edit" is not true anyway, why make it worse? --Gerda Arendt (talk) 10:48, 13 February 2013 (UTC)
  • Oppose, isn't this why we have cluebot? besides, what would all the vandal-fighters do if the ip users couldn't vandalize? on a more serious note, I have seen ip users make good contributions, and I really don't see a problem with the current method of vandalism protection and prevention. Aunva6 (talk) 06:47, 14 February 2013 (UTC)

Stopping worst IP edits

(Another spin-off of thread "#Banning IP editing").

We should discuss what editors can do about protecting more templates. Just yesterday, another horrific IP edit hacked a template #redirect which could foul 20 major articles for 7 hours: garbled text "water is never pure in a chemical sense..." was put in shortcut Template:Ety (edit | talk | history | links | watch | logs), used to show the etymology after an article title, but reverted at 12:18 on 9 February 2013, within 8 hours as the only 2 edits by that IP editor. The nonsense intro (in 20 major articles: Science, Planet, Ocean, Disease, Gorilla, Epidemiology, Number, Neologism, Photographer, Epistemology, Egalitarianism, Vivisection, Cybernetics, Bazaar, etc.) was likely reduced because articles are slow to reformat after a shared template is changed. However, redirects still pose a loophole, where the popular original template gets semi-protected, but the redirect is left vulnerable because edits are so rare, and no one can "show a pattern" of recent vandalism as evidence to protect the tiny redirect, and then, "bam!" it gets hacked when many articles have used the shortcut name, rather than the protected template name. Some issues to consider:

Q1. Do we have a rule now to say, "Semi-protect the redirects of a protected template"?
Q2. Is there an anti-vandalism group who seeks to protect popular templates?
Q3. Do we have a page-impact formula (#transclusions × #pageviews) to trigger protections when used in only 40 (but major) articles?
Q4. What are some other "worst IP edits" and steps to stop them?

The fact that a horrific IP edit could zap the lede sections, of so many major articles for hours, while we were discussing IP-edit bans, is not mere coincidence, but rather more evidence that IP edits are still a massive, major problem (in templates), where more could be done to reduce the impacts. Meanwhile, we need to remove templates where non-protected in the lede of major articles: {{ety|la|scientia|knowledge}} = from Latin scientia, meaning "knowledge". To err is human, but to really trash Wikipedia requires contempt-lates. -Wikid77 (talk) 19:40, 10 February 2013 (UTC)

This looks like it could be a good faith edit; certainly whether in good faith or bad, a new account could have done it also. You're telling us that most of the vandal edits on Wikipedia come from a small, easily recognizable subset of the edits made - and you're complaining about that. Think about that. (But in the meanwhile, protecting or completely eradicating these protected template redirects isn't a bad idea) Wnt (talk) 19:50, 10 February 2013 (UTC)
provided can be fixed, would it make sense to extend cascading protection to redirects to the page? Or define a new protection form that also automatically protects redirects? Martijn Hoekstra (talk) 02:55, 11 February 2013 (UTC)
for indefinite protection a gadget could well work. I'll see if I can make something when I get home. Martijn Hoekstra (talk) 03:23, 11 February 2013 (UTC)
  • At least, if we had a list of unprotected redirects to popular templates, we could spot the weak ones. As I noted above, when 20 articles are major, then "20 transclusions" is reason enough to semi-protect a template. -Wikid77 12:40, 11 February 2013 (UTC)

Why not just ban IPs from editing templates? The arguments for allowing IPs to edit articles apply much, much more weakly to templates. Rd232 talk 12:16, 12 February 2013 (UTC)

Agree with Rd232. All we have to do is semi-protect all templates, and the problem is reduced by a considerable percentage. Jusdafax 22:28, 12 February 2013 (UTC)
Now that idea may have merit. If there isn't support to ban IP editors outright project-wide, perhaps they can be confined to article and user-space only. Tarc (talk) 22:38, 12 February 2013 (UTC)

perhaps semi-protect all templates? articles are fairly well protected by bots and vandal-fighters, so that's not much of a problem. Aunva6 (talk) 06:52, 14 February 2013 (UTC)

Problems with a moderator in Wikipedia

Hello, i'm a big fan and old user of wikipedia. From 4 months ago i started with a porject with some of my professors from university that could help some people. We are developing videos of all the philosophers, in an easy language, just with the term of present them. Maybe it happen in us or other countries, but there isn't ant relevand and reliable material in portuguese. So, we started to make this integrated with a magazine (revistaligados) and a blog (ligadosfm), to show how safe it is. And as the protocol of Wikipedia is not to be inventive, ilegal, or out of the canone, we started to link the posts fo the blog with the video of each philosopher in each page correspondent. However, a moderator Yanguas is saying that is spam, and refuses himself of see the video analise it. - That is terrible and now i'm afraid of lose my account here and can't comtribute with Wiki.

links: ; ;

Please contact-me :)

--Felipobellini (talk) 21:26, 12 February 2013 (UTC)

Seems to relate to this and nearby messages on that talk page. --Demiurge1000 (talk) 23:07, 12 February 2013 (UTC)

yes, it is. and as you can see, all these spans advices just apeared yesterday, what look's as a inapropriate case of redicalism.

--Felipobellini (talk) 17:50, 13 February 2013 (UTC)

Does the WMF have a position on meta:WebCite?

Hi Jimbo, I was wondering if the Foundation had a position on the proposal at meta:WebCite to take over the service we use to archive our references and citations? I have not seen any comments there from any WMF personnel. Thanks for your input. (talk) 15:28, 13 February 2013 (UTC)

I don't know. You might want to ask Philippe to track down a comment from someone. :-) --Jimbo Wales (talk) 18:00, 13 February 2013 (UTC)
OK. Thanks. (talk) 19:01, 13 February 2013 (UTC)