Talk:Reliability of Wikipedia

This is not the page to discuss whether a source in an article is reliable.
	If you want to do that, go to WP:RSN or the talk page of the article in question.

Wiki policy encourages unreliable articles

Because wiki policy calls for secondary sources rather than primary sources, additional sources of errors are introduced into the editorial process. Primary sources are not just useful for research; there are technical areas where the only reliable information is in the primary sources, e.g., reference manuals for computers.

Yes, there are lots of 3rd party books on computers, languages and operating system, but every one that I have read has had at least one serious error. I understand the issue of bias, but for technical topics the use of primary sources is essential if the articles are to be accurate. Shmuel (Seymour J.) Metz Username:Chatul (talk) 15:14, 5 October 2010 (UTC)

Per WP:TALKNO, this talk page is not the place to discuss your personal opinions of Wikipedia policies. If you have reliable sources to back up your claims, then we could discuss putting the claims in this article. If you wish to propose a change to the Wikipedia policy you mention, then your posting should go in Wikipedia talk:No original research. -- JTSchreiber (talk) 04:00, 6 October 2010 (UTC)

Yes agreed wikipedia is very unreliable at times since anyone can edit the pages, most of the times pages tends to change due to too many different opinions, my source usually implys to just about every school I have been to, was one first things I was told at College/University, any other lecturer would say the same thing. This may be a worthwhile read: http://teachpaperless.blogspot.co.uk/2011/11/wikipedia-dilemma.html --Ronnie42 (talk) 01:23, 12 June 2012 (UTC)

Comparison of Wikipedia and other encyclopedias for accuracy, breadth, and depth in historical articles

Any reason this [1] isn't used? William M. Connolley (talk) 21:10, 16 October 2010 (UTC)

Omission? The article was critical of Wikipedia in 2007. It chose nine articles semi-randomly: Badlands, Sand Creek, The Chautauqua movement, the Free Soil party, the Niagara movement, the Harper's Ferry raid, the Mexican-American War, William Kidd and Harriet Tubman. These were compared with Encyclopaedia Britannica. The conclusion was that Britannica had a 96% accuracy rate in this study (96.5% excluding unverifiable facts) while Wikipedia had an 80% accuracy rate (88% excluding unverifiable facts). It went on to state that there were unattributed quotes and five cases of plagiarized content. Then complained about the quality of writing. Then summarized saying more research is needed (i.e. send more funding). (69.138.58.28 (talk) 05:07, 22 February 2011 (UTC))

Sounds like you're the biased one, 69.138.58.28 . We shouldn't remove things just because they criticize us. --69.246.249.158 (talk) 01:50, 3 March 2011 (UTC)

No, you're the biased one, 69.246.249.158, because now you're- wait, that only works once per argument. Darn! On a more serious note, I agree with you. Adding in critical sources that in a sense make Wikipedia less credible will actually make Wikipedia look more credible, as it is often interpreted as a very bold and noble move to criticize yourself in order to make your argument less biased. 98.250.105.170 (talk) 18:34, 12 November 2011 (UTC)

Kosovo

Hi, I find that the english wikipedia has a pro serbian slant to it on the issue of kosovo. it is not reliable for facts. Most of the facts about the names of the places in albanian have been removed. there is a constant unbalanced editing going on. Everything on kosovo needs checking, it is very annoying. James Michael DuPont (talk) 07:56, 7 December 2010 (UTC).

So Much Pun..

Not only is it ironic that the leading authority of wikipedia quality standards is wikipedia itself, but also that the top of the pages hints at the fact that the said article might be unreliable. — Preceding unsigned comment added by Conrthomas (talk • contribs) 03:11, 20 December 2010 (UTC)

It can not be stated that the entire wikipedia is either reliable or it's not. The reliability among the articles varies. As a consequence, one could consider this article as reliable while he's not sure of other articles. This article also shows which self-image wikipedia wants to convey. Deepfloe (talk) 15:27, 30 November 2011 (UTC)

POV issues

As Conrothmas pointed out, this article, especially its introduction, is cherry picking studies and fails to properly address criticism of Wikipedia's reliability, merely showing support of Wikipedia's reliability. — Preceding unsigned comment added by Ummonk (talk • contribs) 05:46, 29 January 2011 (UTC)

The article contains a number of examples in the Notable incidents section describing how incorrect, even deliberately dishonest material existed on Wikipedia for a period of time. It acknowledges that this is difficult to totally prevent. So it's not "merely showing support". What else would you like to see? HiLo48 (talk) 06:00, 29 January 2011 (UTC)

I agree that the lead section (introduction) is overly positive and needs to be made more neutral per NPOV. -- JTSchreiber (talk) 23:46, 30 January 2011 (UTC)

For it to be cherry picked, what other studies are there that show the converse? If there are other studies, then I want them in the article, but i'm not away of any. In the absence of any others, the article would be balanced, if these are the only ones.Rememberway (talk) 23:14, 28 February 2011 (UTC)

In the "Comparison of Wikipedia and other encyclopedias for accuracy, breadth, and depth in historical articles" section above, there's one study which is somewhat less flattering to Wikipedia. Any opinion on that study? -- JTSchreiber (talk) 05:20, 3 March 2011 (UTC)

I noticed my tag was removed. I will add studies such as the study above when I get the time. Until then please do not remove the tag without first fixing the article. Ummonk (talk) 22:35, 15 April 2011 (UTC)

As Rememberway said. I track this article and if I come across significant studies or views I take account of them in the article. Since 2007 mainstream academic studies, papers and coverage have almost entirely come to the same conclusion. Between 2005 and 2011 coverage has become gradually more and more similar, and less skeptical - the routine comments are that Wikipedia is surprisingly accurate, can be used for a first step, is reasonably reliable, but omissions and poor readability are still issues and errors and vandalism can occasionally happen. Todays latest paper is yet another saying exactly the same, but in politics rather than science.

If there is significant missing information please do list it - it will get a careful review. But a POV tag does not mean the mainstream views weren't as expected. Unless there is considerable coverage missed, the article is fundamentally, balanced. However as requested leaving the tag on in case it misses a significant number of contrary academic studies or significant views, or some aspect is unbalanced. FT2 ^{(Talk | email)} 10:39, 19 April 2011 (UTC)

I believe the studies done have made the cherry picking by selecting a few articles of say A-class or better, while me myself know of quite a few really trashy articles here and there. The perceived POV is not a neutrality issue per Wikipedia policies, but rather by the editor's gut feeling, I think. And of course I agree – probably that's one of the most important reasons why I'm still here. Rursus dixit. (^mbork³!) 05:48, 6 May 2011 (UTC)

The section Wikipedia editing model should be checked for structure and use of peacock wording f.ex. "Wikipedia is a pioneer in", it is pretty acceptable but needs some language trimming and structuring, the two last paragraphs being not quite neutral.

By the way: the Encyclopedia Britannica vitriolic attack on the Nature survey should mostly be compared to the sometimes (a few times) very hot debates going on on the talk pages. Rursus dixit. (^mbork³!) 07:17, 6 May 2011 (UTC)

American Renaissance (magazine)

Per this revert diff of The racialist magazine American Renaissance asserted that Wikipedia has a strong liberal bias in racial topics. with reference: Nicholas Stix: Wikipedia on Race – ‘World’s biggest encyclopedia’ serves up propaganda. American Renaissance. July 2007. Per: Wikipedia:RS#Self-published_and_questionable_sources. Is it really necessary to go to the WP:RSN noticeboard about it? I've seen far more credible organizational publications deleted as refs from articles. [Later apologies for not checking out publication thoroughly. This array of 2007 articles and this Jared Taylor article on Jews show they are blatantly racist and not very subtly antisemitic.] CarolMooreDC (talk) 18:35, 2 February 2011 (UTC)

It is not self-published. The article by Nicholas Stix was published by AR that also promote it on its frontpage. The source is also not questionable. What part of the sentence is questionable for you? AR is one of the leading representants of racialist and white nationalists point of view and information is not presented like fact but fact about AR´s opinion. --Dezidor (talk) 18:44, 2 February 2011 (UTC)

Geez, all the times I've been accused of antisemitism for use as a reference some relatively innocuous lefty anti-Zionist publication - or for once posting a link to a mainstream article now only available on some alleged Holocaust denial site which I used without paying sufficient attention - and then had nasty explanations thrown in my face like "Racist (or racialist)/antisemitic/bigoted sites aren't WP:RS!!" -- I'm just amazed that no one has challenged this. I'm not going to debate it with you since I'm asking the question myself now. If no neutral editor explains it in 24 hours, I will bring it to WP:RSN because I just feel like there's some double standard being applied, or someone isn't paying attention.

As for the gender bias stuff, there's a lot more older and new criticism on all that I'm researching so can rewrite with better description of who said what by tonight or tomorrow morning. First draft less than perfect, but a start. CarolMooreDC (talk) 16:44, 3 February 2011 (UTC)

It's a reliable source for the opinion of AR. It's attributed to AR. Of course its not a reliable source for any kind of fact (except maybe the page number), but as an attributed opinion its fine from a WP:RS standpoint. Another question is if the opinion of a small right-wing crackpot publication is important enough to be included, per WP:UNDUE. From an international point of view, when I read "American Renaissance" my gut reaction is "who? what?", so I have serious doubts. --Stephan Schulz (talk) 17:24, 3 February 2011 (UTC)

OK, let me quote all of Wikipedia:RS#Questionable_sources, esp since only the self-published was addressed:

Questionable sources are those with a poor reputation for checking the facts, or with no editorial oversight. Such sources include websites and publications expressing views that are widely acknowledged as extremist, or promotional in nature, or which rely heavily on rumors and personal opinions. Questionable sources are generally unsuitable for citing contentious claims about third parties, which includes claims against institutions, persons living or dead, as well as more ill-defined entities. The proper uses of a questionable source are very limited.

The last two sentences are operative ones. Let's face it articles on organizations or publications with far more mainstream WP:RS than American Renaissance magazine have been deleted. (Like my former DC peace group.) I can only speculate why the "antiracism" faction hasn't tried to. (Not that I in anyway support the threats of violence that closed down the conference by the way.) So I guess I'm getting an opinion together after all. CarolMooreDC (talk) 20:09, 3 February 2011 (UTC)

The only argument against American Renaissance is WP:IDONTLIKEIT. Wikipedia should present all relevant points of view not just what liberal New York Times write about so-called "gender bias" etc. Claim that some editors violetes neutral point of view and promote mainstream point of view without mention opposing points of view in another cases is not valid argument. Claim that some Zionists at Wikipedia accuse left-wingers of antisemitism wherever they can is also not argument to fight against pro-white source like they fight against lefty anti-Zionist publication you like. Wikipedia is not form paper and we should be tolerant to all relevant points of you not close minded people who delete everything they personally do not like. --Minlesa (talk) 11:50, 4 February 2011 (UTC)

Frankly, only reason I haven't gone to WP:RS yet is this might be about the only article where American Renaissance's opinion on 3rd parties would be allowed to be used because it is a criticism of Wikipedia itself. However, since I don't think I'll get sufficient NPOV editors here, I'll go to WP:RSN Monday and see if that might possibly be the reason it has survived so long. CarolMooreDC (talk) 02:23, 5 February 2011 (UTC)

Minlesa: No, WP:IDONTLIKEIT is not the only argument against it. Firstly it is a valid question whether it is a reliable source (however for the opinion of the organisation it is probably valid), secondly, and even more relevant, is the question of undue weight. To my mind the answer to the last question is clearly: yes, it is definitely undue weight to include the opinion of a fringe American magazine here. --Saddhiyama (talk) 08:53, 5 February 2011 (UTC)

FYI: Editors discussing that source should know that Wikipedia has an extensive history with Nicholas Stix, who likely made numerous self-promotional edits to the project, including posting a biography and adding lots of links to his blogs, in addition to engaging in widespread incivility.[2]   Will Beback  talk  10:38, 5 February 2011 (UTC)

An example of his criticism of Wikipedia while still an active editor:

• [Wikipedia] is a false flag operation designed to spread leftwing lies among readers who know nothing of your agenda, while at the same time suckering real encyclopedists into it and seeking to waste their time and exhaust their energy, and stalk, censor, and intimidate them. In other words, it's a mirror of the humanities and social sciences in today’s university.[3]

Make of that what you will.   Will Beback  talk  10:55, 5 February 2011 (UTC)

@Saddhiyama: One short sence is obviously not UNDUE. This is UNDUE but it contains opinion that some group of editors really likes so they revert it again and again. It looks that articles about liberal bias are truthful. --Dezidor (talk) 12:19, 5 February 2011 (UTC)

Any coverage of a truly fringe POV is undue. Even one short sentence in the biography on George W. Bush about the belief that he is a reptilian would be undue, for example. See WP:NPOV and WP:FRINGE.   Will Beback  talk  12:36, 5 February 2011 (UTC)

But AR is well known pro-white group that cooperate with notable people like: American_Renaissance_(magazine)#Notable_contributors_and_speakers (some of them are university professors or notable writers) not fringe group. That´s why it can´t be UNDUE. --Dezidor (talk) 12:56, 5 February 2011 (UTC)

Fringe POVs can be held by well-known and notable groups or individuals. David Icke, to use my earlier example.   Will Beback  talk  13:00, 5 February 2011 (UTC)

Why do you call AR "fringe"? Just because you DONTLIKE their position. --Dezidor (talk) 13:03, 5 February 2011 (UTC)

I call David Icke fringe because his ideas are not shared by reputable sources. I kind of like the idea of Queen Elizabeth being a reptile. ;)   Will Beback  talk  13:09, 5 February 2011 (UTC)

And AR ideas are shared by reputable sources such as professors (Rushton, Levin, Gottfried, Lynn...), authors and politians. Yes, they are in minority now, but are obviously not "fringe". --Dezidor (talk) 13:17, 5 February 2011 (UTC)

The issue here is not AR's views on race and intelligence. The question here is the Stix article about Wikipedia published by AR. Have you seen similar views espoused by reputable sources?   Will Beback  talk  13:24, 5 February 2011 (UTC)

The issue here is AR's views on race and and way how Wikipedia present this issue. They conclude that it has strong liberal bias. It is not just Stix because they promote that article at frontpage for long time. Tim Anderson, a senior lecturer in political economy at the University of Sydney, conclude that Wikipedia has an American and corporate bias and he has one paragraph in this article. AR statement is just one sentence in paragraph where many sources claim a liberal bias of Wikipedia, and the issue of race is just one piece next to the global warming, abortion, immigration etc.--Dezidor (talk) 13:51, 5 February 2011 (UTC)

• I think the issue is is AmRen mag a WP:RS for opinion about 3rd parties, which I doubt the community would support. This page says AmRen mag published Nicholas Stix article and his blog seems to confirm he might have permitted its use. However, given questions Will Beback brought up on AmRen talk page, I guess better evidence might be necessary.
• I think it possibly would be ok to have a section on "fringe group" criticisms since there are so many such groups, if enough of them commented on Wikipeda. Then we only would have to argue over whether The Final Call, Bitch (magazine), Front Page Magazine, Libertarian Party News, etc. were fringe enough or not for that section. However, just having one such source sticking out so noticeably there is vs. WP:RS not to mention WP:UNDUE.
• FYI, thanks for reminding I meant to put a WP:BLP tag on the list of Am Ren contributors since under WP:BLP you would have to provide a link proving those individuals had written for the publication, be it a link to the article or a link to a list of articles the magazine says it has published at the top of the section. (And I've done this a number of times over the years, often cleaning out nonsourced individuals in controversial claims.) I see only one has a ref. Otherwise there's no way of knowing if there is an error or if some vandal stuck something in there.
• Also note, if Am Ren ends up being quoted as a credible source in as many mainstream publications as SPLC, it might also become more reputable here. Of course, many Wikipedia ditors also question SPLC's financial and other motives, as well as its research, in some of its reports. CarolMooreDC (talk) 13:44, 5 February 2011 (UTC)

Fyi, just put this up here: Wikipedia:Reliable_sources/Noticeboard#American_Renaissance_.28magazine.29 CarolMooreDC (talk) 12:54, 7 February 2011 (UTC)

OK, Mr. Beback. I deleted the entry last time. Do you want to volunteer this time or do we need more consensus? (I did put a notice on a relevant Wikiproject but not sure how active it is.) CarolMooreDC (talk) 00:09, 10 February 2011 (UTC)

This strikes me as not so much a RS issue as a weight and notability issue. Clearly the publication is a reliable source for its own views. Whether that view is notable enough to merit inclusion in an article would seem to be to depend on whether any other unrelated, reliable secondary sources reported AR's views on the matter. The answer appears to be "no". In many, if not most, of the instances of criticism cited in this section, an independent reliable secondary source is reporting on the criticism. Those sources are making the determination that the particular criticism or the particular critic is notable enough to merit reportage. Unless this article is intended to simply be a list of every bit of published criticism of Wikipedia regardless of source (and I can see a perfectly valid argument for that approach) I would not think that the views of this publication are notable enough to merit inclusion in the article. I would suggest that, if the direction of the article is not to simply be a list, that the same criteria be applied to the other sources being cited as well. Fladrif (talk) 20:47, 10 February 2011 (UTC)

At this diff at the WP:RS Noticeboard Will Beback presented a bunch of WP:RS calling American Renaissance an extremist group. Wikipedia:RS#Questionable_sources says we don't use references that "include websites and publications expressing views that are widely acknowledged as extremist". So I'd say the matter is settled. CarolMooreDC (talk) 22:54, 10 February 2011 (UTC)

Copy and pasting

Was just wondering if Criticism of Wikipedia is being deleted or something? Just odd to see so much repeated text. We have things pasted over here like this huge copy and past the majority of this is about "criticism" not "reliability". Whats going on here?Moxy (talk) 04:58, 3 February 2011 (UTC)

Yes, Criticism of Wikipedia is being replaced with a redirect. The reasons for this are the POV of the article's title/content and the redundancy between Criticism of Wikipedia and other articles. The discussion for this is on the NPOV noticeboard. -- JTSchreiber (talk)

This merge process is creating some duplication within the article. For example, in Reliability_of_Wikipedia#Academia we have:

Academic circles have not been exclusively dismissive of Wikipedia as a reference. Wikipedia articles have been referenced in "enhanced perspectives" provided on-line in Science. The first of these perspectives to provide a hyperlink to Wikipedia was a 2002 paper called "A White Collar Protein Senses Blue Light", and many enhanced perspectives have provided such links since then. However, these links are offered as background sources for the reader, not as sources used by the writer, and the "enhanced perspectives" are not intended to serve as reference material themselves.
[...]

Some academic journals do refer to Wikipedia articles, but are not elevating it to the same level as traditional references. For instance, Wikipedia articles have been referenced in "enhanced perspectives" provided on-line in the journal Science. The first of these perspectives to provide a hyperlink to Wikipedia was "A White Collar Protein Senses Blue Light", and dozens of enhanced perspectives have provided such links since then. The publisher of Science states that these enhanced perspectives "include hypernotes – which link directly to websites of other relevant information available online – beyond the standard bibliographic references".

Once all the merging is done, someone needs to go through the whole article again and clean up things like this. Mitch Ames (talk) 10:33, 4 February 2011 (UTC)

Thanks. I've removed the redundant info you mentioned. (The non-bolded parts of the paragraphs were similar enough that I decided to delete a whole paragraph.) I have been planning to do clean-up after the merge is done, but I don't know if I would have caught this particular problem. It helps to have someone else looking at it. I do know that there still is some significant overlap between the "Coverage" and "Notability of article topics" sections. -- JTSchreiber (talk) 04:37, 5 February 2011 (UTC)

Well, better to have redundant information than less information! 98.250.105.170 (talk) 18:37, 12 November 2011 (UTC)

"Liberal bias" section and "American and corporate bias" section. Oh the irony

It's ironic that the "American...bias" section immediately follows the section where the word Liberal is used in a very American sense. Can we get rid of this American bias please? HiLo48 (talk) 07:02, 6 February 2011 (UTC)

What wording would you suggest using to replace "liberal bias"? -- JTSchreiber (talk) 05:26, 16 February 2011 (UTC)

"Too great factual accuracy". --87.78.29.96 (talk) 10:21, 22 February 2011 (UTC)

Sorry, that's not a neutral wording. -- JTSchreiber (talk) 05:44, 23 February 2011 (UTC)

(laughing) --UnicornTapestry (talk) 14:10, 4 March 2011 (UTC)

You could just call it a leftwing bias, thats what the rest of the world calls the left! — Preceding unsigned comment added by 89.126.25.46 (talk) 16:39, 17 May 2014 (UTC)

Misplaced para under "False biographical information"?

The last paragraph under Reliability_of_Wikipedia#False_biographical_information is:

Wikipedia considers vandalism "the insertion of false and misleading information in bad faith".[citation needed] The Wikipedia page "Researching with Wikipedia" states:

Wikipedia's radical openness means that any given article may be, at any given moment, in a bad state: for example, it could be in the middle of a large edit or it could have been recently vandalized. While blatant vandalism is usually easily spotted and rapidly corrected, Wikipedia is certainly more subject to subtle vandalism than a typical reference work.

This paragraph does seem to fit in this context. I suspect that it's a copy/paste error. It should probably be somewhere else, but it's not obvious to me where. Alternatively, reworded slightly to mention bio info. Mitch Ames (talk) 10:06, 6 February 2011 (UTC)

This paragraph is not part of the content I've copied over from Criticism of Wikipedia. I might take a look at this later, but the paragraph is related to the section, which is about a specific type of false information. -- JTSchreiber (talk) 04:10, 7 February 2011 (UTC)

Bias

This article is clearly biased in favour of Wikipedia (or at least the header paragraphs). —Preceding unsigned comment added by 99.241.118.40 (talk) 03:06, 14 April 2011 (UTC)

Please see the "POV issues" section above. -- JTSchreiber (talk) 05:23, 16 April 2011 (UTC)

"Conflicts involving Wikipedia policy makers"

The material was cut down to remove the citation and details, which seems inappropriate. Half of the section, as now composed, is about a charge that Wikipedia is a cult. That may be a legitimate accusation, but it is not related to conflicts. Also, we don't mention anywhere the Wikiscanner revelations concerning Dielbold, etc. I propose to retitle the section "Conflicts of interest", to restore the citation, to add the Wikiscanner material, and to move the cult allegation to the Community of Wikipedia article, probably in the "Motive" section. Thoughts?   Will Beback  talk  05:43, 14 April 2011 (UTC)

I agree that the cult accusation fits better in the Community of Wikipedia article. I'm not sure that I agree about merging the Wikiscanner info into the current "Conflicts involving Wikipedia policy makers" section. The article is divided into assessments and incidents. The Wikiscanner stuff in general seems broad enough that it does not belong in the incidents section. -- JTSchreiber (talk) 05:20, 16 April 2011 (UTC)

I tracked down the material removal to this edit by FT2, who stated that the removal was needed per WP:BLP. At the moment, I don't have an opinion about that, so I'll let the two of you discuss it. -- JTSchreiber (talk) 04:57, 20 April 2011 (UTC)

Article takes at face value antiquated conception of reliability

If we were to imagine this article as having been written by a lone author, it could be said with confidence that she had never heard the dictum "never trust an encyclopedia" (or its 21st century equivalent "never trust any solitary source"). The whole thing takes at face value the premise that Britannica, academia, broadsheet journalism etc. are more or less unimpeachably reliable, and that it is of great significance that Wikipedia fails to emulate them. For such an elaborately-developed article, I am very surprised to see no contestation of this notion, when that is the first thing prominent Wikipedia supporters address when faced with questions about its reliability in the press. See Jimmy Wales' comments here for a canonical example.

The point has been made above that the article reads like a Wikipedian's rebuttal of selectively chosen criticisms, and is biased in favour of the site as a consequence. That the article would begin this enterprise by accepting the anachronistic assumptions of Wikipedia critics makes for a very odd read indeed. Skomorokh 05:15, 23 April 2011 (UTC)

Jimmy Wales' opinion is in the article at the end of the "Other" section: "Jimmy Wales, the de facto leader of Wikipedia, stresses that encyclopedias of any type are not usually appropriate as primary sources, and should not be relied upon as being authoritative."

I also read through the lead section (introduction) and didn't see any wording that would suggest that Britannica, academia or broadsheet journalism are fully reliable. Sure, the article compares Wikipedia's reliability to that of other sources, but that doesn't necessarily mean that the other sources are fully reliable. In fact, the lead mentions reliability problems with other sources such as Britannica. Please be more specific about the wording which you think is a problem. -- JTSchreiber (talk) 04:31, 25 April 2011 (UTC)

I think a point that may need to be made in this article (provided a WP:RS reference can be found) is that many of the very difficult topics covered in Wikipedia are far more reliable than the less difficult subjects. (I should probably name this History's Wiki Paradox, but I am sure it is true). For instance, I have found many of the articles on advanced mathematical topics to be pretty well written, comprehensive and error free. For instance, articles on Cardinality of the continuum, Gödel's incompleteness theorems, Algebraic stack, etc. etc. are well written and "can be relied upon as being authoritative" in my expert opinion. In my experience these advanced articles are either hopeless or excellent without too many shades in between. That is because to even say a few sentences about the Cardinality of the continuum one needs to know the basic vocabulary - and that is not common. It would be very hard for a first year math student to edit these articles and introduce errors. They are generally written by experts who love their subjects (and have material already on their computers) and remain stable for long. But the good news often ends there. Many of the introductory topics are not so reliable because "everyone and his uncle" can edit them. Personally I would not even think of editing an introductory article such as algorithm (which gets over 100,000 views a month vs 2500 for Cardinality of the continuum). The talk page for algorithm shows long discussions which can kindly be described as very amateur at best. There is no point for me to even try to improve that article for I will hold no sway over a first year student. So Wikipedia is often "highly reliable" on advanced topics, but the model runs into problems for the introductory subjects. I wonder if the Nature study looked at advanced scientific topics. The results would then make sense. History2007 (talk) 21:35, 3 May 2011 (UTC)

I don't have access to Nature so I can't see what Wikipedia and Britannica articles were compared. You could try asking on the Wikipedia:WikiProject Resource Exchange or some similar place to see if you can get more info on the Nature study. However, I doubt that advanced subjects were used, simply because Britannica has a much lower number of articles and so tends to stick to basic/intermediate subjects. I do think that there is something to the broader idea that some types of articles are more reliable than others, but, as you say, a reliable source would be needed before that could be put in the article. -- JTSchreiber (talk) 05:02, 4 May 2011 (UTC)

I could not be bothered with Britannica right now. There is an old joke that "this is my brother Ed. He died three weeks ago, I have not had the heart to tell him yet." In my view, Britannica died three years ago, no one has had the heart to tell them yet. The default heir apparent is Wikipedia which shines in accuracy on very advanced topics such as Quantum mechanics and lowly entertainment topics such as celebrity rehab visit facts, but has spotty reliability on introductory scientific matters. Any statistician will tell you that the distribution of Wiki-reliability is unlikely to be uniform. So the best course is to make Wikipedia achieve uniformity and become the very best encyclopedia ever. The breadth of coverage is phenomenal, it is the consistency that needs improvement. I am working on a few things to clarify matters - check back in a month or two. History2007 (talk) 12:01, 4 May 2011 (UTC)

I'm curious about History2007's idea that Wikipedia's articles on introductory topics are less accurate than Wikipedia's articles on advanced topics. Here are the 42 articles used in the Nature study:

If anyone who hasn't read the results wants to make a call on which are the, let's say, 5–8 most introductory articles and the same number of most advanced articles, I can compare Wikipedia's error rates in the two article sets. Adrian J. Hunter^{(talk•contribs)} 14:47, 4 May 2011 (UTC)

Did they just use 42 articles? Boy, that was way too small a sample to be useful in my view. And I saw that some of them were biographies of scientists. So I am not sure if I would personally accept the results of the Nature study. Those Nature guys should have read the articles on stratified sampling more carefully in both encyclopedias! And partitioning their small sample further to do a second comparison here would even be less reliable. So I would not buy the results of a 10 sample analysis. Anyway, as I said above, please give me a month or two and I will say more, after doing some homework on this topic. History2007 (talk) 19:19, 4 May 2011 (UTC)

Fair enough; point well taken. Adrian J. Hunter^{(talk•contribs)} 14:33, 5 May 2011 (UTC)

Ok, fine. But now that the topic is still fresh, let me just mention that I would rather forget the Nature study, given that it was performed in 2005. In Wiki-years that is a century ago and in the meantime Wikipedia has grown and changed so much that the Nature study is an item in yesterday's newspaper. And that leads me to my "Second Wiki-Paradox", namely:

• "Statements made about Wikipedia in WP:RS sources may have become outdated by the time they are published, given the dynamic nature of the online encyclopedia."

So who cares what Britannica says about a 2005 study? That is yesterday's news. I DYKed this to let people know about what the future may be like, and the future is certainly automated, not based on 50 hand picked articles. Wiki-Watch is good as a first release software, but will not be the last reliability tool. I am sure in 5 years we will have a number of other good tools for assessing and improving Wiki-reliability. That will be the only way to go.

In any case, in a self-referential form, the authors of this article need to be aware of the fact that their search for WP:RS sources is likely to hand them outdated facts.History2007 (talk) 13:31, 6 May 2011 (UTC)

Regardless of what you think of WP:RS sources, it is still Wikipedia policy to rely on them rather than base articles on original research. Are you calling for a change in policy?

As far as the Nature study, I would strongly object to removing it from the Reliability of Wikipedia article. First, this article is not just about the current state of Wikipedia reliability but also covers the history of Wikipedia reliability. Second, the history provides insight into the quality of Wikipedia's process, and that insight is what gives people an idea of what to expect with Wikipedia's current reliability. For example, the 42 articles studied by Nature were fairly close to Britannica reliability in 2005, but the Seigenthaler incident showed very serious reliability problems from one article in Seigenthaler in 2005. Similar highs and lows of reliability can be expected today, though not necessarily with those same articles.

I would also strongly object to removing Britannica's response to the Nature study. First, your opinion that Britannica is obsolete does not seem to be represented in reliabile sources (such as [4][5]), which still seem to consider Britannica relevant. In its approach to trends, Wikipedia is reactive (following what reliable sources have said) rather than proactive (trying to predict what reliable sources will say and trying to get there ahead of them). Second, even if reliable sources were to consider Britannica irrelevant, that encyclopedia's opinion might still have some historical significance. Third, Britannica's response helps provide a neutral point of view for the Nature study. Such responses may not be available for more recent studies. This is also a reason to keep the Nature study in the article. -- JTSchreiber (talk) 05:27, 9 May 2011 (UTC)

Who said remove the Nature study? I said ignore it and move on - no point debating it any more. It is part of history, and needs to be remembered and mentioned, but no longer impacts anything. As for WP:RS sources, I think of them as totally essential for Wikipedia. However, the paradox I mentioned exists and has no immediate solution. I see zero need for (and zero chance for) changing the WP:RS policy in general. However, readers need to be aware of, and somehow cautioned about the paradox. So I think the article needs to pay less attention to yesterday's news (e.g. the Nature study) and have more focus on what there is to come - e.g. Wiki-Watch type reliability tools. Wikipedia is a knowledge juggernaut that can not be stopped, or managed by people. The best (and only way) will be through automation via reliability tools - the first generation being Wiki-Watch type systems.

But now that we are on the topic of WP:RS, let me tell you about the third paradox:

• In time Wikipedia will begin to influence the sources it uses, and will merge with all human knowledge.

I realized that when I saw a Wiki-article I had written appear verbatim (I really mean verbatim) in a book by a less than famous publisher. But in time, modified versions of Wikipedia content will show up in many books and it will be hard to know where material came from. So it is essential that a great deal of care goes into tools for assuring Wiki-reliability. The article needs more focus on what Wikpedia is doing about reliability, while mentioning historical items such Nature study as part of the past. So: more focus on what is being done about the future, less about the past. History2007 (talk) 07:43, 9 May 2011 (UTC)

While you meant to say that you wanted to ignore the Nature study and move on, what you wrote was vague and could be interpreted other ways. Please try to be more precise.

As far as alerting readers to your paradoxes, this can only be done if the paradoxes can be verified by reliable sources. What sources do you have? If you don't have any, the paradoxes are just original research which does not belong in the article.

As far as giving tools more attention in the article, we need to be careful about giving undue weight to tools. Undue weight is part of the NPOV policy, which is non-negotiable. Do tools really get enough coverage in reliable sources to justify a significant change in focus toward tools? -- JTSchreiber (talk) 05:31, 11 May 2011 (UTC)

Thank you ever so much for you continuing advice JT. It is so touching to be told "be more precise", do this, do that... repeatedly. Makes me feel like a toddler again. I am sure your advice will be useful when I start to edit Wikipedia. Now let me give you a hint: it is a "paradox" because it refers to inclusion based on WP:V but can not be subject to it, as it states. It is a simple variant of Russell's paradox: Let w be the predicate: to be a predicate that cannot be predicated of itself. Can w be predicated of itself? So there you go.

Now how about planning a good sized section on what Wikipedia is doing to assess its own reliability? That is where the tools will come in. I must say that I was surprised when the DYK of Wiki-Watch received only 4,000 hits. I had expected it to be a lot more. So there is significant need for increasing awareness of reliability measures within Wikipedia. History2007 (talk) 07:00, 11 May 2011 (UTC)

It doesn't matter how many paradoxes and hints you can provide or how much you think that these demonstrate weaknesses in Wikipedia policy. It also doesn't matter what you think the need is for a tools discussion. WP:WEIGHT determines what size the tools discussion can be in the article. Again, do tools really get enough coverage in reliable sources to justify a significant change in focus toward tools? -- JTSchreiber (talk) 05:21, 13 May 2011 (UTC)

No, I do not agree with your characterization of the situation. Does the existence of a picture of man within the Wikipedia logo get coverage in RS sources? No, because there is no picture of a man there. Hence, the fact that there are very few tools used to measure the reliability of Wikipedia gets little coverage because there are so few. So that issue is simple: there are very few tools (as I said below one just started a larger trial). So that glaring fact is not going to get coverage. But you did eventually make a correct observation: I am not questioning the reliability of Wikipedia, but the reliability of the policies towards reliability. The issue is not the content, but the attitude towards improving the content. History2007 (talk) 06:43, 13 May 2011 (UTC)

Since we are not heading towards consensus, I have posted the issue in this section on the NPOV noticeboard. -- JTSchreiber (talk) 05:25, 15 May 2011 (UTC)

As I said above, you need to wait a month or two for me to figure out the issues. In the meantime, the MediaWiki people seem to be making good progress on the feedback tool. So in time the issues will become more clear. History2007 (talk) 08:23, 15 May 2011 (UTC)

@History2007: You seem to be such a convinced fan of Wiki-Watch. I only want to inform you that this self decribed project for transparency and reliability of Wikipedia, who's founder also runs a PR company, was found to be connected to a small army of sock-puppets that put massive POV in articles related to pharmaceutical and evangelical topics. Meanwhile, this story even made it into major national news-outlets in Germany ( Hier prüft der Bürger das Insulin noch persönlich: Die dubiosen Verstrickungen von Wiki-Watch, faz.net vom 1. Juli 2011) (sorry, only german). --MTYM (talk) 01:17, 2 July 2011 (UTC)

Interesting. I did not know their background, except that they were professors. But that information should be added the to the page about them - one of them has a Wikipage. Please feel free to do that - it will help. However, the concept of an automated tool will outlive Wikiwatch and the activities of these authors. I still like it and would like to see 12 other systems of that type. Checking things by hand is a pre-industrial revolution approach. Remember that the word computer used to apply to human beings who would sit in a room with 30 other people who added numbers by hand. That type of approach has to give in to automation. History2007 (talk) 04:08, 2 July 2011 (UTC)

If it is run by a transparent organization I agree that an automated tool can be of great help. Obviously, that is not the case with Wiki-Watch. --MTYM (talk) 02:38, 5 July 2011 (UTC)

Article feedback tool

I just found out that there is an Article feedback tool only because I was prompted by it as I was reading a page. It is getting wider use as of May 9, 2011. This article does not mention that tool, or other tools Wikipedia may be working on to rate/assess or improve reliability. Are there other tools like that in the works? A section on that specific tool and those types of project would be a good idea. History2007 (talk) 21:14, 11 May 2011 (UTC)

English language only?

In this edit [6], History2007 restrictied the scope of the article to the English language content of Wikipedia. History2007 thought that there was "nothing about French, Italian, Chinese Wki etc. in article". This is not quite true. While the vast majority of the article is about the English language content, there are facts about content in other languages. I found the following examples:

• In the "Comparative studies" section, two studies of German language content are described.
• In the "Tools for testing the reliability of articles" section, one sentence states that Wiki-Watch can be used to analyze German language content.
• In the "False biographical information" section, one incident is described for Wikipedia’s French language content.

The article is now inconsistent. The options I see are:

• Remove the qualification in the first sentence limiting the article to English language content.
• Remove the content dealing with languages other than English from this article. In this case, it would be best to rename the article to make it clear that the article is limited to English language content.

I prefer the first option, but will wait for others' opinions before making a change. -- JTSchreiber (talk) 05:04, 23 May 2011 (UTC)

Yes, but you had to fish to find those examples. And still there is nothing about Chinese, Italian, Portuguese or other Wikipedias. The mentions of the other languages are just token references. And that actually brings about an interesting point. What does Wikipedia mean in general? I the early days the English version was the key item, and it still remains so. But now there is a wealth of content elsewhere - specially on local topics such as cities, buildings, artists etc. I was going to discuss "cross-language" consistency later, so for now let us just refer to what the article does.

The best way will probably be to say at the top that this article "primarily" refers to English-Wiki - because that is what it does. So there is need for clarifying what Wikipedia means. As far as I can see (and as far as I know) the Japanese Wikipedia has not been tested. Now, does it have accurate info? Who knows? Does it have valuable info? In some cases, probably.

And we should remember that the other Wikis are no longer isolated islands, because people do look up info in them and add to the English language Wiki. So the reliability of those will affect the English Wikipedia through reliability osmosis. And that does not mean a reduction in reliability, but can also enhance it. In the area of art history, life of artists etc. the other Wikis have pretty good information - but I have not compared what they say about physics.

The long and short of it is that the other Wikis will not be forever ignored, because they have useful info on many topics, at times more info than the English version. The eventual very interesting topic will be to wait until someone does a "cross-wiki" study, information osmosis analysis, etc., then we will see. The eventual goal will be to have harmony between the top 10 Wikis - they can certainly benefit from each other. History2007 (talk) 06:56, 23 May 2011 (UTC)

Academic references ?!?

Does anyone know how those numbers were obtained? I tried to replicate them so I could update the table but I just can't get the values for the years 2003-2009 to match up. If this isn't resolved, the table should be deleted or remade with defined search paramaters and reproducable results. The current situation is simply unacceptable.--U5K0 (talk) 15:29, 7 June 2011 (UTC)

Definition of reliability

In the section above called "Article takes at face value antiquated conception of reliability", user:Skomorokh made a number of comments which in the end received no final answer on this talk page. I have been thinking about that, and although at first reading that may seem like a radical statement, once one thinks about it there is substance there that needs attention.

So let me ask a couple of obvious questions here, to go back to that issue:

• What is the definition of "reliability" used in this article?

Clearly the link to Reliability is not adequate at all. So what does reliability mean here? The article mentions "Comparative studies" and "Expert opinion" as two approaches. Are they the only two methods of assessing reliability? If so that should be stated. But then what about the rest of the article, where Mr X who is alive disputes the assertion of his own death? That is neither expert opinion, nor a comparison to Britannica. And what about the term "bias" discussed in the article. Is tone a component of reliability? What about "internal consistency"? What if the Wikipage World's busiest airports by passenger traffic has different number of passengers for some airports than the pages about the airports (check it, you may be amused)? In any case, as a reader I would like to see a better definition of reliability.

• What is Wikipedia? Does it include Wikisource? Does it include WikiCommons?

I ask that because I find Wikisource such a valuable resource. If Wikisource has the actual text of what Isaac Newton wrote, and I can just look it up, and compare to the article, then I personally do not care what Britannica says. I will believe Wikisource before Britannica. I think what Skomorokh implied was: Britannica is losing its relevance really fast, and I think Wikisource validates that sentiment in man cases.

Moreover, a comparison of Wikipedia with other encyclopedias is like comparing the movies made in the United states to those made in Luxembourg. There is so much more in Wikipedia that it is impossible to get a broad assessment by comparing its content to encyclopedias which have no equivalent articles. So that definition of reliability needs further attention.

And regarding Skomorokh's comment about the reliability of venerable publications such as the New York Times and The Washington Post , I think we should remember the awards given to Janet Cooke, Jayson Blair, etc. and United States journalism scandals. Given the number of articles in Wikipedia compared to the general newspapers, are the number of Wiki-scandals not low? So that comment may also have substance and the definition of reliability needs to address the ratio of articles to the false assertions.

Anyway, I think this article needs help for the definition of reliability it uses is unclear, and as Skomorokh said probably outdated. Comments will be appreciated. History2007 (talk) 20:56, 1 July 2011 (UTC)

There's too many issues here to discuss all at once. I'll just start with a couple simple ones.

First, the opening sentence of the article is defined too narrowly and it should be made inclusive of all ways of determining reliability that are mentioned in the article. For example, the opening sentence might come across as saying that all ways of determining Wikipedia's reliability involve comparison against the reliability of other sources. Not all ways of determining reliability in the article involve such comparisons.

Second, you wrote, "And what about the term "bias" discussed in the article. Is tone a component of reliability?" Is this one question or two? In other words, do you consider bias to be just about tone, or do you consider these two separate issues? -- JTSchreiber (talk) 05:29, 4 July 2011 (UTC)

First, yes, the definition is way too narrow; and in fact I do not see it as a clearly stated definition anyway. Hence this section.

Second, the discussion of "bias" is likely to be so complicated that I would prefer to postpone it for now. I see tone as a component of bias, but I have no solid science behind that at hand - although there is probably material on that somewhere. And I am no expert on bias - what I say on that issue is just what a typical reader may think.

But just the issue of the reliability of the factual accuracy of "hard statements" (such as the number of passengers for an airport or whether a person exists at all) needs to be clarified before the "soft issues" such as the quality of the work of an artist are handled. Bias will vary between Wikis, just as it does in newspapers.

I think a reader who reads this article will probably think of reliability in intuitive terms by asking "can I trust what I read to be true" which then implies a notion of truth. For numerical "hard facts" such as the year of death of Isaac Newton the issue can probably be handled. And suitable definitions for those need to be made here first. And again that brings in Wikisource because it is in itself a "verification method" of some sort for many readers such as myself.

But I think we must accept, and the article at least needs to mention, that there are artistic and (and even business related) statements that are inherently subject to debate regardless of what Wikipedia says. E.g. the statement "The yuan is subject to active government manipulation" will be accepted in a New York minute if appears in the Wall Street Journal. But I wonder how that statement will work in Chinese Wikipedia, given that stating that in a Chinese newspaper will lead to house arrest - at the very least.

So there are many issues involved here, as you said, but we should probably start with the obvious issues that the article has not mentioned, e.g. the analogy between Wikipedia and Britannica articles as the movies made in the US vs Luxembourg and the fact that venerable publications such as the New York Times have "proportionally" more Jayson Blair type incidents than Wikipedia. History2007 (talk) 07:53, 4 July 2011 (UTC)

The least complicated way to improve the opening sentence would be to make it better summarize the current content of the article. Bringing up new content, such as Wikisource and the Luxembourg film analogy, will only add further complexity to the discussion. New content can always be discussed later.

I'm fine with postponing some of the discussion on bias, but since you already wrote some things about the subject, I will respond to what you wrote. First, why are you mentioning the quality of work of an artist? I don't see anything like that in the "Susceptibility to bias" section of the article. Are you proposing that artistic bias be added to the section?

Also, I'm not too impressed with your example of Chinese Wikipedia. Some Chinese language newspapers are not under the control of the PRC, including newspapers in Taiwan and the Apple Daily in Hong Kong. Also, since the PRC government has blocked access to Chinese Wikipedia at times, I don't think that Chinese Wikipedia has taken a pro-PRC stance. -- JTSchreiber (talk) 05:29, 7 July 2011 (UTC)

It is better to ignore tone and bias for now - I am not sure what they buy before factual issues have been better understood. Art/politics/etc. are just examples in a general case and the Chinese papers still make it appear there are conflicting views. But anyway, factual issues need to be addressed first. However, the fact that Wikisource is not there now does not mean it should not be mentioned. I think it is important to add that to the article, because people use it. History2007 (talk) 07:21, 7 July 2011 (UTC)

Politics is mentioned pretty often in the "Susceptibility to bias" section, so it makes a good example. There is no need to mention art again unless you are intending to add comments about artistic bias to the article. We can save the rest of the discussion about bias for later.

The fact that Wikisource is not in the article is a reason to leave it out of the lead section. Per WP:LEAD, it's the Wikipedia standard to leave it out of the lead if it's not in the article body. Again, the least complicated way to improve the opening sentence would be to make it better summarize the current content of the article. That doesn't mean that this would be a permanent version which couldn't be expanded later. In fact, it may well have to be expanded when bias is discussed later. -- JTSchreiber (talk) 05:31, 9 July 2011 (UTC)

Actually the lede should be driven by the content, not wag the content. So I think a decision needs to be made about the term Wikipedia (with Wikisource or with commons) then the lede reflects that. E.g. as an extreme case suppose commons has a picture of Newton, calling him Einstein and an article uses that - is that reliable? Or a picture of London, calling it Paris, etc. or a diagram for a molecule with the wrong attribution. So that decision comes firs, content gets affected then lede summarizes it. History2007 (talk) 13:18, 9 July 2011 (UTC)

It would have been simpler to leave Wikisource out of this for the time being, but since you insist on discussing it, I will set up a subsection for that discussion. -- JTSchreiber (talk) 05:25, 12 July 2011 (UTC)

Wikisource

On 1 July 2011, you (History2007) wrote

What is Wikipedia? Does it include Wikisource? ... I ask that because I find Wikisource such a valuable resource. If Wikisource has the actual text of what Isaac Newton wrote, and I can just look it up, and compare to the article, then I personally do not care what Britannica says.

To begin with, I'll just discuss the issue of whether Wikipedia includes Wikisource. Looking at how Wikisource defines its relationship with Wikipedia([7][][8]), I see nothing that suggests Wikisource should be considered part of Wikipedia. What reliable sources do you have to say that Wikipedia might include Wikisource? I did some Google searching and didn't find anything, but perhaps you have something. -- JTSchreiber (talk) 05:38, 12 July 2011 (UTC)

Given that Wikisource is "operated by the Wikimedia Foundation" it is part of the same knowledge family anyway and Wikipedia articles often use it as links. I do not think there are WP:RS sources either way that say it is or is not part of Wikipedia. And again, that was the part of Skomorokh's point I think. So a definition must be made that the term as used here has certain boundaries, and as a reader if those boundaries do not include the sister projects of Wikisource and Commons, I would find the article "unreal" because as a Wikiuser I rely on them so heavily. So a discussion that excludes them would seem unreal to me as a reader. History2007 (talk) 09:29, 12 July 2011 (UTC)

What do you think was part of Skomorokh's point? Also, how did you determine what Skomorokh might think about Wikisource when s/he never mentioned Wikisource on this talk page? -- JTSchreiber (talk) 05:38, 14 July 2011 (UTC)

I meant that the definition and approach in this article is outdated, as is this ping-pong between the two of us. Given that there are no other comments, and that in general the two of us will probably not even agree on the time of day, I will suspend Wikisource discussions. Next topic, WikiCommons. So let us waste time playing ping-pong on that and see if that should be part of the definition. History2007 (talk) 07:36, 14 July 2011 (UTC)

While Skomorokh complained about the definition of the article, any assumption of what s/he thinks about Wikisource is reading too much into what Skomorokh wrote. We can suspend the rest of the Wikicommons discussion and move on to WikiCommons, as you like. However, I think that Wikisource and Wikicommons should each have their own subsections. JTSchreiber (talk) 05:20, 16 July 2011 (UTC)

WikiCommons

On 1 July 2011, you (History2007) wrote

What is Wikipedia? ... Does it include WikiCommons?

On 9 July 2011, you wrote

E.g. as an extreme case suppose commons has a picture of Newton, calling him Einstein and an article uses that - is that reliable? Or a picture of London, calling it Paris, etc. or a diagram for a molecule with the wrong attribution.

As with WikiSource, I'll start by discussing the issue of whether Wikipedia includes WikiCommons. Looking at the WikiCommons main page ([9]), it doesn't define WikiCommons as being part of Wikipedia. Do you have reliable sources to say that Wikipedia might include WikiCommons? Without such a source, the article cannot state that WikiCommons is part of Wikipedia. -- JTSchreiber (talk) 05:38, 16 July 2011 (UTC)

I thought you would ask that. Now, is there a source that says images used in Wikipedia are not part of the reviews they have performed? So the first obvious question is:

• Are the images used within Wikipedia part of the definition of Wikipedia in this article, or is this article purely about text?

That needs to be answered first. If it is only about text it should mention that right at the top: In this article Wikipedia only means textual information and nothing said here refers to images. Should that get added? History2007 (talk) 06:26, 16 July 2011 (UTC)

I don't know that the referenced sources are that specific. I looked at the 2003 IBM study and the 2005 Guardian study. Neither mentions images at all. However, I think it may be reading too much into them to say that they looked only at textual information. If the sources are vague, I think that the Wikipedia article should also be vague on this issue. To do otherwise may well be original research. -- JTSchreiber (talk) 05:19, 19 July 2011 (UTC)

At the very least, it must be said that there is no clear definition in the studies presented regarding images and that comparison on images may or may not have taken place. That will show the narrowness of the studies. Some of the other discussion in those articles is about topics like "Mr A is dead" when he is not. But that requires no image, unless it is a photo of the funeral.

However, WP:OR will only apply if we make inferences and assert facts based on deductions. It does not apply to summarizing statements about what there is or is not; nor does it apply to "definitions".

Again this brings up the "detachment from reality" problem about the way this article is written. As a reader of this article, and as a user of Wikipedia, this article does not help me and seems to be trying to discuss the geography of the world while confining itself to Vermont. As a reader, it misses so many, many issues that it is of very little value to me.

I was actually trying to see how to remove the tag from the top of this article, but the more we continue this melancholy dialogue the more I feel like another tag is even needed. In my assessment, the article is so myopic that it tells me very little and short changes the wealth of information that I use to get reliable answers from Wikipedia.

As a knowledge source, Wikipedia/Wikisource/Wikimedia are actually far more powerful than this article presents them to be, and the treatment here is hopelessly narrow and inadequate. And the real issue of reliability is ignored by the text here. The article short-changes the breadth of knowledge available in Wikipedia/Wikisource/Wikimedia, and the comparative reliability with other sources. And this melancholy dialogue is not going anywhere that could remedy the presentation here. History2007 (talk) 06:05, 19 July 2011 (UTC)

The sources used to describe the studies in this article are mainly primary sources. WP:OR says, "Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so." I think that the wording you want involves some analysis and evaluation, especially when you say the purpose of the wording is to show the narrowness of the studies.

WP:WEIGHT also applies here. If 0% of the content in the reliable sources mention images, then 0% of the content of this Wikipedia article should mention images. To even mention images would give undo weight to them.

Without reliable sources to back up your personal opinions, those opinions (about whether this article is myopic and the usefulness of Wikisource/Wikimedia for evaluating Wikipedia reliability) are irrelevant arguments for making changes to this article. It's not uncommon for Wikipedia editors to experience disappointment and frustration because there are no reliable source to justify content additions which are important to those editors. -- JTSchreiber (talk) 05:24, 22 July 2011 (UTC)

I do not agree, but then that is not surprising is it? History2007 (talk) 07:25, 22 July 2011 (UTC)

It's not surprising that we still disagree. -- JTSchreiber (talk) 05:16, 24 July 2011 (UTC)

What is surprising is that this article can sit there for so long with a big NPOV tag on it and no resolution. But I guess that is how Wikipedia works, or should I say does not work? History2007 (talk) 09:08, 24 July 2011 (UTC)

two court cases about wikipedia not being accepted in US court

I added the following:

In the United States, the [[United States Court of Federal Claims]] has ruled that "Wikipedia may not be a reliable source of information."<ref>Nordwall v. Secretary of Health & Human Services, No. 05-123V, 2008 WL 857661, at *7 n.6 (Fed. Cl. Feb. 19, 2008) as cited in [http://docs.justia.com/cases/federal/district-courts/california/candce/5:2008cv00904/204116/62/ ''Capcom Co., Ltd, et al. v. The MKR Group, Inc.'', No. C 08-0904 RS]</ref> and "...Articles [from Wikipedia] do not - at least on their face - remotely meet this reliability requirement...A review of the Wikipedia website reveals a pervasive and, for our purposes, disturbing series of disclaimers..."<ref> ''Campbell v. Sec'y of Heath & Human Servs.'', 69 Fed. Cl. 775, 781 (Ct. Cl. 2006)</ref>

As I was adding the case site in the capcom case, I came across: Wikipedia:Wikipedia_in_judicial_opinions which is a second United States Court of Federal Claims case. The section I added this to seems overly postive about wikipedia, so I hope my contributions remain as a well sourced addition. Igottheconch (talk) 05:05, 19 August 2011 (UTC)

I think that was a good addition. And the fact that there are varying opinions is not surprising. Intuitively speaking, the issue is that in some cases Wikipedia has very high quality material, in other cases very low quality items, so although the average quality may be good, there is no guarantee of a minimum threshold. But then those who do not like it can immediately ask for a refund of the purchase price. However, in time, with suitable (and as yet lacking) quality initiatives things can improve. History2007 (talk) 05:21, 19 August 2011 (UTC)

Igottheconch, thanks for the addition! You are correct that the section was overly positive and your content does help. This is the type of addition (properly sourced content about negative views of Wikipedia reliability) that is needed to help address certain complaints about the article's neutrality. -- JTSchreiber (talk) 04:34, 21 August 2011 (UTC)

No, don't tell me we are agreeing on something JT! Who would have thought.... History2007 (talk) 04:53, 21 August 2011 (UTC)

Wow, thanks for the comments guys! Big welcome surprise. Just shows that sometimes piggy isn't killed by his classmates :) (my username is a reference to Lord of the Flies) Igottheconch (talk) 04:41, 1 September 2011 (UTC)

You're welcome. -- JTSchreiber (talk) 05:35, 1 September 2011 (UTC)

Public Relations

Try as I might, I have no idea what this sentence means: "However, omissions sometimes remained an issue, at times due to public relations removal of adverse product information and a considerable concern for fields such as medicine," specifically, public relations removal of adverse product information. Could anyone make that a little clearer?

You might also consider revising the entire tone of the article, which reads like a marketing piece. — Preceding unsigned comment added by 72.37.171.204 (talk) 20:58, 22 November 2011 (UTC)

I removed the last phrase in the sentence you quoted, as I think that phrase is problematic in other ways besides readability. That's a step toward making the sentence more understandable. As far as your concern about the article having a public relations tone, it would help if you provided examples of this from the article. -- JTSchreiber (talk) 06:24, 1 December 2011 (UTC)

I do not see the article as having a serious public relations tone, but it has some elements that are just out of The Twilight Zone. But if the IP wrote that the article has a PR tone, it was probably because he "felt it". And I went on and read some parts of the article again and I must say I just chuckled when I read quotes like:

• Dan Gillmor, a Silicon Valley commentator and author commented in October 2004 that, "I don't think anyone is saying Wikipedia is an absolute replacement for a traditional encyclopedia. But in the topics I know something about, I've found Wikipedia to be as accurate as any other source I've found."

I said to myself, it looks like Gillmor has been just reading the quotes on Starbucks coffee cups. I know, I know, I know JT, that statement "is well sourced". But sourced is one thing, a chuckle is another. Then I saw:

• Sheizaf Rafaeli and Yaron Ariel report how "most people agree that at least the English version of Wikipedia is approaching critical mass where substantial content disasters should become rare"

Then I remembered how aghast I was today as I started to clean up the articles on computer cluster. Look a the talk page Talk:Computer_cluster: no discussion since July 2010, except my note today. I had to Afd two of the article as a start because they were so incorrect and non-notable. I will probably have to delete 80% of the text as I clean the series up within the next 3 months. If that is not a "substantial content disaster" then I do not know what is. And the entire series on digital signal processing is just as bad, if not worse. So when people see these things, then read "sourced quotes" by Gillmor, they just feel there is something funny going on. But do I have a WP:RS source hat I feel there is something funny about Gillmor's quote? No, I do not. But the IP did feel that there was something that did not fit, as I did when I read the Gillmor quote and then the other one. It just does not add up to people's own experiences - so they just chuckle and feel it is PR. And I do not blame them. History2007 (talk) 01:33, 21 December 2011 (UTC)

Thanks for providing some specifics. I don't have a big problem with the Gillmor sentence, although it is quote-heavy. Perhaps it could be paraphrased and/or condensed. The sentence about Rafaeli/Ariel is another matter. It has problems with WP:SAY, WP:WEASEL and WP:CRYSTAL. The first issue is easy enough to fix, but the other two are more serious. We don't report what most people think is going to happen. For example, we don't report what movie most people think is most likely to win the best picture Oscar in 2012, and we don't report who most people think is going to become the democratically elected leader of Egypt. The fact that the quotation about Wikipedia is attributed to a couple scholars doesn't fix the problem. I think that this is probably enough justification to remove the sentence from the article. -- JTSchreiber (talk) 05:53, 21 December 2011 (UTC)

Ok, touching those up would be a start. And by the way, I noticed that the WikiWatch item is out of date with respect to the German news fiasco, charges of deliberate manipulation, etc. if you want to touch that one up too. I no longer even watch the Wikiwatch page I wrote - it is such a mess.... But it does point out that there may be paid manipulators out there. Yet another issue. History2007 (talk) 07:56, 21 December 2011 (UTC)

The broad assessment section

Now that the Rafaeli quote is gone based on the discussion above, I looked at that section again. So I asked myself, "how does one edit that section so people do not laugh when they read it?" The problem I see is that it is basically a "quote-based section" which uses quotes within the context of "assessment". However, absolutely no basis is presented for how the said assessments were performed. And no basis is provided for the ability of the people making the assessments to be able to provide reliable assessments. So as a reader, I am surprised such a low reliability section exists within a reliability article. Of course if something has a WP:RS source one can just use that as a justification for its existence in a Wiki-article, but there is no point in making a "RS soup" of RS quotes that are free of logic.

The other problem is that part of what those quotes say is valid. There are specific, and very large, patches in Wikipedia which are very high quality and this type of unfounded "broad assessment" overlooks the high quality present in many parts of Wikipedia and is just risible when applied to many other parts. Most users who think realize this. That section needs other quotes, better logic or is best left out. I do not know how to fix it now without further research and do not have the time to do the research now. But whoever has the time should work on it.

The specific problems with the quotes are:

• What Gillmor said 7 years ago is ancient history in web-time. No justification is provided whether the reliability of Wikipedia-2004 is in any way comparable to that of Wikipedia-2011. It may be much higher or much lower. Your guess is as good as mine. So that statement is like knowing the blood pressure of a patient from 7 years ago, then trying to diagnose the patient. Seven years is a long time. And the Larry Sanger statement from 2001 is just too far out to require comment. That was 10 years ago and unless Mr Sanger is a clairvoyant as well as a philosopher he has no basis for speculating on events in 2011 ten years before the fact. Again that statement just makes the section look bad and throws the baby out with the bath water.

• The Joi Ito statement is not as bad, but will again lead to objections from users such as myself. First, exactly what qualifies Ito as an authority? None that I can see. And has Mr Ito read any of the articles on biochemistry, law, physics, computing, etc, in detail to be able to comment on them? No, he made a shotgun statement. As is articles such as distributed operating system and intangible property flatly invalidate what he said because one is so high quality, the other so low quality. The high quality article was written by a single user who disappeared after writing it as his single contribution, and the low quality article has many footprints and no major author. Wikipedia readers are aware of these issues and just do not believe Mr Ito's viewpoint.

So the broad assessment section just seems illogical, despite having WP:RS sources. The Ito statement, is however, valid in some cases. As a reader, I would trust Wikipedia articles on London, Vienna, Chicago or Boston far more than I would any of the Fodor's guides. Those pages are certainly subject to the Ito classification because so many people have checked them and would be more reliable than a travel guide written by 2 people. But the other examples above show the failings of that statement.

Of course, logically speaking, the fundamental problem is that via existential quantification it is much easier to find problems in Wikipedia and any statement about "overall assessment" requires universal quantification which is harder to achieve.

Anyway, that section needs help. But I will stop for now. History2007 (talk) 14:26, 22 December 2011 (UTC)

Many of the statements in the broad assessment section are opinions rather than objective summaries of statistical studies. This doesn't necessarily mean that the opinions should be taken out, but it at least raises the question of whether the opinions should be moved to the expert opinion section. If that's done, then there may or may not still be a need for a separate section called broad assessments.

As far as Gillmor's and Sanger's statements being old, we've discussed the issue of statement age before. We're not getting any closer to agreement on this, so I don't see a point to keep discussing the issue.

I think that the Larry Sanger and Joi Ito statements are there to represent points-of-view of prominent people who are related to Wikipedia. (Ito is the chairman of Creative Commons, which is tangentially related to Wikipedia.) Sanger's statement in this article is problematic, because he has drastically changed his opinion of Wikipedia in the last ten years. This needs to be mentioned with an RS, so I'll look into doing that. I'm not talking about replacing his old opinion, but mentioning both of his opinions. -- JTSchreiber (talk) 06:38, 30 December 2011 (UTC)

Well, in Wiki-language that section needs an "out of date" tag. But it would also be good to make those an "opinions" section rather than assessment section. They have included no formal study whatsoever, and are just stated as opinions pulled out of thin air with absolutely no basis. In other words those are not "reliable assessments".

What will also seem strange to a reader is that there are no recent opinions or broad assessments. Again, on that front the article does not help Wikipedia establish itself, but seems like a half-hearted attempt. The article would be better served to not call them assessments but opinions, as a start. Then find some new opinions as well. History2007 (talk) 07:33, 30 December 2011 (UTC)

I agree that "opinions" is more a specific description for the section than "assessments", so I renamed the section. That leaves us with two opinion subsections underneath assessments, which seems odd. We should still look into merging some or all of the "broad opinions" content into the "expert opinions" section.

I'm not opposed to adding newer opinions, but I disagree with the need for the "out of date" tag. I think that would only be used if we knew that newer RS opinions were substantially different than what is in the section. -- JTSchreiber (talk) 06:19, 2 January 2012 (UTC)

It looks better now. I would agree with a merge into "expert opinions". But the whole purpose of this discussion is to bring it closer to "what people feel" so they will not chuckle as they read it. Skeptics will wonder why there are no opinions there since 2005, given that the number of articles has gone up. And supporters will feel that it does not do justice to the more recent high quality articles in Wikipedia. So more quotes must be sought. Again, some patches such as Art History have very high quality articles (take anything on van Gogh) while many scientific articles (e.g. computing) are low quality. At some point if topic specific assessments are found, that would help. History2007 (talk) 06:42, 2 January 2012 (UTC)

There actually are a couple opinions from 2008 in the section already. I will edit the text of the section to make that more clear. If you feel so strongly that additional quotes from after 2005 are needed, why don't you look for some? -- JTSchreiber (talk) 06:28, 4 January 2012 (UTC)

Number of articles in the English Wikipedia

Actually, contrary to popular belief, I do not stay up at nights worried about this article. However, I do think that an encyclopedia lives and dies on its credibility. So I would like to see Wikipedia maintain its credibility, and increase its reliability. To that end, I still think that the Gillmor quote will make thinking readers laugh, given that when he made his comment there were about 400K articles, and now there are about 4 million. So his comment misses 90% of Wikipedia content - at least. It is like quoting someone bout the financial health of a company 7 years ago. The company may have expanded and succeeded, or may have gone under. Statements from that long ago mean nothing. Yet they are used in the article and make people like me shake their head in disbelief that such logic is used in a key article like this. And the same applies to the 2005 Nature study. At that time Wikipedia had less than 1 million articles. So any study in 2005will have missed 3 million articles overall, and I am not sure how many science articles. And again that type of statement is like looking at the blood pressure and EKG chart for a patient 6 years ago and assessing his health today. No sane MD will ever do that. History2007 (talk) 01:47, 11 January 2012 (UTC)

We already had a discussion like this back in May. Basically, I said that Wikipedia's process hasn't changed much over the years, so similar highs and lows of reliability can be expected now as has happened in the past. Individual articles have changed, some for the better and some for the worse. Many new articles have come and a few old ones have gone, but I don't see evidence that the overall reliability has changed, so I think the old opinions still have relevance.

As far as the Nature study, I still strongly object to removing it from the Reliability of Wikipedia article. It was a landmark study that has received a lot of attention from reliable source, including some attention years later [10].

I am still looking into selecting a more recent Sanger opinion. If you wanted to find some more recent non-Sanger opinions, I would consider replacing some older opinions with newer ones, but I don't see a reason to delete the older ones without replacements. -- JTSchreiber (talk) 06:40, 13 January 2012 (UTC)

But the statement "similar highs and lows of reliability can be expected now as has happened in the past" is pure, absolute guesswork. There is zero, and I mean "absolutely zero" scientific basis for that statement. Do you have a basis for that statement apart from "a personal guess"? History2007 (talk) 09:36, 13 January 2012 (UTC)

My statement is based on the work of W. Edwards Deming and others who state that process is the most driver for product reliability. If the process is not changing much, then you can expect the reliability staistics of new stuff to be simlar to that of old stuff. This is a significant basis for my statement, rather than "pure, absolute guesswork". Where is there any evidence to show that newer articles have a different reliability than older articles?

Also, I find your medical analogy to be dubious. A patient can appear fully healthy on medical tests one day and be in critical condition from a heart attack the next day. I am aware of no data that overall reliability of Wikipedia's articles can change so drastically from one day to the next. Sure, an individual article can change drastically in one day, but individual article reliability is not what the Nature study or the Gillmor quote are talking about. -- JTSchreiber (talk) 06:19, 17 January 2012 (UTC)

How do you spell WP:OR? You are taking the Deming's process-based statement, applying it to a online edit setting? That is pure, pure, pure WP:OR. And may I start the usual refrain of this talk page now: Do you have WP:RS sources that say Deming's statement is applicable here? Of course you do not.

And the issue is not if "you are aware" of how reliability can change. That is beside the point, unless you happen to be omniscient. What you "happen to be aware of" matters not, unless you have total knowledge. That would go well next to the Sanger prediction. Clairvoyance predicts things 10 years ahead, and omniscience leads to rationale for content... Way to go...

There is "zero basis" for any assertion that Wikipedia reliability in 2005 is the same as or different from 2011. My medical analogy did not say one day, it was about 6-7 years. And I still maintain that. No one has any idea if overall Wikipedia reliability has changed since 2005 because there have been no overall comprehensive measurements. Any statement about overall improvement or degradation of reliability is guesswork. Pure guesswork. Zero science, 100% guesswork.

As a side point, just as a matter of common sense, while the Deming process-based statement does have validity in various traditional manufacturing and service settings, it is easy to say that it does not apply to sausage making or Wikipedia. A quote (attributed, but probably not due to) Bismark says: "No one should see how laws or sausages are made." And that may also be true of Wikipedia content. No one has any idea where it came from. But going back to "the process" while the process of sausage making may remain the same during a 7 year period, the quality of the end product is highly dependent on the ingredients. If there is a dramatic change in ingredients, the quality will change. In a Deming scenario, to safegaurd against that scenario, the process should include an initial measurement of the ingredients. While the Wiki-editing process may not have changed much (but it has, it has) the editor profile is unknown. Apart from a few simple charts, we have no concrete idea how editor characteristics have changed. Most editors are anonymous.

When one mentions change in reliability, Wiki-standard bearers usually get alarmed, for they assume it means reliability has declined. That may not be the case. It may have improved. We just do not know.

My goal in this discussion was to reduce the risibility factor for this page. This discussion has not helped. I have been chuckling all along. Now let me bring up the 2001 Sanger quote again. A thinking reader would ask: "Are they using Sanger as an authority to predict things?" Is this not the same fellow who thought Citizendium was going to work? How can he predict things.... Is a chuckle in order here? History2007 (talk) 08:04, 17 January 2012 (UTC)

You wrote, "That is pure, pure, pure WP:OR." Please explain. Are you saying that my comment about Deming above violates the WP:OR policy? -- JTSchreiber (talk) 06:40, 21 January 2012 (UTC)

Yes, yes, yes. To go back to the usual page refrain, you have no WP:RS source that says that Deming's views apply to this case. Do you? When Deming stated those views there was no Wikipedia and personal computers were mostly unheard of. Might as well quote Plato. You made "the deduction yourself" about the applicability of Deming's views to Wikipedia. Talk page matters not, but it will be 100% original research if used in the page itself. Classic WP:OR. History2007 (talk) 08:06, 21 January 2012 (UTC)

What made you think I was intending to put something about Deming in the article? We were discussing my statement from May about "similar highs and lows of reliability", which was part of a discussion about whether old sources should be deleted. You wrote that my statement had no basis and I responded with the basis. It was all in the context of whether old sources should be deleted and I see nothing in the discussion to indicate that the context had changed before you brought up WP:OR. -- JTSchreiber (talk) 06:29, 23 January 2012 (UTC)

I still say that there is "no basis" for applying Deming to this case. Might as well apply Plato. And apart from your "personal opinion" that Deming applies you have no other support/basis/etc. It is just your personal guess/opinion. In any case the "out of date" tag is there for a reason. And that is another issue with respect to time, an article that may have looked OK in 2005 may just be useless today, as was/is the case here. I must really go and fix that one - instead of yapping here about Deming (for Heaven's sake!) but talking of the risibility factor here, how do you like Conan's latest on this topic? Accept it, the jokes are out there.... History2007 (talk) 08:39, 23 January 2012 (UTC)

I'm well aware that jokes are out there. There's no need to ask me to "accept it". -- JTSchreiber (talk) 06:24, 27 January 2012 (UTC)

Therefore keeping the outdated quotes, sans any support that they are of any applicable value to the current measure of reliability makes that section risible given the general mood of some of the public such as the IP above and myself. So why increase the risibility factor by keeping outdated material whose presence belittles the perception of how Wikipedia content addresses its own reliability? History2007 (talk) 08:58, 27 January 2012 (UTC)

The IP above? The IP editor never said that the content was funny. The IP editor never specifically mentioned any of the statements you are complaining about. Please don't put words into the mouths of other editors.

As far as the rest of you comments, we don't agree on whether the quotes are outdated, and you haven't provided reliable sources to show that the quotes are outdated. We don't agree on whether there's any significant risibility in the quotes. Where's the proof than anyone else besides yourself laughs at them? Conan's joke is about Wikipedia content that contradicts what's found in reliable sources. That's completely different than the content you are complaining about in Reliability of Wikipedia. -- JTSchreiber (talk) 06:38, 31 January 2012 (UTC)

I could put an outdated tag on it and start an Rfc etc. But why bother... The Rfc would be commented on "by Wikiedians" and not the world outside. So the issue is simple: an encyclopedia by Wikipedians for Wikipedians, etc. In the meantime, the world-wide chuckling continues outside ... History2007 (talk) 11:51, 31 January 2012 (UTC)

Banned in schools?

I happened to come across this from early 2011 and was somewhat surprised. Are some schools still banning Wikipedia? It would be interesting to know if the ban is still continuing, and how widespread it is. The article does mention a ban that started in 2007. I does not state if it is still in effect, and how widespread it is. Does any one have any more information to add to that section? History2007 (talk) 00:50, 21 December 2011 (UTC)

Yes, my niece's school bans it currently. So did the high school my kids attended. 155.213.224.59 (talk) 19:21, 5 June 2014 (UTC)

Why? HiLo48 (talk) 22:29, 5 June 2014 (UTC)

Her/their instructors do/did not believe it provides accurate information sufficient for scholarly research. 155.213.224.59 (talk) 17:40, 6 June 2014 (UTC)

Wikipedia is a great jumping stone for researching....put it simply its a great place to start. See Wikipedia:Researching with Wikipedia for more info. See also Wikipedia:Risk disclaimer. -- Moxy (talk) 00:05, 6 June 2014 (UTC)

Yes, as a high school teacher that's precisely what I tell my students. HiLo48 (talk) 00:20, 6 June 2014 (UTC)

Study about Psychiatry coverage

Someone just pointed out this study on Jimbo's talk page, it seems highly relevant to this page. The abstract states that Wikipedia stacks up well against the EB's Psychiatry coverage in a couple key areas. Mark Arsten (talk) 15:03, 22 December 2011 (UTC)

Interesting link. And certainly an item to add here. The fact is that on many patches Wikipedia reliability is quite high, on others quite low. That is why a "broad assessment" seems illogical. Wikipedia quality is far, far from uniform. I think if someone is tasked with comparing the articles on Mathematical logic with other sources, specially those edited by user:CBM, you will find really high quality articles. In many cases, scholarly quality seems to come from "super editors" who write many high quality articles and the way to improve scholarly quality will be to find that type of user. History2007 (talk) 15:11, 22 December 2011 (UTC)

USA bias

Can we have more about American Protectionism here, as we all know Wikipedia likes to print USA version of things only.

Great example being game of Football wich is played by 7 billion people with the name of "Football", but is presented in wikipedia as "association football" or "soccer", names that just don't exist to those 7 Billion people.

It's clear that biased people are writing things like these, there's also strong believe that NFL is paying wikipedia for the name change.

02:52, 30 December 2011 (UTC) If there's Lies, why would anyone donate to you? — Preceding unsigned comment added by 212.149.211.61 (talk)

Do you have any reliable sources that mention the claims of bias you are talking about? If so, I can look into putting these sources into the article. If not, these claims are original research which is not supposed to be found in Wikipedia articles. Also, please be careful not to use this talk page as a forum to discuss your personal opinions about Wikipedia. -- JTSchreiber (talk) 06:44, 30 December 2011 (UTC)

It ain't easy. I live in a city where four different sports known as football to (at least some of) their fans are played professionally. At least two others are played by enthusiasts. The one our IP loves is not the most popular. To call it football in my city would be quite inappropriate. Unique identifying names are the best. HiLo48 (talk) 07:40, 30 December 2011 (UTC)

I do not think the "NFL is paying wikipedia" type statement even deserves attention. That would be the day... And the 7 billion people are not English speakers by the way, and per WP:Commonname it is the usage in the US/UK/Australia/etc. that matters for English Wikipedia. What the game is called in Germany has no value for English Wikipedia, but it has for German Wikipedia. And the encyclopedic content does not get affected by minor naming issues anyway. History2007 (talk) 07:48, 30 December 2011 (UTC)

If the NFL is paying, why is the sport also called Soccer in Australia? (We really don't care much about the NFL here?) HiLo48 (talk) 10:56, 30 December 2011 (UTC)

Its called soccer in Australia because they called Ozy rules football football. You are really just supporting the idea that wikipedia is US biased by not knowing that! http://en.wikipedia.org/wiki/Australian_rules_football — Preceding unsigned comment added by 89.126.25.46 (talk) 16:42, 17 May 2014 (UTC)

I actually think this whole thread does not merit much discussion, and the NFL claim is just empty. Let us just move on. History2007 (talk) 13:41, 30 December 2011 (UTC)

proposals to improve assessment of reliability of wikipedia articles- making a quantitative and visible index

Currently, assessing the reliability of a given wikipedia article is difficult.

Problems: 1.Some heuristic methods are intuitively developed by people to make this assessment (length of article, number/quality of references, etc..). However, this heuristic is not made by most people, who often remain confused on this matter. This erodes the overall credibility of the encyclopedia.

2. Articles are sometimes flagged, or put in some categories, (and on the other hand, very few articles are classified as "Featured/good articles). But it must be recognized that this organization is very rudimentary. More efficient tools are needed and possible.

Proposals:

1. build a quantitative index of assessment of the probable "reliability" of an article. This index includes: length of article, number/quality of references, people's judgment about the reliability, etc..

2. Improve visibility of this index: adopt a color code. Instead of a uniformly white article, color each article and/or each paragraph, and/or each sentence with (the "whiter", the more "possibly reliable" the information is). Page ratings are not widely used nor visible.

It will also give incentives to good participation, as their changes will be more visible.Mokotillon (talk) 09:25, 25 January 2012 (UTC)

Yes, a formal methodology is needed. At the moment the "rating" assigned to an article is but a joke. Any 12 year old can go and change the rating on an article on advanced physics. On more visible articles that may/will get challenged, but on smaller articles no one will notice. However, I do not see an immediate solution, given the multitude of issues involved.

As an aside, even glaring errors are not noticed on smaller articles, e.g. look at this. My own copy/paste error, and it sat there for three weeks until I noticed it by chance. On these smaller articles with low page views, reliability is totally unknown. There are not enough people to check them, etc. The situation on Chicago is different and those articles do get checked. Yet I do not see an easy or logical way of assessing reliability on smaller articles anytime this decade unless Buffet leaves $1 billion to Wikipedia and librarians are paid to check it. Maybe someone should give him the idea.. History2007 (talk) 04:35, 26 January 2012 (UTC)

Mokotillon, thanks for thinking of new ideas for assesing reliability. However, this talk page is only for discussing improvements to the Reliability of Wikipedia article, not to propose changes to Wikipedia. A more appropriate place to post your ideas is the Village pump. -- JTSchreiber (talk) 06:32, 27 January 2012 (UTC)

Possible editorial bias in project pages list

I don't think it is an issue to list a few pages to the Wikipedia namespace, but a some of these seem to be almost promoting Wikipedia and its reputation. I removed two links that really seemed to be going against the site's neutrality. Wikipedia:Replies to common objections is a bit puzzling to see here too due to its slight lack of relevance and its POVish nature, but I left it. —WP:PENGUIN · ^{[ TALK ]} 19:56, 2 March 2012 (UTC)

Gender Bias

This section is lacking, there is an obvious female bias on wikipedia, which is in line with the liberal bias. This section needs to be expanded to include this criticism. — Preceding unsigned comment added by 27.32.222.133 (talk) 00:29, 12 May 2012 (UTC)

Evidence please. HiLo48 (talk) 00:34, 12 May 2012 (UTC)

Per WP:V, the section should not be expanded beyond what reliable sources have said. Do you have such sources? -- JTSchreiber (talk) 04:10, 12 May 2012 (UTC)

Proposed capture

Editors faults, page 2. [11].--Vagrand (talk) 01:19, 25 October 2012 (UTC)

Problems with citing the 2005 Nature study

As long as the fundamental articles in the chemistry and other basic areas are in as bad shape as they are—continuing in need of expert attention—anyone practicing in the fields can only sigh or laugh at this article, and its apparent self-affirming aims. In any case, we will not be sending students here to use the science materials. — Preceding unsigned comment added by 67.175.244.80 (talk) 01:31, 10 January 2013 (UTC)

You are free to improve them, if you feel they are badly done. Toccata quarta (talk) 08:55, 10 January 2013 (UTC)

Who would want to edit them, if we are not paid to do so? Scientists won't even bother questioning the reliability of Wikipedia, they already knew how unreliable Wikipedia is. 119.237.27.87 (talk) 23:36, 28 October 2013 (UTC)

There are more than 77,000 active contributors working on more than 22,000,000 articles in 285 languages. As for reliability you are correct as per Wikipedia:General disclaimer "Wikipedia cannot guarantee the validity of the information found here." That said Wikipedia can be a great tool and stepping stone for learning see Wikipedia:Researching with Wikipedia for more info. -- Moxy (talk) 23:51, 28 October 2013 (UTC)

Who would want to edit them, if we are not paid to do so? Um, people who want to make the sum of the world's knowledge available to everyone? People like the thousands of people who've been editing here since 2001? Nightscream (talk) 03:41, 29 October 2013 (UTC)

And people often contribute because they are genuinely interested a topic or genuinely motivated to bring knowledge to the public. That is the best motivation. --Erik den yngre (talk) 16:27, 29 October 2013 (UTC)

It fails in both theory and practice

The sad fact is, that the wikipedia concept is one conducive to an intrinsic lack of reliability. The English language version is an auspicious display of ego over integrity, editorial elitism, and an anal retentive preference for "rules" and the personal bias of "Sr editors" over accuracy and the comprehensive nature of the information contained. Depending solely on sources which have been arbitrarily declared acceptable, and disallowing information based on the expertize of casual editors is a sure fire recipe for a lack of accuracy and reliability. Fact and the inclusion or exclusion of information by virtue of consensus and compromise is the ultimate in anti intellect, and unfortunately that is the core of the wikipedia model. Facts are neither subjective,nor determined by the one who can win the favor of the most editors to support a point of view, and the idea that this process will boil down through multiple correction and edits into a reliable information source is preposterous. A comedian once said, "The good thing about Wikipedia is anyone can edit it, but the really bad thing about Wikipedia is that anyone can edit it. I grew up with a TV producer. We were born the same year. Wikipedia had him born 3 years later. When I corrected the page I was told mine was "original research". I then cited ImD, and was told it was not a "reliable source". When last I checked, the wrong date is still on the page. Cosand (talk) 22:15, 14 January 2013 (UTC)

Please see WP:NOTAFORUM. Toccata quarta (talk) 22:18, 14 January 2013 (UTC)

My comments addressed the topic to which the talk page was attached, it was not a random expression of personal opinion. Cosand (talk) 02:05, 17 January 2013 (UTC)

Prepare for error, and the error can be controlled. Unknown 11.39 25 April 2014 (UTC) — Preceding unsigned comment added by 109.148.158.57 (talk)

Information Loop and Wikipedia Editors

More needs to be made of the information feedback loop as it relates to Wikipedia page editors. An information loop is also formed when a Wikipedia article become an authoritative source, usually on a relatively obscure topic. The content of the Wikipedia page is then disseminated by journalists, corporations, government employees, websites and others who consult Wikipedia without citation. This reinforces what was written in the original Wikipedia article. The editors of the Wikipedia page then cite these sources to defend the page against corrections, further edits and other points of view. This "Wikipedia loop" makes many articles on Wikipedia highly resistant to change, even if errors or disagreements exist. The loop also grows stronger over time. This is a powerful force and can be abused by dogmatic Wikipedia page editors who want to transform their view into "fact."--CaperBill (talk) 15:05, 12 April 2013 (UTC)

In terms of internal Wikipedia policy, that is addressed by WP:MIRROR. However, I don't know offhand of any secondary sources that address this that could be cited in the article itself. Do you know of any? Nightscream (talk) 21:46, 12 April 2013 (UTC)

It's not mirrors that I'm referring to. The "Wikipedia Loop" is much more subtle than that. For example, consider a Wikipedia page lists the origins of a geographical place name. The history of this name as stated on the Wikipedia page might be a complete fabrication, merely incorrect or just one among many competing origins. However, if the origins of that name appears as "fact" in Wikipedia, over time it starts to be reproduced in government tourist literature, websites of tourist operators, newspaper articles, municipal websites about the area and so on. If a government communications officer (who knows nothing about the topic) has to write a paragraph on the subject, they most likely turn to the Wikipedia article without citing it. It is important to note that only the "fact" is reproduced, not the whole article or any prose (so it is not a mirror). After this "fact" circulates to reliable sources it becomes self-reinforcing. Since it is impossible to prove any link between the Wikipedia article and those who silently draw upon it, these supporting references are as always viewed as valid. The Wikipedia page editor can then use these sources as a reference to defend their position and resist change.

A well meaning page editor might not even realize that there is a causal link between the article and the sources that have silently used it for research. Wikipedia can be extremely powerful in shaping discourse on many subjects. Unfortunately, because Wikipedia is "drawn on" infinitely more than it is "cited," the true influence of Wikipedia can never be measured. On relatively obscure topics, only sources older than the Wikipedia page were written by people who did not read the Wikipedia page in question.CaperBill (talk) 16:25, 15 April 2013 (UTC)

Obviously, what you say is a valid potential problem, but like you said, how do we know that an external source that states the assertion in question derived it from Wikipedia, particularly in instances in which this is not explicitly disclosed? How do we know how pervasive this loop is? For my part, this is why I'm completely against the addition of unsourced material to articles, and tend to favor removing it, and not just the material that some say is "contentious", "controversial" or "likely to be challenged", since those criteria are completely subjective, and circular. Nightscream (talk) 18:04, 15 April 2013 (UTC)

Historians have a similar problem. Each time a scribe copies a manuscript, inferences, folklore, and the authors own biases get layered on, to say nothing of mistranslations. It's a game of "telephone" played over thousands of years. I suspect this is where most thing on the List of common misconceptions come from. For Wikipedia editors, historians, (and any other researcher, for that matter) it's always best to go back to the primary source. --69.205.190.188 (talk) 05:20, 19 November 2013 (UTC)

Information loops is clearly a potential problem as news media more and more use WP as source for facts. That is one reason to be skeptical about news media as sources for WP content, also a reason not to include any unsourced material. I think that editors should only use news media as source when there is first-hand reporting, that is when journalists have gathered "primary data". --Erik den yngre (talk) 15:20, 26 April 2013 (UTC)

The problem CaperBill describes has been informally dubbed "Citogenesis" and is depicted in this webcomic: http://xkcd.com/978/ See Talk:Wholphin#Grey Beast? for an actual case in which a bit of Wikipedia vandalism was repeated without credit in a book, and that book was later cited as a source for the vandalism. Adrian J. Hunter^{(talk•contribs)} 00:57, 27 April 2013 (UTC)

This problem is becoming more pervasive as the influence of Wikipidia grows. See Wikipedia talk:WikiProject Languages#Languages of the World: An Introduction for a book from by a respecteded academic publisher, used as a textbook at major universities, that contains text copied from Wikipedia. Kanguole 08:25, 22 April 2014 (UTC)

Amount of quality flaws

In a recent doctoral thesis [12] (M. Anderka, Analyzing and Predicting Quality Flaws in User-generated Content: The Case of Wikipedia, 2013), cleanup tags are employed to quantify the amount of human-tagged quality flaws in the English Wikipedia. Important findings are, among others:

... 26.86% of the English Wikipedia articles are tagged to contain at least one quality flaw, whereas 23.42% of the tagged flaws refer to a certain text fragment and 76.58% to the whole article. Furthermore, 50.1% of the tagged quality flaws concern article verifiability ...

The actual frequency of the flaws is even higher, it is likely that one out of five articles does not cite any references or sources.

Should this be mentioned in the article? NiPu (talk) 08:28, 12 July 2013 (UTC)

Tags are not relevant to this article; what's important is Wikipedia's factual accuracy. Toccata quarta (talk) 08:38, 12 July 2013 (UTC)

And sadly the corelation between what ought to be tagged & what is tagged is relatively low, but you won't get that from running stats on the database. Johnbod (talk) 23:32, 26 October 2013 (UTC)

Why isn't there a separate article?

The liberal extreme liberal bias is inherent in Wikipedia, and I believe it was designed that way. Look at the "you guys figure it out" mentality concerning definitions of words like "reliable source" and "notable." In practice this leads to a few editors being allowed to control given pages because they, though "consensus"(i.e., majority rules), get to define these words, and usually they are liberals. I've seen even major media outlets like Reason.com and the National Review described as "unreliable." Since most of the politics-related articles are simply lists of "this person has this opinion," it becomes paramount to determine whose opinion is "notable" and whose isn't. Because most of Wikipedia is written by liberals, and because it is, it is mostly the opinions of the liberals that are considered "notable." Just look at a typical politics article, ask yourself how many of these opinions are those of liberals, probably about 90%. And they are mostly one type of liberal. The fact that there is not a separate article for Wikipedia political bias(this article is mostly about factual inaccuracy, which is not the same as systemic bias), but there is an article for "genderfuck," is just one more example of liberal bias. Why isn't there a separate article, why is there not more than a few paragraphs? Undoubtedly much has been written about it, but because liberals get to control the definition of the words "reliable" and "notable," that stuff isn't reliable and notable! This is why I feel that editing Wikipedia is a waste of time.Train Trane (talk) 20:42, 25 August 2013 (UTC)

There isn't another article with a title containing "political bias" - political bias itself is a red link - or one generally about bias in Wikipedia. Maybe those should come first. Peter James (talk) 23:10, 26 October 2013 (UTC)

Interesting point (but possibly off-topic here), and similar but less newsworthy problems appear elsewhere... believe it or not, some of us find a particular POV evident in chemistry articles! And if Wikipedia had been around in the era of Galileo I guess he'd have been called a pseudoscientist. But the Wikipedia we have is far better than nothing, that's why I edit, and I guess why you do too. See WP:creed. Andrewa (talk) 19:13, 31 January 2014 (UTC)

Not really. Galileo was a very well-known and respected figure who had moved smoothly between teaching posts at two top universities before he became "controversial" in his fifties, and he always had a large proportion of contemporary opinion behind him. Johnbod (talk) 21:10, 31 January 2014 (UTC)

Well... we do seem to regard the IUPAC Gold Book as a not-so-reliable source. So it would have depended on whether that large proportion of contemporary opinion was regarded as reliable by the Wikipedians of the day. Don't get me wrong, I love Wikipedia. Andrewa (talk) 03:24, 1 February 2014 (UTC)

I think you may be looking for Conservapedia. The use of the term "extreme liberal bias" says rather more about you than about Wikipedia. Guy (Help!) 08:43, 6 June 2014 (UTC)

Unreliability of this article

I just noticed this article says, "Encyclopædia Britannica also argued that while the Nature study showed that the error rate between the two encyclopedias was similar..." but in the source (Britannica's repudiation of the Nature study) they vehemently deny this and criticise Nature for trumpeting a similar error rate in their heading while burying the one third more errors on Wikipedia's part:

While the heading proclaimed that “Wikipedia comes close to Britannica in terms of the accuracy of its science entries,” the numbers buried deep in the body of the article said precisely the opposite: Wikipedia in fact had a third more inaccuracies than Britannica.

Anthonyhcole (talk · contribs · email) 20:31, 21 February 2014 (UTC)

This is confusingly written & will be very unclear to those who haven't read the originals, but EB start by quoting Nature as saying: "the difference in accuracy [between Britannica and Wikipedia] was not particularly great.” The original Nature article used to be online, and should be reffed directly. Johnbod (talk) 22:08, 21 February 2014 (UTC)

How's this? [13]

Encyclopædia Britannica also pointed out that while the Nature report claimed a similar error rate for both encyclopedias in its heading, Wikipedia in fact had a third more errors and its mistakes were more often the inclusion of incorrect facts, while the mistakes in Britannica were more often errors of omission, making "Britannica far more accurate than Wikipedia."

I think it's clearer. I've clarified that ED doesn't think the study shows similar error rates, and I've added the one third difference to their case for EB being "far more accurate." --Anthonyhcole (talk · contribs · email) 22:42, 21 February 2014 (UTC)

I don't know. 162 vs 143 (was it) is not quite a third in fact. The whole thing got messy - see Nature's responses though you apparently now have to pay for the original article. I don't quite know why there is so much on it in the lead. I think the article is balanced - it doesn't for example trumpet the startling finding that American toxicologists thought WP less inaccurate than the FDA. Johnbod (talk) 23:04, 21 February 2014 (UTC)

You're not addressing my point. We say EB says that "the Nature study showed that the error rate between the two encyclopedias was similar." EB does not concede that. It explicitly takes issue with the claim the error rates are similar, pointing out that the study showed a significant dissimilarity. 166:123 (35%) is my count but I don't care if it's a third or not. I'm pointing out what they explicitly say about the "similar" claim - they explicitly repudiate the use of "similar" to characterise the error rates claimed by the study. We say they go along with the "similar" claim. --Anthonyhcole (talk · contribs · email) 08:32, 23 February 2014 (UTC)

The article says "The study by Nature was disputed by Encyclopædia Britannica,[3] and later Nature replied to this refutation with both a formal response and a point-by-point rebuttal of Britannica's main objections." How is this "We say they go along with the "similar" claim"? Johnbod (talk) 13:27, 23 February 2014 (UTC)

"Encyclopædia Britannica also argued that while the Nature study showed that the error rate between the two encyclopedias was similar..." --Anthonyhcole (talk · contribs · email) 14:36, 23 February 2014 (UTC)

Ok, removed from "while", which is not needed anyway [14]. Johnbod (talk) 15:56, 23 February 2014 (UTC)

Wow, this is way too long and dense

Just a heads up wiki people, but this guide is way, way to long for the average contributor to ever fathom.

If wiki is a sum of its parts, and ostensibly wants new blood, things like this are the rate limiting step.

Even I lost interest, and I came here looking for an answer. — Preceding unsigned comment added by 70.50.217.198 (talk) 17:39, 22 February 2014 (UTC)

Semi-protected edit request on 5 March 2014

Wikipedia is not a reliable source considering i am changing it right now. If you are reading this do not trust Wikipedia, if not then it is reliable. The reliability of Wikipedia (primarily of the English-language edition), compared to other encyclopedias and more specialized sources, has been assessed in many ways, including statistically, through comparative review, analysis of the historical patterns, and strengths and weaknesses inherent in the editing process unique to Wikipedia.[1] Several studies have been done to assess the reliability of Wikipedia. An early study in the journal Nature said that in 2005, Wikipedia scientific articles came close to the level of accuracy in Encyclopædia Britannica and had a similar rate of "serious errors".[2] The study by Nature was disputed by Encyclopædia Britannica,[3] and later Nature replied to this refutation with both a formal response and a point-by-point rebuttal of Britannica's main objections.[4] Between 2008 and 2012, articles in medical and scientific fields such as pathology,[5] toxicology,[6] oncology,[7] pharmaceuticals,[8] and psychiatry[9] comparing Wikipedia to professional and peer-reviewed sources found that Wikipedia's depth and coverage were of a high standard. Concerns regarding readability were raised in a study published by the American Society of Clinical Oncology,[10] a study published in the European Journal of Gastroenterology and Hepatology[11] and a study published in Psychological Medicine (2012).[9] Wikipedia is open to anonymous and collaborative editing, so assessments of its reliability usually include examination of how quickly false or misleading information is removed. An early study conducted by IBM researchers in 2003—two years following Wikipedia's establishment—found that "vandalism is usually repaired extremely quickly—so quickly that most users will never see its effects"[12] and concluded that Wikipedia had "surprisingly effective self-healing capabilities".[13] A 2007 peer-reviewed study stated that "42% of damage is repaired almost immediately... Nonetheless, there are still hundreds of millions of damaged views."[14] Several incidents have also been publicized in which false information has lasted for a long time in Wikipedia. In May 2005, in the Wikipedia biography controversy, a user added several false and defamatory statements to the biographical article John Seigenthaler.[15] The inaccurate information went unnoticed until September 2005, when they were discovered by a friend of Seigenthaler. After the information was removed from Wikipedia, it remained for another three weeks on sites that mirror Wikipedia content.[16] A biographical article in French Wikipedia portrayed a "Léon-Robert de L'Astran" as an 18th-century anti-slavery ship owner, which led Ségolène Royal, a presidential candidate, to praise him. A student investigation later determined that the article was a hoax and de L'Astran had never existed.[17] — Preceding unsigned comment added by Gavin137 (talk • contribs) 18:35, 5 March 2014 (UTC)

Removal of abbreviation

In the paragraph citing the "Guide to Military History on the Internet", is used the abbreviation "WP" (that I think refers to Wikipedia).

"When rating WP as the No. 1"

I believe that the abbreviation makes no sense in this context, as all the nearby references to Wikipedia use the full word.

To solve this problem, I suggest that someone that can edit the page replace "WP" with "Wikipedia". — Preceding unsigned comment added by 82.123.206.30 (talk) 14:32, 9 April 2014 (UTC)

WP:SELF, archives, and this page

An editor recently reverted another editor here, citing two reasons. Both were applied incorrectly.

Regarding WP:SELF: The point is to avoid gratuitous or casual self-reference, so that any project fork or repackaged version or the like still makes good sense. An internal poll of Wikipedians regarding preferred web browsers, say, does not belong in an article on web browsers. But in this article, the point is to talk about Wikipedia itself. The only aspects of WP:SELF that applies is that we should talk about ourselves in the third person, so that the article makes sense if bodily lifted to somewhere else (say a book on the ups and downs of on-line collaboration in general), and perhaps certain technical details mentioned WP:itself should be applied.

Regarding archives: They were certainly created for internal use. What, exactly, is improper about using them for something else? Nothing whatsoever. For most articles, of course, one uses ordinary wikilinks to allow for updating of information. In this case, however, we definitely want the dated reference/information, and so should definitely link to the archived version. Choor monster (talk) 11:31, 30 April 2014 (UTC)

As an example of correctly applying WP:SELF, my own inclusion of the Chihiro number hoax, as an example of something reaching the DYK page, was correctly reverted by another editor, who pointed out nobody outside WP seems to have noticed or cared, ergo, it's not notable. If down the road it achieves real-world notability, it can come back, along with appropriate internal references. Choor monster (talk) 11:40, 30 April 2014 (UTC)

The New York Times: Wikipedia Emerges as Trusted Internet Source for Ebola Information

• Cohen, Noam (October 26, 2014). "Wikipedia Emerges as Trusted Internet Source for Ebola Information". The New York Times. Retrieved October 28, 2014.

Might be a useful source for use in this article. — Cirt (talk) 03:09, 28 October 2014 (UTC)

Wiki-storm

I recently came across this from a couple of years ago about U.S. American colleges offering their students college credit, I A) don't know the reliability of the paper and B) question if this ever happened and if so ¿what did administrators do against it? (as it would be college-funded bias that might harm the reliability of this site), anyhow this can be a non-sensical article by some content farm to get headlines, anyhow ¿is this paper correct and were these programmes eventually condemned if they ever existed? Sincerely, --Namlong618 (talk) 21:12, 24 February 2015 (UTC)

Actually there are much more both feminist and college projects:

• Wikipedia:WikiProject Countering systemic bias
  • Wikipedia:WikiProject Countering systemic bias/Gender gap task force
• Wikipedia:Student assignments

And more. But I don;t understand how this is related to article subject. Do you have any reliable sources which suggest that such efforts must be condemned? I understand that giving credits for writing wikipedia articles may seem to be dangerously close to WP:COI, but I don't remember this claim discussed in the media (or in WP). Staszek Lem (talk) 21:51, 24 February 2015 (UTC)

I was wondering if it ever happened, and if it's considered to be harmful for the reliability, I'm a Socialist but I would never re-write any economic theories on Wikipedia to fit my ideology. Anyhow I was asking if it would be condemned.

Sincerely, --Namlong618 (talk) 10:19, 25 February 2015 (UTC)

Suggestion for a case study

I'm aware that there exist a bunch of studies on the spread of misinformation online (e.g., 1, 2 and 3). And Wikipedia's revision history offers a lot to answer various research questions. Here I suggest an interesting case for analysis. The incident involves a set of mysterious flags (A, B and C). I think it deserves in-depth analysis for the following reasons:

1. A wide circulation. These images have persisted for more than a decade (this file was uploaded in 2004 but there might be earlier copies). Before I started taking action on April 2, 2015, they appeared at literally thousands of pages of Commons, Wikidata, multiple Wikipedias and a Wiktionary.
2. Images as a loophole of verificability checking. Wikipedia has stressed verifiability, but these images have slipped though the checks.
3. Diffusion by translation. English Wikipedia serves as the hub while other Wikipedias act as spokes. The translators did not validate the reliability of English articles.
4. Language barrier hampers correction. This is a Japan-related topic. Almost all sources are written in Japanese. Even though Japanese Wikipedians identified misinformation seven years ago (ja:ノート:琉球王国#琉球国旗について), they failed to contain misinformation globally.

--Nanshu (talk) 14:34, 12 April 2015 (UTC)

This would be a poor case study, as Nanshu labels the flags as a "Wikipedia hoax". For them to meet this label, they would need to have been created on Wikipedia with the intent to trick others, which is not the case. This source from 1854 shows that one of the images has "persisted" for at least 161 years, while the other two existed since at least 1946. These flags were not created on Wikipedia and are therefore not "Wikipedia hoaxes". ミーラー強斗武 (StG88ぬ会話) 18:13, 13 April 2015 (UTC)

I believe these "mysterious" Ryukyu flags would indeed be a good case study. The inclusion in 2013 of a fictitious Tibetan flag on the List of Chinese flags by a one-time user blocked as a sockpuppet a week earlier on commons is another good case study regarding invented flags. The flag, created in 2011 based on an illustration appearing on a chocolate manufacturer collector card (with no mention that it would related to a Tibetan flag), rose suspicions a few days after its creation, but made its way into wikipedia projects in different languages and remained for one and half year on the English wikipedia until I removed it a few days ago. Hoax, test, or simple ignorance? It is difficult to assess, but it shows that we do have some serious reliability issues at WP, and that it is always easier and faster to add than to remove ivented content.--6-A04-W96-K38-S41-V38 (talk) 11:06, 16 April 2015 (UTC)