Wikipedia:Reliable sources/Noticeboard/Archive 96

It would be personal interpretation to say that it's an official statement, really. It could just as easily be interpreted as an artistic statement about Wikipedia. At best, though, even if it is an official statement, they are admitted that their source for their statement is Wikipedia, and we're not a reliable source by any means. Ian.thomson (talk) 19:04, 4 May 2011 (UTC)

I've commented there, I'll agree here. How can a screenshot of an outdated Wikipedia page be used as a reliable source? If that's the case, then we are saying the producers of the show are endorsing every single thing on that page. That seems to be a fairly large assumption. Dayewalker (talk) 21:01, 4 May 2011 (UTC)

The only thing the screen shot is reliable for is a very basic descriptive statement... saying that a version of the article was shown in the opening scene of an episode. Neither the screen shot, nor the fact that it appeared in the episode is in any way an endorsement of the information that was in the article at the time it was filmed. Blueboar (talk) 21:36, 4 May 2011 (UTC)

The only thing that we, as Wikipedians, are allowed to deduct is that "the words appeared on a TV series, therefore serve as a direct source unless can be proven otherwise". Their source is their own editing team, choosing to put that particular text that pertains to their own creation, in their own creation. The source is not Wikipedia. It happens to be a Wikipedia page, but they put it there on their own – this is, therefore, their own statement. Unless, of course, someone can come up with a reasonable proof that the makers of the series were "too high to notice" (an actual quote of one of the opposing voices on the talk page, just to display the amount of adequacy here) or otherwise oppose to the statement, we cannot make these kind of almost libelous assumptions, especially when they go against

common sense. Hearfourmewesique (talk

) 22:04, 4 May 2011 (UTC)

No... we are not allowed to make any deductions. We can say "Episode X opened with a shot of what appears to be the Wikipedia article on the TV show Weeds"... that does not require deduction and is purely descriptive. But that is as far as we can go. We can not draw any conclusions based on that shot. Doing so is OR. Blueboar (talk) 22:13, 4 May 2011 (UTC)

I'm not sure we can even say that. Neither Wikipedia's name nor its logo nor anything that can be iconified (officially or unofficially) with Wikipedia beyond the page's style appears on the page. As such it could be any wiki, including one designed specifically for the show. It might not even be a wiki at all; just a good job at photoshopping.陣内_Jinnai 22:19, 4 May 2011 (UTC)

Have you ever seen an actual Wikipedia article that doesn't say "From Wikipedia, the free encyclopedia" right under the title?

talk

) 19:31, 6 May 2011 (UTC)

This is clearly the start of our article (as of April to June 2009) with a photoshopping job or equivalent to get rid the menus, change the title, and add the photo. The only question is whether it's noteworthy that they cited us in this way. Hans Adler 12:11, 7 May 2011 (UTC)

Xeworlebi, can you help me out here with a simple summary (or a diff or three) that shows exactly which sentence this image is supposed to support? If it's something like "a modified screenshot from Wikipedia was shown in one episode", then it's probably okay. If it's something like "the meaning of the title is...", then it's not. (I just reverted that, BTW: http://forums.jokersupdates.com/ubbthreads is not a reliable source for any purpose more significant than "somebody posted something on the Internet"). WhatamIdoing (talk) 22:20, 4 May 2011 (UTC)

The sentence is "The show's title, according to Kohan, refers "to a lot of things", including marijuana and widow's weeds;^[11] however, it mainly alludes to "hardy plants struggling to survive."^[12]", this is the latest revertal/addition of that sentence, but if you take a look at the article history, 6 of the seven last edits by Hearfourmewesique were adding it some form or another, with the episode as source, the forum as source or both. The image used as the reference displays the following text: "The title is a play on words, referring to both the slang term for marijuana, and widow's weeds^{[citation needed]}." The forum was another source Hearfourmewesique introduced to support this, which also met with opposition at the talk page, but Hearfourmewesique believes this to be a good source because "One has be a real nutcase to go through the trouble to fake an interview...". _Xeworlebi ^(talk) 04:47, 5 May 2011 (UTC)

This is still a project to build an encyclopedia. If you have described the situation correctly, the editor should be blocked indefinitely if he or she does not desist immediately, as this is clearly either a case of

WP:COMPETENCE problems. The slogan that everybody can edit Wikipedia has some obvious small print that is too often ignored. Hans Adler

22:25, 4 May 2011 (UTC)

Thanks for confirming that it really was that bad... I just couldn't bring myself to believe it yesterday. I'm tending to agree with Hans. I'm amazed that anyone would think that a screenshot from Wikipedia is a reliable source for the facts presented in the screenshot. The second sentence of CIRCULAR addresses it directly: "do not use sources that present material originating from Wikipedia [you know, like a screenshot of an article] to support that same material in Wikipedia".

If this doesn't stop real soon now, we may need to get an admin to relieve the project of this timesink, so the rest of the folks at that page can actually do something useful. WhatamIdoing (talk) 23:23, 5 May 2011 (UTC)

Am I right in thinking that this TV programme is a work of fiction? Surely nothing presented in this show can be a reliable source for anything in the real world, as opposed to the fictional world it portrays! Sergeant Cribb (talk) 19:20, 6 May 2011 (UTC)

The only thing that's circular here is the amazing lack of attention to the following detail: it's a screnshot from the series that talks about the series. It's not a screenshot of a Wikipedia page. It's the screenshot of the opening credits of an episode that happens to have been taken from a Wikipedia page. It's basically saying that the makers of the series don't know what they are or what their statement is, so their source for facts pertaining to themselves is Wikipedia. Is anyone else finding this logic outrageously fallacious??? Hearfourmewesique (talk) 23:28, 6 May 2011 (UTC)

You are really trying hard to become one of the first editors explicitly blocked per

WP:COMPETENCE

, aren't you? Or is there any reason to believe that of all the possible reasons for putting a Wikipedia screenshot into such an episode the reason was that the creators of the series meant to endorse every single statement seen in the article, even including those that have "citation needed" behind them?

Sourcing is not some abstract game. It's about finding sources which make it likely that the claimed statements are correct. Sources that are clearly not endorsing a statement, e.g. because they are just citing it as this one does with the Wikipedia page, are useless for this purpose. Hans Adler 23:38, 6 May 2011 (UTC)

A block for

WP:LETGO is far more likely. WhatamIdoing (talk

) 02:48, 7 May 2011 (UTC)

I'm with Hans here. It's merely a humorous shot in the title sequence, Hearfour, like other humorous shots the series has used. The Wikipedia text shown is not anywhere near a reliable source for the intentions of the show's creator. This interview, now cited, is an acceptable source, so the issue is moot anyway. --JN466 16:33, 8 May 2011 (UTC)

Add mee to the "This is bloody sill"-camp. We can't use that text in the screenshot to reference itself. --Conti|✉ 16:45, 8 May 2011 (UTC)

The image is of a sign describing the remains of an ancient mosaic in an archeological site/museum in Aegina, Greece. (I took the photo.) The sign says: "Mosaic floor of a Jewish synagogue. The mosaic decorated the floor of the Jewish synagogue of Aegina. Discovered in 1829 to the East of the "secret harbour" in the midst of an important quarter of the ancient city, the building has been excavated only in parts because of its modern superstructures. During the investigations remains of an older predecessor have been identified (hatched walls). Through the entrance in the West and an antechamber people reached the main hall which measured 13.50 to 7.60 m. To the East - oriented towards the city of Jerusalem - a raised apsis was annexed in which the shrine for the Torah was stored. Between apsis and main hall stood the reading-desk (white recess in the eastern mosaic-frame). For the purpose of preservation the mosaic has been removed from its original context to the actual position. It displays the typical decoration-patterns of mosaic-floors of the 4th century AD consisting of artistic geometric designs in a rich scale of colours. At the west side near the entrance two donation inscriptions are visible. They refer to the erection of the synagogue using revenues of the community and donations of its members when Theodoros was head of the synagogue for four years. The completion of the synagogue and the bedding of the mosaic took place when his homonymous son held the same position. The Jewish community of Aegina has been established probably when Jews and other refugees resorted to the island in the course of the Barbarian invasions into Greece at the end of the second and during the third century AD. The abandonment of the synagogue is testified by several Christian graves in the foundation walls of the building (dotted rectangles). The mosaic has been created between 300 and 350 AD."

My question is, Can my image be considered an RS for editing the Aegina article to say, for example, that Jews are believed to have settled in Aegina late in the second century and in the third century when the Barbarians invaded Greece? Also, what format would the ref take in such a case?—Biosketch (talk) 19:08, 4 May 2011 (UTC)

An interesting question. I'd think that the image cannot really be considered as a reliable source, given that we can't actually confirm where it was taken, and that it hasn't been photoshopped (not that I'm suggesting it has, needless to say). This raises the question as to whether the sign itself is a reliable source. It is verifiable in that presumably anyone else can check it if they visit the site, and unless there are good grounds to doubt the academic credentials of the museum, It would seem to qualify as a 'source'. I don't think it would qualified as 'peer reviewed' though, so I'd be wary of using it as a sole source for anything controversial. As for exactly how one would cite a sign in a museum, I've no idea.

Sadly, I think that the best solution might be to try to find another more conventional source that says the same thing. AndyTheGrump (talk) 19:39, 4 May 2011 (UTC)

Both the original sign and the photograph of it are acceptable. Use {{cite sign}} if the article uses the most common family of citation templates.

Per Wikipedia:Images#Pertinence_and_encyclopedic_nature, we do not require editors to provide a published reliable source that they aren't lying about the contents of the images they upload. WhatamIdoing (talk) 19:51, 4 May 2011 (UTC)

Going to be annoying, and point out that the image itself is a violation of the signwriter's copyright and needs to be deleted; Greece

iridescent

20:30, 4 May 2011 (UTC)

...deleted from commons but it could still be used here with a decent fair use rationale, no ? Sean.hoyland - talk 20:36, 4 May 2011 (UTC)

I don't think you can cite your picture of the sign ... But there is nothing wrong with citing the sign itself. The sign is verifiable by anyone by simply going to the location and reading it. Blueboar (talk) 20:42, 4 May 2011 (UTC)

Wikipedia doesn't need to have a copy of the image for you to cite the sign; we only need a copy if you want to include the image directly in the article, as part of the article's contents. (In fact, we'd normally send copies of sources over to Wikisource.) WhatamIdoing (talk) 22:23, 4 May 2011 (UTC)

Thank you for the replies. It's since come to my attention that others before me uploaded similar images to Commons – just no one edited the

You don't actually need another reliable source.

Sources do not have to be available online. You can cite the sign directly. The sign itself is sufficient. WhatamIdoing (talk

) 23:27, 5 May 2011 (UTC)

It's preferable if you can find something more accessible but I agree the sign does satisfy the requirements of a reliable source and if you haven't a particular author could e credited to the museum since they have approved it being in plain view. I think it also counts as a secondary source as giving commentary rather than a primary source but someone else might disagree on that.

Dmcq (talk

) 06:11, 6 May 2011 (UTC)

You mean "something more accessible of equal quality, of course. A blog post might be

free full text on the net, but it is certainly not a preferable source. WhatamIdoing (talk

) 02:53, 7 May 2011 (UTC)

As someone trained in Museum science I have point out that not all museums are created equal. Funding and resulting the quality of the staff are key factors in determining quality. A nearly all volunteer museum (even if is funded by the government) is going to have problems an all professional museum will not. Then you have private museums whose quality can be all over the map. The sign itself is not reliable without knowing more about who put it there.--BruceGrubb (talk) 21:15, 8 May 2011 (UTC)

If there is only one reliable source for the information (the sign) I would think the material should be attibuted to it. Ward20 (talk) 06:42, 6 May 2011 (UTC)

On the West Ridge Academy page, an editor is removing [24] the information "The Utah Boys Ranch was founded by Lowell L. Bennion and a group of his associates in 1964." That information was sourced to this newspaper article [25], which the editor claims is taken directly from Wikipedia. As you can see from the talk page, the editor shows that the sentence appeared previously on Wikipedia, and draws the conclusion that the article must have come from the WP page. More eyes and opinions would be appreciated. Dayewalker (talk) 04:40, 5 May 2011 (UTC)

Just to add a few more relevant details to this request: the exact sentence "The Utah Boys Ranch was founded by Lowell L. Bennion and a group of his associates in 1964." was added to the original Wikipedia article in 2009. It appeared, verbatim, in a newspaper article written in 2010 (referenced above). It appears no other place prior to 2009, other than on a blog (where it seems the sentence originated). Every other primary source about Lowell Bennion, such as his biography and the other sources about his life, completely contradict the claim that he had anything to do with the Utah Boys Ranch / West Ridge Academy. See the talk page for more the sources. --EarlySquid (talk) 04:51, 5 May 2011 (UTC)

If the preponderance of sources contradict this exceptional claim in a newspaper, then I would think it would be better that this newspaper not be used as a source. TimidGuy (talk) 11:19, 5 May 2011 (UTC)

It looks like FyzixFighter has provided links to a handful of sources from the 1980s (=sources that cannot possibly be copied from Wikipedia) that support this. I therefore believe that the preponderance of sources do not contradict this claim, and that Bennion should be mentioned in the article as a founder or co-founder. WhatamIdoing (talk) 03:02, 7 May 2011 (UTC)

A quick search through Google books produced a pre-2009 work that supports the claim:

Bradford, Mary Lythgoe, (1995) Lowell L. Bennion: Teacher, Counselor, Humanitarian; Page 214-5

Since wikipedia launched in 2001 there is no way this information came from wikipedia--BruceGrubb (talk) 21:31, 8 May 2011 (UTC)

I don't know why you people insist Leroy was Filipino. We are his family and he is 100% Portuguese. Documents are published by people not by our family. Please correct your article it is incorrect. You can speak with my wife and get the truth. Please stop, it does no one any good printing false hoods. Let him rest, he has given enough.

Note to closer: Involved

WP:OR; Wikipedia is the end of the stream for correction, not the top. Dru of Id (talk

) 09:16, 9 May 2011 (UTC)

As can be seen [26], the Factbook gives a listing on a per-country basis, with 'ethnic groups' in turn normally listed in decreasing percentages. The page gives no general indication of sourcing for the data, though a few have an indication that they are from a census, giving a date. For reasons why I consider the Factbook figures problematic, let us consider a few of the more obvious examples:

Albania gives two different figures for the 'Greek' population: "1% (official Albanian statistics) to 12% (from a Greek organization)".

Argentina gives a figure of "white (mostly Spanish and Italian) 97%", which all those involved in the long-running White Argentine/Argentines of European descent debate have acknowledged is implausible, and contradicted by other sources.

Neither Denmark, France, Madagascar, Malawi nor Maldives for example give percentages for the various groups named - there are other examples, but listing them all seems pointless.

Italy lists "Italian (includes small clusters of German-, French-, and Slovene-Italians in the north and Albanian-Italians and Greek-Italians in the south)". Are we to understand that there are no other ethnic minorities in Italy?

Portugal states "homogeneous Mediterranean stock; citizens of black African descent who immigrated to mainland during decolonization number less than 100,000; since 1990 East Europeans have entered Portugal".

...Which contrasts nicely with Spain with a "composite of Mediterranean and Nordic types" - since when have 'types' been an ethnic group?

...And then there is Switzerland, which the CIA suggests is "German 65%, French 18%, Italian 10%, Romansch 1%, other 6%" - one can only assume the CIA does not understand the difference between a linguistic category and an ethnic group.

The United States entry, which one might at least expect the CIA to get right, is clearly based on US Census Bureau data - but then has to admit that it omits data for 'Hispanics', "who may be of any race or ethnic group" - even the data provided for the US is incomplete.

One can only reasonably conclude from this that the data is derived from multiple sources, using differing criteria of 'ethnicity' and of differing utility. One could be charitable and describe this as a tertiary source - except that it rarely gives any indication of where the data is from. The better examples are clearly derived from relevant national census data, but such data is usually simple enough to track down and cite properly. The poorest examples, as I have indicated, seem to consist of confused 'linguistic' and 'racial/descent'(?) categories rather than anything related to ethnicity as such, or simply list groups with no percentages at all. On this basis, I'd argue that one can only rely on the CIA figures when they can be verified against properly-collated census data - which is another way of saying that the Factbook is no good as a citable reference. AndyTheGrump (talk) 22:30, 5 May 2011 (UTC)

The other problem with the data in the Factbook is that it gives no mention of sources. For some countries for which the factbook gives information about race the official censuses of the country does not include information about race. How does the factbook then get the information? In the case of Mexico the last census that took down data on race was made in 1921. This[27] study of demography of ethnic categories discusses the reliability of the Factbook. It is quite clear that the factbooks data cannot be used as a reliable source without corroborating and correcting with other more reliable sources. If the source cannot be used by specialists without doing specific research, analysis and fact checking then it obviously cannot be used uncritically for wikipedia purposes either.·Maunus·ƛ· 22:47, 5 May 2011 (UTC)

As a point of fact, Wikipedia does not require reliable sources to name their sources, so all of this about "it gives no mention of sources" is completely irrelevant.

IMO the CIA Factbook is clearly a reliable source for this information. They appear to update and correct things over time, and they have a process of editorial oversight. That—and not the provision of a bibliography or an explanation of their census method—is the definition of a reliable source on Wikipedia.

It is also not the only reliable source, and if there is any disagreement, then editors should consider a range of reliable sources, and if necessary report that reliable sources directly disagree: e.g., that the country's official census says X, the CIA Factbook says Y, and the political opposition says Z. WhatamIdoing (talk) 23:36, 5 May 2011 (UTC)

No wikipedia does not require reliable sources to name their sources, but the question of whether a source is reliable or not hinges on whether they can be believed to purvey credible information, which again (as for all claims to scholarly credibility) hinges on the transparency of the factual bases for their statements. No source is apriori reliable just because it is governmental. The editorial oversight of the factbook does clearly not live up to the expectations of peereview, as it is clearly not experts who review the information of ethnicity as expert testimonies about the usefulness of the factbook makes clear.·Maunus·ƛ· 23:58, 5 May 2011 (UTC)

You have missed the point: Your personal belief that a source purveys (or doesn't) credible information is irrelevant. We've got lots of folks who think that gold-plated reliable sources have nearly everything wrong. "I believe the information in the source is right" is not one of Wikipedia's standards for determining whether a source is reliable. "Transparency" might be favored among academics, but it is not one of Wikipedia's standards for determining whether a source is reliable. "Living up to the expectations of peer review" is also not one of Wikipedia's standards for determining whether a source is reliable.

Wikipedia requires some notion of editorial oversight (you can't just post whatever you like whenever you like it) and some notion of fact-checking (you don't routinely get everything wrong, and you might correct stuff if you get it really, really wrong). Wikipedia follows the journalism standard, not the academic standard. Wikipedia accepts routine newspaper articles as reliable sources for the facts presented in them, despite newspapers being non-transparent and not even pretending to academic standards. The CIA World Factbook is no worse—and probably no better—than citing a newspaper article on this point. Both newspapers and the Factbook are reliable sources for this information.

NB that I'm not saying they are the "best possible" sources for this information. I only say that they meet the minimum standard for reliability, and therefore may be used. WhatamIdoing (talk) 19:43, 6 May 2011 (UTC)

You have missed the point. I have presented a reliable source demonstrating that the factbook is not reliable for studies of ethnic groupings in most of the world.·Maunus·ƛ· 22:48, 6 May 2011 (UTC)

So your position is that they are 'reliable' even after it is demonstrated that they aren't? AndyTheGrump (talk) 19:47, 6 May 2011 (UTC)

My position is that any source can be

WP:RS isn't a redirect to dictionary.com: if we were using plain English, we wouldn't have long policies and guidelines defining what wiki-reliability is and how wiki-reliability differs from, say, AndyTheGrump-reliability. WhatamIdoing (talk

) 21:26, 6 May 2011 (UTC)

"The reliability of a source depends on context. Each source must be carefully weighed to judge whether it is reliable for the statement being made and is the best such source for that context.":

WP:RELIABLE. Do you not consider my comments, and those of Maunus, constitute "carefully weigh[ing] to judge" the reliability of the source in question? Can you show us how you arrived at your judgement? Or are you going to ignore this part of WP:RELIABLE and just repeat the mantra that a source is reliable because it is a reliable source? AndyTheGrump (talk

) 21:38, 6 May 2011 (UTC)

Agree with WhatamIdoing. The CIA World Factbook is definitely a reliable source, and eligible for inclusion. As discussed in various places though, doing good research and constructing neutral articles does not necessarily mean giving all sources equal weight. For some facts there will definitely be better documented and more transparent sources than the Factbook. If there are good reasons to believe that the Factbook is less than accurate (or outright wrong) for particular points than you should at least present the opposing view. In some cases, one may even be right to omit the Factbook data if the majority of scholarship on an issue clearly disagrees. Saying that the Factbook is a reliable source, in general, is not the same as suggesting we should parrot everything they say. It is a starting point, and often a very useful one when few other sources exists, but if better sources exist then article writers should obviously take note of that and respond accordingly. Dragons flight (talk) 20:06, 6 May 2011 (UTC)

I am not asking whether the CIA Factbook is reliable in general. I am asking whether it can be considered

Wikipedia:Identifying reliable sources requires: that sources be "source must be carefully weighed to judge whether it is reliable for the statement being made" - they is no absolute yes/no answer regarding whether a source is always reliable. My point is that all indications seem to suggest that regarding ethnicity, the Factbook isn't. AndyTheGrump (talk

) 21:44, 6 May 2011 (UTC)

I stand by my statement in the context of ethnicity. If you want to replace Factbook statements with something else, then there may be good cause for that in many cases, and such issues should be discussed. (You don't seem to have made any general suggestions for alternative sources.) If you want to introduce a general prohibition on Factbook-based ethnicity citation, then I don't see that you've made a sufficient case for that. Ethnicity is a fuzzy concept (as I'm sure you know). As a whole, Factbook seems to strive for accuracy (and hence reliability), though at times they are undoubtedly limited by the poor availability of data. That's a reason to discuss the uncertainties, and the shifting and fuzzy definitions for various groups. But unless you've got something better, I don't see it as a reason to exclude the Factbook in general. Dragons flight (talk) 22:28, 6 May 2011 (UTC)

Perhaps let me put it another way. It think Factbook aims for accuracy in their reporting of ethnicity, and probably does as well (or even better) than other traditionally reliable sources such as newspapers. However, because ethnicity is hard to measure, and often measured in ways that are inconsistent for one study / census to the next, the accuracy of statements about ethnicity should be expected to come with large uncertainties, even from reliable sources. Hence we should look for the best sources we can to try and reduce the uncertainty and put information on an equal standard for comparison. (Though, we should do that always anyway.) But if we can't find better information than I think we should still continue to use sources like the Factbook, but make a point of making people aware of the uncertainties and difficulties involved. Dragons flight (talk) 22:42, 6 May 2011 (UTC)

How do you base this opinion in the face of reliable sources saying that the factbook does not do a good job at this particular task? It doesn't impress you that scholars have written papers about the unreliability of the factbook for demographic breakdowns by ethnicity? Why does your opinion of the factbook override the published testimony of an expert?·Maunus·ƛ· 22:51, 6 May 2011 (UTC)

Read what I am saying. Use better sources, if you have them. If you don't have them, I would still regard Factbook as better than nothing. I'm sure many experts would tell you that newspapers are horribly prone to factual errors, but we still rely on them all the time. A source can be "reliable" by Wikipedia standards without being particularly good. This is why we should look for the best possible sources, but there are many forms of low-quality sources that are allowable in Wikipedia at least until something better is provided. Dragons flight (talk) 23:08, 6 May 2011 (UTC)

I don't think the "journalism standards versus academic standards" is based on policy. We often need to use non-academic sources but we recognise that there are problems with newspapers, not least the "recentism". The points made about errors in the Factbook are well worth considering. I don't think we can rule it out completely though, because sometimes it is the only source we have. For example, even those 100% politically opposed to it used it for estimates of ethnic groups in Vietnam during and immediately after the war. It is often out of date. Data from a national census website is definitely preferable. The immediate question seems to be about whether the category "white Latin American" makes any sense at all. The CIA use such a category but national censuses don't. It's a tricky one. A parallel is for data on France. In France it is widely seen as unethical to collect data on ethnic identity (not nationality), because "everyone with French nationality is French". Non-French social scientists have no such qualms and would like to see estimates for the different ethic groups, in order to compare the ethnic composition of France to other countries in Europe. What should we do, follow the French lead, or that of the Americans and other countries? It probably needs to be resolved case by case. Itsmejudith (talk) 20:00, 6 May 2011 (UTC)

Actually, the CIA doesn't use the term "white Latin American". The Factbook gives widely-differing categorisations for ethnicity for each country in the area, sometimes with no separate 'white' category, and for Venezuela gives no percentage figures at all. AndyTheGrump (talk) 20:11, 6 May 2011 (UTC)

Itsme, the actual standard described in the policies and guidelines is a very typical journalistic standard: you need editorial oversight and a hope of fact-checking—the two basic standards at any journalistic endeavor. Academic publications set the bar much higher: there, you get editorial oversight, significant fact-checking, critical evaluation through peer review, a transparent list of sources you've drawn on, and compliance with discipline-specific rules. (For example, we've had to go to some trouble to convince hard-science academics that they really are supposed to use and cite secondary sources on Wikipedia—even though doing so is actually prohibited in most hard-science academic publications.) We've chosen the lower standard as the minimum for our sources. This source meets our minimum standard. WhatamIdoing (talk) 21:26, 6 May 2011 (UTC)

To you(Andy) and Maunus it is not reliable but to the Encyclopedia it is or what next the New York Times is not a reliable source either because you possibly disagree and deem them not credible on a certain subject,we can't cherry pick. Andy and Maunus here are POV pushing editors trying to push there own view of race/ethnicity instead of letting reliable sources guide articles --Wikiscribe (talk) 19:57, 6 May 2011 (UTC)

I also don't like you very much.·Maunus·ƛ· 22:54, 6 May 2011 (UTC)

Thank you for your totally irrelevant personal attack. AndyTheGrump (talk) 20:01, 6 May 2011 (UTC)

Wikipedia exists for the general public. This public uses sources like the CIA World Factbook (and newspapers, as others have stated); so why shouldn't the Factbook be among the sources from which WP article content is drawn? Maunus and Andy want a far stricter, narrower set of sources to be used. They not only want them to be exclusively academic ones (as if these are necessarily perfect and bias-free), but often give the impression that they want to use only the work of those academics with whose POV they agree.

Maunus, regarding that study that questions the Factbook's reliability: Why should that study be used to determine the Factbook's reliability all of a sudden? Wouldn't that constitute giving the study undue weight? Yet the fact remains that that study notwithstanding, the Factbook has continued to be accepted as reliable by this community.

Answering Andy's points:

What would you prefer, Andy: that the CIA, despite knowing that sources vary over the percentage of Greek population in Albania, give you one sole figure and pretend that no variation exists? Maybe that would make you feel better, but the fact is that you'd be less informed, not more. Kudos to the CIA for being so open (wink).

Do you actually expect that a full list of all the minority groups present in Italy is feasible or even meaningful? Obviously the point is to include the largest minority groups, not all of them. I'd be completely schocked if a developed country like Italy didn't have some people from nearly a hundred countries or even more. It would be senseless to list them all in the Italy entry—and then there are the more that 200 other countries listed in the Factbook.

Regarding Switzerland, it so happens that "German", "French", "Italian", and "Romansch" are demonyms/ethnonyms as well names of languages (!). Not to mention that in Switzerland, as in many countries, ethnic identity is connected to language.

The fact is that "Mediterranean" and "Nordic" *are* terms used in ethnology.

There is nothing incomplete about the US entry. The info given is for what is officially termed "race". Since Hispanics and Latinos are not a race, but an ethnic category, there's no Hispanic/Latino entry; instead, Hispanics and Latinos are included in the percentages given for the races (most are included in "white 79.96%"). (For more about US race and ethnic classification, a good starting point is

Race and ethnicity in the United States Census

. (Full disclosure: it is an article which I've edited much, though not in recent months.)

Andy would want the CIA's info about ethnic groups to be limited to numbers. But is the fact that it isn't (for example, it mentions "types" in one instance), and the fact that numbers are given for some countries and not others a good enough reason for ignoring the numbers that are given, in such a widely used source?

All sources are supposed to be balanced, and that goes for anything taken from the Factbook. I propose that adding balance, not keeping the Factbook out, should be Andy's and Maunus' mantra. SamEV (talk) 02:01, 7 May 2011 (UTC)

"that study notwithstanding, the Factbook has continued to be accepted as reliable by this community". Really? So the 'community' was aware of the study? And the 'community' cannot change its mind? Don't kid yourself that getting this accepted as 'reliable' (which seems unlikely, unless you define 'reliable' as 'not good, but marginally better than nothing') is going to justify the dubious

white Latin American article in any case. I only brought this here first to bring general attention to the questionable reliability of the source, and will deal with that issue later. AndyTheGrump (talk

) 02:40, 7 May 2011 (UTC)

The figures should be cited in the article as they appear in the Factbook. The Factbook lists them under "ethnic groups": is there a reason why they should not be cited as "ethnic group" figures, then? SamEV (talk) 03:30, 7 May 2011 (UTC)

Andy and Maunus's points are serious. One other way of seeing it is as a tertiary source, a compilation of facts gathered from national censuses and other surveys. Can be used with caution where it is the only source. Does not have to be used in every case. Where it conflicts with other sources, give those too. Keep an eye out for where it is not up to date. Itsmejudith (talk) 10:24, 7 May 2011 (UTC)

It is rather difficult to assess whether figures are 'out of date' when there is no source stated for the data. I think a consensus is emerging that the Factbook cannot be considered a 'reliable source' regarding ethnicity, except in the loosest possible terms - that if no other source is available, the data is better than nothing, but it cannot be taken as authoritative where there is better information to be found elsewhere. I think one can conclude from this that the way Factbook data is being used to arrive at an estimate for a 'white Latin American' percentage figure is highly questionable, and that the source should not be used in this way. AndyTheGrump (talk) 12:38, 7 May 2011 (UTC)

Use it. It's authoritative. It's like saying don't use Gray's Anatomy as a source. Yes it is a tertiary analysis of many sources. If we want to do a section about academic debates over ethnicity, then we can go ahead and cite academic journals. This debate sounds oddly familiar; often somebody doesn't like one label or another, or distrusts the U.S. to make the estimate. The alleged "errors" are actually spot-on. Argentina actually is mostly of Italian and Spanish ancestry. Italy might have a few other peoples besides those listed, but are they numerous enough from a foreign-policy perspective? And in the U.S. Census, "Hispanic" is considered a cultural/language identity, not a race, and we have Hispanic people of many different backgrounds. By all means use the source, but feel free to compare/contrast with other sources. Squidfryerchef (talk) 17:45, 7 May 2011 (UTC)

"Argentina is mostly of Italian and Spanish ancestry". No, or at least not except in as much as migrants from Italy and Spain have made the largest single contribution to the Argentine 'Gene pool'. As our Argentines of European descent article notes, there have been several studies of the genetic makeup of the Argentine population, all of which suggest that they have a significant contribution from Amerindian populations too. The 'European' proportion is the largest (excepting that for Mitochondrial DNA, which is transmitted solely on the maternal line), but most Argentines have at least some Amerindian ancestry.

That is genetics of course, and the debate is (or should be) about ethnicity, which is a social construct, and only has a loose correlation with genetics, or with actual descent. This then raises the question, what do the Argentines themselves consider their ethnicity/'race'/whatever to be? Fortunately, we have access to a survey which answers this (though in Spanish, and reporting the survey as a secondary source - possibly this gives it more credibility though) I refer to Étnia, condiciones de vida y discriminación [28] by Simon Schwartzman (a notable Brazilian social scientist). In this, he cites an ECosociAL, 2007 survey which gives the percentage IDENTIFICACIÓN ÉTNICA for seven Latin American countries. For Argentina the figures are as follows: Blanco 63, Negro 1, Indígena 1, Mezcla de blanco con negro 8, Mezcla de blanco con indígena 9, Mezcla de negro con indígena 0, Mezcla de todo 16, No sabe, no contesta 3. Clearly, the Blanco figure of 63% is incompatible with Factbook data, which describes the Argentine population as"white (mostly Spanish and Italian) 97%, mestizo (mixed white and Amerindian ancestry), Amerindian, or other non-white groups 3%". Both genetic data, and surveys of perceived ethnicity, conclusively demonstrate that the Factbook is wrong. AndyTheGrump (talk)

Itsmejudith: "Where it conflicts with other sources, give those too." ". Squidfryerchef: "By all means use the source, but feel free to compare/contrast with other sources."

Exactly.

BTW, the preponderance of sources place the Argentine white percentage at 85 or higher.

Andy, you've found the Schwartzman source. That's the point: use it to balance the Factbook. Your problem is solved. Quit expending so much energy trying to have the Factbook banished. Quit being so intolerant of information that is contrary to that which you prefer. SamEV (talk) 19:39, 7 May 2011 (UTC)

I'd remind you that it was your insistence that it was legitimate to use the CIA Factbook figures to arrive at an overall 'white Latin American' percentage figure that led to me raising this issue here in the first place. I considered this

WP:OR, and said so. If you are now proposing that we should somehow combine the CIA figures with that from other sources to arrive at this percentage, we are going even further into WP:OR territory, and I cannot believe for one minute that this would be acceptable. The whole exercise is pointless anyway, as perceptions of 'whiteness' vary markedly across Latin America, as I'm sure you are aware, and any 'percentage' would be practically meaningless. AndyTheGrump (talk

) 19:53, 7 May 2011 (UTC)

Yes, it is my insistence that we can add up the figures to arrive at a regional figure.

Of course I'm not proposing that the CIA figures be combined with any other. I'm saying that both should be mentioned, separately, for balance.

Let it be 'meaningless'. Just make sure to cite a reliable source that explains why it's meaningless, so it's not just coming from you. SamEV (talk) 21:41, 7 May 2011 (UTC)

On the contrary, since you are supporting the inclusion of your figure, it is for you to provide a reliable source that suggests it will have any meaning. You are proposing to add up data from sometimes-unknown primary sources, using widely-differing ethic terminology, and lacking any usable data at all from several countries, to arrive at a 'percentage' for the 'white Latin American' population, in a context where there is no consistancy whatsoever in regard to how the term 'white' is used. This isn't just

WP:OR it is unscientific, bogus, and based on an entirely false concept - that 'white Latin Americans' consist of something other than an 'ethnic category' created by outsiders. Nonsense like this should have no place in Wikipedia. AndyTheGrump (talk

) 22:55, 7 May 2011 (UTC)

Can I ask what you are basing your 'HO' on? This is neither a vote nor a !vote, and a little more elaboration might help... AndyTheGrump (talk) 02:12, 8 May 2011 (UTC)

Andy, regarding meaningfulness: this is an encyclopedia, so providing information, providing facts (you know, numbers and such) from reliable sources is part of what we're supposed to do. Merely adding up the numbers (same source, same type of data, same region) is trivial. SamEV (talk) 23:28, 8 May 2011 (UTC)

You think so? Then I suggest you give us a complete breakdown of how you arrive at the figure, and then we can have an RfC on whether it is

WP:RS though. AndyTheGrump (talk

) 23:35, 8 May 2011 (UTC)

It's really simple, Andy: If Latin America consists of countries B and C, then the total white population of Latin American consists of the white populations of B and C.

"It still hasn't been demonstrated that there is anything like a consensus that the Factbook ethnicity data is

WP:RS

though."

I'm convinced your denialism has no end. I refer you to every comment that's been made in this thread by anyone not named "Maunus" or "AndyTheGrump".

I'm sorry, Andy, but your objections and your obstinacy are becoming embarrassing. SamEV (talk) 23:46, 8 May 2011 (UTC)

Are you going to show us the figures, or not? You certainly cannot include them without indicating how you arrived at them. AndyTheGrump (talk) 01:25, 9 May 2011 (UTC)

Andy, I'm not sure that you're understanding the consensus here: Everyone is telling you that this source may be used. Nobody has said that it must be used. Using this source is permitted, but it is not required. If you have a better source for the same type of information, then you should normally use the better source. (Equally, if you find a worse source for the same type of information, then you should normally replace the worse source with this one.)

What you may not do is reject this source entirely or demand the removal of information supported by it. It may not be the best possible source, but it meets the minimum standard for reliable sources. We do not want the best to become the enemy of the good; perfectionism is not helpful. WhatamIdoing (talk) 20:33, 9 May 2011 (UTC)

I have two questions:

(1) Is Clearfield Publishing Company, Baltimore, a publisher of reliable sources that meet BLP standards? The website [29] claims: "Our genealogy books are authoritative and timeless, found in every library in the country with a genealogy and family history collection, and are widely respected by amateur and professional genealogy researchers alike." It's true that a lot of American libraries have some of their books, but I couldn't verify the part about reputation among "amateur and professional genealogy researchers".

(2) Are publications by this specialty publisher sufficient to establish notability/noteworthiness? The website says: "In addition, we publish hundreds of volumes of genealogy source records, most of which go all the way back to the colonial period of American history. These books save researchers the trouble of traveling to faraway repositories in search of materials on specialized topics such as immigration and royal and noble ancestry, while also supplying information on the more conventional records of birth, marriage, and death, will and probate records, land records, and census records." Their first example is "New England Marriages Prior to 1700", which sounds like a collection of primary source material to me.

The context of my question is

Yes, I think the publisher is credible and their books typically reliable sources for such information, and, no, I don't think that it would be possible to delete that list even if the particular source had never existed, so whether it is "sufficient to establish notability" is unimportant. WhatamIdoing (talk) 23:22, 7 May 2011 (UTC)

The point about notability/noteworthiness is not the existence of the list but its incredible length. Encyclopedias are supposed to summarise. They don't usually copy and update an entire book full of information that is not to be found elsewhere in that form. And this seems to be primarily a publisher of primary information. I am sure an article listing all documented marriages in New England before 1700 would have no chance to survive an AfD. What's different about an article listing all people who are in line for the British throne, including those who are normally never mentioned -- a book by this publisher being the only exception?

As to reliability, the publisher's home page does not look professional at all, and they seem to be primarily in the business of providing all kinds of information and services to Americans who want to find their ancestors. That's why I asked. Hans Adler 11:43, 8 May 2011 (UTC)

It is not the sole source for the page, indeed it seems to have been used to back-fill some references. The rules of succession are quite clear, and the line of succession is also limited (to certain descendants of Sophia Electress of Hanover, IIRC), otherwise the page would have been cut short long ago I would say, leading, as it has, to certain technical problems. The line of succession is certainly notable, the only question is whether the last 2000 places (say) are significant. I would be inclined to say yes, simply because an exhaustive listing is easier than deciding on a cut-off point. Rich Farmbrough, 16:00, 8 May 2011 (UTC).

Hans, it doesn't matter if this publisher produces primary sources. The line of succession would be notable from other sources, even if this publisher had never existed. This publication is not what makes the line of succession a notable subject. We're using this book to fill in details about an already-and-separately-proven-to-be-notable subject, not to prove the subject's notability. That is a perfectly legitimate and appropriate use of a primary source on Wikipedia. WhatamIdoing (talk) 20:42, 9 May 2011 (UTC)

Gezim Dimnaku (born 14 february 1972) in Shkodra Albania , was a albanian athlete who competed in triple jump and long jump ,was a universal athlete :nine time Albanian champion1989-1998 ,bronze medalle junior track and field Europe champions clubs (Athene) Greece 1992, bronze medalle Europe champions clubs (Istambul)1995, three time west Germany champion 2001-2002-2003, (15,70cm) third place meeting germany (vesel) 16.06 cm. hi retired in the winter 2007 .

Thanks for this information. Wikipedia relies on published sources for information. If you'd like to create a Wikipedia article on this person, you'll need to cite published sources. TimidGuy (talk) 11:33, 9 May 2011 (UTC)

Anything that's only from NewsCorp affiliates is best just discarded. For example:

http://www.theaustralian.com.au/business/aviation/elusive-warbird-f-22-raptor-makes-its-australian-debut/story-e6frg95x-1226003947149 The decision, which capped F-22 production at 187 aircraft, has been called into question by the public emergence of China's J-20 and a co-operative program between Russia and China to build a similar aircraft.

Of course there is no joint Russian Chinese fifth generation fighter program (if you discount the mostly Russian tech used in the J-20, starting with the engines) and the first and best clue that the article is bogus is given with the big warning label on top that says "With The Wall Street Journal", which is a sure mark of garbage. Hcobb (talk) 14:45, 8 May 2011 (UTC)

That being said, it's very unlikely we will ban or restrict NewsCorp papers. --Wehwalt (talk) 14:51, 8 May 2011 (UTC)

Especially the Wall Street Journal (which despite Hcobb's assertions, actually has a very high reputation for fact checking and accuracy). Blueboar (talk) 14:58, 8 May 2011 (UTC)

Then explain why they continue to lend their authority to this article many months after it was put out. Hcobb (talk) 16:07, 8 May 2011 (UTC)

Where would you look for an update?

TEDickey (talk

) 16:19, 8 May 2011 (UTC)

How about the same webpage? Hcobb (talk) 18:34, 8 May 2011 (UTC)

Errata are usually published in followup issues. I haven't seen any media which edits back-issues of articles in the manner which you're suggesting. They're not Wikipedia.

TEDickey (talk

) 18:40, 8 May 2011 (UTC)

(that distinction is why Wikipedia is not a reliable source - because its content can change unpredictably, and the same statements cannot be found reliably for verification.

TEDickey (talk

) 18:42, 8 May 2011 (UTC)

Hcobb, your opinion seems to be a minority one in light of the fact that the Wall Street Journal has the largest circulation of any paper in the US (http://en.wikipedia.org/wiki/List_of_newspapers_in_the_United_States_by_circulation). Wikiant (talk) 20:14, 8 May 2011 (UTC)

Your metric is of dubious value, given this newspaper has the largest circulation of any in the UK.

talk

) 23:20, 8 May 2011 (UTC)

What has all this got to do with the Wall Street Journal? That piece was written by a staff reporter of the Australian. It says Business with the Wall Street Journal, not by.

Dmcq (talk

) 20:30, 8 May 2011 (UTC)

In a word, nothing. Rich Farmbrough, 12:32, 9 May 2011 (UTC).

It's really trivial to post vast lists of WSJ errors:

• http://online.wsj.com/article/SB10001424052748703445904576117773666680878.html Bahrami had no known history of drug use or criminality of any kind.

See Zahra Bahrami

• http://online.wsj.com/article/SB20001424052748704662604576256952204415130.html UBS's Cost to Settle Lehman Case: $10.75 Billion
• http://online.wsj.com/article/SB10001424052748704129204575506132565489538.html In addition, this listing erred by identifying Mr. Young as a former heroin user; the Journal has no reporting to support that Mr. Young used the drug.

So there is no need to use a NewsCorp link ever. If they are a sole source then that is because they are wrong. If they are referencing somebody else then they usually misquote. Just say no. Hcobb (talk) 21:48, 8 May 2011 (UTC)

You suffer from observation bias. Give me counts of (a) all the errors the WSJ has published, and (b) all the non-errors it has published, then tell me what is the ratio of (a) to (b). WSJ publishes tens of thousands of statements in a single issue. If they printed ten errors every day, their accuracy rate would still above 99.9%. Wikiant (talk) 01:43, 9 May 2011 (UTC)

I agree; on the whole, they're very reliable. In addition, the question seems similar to one I asked a few days ago, asking if, having established a publication as reliable, that meant the articles it carried automatically became reliable. The answer I got was that the consensus was to consider all the particulars. I think that makes sense, that there's some common sense and a judgment call involved, that we should consider the merits and do it case-by-case. In the case of WSJ, I'd accept them as 100% reliable in reporting GM's profits or the possible cost of Sony's identity theft scandal, for example. But based on their editorial position, I'd accept them as reliable but less so reporting an appeals court ruling on the healthcare mandate. I'd consider them even less reliable explaining the difference between the Republican and Democratic plans for the Medicare. But that's just me, or at least, how I might be expected to behave if the WSJ was offered as a source. Hope that helps. Cheers, Msnicki (talk) 02:03, 9 May 2011 (UTC)

Actually, I would say the WSJ would probably explain the difference in the two plans very accurately and reliably. What I would not use them for is a statement of fact as to which plan was best (they would be fine for a statement as to the WSJ's opinion as to which plan was best). Blueboar (talk) 02:13, 9 May 2011 (UTC)

I understand your distinction and I agree, it's a good one. But (just my opinion) I find that while their editorial bias rarely slips into their reporting of political issues like healthcare in the form of misstatement, it does appear in the form of selection, i.e., which facts to report. Case-by-case, I would need to see the article and how it's being used to give you an opinion of whether I thought it was reliable. But I must also come back to Wikiant's point: On the whole, the WSJ is very reliable. Msnicki (talk) 02:30, 9 May 2011 (UTC)

To say that any information from a certain source is not correct, due to disagreement of the sources alleged political leaning, smacks of violating

If you want others to agree that a source is unreliable, I doubt it would ever be sufficient simply to argue that that's just your personal opinion and that's all there is to it. I think you'd need to give actual reasons. For example, if think editorial bias has crept in as a selection bias in the reporting, you might point to actual evidence, e.g., that many other reliable publications did report the missing facts and argue that the omissions are significant. But again, on the whole, I consider the WSJ very reliable, in the top tier of all periodicals. Does that help? Msnicki (talk) 03:23, 9 May 2011 (UTC)

Btw, if you'd like to see an example of what to me looks like a selection error in a WSJ article, even if not necessarily an example of actual bias, consider their Jan 31, 2011 coverage of the arrest of Raymond Davis in Pakistan. The US claim that Davis is diplomat is repeated several times. What's not reported is the Pakistani claim, extensively reported in Pakistan the previous day, that Davis had not entered the country as a diplomat but on a business visa. The WSJ is very, very good, but even they make mistakes. They're human like we are. Msnicki (talk) 06:16, 9 May 2011 (UTC)

Or it may have been poor journalism under the guise of patriotism, accepting the US claim without question.--Wehwalt (talk) 06:28, 9 May 2011 (UTC)

But that's my point, that they're reporting on a dispute between the US and Pakistan but incredibly, it appears it didn't occur to them to ask the Pakistanis for their side of the story. They selected the facts by selecting who they interviewed. Without trying to pick this apart too far, my point is that even very, very reliable news organizations can make mistakes that result in unreliable articles. We're all human. Msnicki (talk) 06:47, 9 May 2011 (UTC)

Newspaper biases are fairly well known, though. You aren't going to go to the Washington Post for a balanced view of Richard Nixon. You have to allow for that sort of thing or you'll never write anything.--Wehwalt (talk) 06:54, 9 May 2011 (UTC)

I like the money papers because their biases are so straightforward. They're not very interested in your opinions or converting you to their view of the world, they're interested in reporting things accurately so people with money can make informed decisions what to do with it.

Dmcq (talk

) 09:30, 9 May 2011 (UTC)

In article Our Lady of Medjugorje, and elsewhere(?), it is claimed that Pope Benedict XVI publicly blessed a statue of Our Lady of Medjugorje. The source for the claim is a YouTube video[30] made by people (Movimento Gospa)[31] who believe the apparitions are real. See footnote 11 and click on YouTube to see the video. Does the video meet our reliability standards? Kenatipo ^speak! 17:03, 9 May 2011 (UTC)

We usually don't consider YouTube reliable (we have no way to know if the footage shown has been manipulated in some way) - the exception to this is when the video is hosted on a dedicated media sub-channel (such as the BBC's youtube channel). I am sure that the pope did in fact bless the statue... but we need a reliable source to say so. So... we should find another source for the statement. (I would be willing to bet that the Vatican issued an itinerary for the Pope's activities that day... perhaps in the Observatore Romano?)

That said, I have to ask why we mention the blessing in the first place? The statement is part of a section about the Church's "official position" on the apparitions... and in this context, there is an implied connection between the Pope's blessing and the Church's position. Is there a source that connects the blessing to any sort of "official position"? if not this may be Original Research. Blueboar (talk) 18:15, 9 May 2011 (UTC)

I've looked into this a bit, and I agree that YouTube is typically not an appropriate source because the footage can be manipulated. And in fact, I agree with Kenatipo's concerns on the talkpage of the article that this particular video may have been manipulated, with the framing of the photos in the video altered to exclude the other statue seen in the video footage following the stills shots. It's actually worth looking at to see.[32]

Note that Pope Benedict is a living person and per BLP we need to use the highest quality sources; a youtube video, especially one's whose accuracy is questioned, is not one. There is an article in the Croatian Times that mentions the blessing, but it is review of a book written by believers in the visions and appears to me to simply be quoting them. In fact a poster in the comments section also questions the accuracy of the statement about the statue blessing. In my view neither source is reliable for a claim that a living person (and by extension the Catholic Church) has to some extent have accepted the apparitions, which they clearly have not, to date. I will be removing the content per BLP, RS, and OR. --Slp1 (talk) 11:57, 10 May 2011 (UTC)

The Croatian Times was, unfortunately, quoting our article with the word "publicly" removed. (Things done in St Peter's Square by the pope tend to be public). I read the Holy Father's remarks at the audience that day and there is no mention of Medjugorje or the Queen of Peace. His address was about Pope St Gregory the Great (part 2). VISnews also did not report anything about Benedict blessing statues before or after the audience that day. I haven't researched Osservatore Romano yet. I agree that the sentence in our article should be removed as a violation of RS and therefore, also BLP. Kenatipo ^speak! 16:29, 10 May 2011 (UTC)

Actually, I don't think it is correct that the Croatian Times copied Wikipedia, since the newspaper article was published before the information was added to WP. Nevertheless I am not convinced that the Croatian Times has enough of a reputation for independent fact-checking to be the sole source for this information. --Slp1 (talk) 21:25, 10 May 2011 (UTC)

The offending sentence was added to our article by an IP on 28 August 2009, removed, and re-added on 4 October 2009. The Croatian Times article is dated 9 January 2011. (Oh, and thanks for beating up that big bully for me!) Kenatipo ^speak! 06:24, 11 May 2011 (UTC)

We have hundreds, perhaps thousands, of articles solely based on geographic.org [33], an ad-filled database of geographic names, supposedly coming from the National Geospatial-Intelligence Agency. While this is a government institution, the database in itself is highly unreliable (for at least the US, Belgium and Aghanistan, I haven't checked their entries for every country). I have listed some examples at Wikipedia:Articles for deletion/Dara-I-Pech.

The question is whether I am correct to believe that given the number of obvious errors, we shouldn't consider geographic.org as a reliable source.

To add another example: Khvoshi. I can't find a single reliable source for this location which isn't based on the same database. I can find sources for Khoshi, also known as Khushi, which seems to be an administrative center of Khoshi District, and which is located in Bala Deh. I presume Khvoshi is an error, and this is supposed to be Khoshi/Khushi. 13:31, 2 May 2011 (UTC)

Perhaps a more directed question—with a more actionable answer—would be, "is geographic.org sufficiently reliable to be the sole source for the existence and location of populated places?" If I'm not mistaken, the issue arises with respect to stub articles about populated places, where no other references are cited at all (not merely no other place data is cited, but no other reference is cited even with respect to the existence or type of the place).

It would be helpful for editors wishing to formulate opinions to have a brief summary of evidence for the ways in which the source suspect, or reasons to believe the source is reliable. ^Bongo_matic 13:44, 2 May 2011 (UTC)

Allright. The reason I think the source is suspect is because, when checking some entries, I couldn't verify them. This could be explained away for e.g. Afghanistan with a lack of online sources, and transcription problems. However, when I did the same for the US, I noted that it clearly contains glaring errors, like the populated place "A Sherton", which is an error for

Fram (talk

) 14:05, 2 May 2011 (UTC)

Just a quick interjection. Based on the edit summaries here, terms are being confused... "primary" does not = "unreliable". Some primary sources can be extremely reliable, and some secondary sources can be completely unreliable. Blueboar (talk) 16:31, 2 May 2011 (UTC)

No one discussed this distinction, as far as I am aware, and I don't think it is relevant here.

Fram (talk

) 07:26, 3 May 2011 (UTC)

Fram states: the database in itself is highly unreliable yet other than antedotal onesy-twosy examples, there is no criteria or statistical evidence that supports that. Geographic.org claims 8 million + geographic names. What % of those must be in error for it to be highly unreliable? And indeed, what % of the total data is in error? And how are those errors being validated, against what other Reliable source. I can say for sure that GNIS and UKGeonames contain some % of errors. Whether or not we should be sourcing articles from Geographic.org is a good question, but lets establish some realistic and verifiable criteria for highly unreliable before we jump too far. --Mike Cline (talk) 20:03, 2 May 2011 (UTC)

there are, even in very well mapped countries, many geographic names which may or may not represent a distinct place, as compared to an isolated house (frequent on UK Ordnance Survey Maps) or an indistinct district, as I've encountered many times in the US. There are additionally many example all over the world of settlements, some of even with populations of hundreds of thousands, which have exactly the same name, and are located in different provinces or districts, or US states. As an additional difficulty, there are many instances where the same name is used for a settlement, and also the district containing it--New York and New Jersey have particularly confusing systems, so I assume that parts of the world I know less well, such as Afghanistan and Nepal, may be similarly complex. The solution is to give what information there is, and attribute it to what sources there are. No source is completely reliable for all purposes. Typographical and transcription errors occur in all. A reasonable degree of accuracy is all that can be expected in the world. The principle at Wikipedia is that our multiple contributors will fill in the gaps and resolved the difficulties. If we wait until they are resolved, we will have very little content in most subjects. I would not necessarily reject in most subjects a source with even 10% errors, if it was the best available. In my own subject, the very best sources have between 1 and 5% errors. The ones that are compilations from many sources (such as WorldCat) have relatively higher rates of inconsistency, but they also have multiple components to cross-check. DGG ( talk ) 22:34, 2 May 2011 (UTC)

Blueboar, I don't think anyone here has made arguments on either side that touch on primary/secondary/tertiary issues.

Mike Cline / Fram, I agree with Mike that there is no evidence that the database is "highly unreliable"—rather, there are some entries that have been shown to be highly suspect.

DGG, what threshold of error is acceptable for topics where it is possible that no other sources will exist to verify the information? Is this the same threshold as for topics that are likely to be easy to find multiple sources on? Does it matter that here, it is is quite difficult to attribute lack of other sources to (variously) incorrect versus obscure information? ^Bongo_matic 22:48, 2 May 2011 (UTC)

While "highly unreliable" may be too strong, the percentage of problems I have found by spotchecking is too high to be acceptable to me.

Fram (talk

) 07:26, 3 May 2011 (UTC)

Fram, absolutely acceptable (although doing so would be pointy if the sole purpose was to expose errors in the source). Given the existence of multiple reliable sources related to geographic names, geo-locations, elevations, et. al., any error that geographic.org data might contain would be easily rectified by data from another source. Especially considering the online mapping we have today, geo-locations are extraordinarily easy to validate. Geographic.org may not be a perfect (error-free) source for geographic data, but neither is GNIS or PCGN (UK). That said, a great many NY Times articles contain factual and other errors when viewed in the light of history and new evidence, but does that make it an unreliable source? No. --Mike Cline (talk) 13:04, 3 May 2011 (UTC)

The comparison with the NYT is not instructive. NYT is a publication with robust, if imperfect, editorial oversight. geographic.org cannot be said to be subject to editorial oversight at all, so (now I will raise the issue touched on by Blueboar) it's not clear that it even falls within the bounds of a secondary source. ^Bongo_matic 22:53, 3 May 2011 (UTC)

A small random sample

I just did a random sample of various places in different countries where I've spent some time (sorry if that looks like OR; but personal experience is an extra datapoint to add to what atlases say &c).

1. Mkuze: First search result on Geographic.org places it over 1000km away (and in a different country) from the correct location which is attested by multiple atlases &c. Careful reading through search results with the help of a different - reliable - source will get the "real" mkuze but with a variant spelling and a mere 100km from the correct location.
2. Woluwe-Saint-Lambert: This is a district of Brussels. You'd expect a national intelligence agency to have a rudimentary understanding of the city where NATO is based. Geographic.org gives a coordinate in a different district of Brussels.
3. Lytham St Annes: Geographic doesn't know it exists, although it does have reasonable coordinates for "Lytham" and "Saint Annes".
4. Coeur d'Alene: Geographic seems to get this one right.
5. Kornelimünster: Geographic is almost right on this one (compared to Google Maps) - the coordinates are for a field a mile from the town centre. Presumably this is a rounding artefact; if Geographic don't actually have data accurate to a millionth of a degree, the extra digits they present are somewhat misleading.
6. Lacanau: Same as Kornelimünster.
7. Bobo-Dioulasso: Same as Kornelimünster.
8. Dakhla: Same as Kornelimünster (it's a coastal town; Geographic gives coordinates for a location which Google Maps reckons is 10km offshore, and I'm inclined to trust Google Maps in this case)
9. Sabratah
   : Looks reasonably accurate.

So, I would not trust it as a source of geographic information any more than I would trust, say, fallingrain. bobrayner (talk) 14:42, 3 May 2011 (UTC)

Incidentally, I've seen a lot of quirks in transcriptions of placenames from non-English languages &c, as well as the fact that Geographic seems to have a lot of entries for things which aren't actually settlements, per se. Therefore, if Geographic alone says that a settlement exists, I simply wouldn't trust it as a sole source- because there is a significant probability that it's not actually a settlement, or that the name on Geographic is just a variant of some other name which is better evidenced elsewhere (and may already have a wikipedia article &c). Based on my first example, if you took Geograph seriously as a source you could create twenty new articles for places that look a bit like "Mkuze", "Mkhuzi", "Mkuzi-Suid" (a random bit of uninhabited land an hour's drive to the south), &c plus the mysterious "Mkuze Estate" which is - in reality - a random point within a differently-named nature reserve... bobrayner (talk) 15:26, 3 May 2011 (UTC)

Fallingrain mirrored the GNS/GNIS. The rounding "artifacts" are obvious if you simply convert back to deg. min. sec (most that I looked at some years back were accurate but not precise - this is a fairly old database). In terms of using the DB as a source for co-ordinates, it is citable, and in some cases it seems to be the best we have. In terms of using it as the sole source for a stub article, this has been looked at many times, and generally the response has been "well you could, but unless the stub was wanted anyway, it might not be a great idea." Rich Farmbrough, 14:21, 8 May 2011 (UTC).

Poll

Should articles be created using only geosource.orggeographic.org without additional editorial input including searches for other sources?

• No, for reasons articulated above. ^Bongo_matic 22:53, 3 May 2011 (UTC)
• No (assuming you mean geographic.org - or are they related?). This is because we don't have sufficient confidence that any given entry in the database is accurate and unique; too much potential to create duplicate articles for spelling variants, or for other secondary landmarks near the "real" landmark, &c (and an entry in this database is not a very good assurance of notability). bobrayner (talk) 02:05, 4 May 2011 (UTC)
• No, it can obviously be used as a starting point for a personal list of potentially missing articles, but needs checking with other, more reliable sources before an article is created.
  Fram (talk
  ) 09:07, 4 May 2011 (UTC)
  • As another, blatant example of an earlier run of geographic.org articles:
    Fram (talk
    ) 13:09, 4 May 2011 (UTC)
    • That's really quite impressive. We can't rely on such sources for new articles; this is a project to build an encyclopædia, not a race to create as many placeholders as possible, expecting somebody else to come along later and remove all the mistakes. bobrayner (talk) 17:07, 4 May 2011 (UTC)

• No, per
  WP:IINFO, and Wikipedia:Notability There appears to be too many problems for any article to based on this one source and the source seems to used to do an end run around Wikipedia:Notability.--BruceGrubb (talk
  ) 07:12, 6 May 2011 (UTC)
• not a vote, a discussion, but no source not reliable for claims and unable to establish notabilityFifelfoo_m (talk) 07:20, 6 May 2011 (UTC)

Wikipedia:Articles for deletion/Geographic.org

Everyone is invited to participate in

On this page West Ridge Academy a self-published website/blog is being used as a reference (see http://en.wikipedia.org/wiki/West_Ridge_Academy#cite_note-strugglingteens5382-2). I feel like this reference falls into the "self-published" sources category, and should, per Wikipedia policy, only be used to describe itself. I think that the POV of the website is more than obvious, and does not at all contribute to the article positively. Additionally, there is nothing stated in the West Ridge Academy article that needs to be referenced by this single source, and for which a better, primary source is not already available. Your thoughts on this would be greatly appreciated. --EarlySquid (talk) 18:24, 9 May 2011 (UTC)

I believe that the source that EarlySquid is questioning is StrugglingTeens.com. Strugglingteens.com belongs to a company called Woodbury Reports. They say they have been online since 1995; the current domain name was registered in 2001. It's a complex website, including promotion for the business and advertising for other entities. The content that is most often cited on Wikipedia is mostly found in archives of a company newsletter (at http://www.strugglingteens.com/archives/index.html -- the archive spans the period 1998-2011) that includes articles about various special schools and therapeutic programs for teenagers. Since the website names its owners, who are a business with a reputation to uphold, and most articles name their authors, I see the website -- especially the newsletter -- as having the kind of "meaningful editorial oversight" mentioned on

WP:V. When I searched yesterday, I found that strugglingteens.com is currently cited as a source in 35 articles on Wikipedia. In West Ridge Academy it is cited as a third-party source of some pretty basic descriptive information about the school (the newsletter had a "visit report" on the school, which is the specific item cited). In general, it is cited in Wikipedia articles as a third-party source of information about various programs, as well as a republisher of press releases about these programs. I consider it a pretty reliable source. --Orlady (talk

) 20:21, 9 May 2011 (UTC)

I would agree with Orlady... I think that it is a reliable source. Whether it is the best source for the information is another issue entirely. We encourage editors to improve articles, and finding better, even more reliable, sources is one way to do so. So my advice: leave the current source as "good enough for now", but also look to see if you can find a better source for the information... if you find one, replace the current source with the better one where appropriate. Blueboar (talk) 00:38, 10 May 2011 (UTC)

The source is not RS and if RS is not available then we should question if the information is even worthwhile. I would lean towards common sense saying video shows something so we should try to find RS but a quick Google News Archive search came up with nothing, and someone else already expressed concerns about possible manipulation. It all leads back to

WP:VIDEOLINK Cptnono (talk

) 06:53, 11 May 2011 (UTC)

Four books on computing and the Internet use http://irc-junkie.org, an IRC-related blog, by either paraphrasing, citing, listing in a web-bibliography, or explicitly recommending it to readers. In my opinion, per

Per discussion of reliable sources at Talk:Comparison of Internet Relay Chat clients#Article blanking and "sources", I've boldly created WP:WikiProject IRC/Sources listing the sources mentioned, as a rationale for treating http://irc-junkie.org as a reliable source.

Is there any policy or guideline against inheritance of reliability?--Lexein (talk) 01:42, 10 May 2011 (UTC)
Can irc-junkie.org now be considered reliable per above? --Lexein (talk) 01:15, 11 May 2011 (UTC)

• irc-junkie.org has an editorial policy, and an editor. However, its chief author is also its editor. You would need to clearly establish the expert status of Christian "phrozen77" Lederer of 92237 Sulzbach-Rosenberg. You ought to note that the publication's title is IRC-Junkie.org: your daily dose of IRC related news and that its place of publication is Sulzbach-Rosenberg, Germany. For example, to cite the current top item: phrozen77 (pseud.) [Christian Lederer] (20101117) "A Day on QuakeNet" IRC-Junkie.org: your daily dose of IRC related news (Sulzbach-Rosenberg, Germany), unpaginated. Is phrozen77 an expert? Shelly and others (2007) says no, "You can learn more about IRC clients, networks, and channels at sites such as IRC-Junkie.org" is not an expert recommendation. Neither are the other three sources. Scholars regularly cite non-expert material as evidence, scholars have the inherent capacity to transform through analysis unreliable sources into elements of their own reliable texts. You need to find instances of citation which consider IRC-Junkie.org to be equivalent to scholarly or expert works—mere citation as factual evidence is insufficient. Fifelfoo (talk) 02:35, 11 May 2011 (UTC)

• Thanks for the analysis. I see the point about full author name, and full publication name. I agree that Lederer doesn't seem to be cited as an expert by name. I would suggest, however, that the statement in the Shelly et. al. textbook actually is a straight-ahead recommendation of the site, neither weak nor strong, implying that expertise will be found at irc-junkies.org, and it is listed first. Kshetri selected the "Help! I'm being DoS'ed" (from archive) irc-junkies.org article to quote and paraphrase, out of all the available (legion) sources on Denial of Service attacks; to me, this also seems to acknowledge a level of expertise at the site, rather than non-expertise. Given the rationale and our discussion here, would you go so far as to delete content in a list which cites irc-junkies.org and a primary source? --Lexein (talk) 06:22, 11 May 2011 (UTC)
  • Is the textbook (Shelly et. al.) aimed at post-graduate coursework or research students or post Masters certified (or equivalent) professionals? Seek this in the introduction, preface, and bibliographic page. This does matter. Fifelfoo (talk) 06:54, 11 May 2011 (UTC)
    • It's a college textbook searchable under the publisher's Higher Education catalog - there's even a "College Bookstore Wholesale Price = $68.25" listed. It has supplementals for instructors. So it's definitely not targeted for mass-market. Seems to fall under
      WP:SOURCES. What guideline were you alluding to? --Lexein (talk
      ) 08:21, 11 May 2011 (UTC)
      • I can't currently recall. In previous debates on RS/N textbooks aimed at a post-graduate audience have found appreciation as scholarly sources—they are aimed at professional in practice and trainee scholars. This also follows Australia's HERDC collection advice on what constitutes a scholarly book as opposed to a non-scholarly book (undergraduate and schools textbooks are excluded). Being cited in an undergraduate textbook as a worthy resource isn't an indication of expert status to my mind; it is not being cited as a scholar in the literature of scholars. Fifelfoo (talk) 11:21, 11 May 2011 (UTC)
        • Well, at least Dr. Kshetri's collection of essays was scholarly, quoting irc-junkie.org. I'm not convinced the website is unreliable, and feel that it's reliable enough for very careful use in a limited way. Note that its use is limited: it's only needed for one claim, in one list/table; we don't need it for any of the other IRC clients in the Comparison. I appreciate your input, and will certainly apply it to marginal sources in future. --Lexein (talk) 14:58, 11 May 2011 (UTC)
          • To be honest, this source doesn't meet Wikipedia's reliable sourcing policy. However, I believe this source is reliable due to expertise. The rules are standing in the way of the encyclopaedic project, and therefore, following Ignore All Rules I support this source being used very carefully as a reliable source for limited claims about irc. Where available, other sources should be used. Hopefully, eventually, irc-junkie.org will be cited in other scholarly works, further clarifying the expert element. Fifelfoo (talk) 15:04, 11 May 2011 (UTC)
            • Oh, well that's alright then. I feel the same way. Haven't used
              WP:IAR yet, but maybe this is the time. --Lexein (talk
              ) 15:45, 11 May 2011 (UTC)

WP:BLP is pretty clear on this: "Never use self-published sources—including but not limited to books, zines, websites, blogs, and tweets—as sources of material about a living person, unless written or published by the subject." These are clearly self-published websites. TimidGuy (talk

) 10:51, 11 May 2011 (UTC)

That's a good point. Thanks for that. SilkTork *^{Tea time} 00:55, 12 May 2011 (UTC)

Reviews and commentary about films at Film Threat have been frequently cited in deletion discussions as reliable sources for articles about those films, with the result that some films relying on them as a reliable source have been deleted while others have been retained. (For examples, see the current discussion at