Wikipedia talk:Reliable sources

From Wikipedia, the free encyclopedia

Warning To discuss reliability of specific sources, please go to
Wikipedia:Reliable sources/Noticeboard.

This is the talk page for discussing improvements to the Wikipedia:Reliable sources page.


Archives


Contents

[edit] WP:MOS subguideline, anyone?

Imo, this guideline could be far more usefully handled and improved if it where a dedicated WP:MOS subguideline. The applicable policy (WP:V) appropriately handles the required minimum threshold, while this page could explain various scenarious in greater detail than a policy, including the ideal case, or how to proceed in the many suboptimal cases where high quality sources are not easily available etcpp. User:Dorftrottel 14:52, February 15, 2008

[edit] "Emails not sources..." - What about from official sources?

Hello all,

I read that emails are not reliable sources. So, what if I have an email from the National Tramway Museum that is a response to my question about Leeds 602, and the question is is why is this tram off the road?

Would I not be allowed to quote and use this as a source?

The same goes for official forums - we asked the same question, with the answer that it's difficult to drive - could I not source this even though it is from a Board Member?

Please let me know!

Thanks,

BG7 10:52, 6 April 2008 (UTC)

If it's an official forum and the post is from a named board member, it should be okay to cite it. The problem with e-mail is that other readers can't check it for themselves, and the spirit of our sourcing policies is that every reader should be able to see for themselves where our material comes from. SlimVirgin talk|edits 17:28, 6 April 2008 (UTC)
Ok... but I suppose that the forum wouldn't actually be allowed either then, as it's private requiring a TMS membership number.
Is there a way perhaps of me posting the sources on here, say in my userspace? Or perhaps on my website? (I promise they are all true and reliable, and NOT made up - users can feel free to check!)
Thanks,
BG7 17:45, 6 April 2008 (UTC)
Unfortunately, No... your user space or website would not be considered a reliable source. And it would not change the fact that the underlying email is unverifiable. Also, using such an email would probably be a WP:No original research violation. sorry. Blueboar (talk) 15:12, 7 April 2008 (UTC)
Even though the info isn't original research? How about if it's in a book? (which it is... but this book is available through ~1 outlet... the museums...)
Thanks,
BG7 17:06, 7 April 2008 (UTC)
As I understand it, that would be acceptable if the book is one of general publication (not a self-published work). Cite the book as a supporting source (See {{Cite book}}, {{Citation}}, etc.), giving title, author, year of publication, publisher, etc. specify the page number(s) where the relevant info appears. If the book is of limited availability, a parenthesized comment following the citation giving info about its availability (e.g., "(Limited availability — available through the info desk of the Field Museum in Chicago)") would be useful. -- Boracay Bill (talk)
But there's no need to use a citation template. :-) SlimVirgin talk|edits 00:16, 8 April 2008 (UTC)

reset!
Ok, thanks all,
BG7 17:33, 8 April 2008 (UTC)

Quote emails from official sources on the talk page. (Make sure the person is ok with posting his email). "No Original Research" is mostly intended to prevent crackpots from posting that the earth is flat or what have you. If your information is verifiable and reliable, you're ok. --Kim Bruning (talk) 17:47, 8 April 2008 (UTC)
Ok, so would something such as "ran up to 1987 before being withdrawn[1]", with the [1] linking to the refs section, which links to the talk page, such as:
==Email Ref==
"quote" -- From xx at the National Tramway Museum email xxx to verify?
would suffice?
Thanks,
BG7 18:23, 8 April 2008 (UTC)
No.
As follows:
  1. contact official source, ask question.
  2. at same time, ask permission to publish their reply.
  3. when reply received, post e-mail to talk page.
  4. Now you have a valid reference on the talk page.
--Kim Bruning (talk) 18:59, 8 April 2008 (UTC)
Ah ok, that makes sense!
Thanks,
BG7 19:07, 8 April 2008 (UTC)
But the so-called "valid" reference on the talk page may disappear into an archive, and the assertion in the article which it is intended to support may still be removed as unsupported. -- Boracay Bill (talk) 23:59, 8 April 2008 (UTC)
Hmm... it's a tough one isn't it? I doubt the talk page will be archived - it's not that an active page, with me the only content contributor, the rest being copy-editing etc.
I will put it on and then we can see what happens!
THanks,
BG7 11:08, 9 April 2008 (UTC)

Hold on... This advice is wrong. While we can certainly discuss and even copy these emails on the talk page, we can only do so as an aid to our background discussions about the article. We can not use something posted to a talk page as a reference... Wikipedia is not a reliable source, and Wikipedia talk pages are even less reliable than our articles (things like WP:V and WP:NOR do not apply to talk pages). Furthermore... the problem with the verifiability of the underlying emails would remain. Our readers have no way to verify that the emails are in fact "Official". I'm sorry that our policies limit your ability to add factual information, but the policies are clear. You really have only two options... 1) you can contact the official source and ask them to issue a public statement about this information... by issuing a press release, puting the information on their web page, or publishing it in some other publically accessible forum; or 2) assuming that the consensus of all the editors working on the article is that the information is both accurate and important to the article, you can invoke WP:Ignore all rules, and add the information to the article without citation. But one thing you can not do is put the emails on the talk page and cite to that. Blueboar (talk) 13:09, 9 April 2008 (UTC)

Bullshit! In the stated situation, we just got an email from an official source (we can cross-confirm that an email comes from an official source). Now that the E-mail is posted on the talk page, the talk page just became a reliable source.
Something is a reliable source based on -well- how reliable it is, and where you got it from, not what it's written on. If the President of the United States writes something down on a napkin and signs it, it doesn't matter that it's on a ratty paper napkin... it's from the official source (ok, so presidents of the united states are not as reliable as they used to be, but you get the idea). It doesn't get any more reliable than that. --Kim Bruning (talk) 16:07, 9 April 2008 (UTC)
Kim, it isn't Bullshit at all. This goes to the heart of WP:Verifiability. How are readers to verify that the napkin was actually signed by the President and is not a forgery? We can not take the word of a fellow editor, we need a reliable source that confirms it. In the situation that BG is asking about, we have an email... an editor says the email is from an "official source"... but how am I or any other reader supposed to verify it? Email can be faked. What is needed is a reliable secondary source that tells us that the email is valid, or discusses the information on the email.
I strongly object your contention that posting something on a talk page can make it reliable. Our policies are quite clear on this... Wikipedia is NOT a reliable source. We can not cite other wikipedia articles, and we certainly can not cite to a Wikipedia talk page (where the rules on Original Research, Verifiability, etc. are more relaxed).
Furthermore, since what we are talking about is a personal correspondence to a wikipedia editor, discussing the emails would be a WP:NOR violation. Blueboar (talk) 16:59, 9 April 2008 (UTC)
Guys, it's alrite! I will cite the book instead - although it's not available online, or widely, i presume it's ok, as it's the official stock book?
Some more questions however:
  1. Does information taken from the board on the workshop wall count as original research (again, as there's nothing to back it up?)
  2. Does information from a website that practically is orig. res. count?
Thanks,
BG7 19:38, 9 April 2008 (UTC)
Blueboar: Dude: Most newspaper articles and books don't even have a fine signature! Your interpretation would really cut down on the number of verifiable sources we can use :-P (down to approximately...oh... near zero). I am totally going to trust a signed source over an unsigned source. If you start on about forged signatures, I figure you're grasping at rather thin straws by then.
Posting a thing on a talk page does not magically make the thing or the talk page reliable. Having a thing in some book or newspaper article doesn't necessarily make it reliable either (eg. "The head kook of the western kook society says...") The thing must be reliable in-and-of itself. But once established that it's reliable in-and-of-itself, posting it to a talk page doesn't magically make it unreliable either. Alright? Reliable is reliable. --Kim Bruning (talk) 23:35, 9 April 2008 (UTC) If only 100% unforgeable signed documents were considered reliable, we'd only have one or two wikipedia articles left (namely those about people smart enough to use cryptographic signatures :-P.. if even that... what happens if NSA is already fielding quantum computers {which could theoretically forge even those signatures} and just hasn't told the rest of us yet? No my friend... down that road lies madness. Not Sparta, just madness.)
Kim, duuuude, the important idea is that we be reasonably sure the source was published by an entity with a reputation for accuracy and fact-checking, or otherwise such that we can trust the content within. Very simply, a private email doesn't fit into that, as no one can reasonably ascertain that it's authentic. It's different with a book or newspaper, as anyone can go and find the damn thing with enough effort. What are we supposed to do with Goblin's email? Hack his email account so we can see it for ourselves, so we can be sure he didn't make it up? While true, we can never be perfectly sure a source is not faked, if the sole possible assurance that a source is authentic is the word of a single editor, then that source is not reliable. Someguy1221 (talk) 00:08, 10 April 2008 (UTC)
Yo dude ;-) True. But it differs a bit on who you are e-mailing. What if you e-mail a librarian to confirm that a source says something? Anyone else can mail that librarian (or possibly a different librarian, if it's not one of those ancient unique tomes) and get the same answer. If you can e-mail a librarian, you can also e-mail a newspaper to request (for instance) copies of particular articles. Anyone who e-mails that newspaper will get the same article. In the same way, you can e-mail the curator of a museum to confirm a certain item exists in their collection, or certain particulars about that item. Anyone mailing the museum will get the same answer. That's one of the tasks of the museum, isn't it? Just by inquiring after the contents of a library/newspaper archive/museum using e-mail doesn't change the verifiability of their contents. Anyone else can also e-mail/snail mail/visit that same information-storehouse and obtain the same answer. The method used for obtaining information from an information store does not alter the reliability of the information store itself, and the information should be considered reliable.
--Kim Bruning (talk) 00:27, 10 April 2008 (UTC)
That's all fine because the ultimate source of your information is actually the book, and not the librarian/curator/whatever. Just the same as if that curator had registered a Wikipedia account and added the information himself, citing the book, his nature of having actually read the book will not be inherently suspect. In that case, you can certainly point out in the citation (with a link to a copy on the talk page, sub page, archive, whathaveyou) that the information was retrieved indirectly via email correspondence with someone who has access to the book (WP:SAYWHEREYOUGOTIT). But the citation is still to the book! I was speaking on the more general matter of content that isn't actually accessible except via communication with a particular individual. Until it's published in a more permanent manner, it's not reliable. So in conclusion, I think we're arguing over different things :-) Someguy1221 (talk) 00:37, 10 April 2008 (UTC)
Well, BG7 was mailing the curator of a museum I gather, so I guess he(/she/it/other) should be able to get the reliable reference he needs out of it. Perhaps he could formulate his question to the curator a bit more sharply? --Kim Bruning (talk) 11:51, 10 April 2008 (UTC)
The key is to ask the curator/librarian to provide us with reliable sources for the information. Ideally the curator would provide us with full citation information. We could then cite to these sources. The point is, the email correspondence is not cited as the source... it is the means by which we obtain confirmation and verification of the source. Blueboar (talk) 13:21, 10 April 2008 (UTC)
Would the content of the museum itself be a reliable source, according to you? --Kim Bruning (talk) 13:28, 10 April 2008 (UTC)
You mean physical objects?... I would say "Yes... but". Such objects are reliable for basic statements of fact about the object itself ...but... the object would essentially be a primary source, and so all the cautions expressed at WP:PSTS would apply. Any statements about the object would have to be bluntly factual as to the specific object itself. For any interpretation or analysis of the object, or about similar objects, we would need to cite a reliable secondary source. Blueboar (talk) 14:38, 10 April 2008 (UTC)
I'm just going to throw in here, as someone who has used e-mails (and phone calls for verification, not as a source) to get at basic facts, that the guideline is missing a very important qualification to verification: verification does not have to be easy. I used emails to a government to verify almanac style information that differed from almost every major published source (including Brittanica, Encarta, and CIA Fact Book). In subsequent years, EB and Encarta corrected their information, but Wikipedia did it first, and that is our strength. The research wasn't original to me, it came from the government we were writing about - the horses mouth as it were. Anyone could have used the email addresses published by this government to ask these questions, or to ask again to verify. I simply took the effort to do it when we were repeating old, incorrect information.
If a college has logbooks of old sports records, that is a verifiable, reliable source. Source based research is not original research. SchmuckyTheCat (talk)
Yes, but see for example Romblon, Romblon, Talk:Romblon, Romblon, and User talk:Taga guinpucan#Sourcing and copyright status of your additions. (I'm presently trying to improve this but it looks like it'll be a slow process and may not be doable). -- Boracay Bill (talk) 02:29, 14 May 2008 (UTC)

[edit] Foreign Language

Would a source written in something other than English be considered "reliable"? —Preceding unsigned comment added by 71.112.226.201 (talk) 08:05, 11 May 2008 (UTC)

See WP:V#Non-English sources --Francis Schonken (talk) 08:16, 11 May 2008 (UTC)

0

[edit] Blogs as a reliable source.

First, could someone point me to the rule that explicitly states that a blogger is not a reliable source?

Second, if this is really true, I consider this an incredibly short sighted notion in light of the evolution of new media and the slow obsolescences of what used to be called traditional media. More and more journalists and reports now publish blogs these days, in fact quite a few use the blog to augment/supplement their writing, especially in many of the technology based communities. This is even more the case in virtual world environments where the publications are virtual magazines produced in their environments, but dual published to blogs like a newsletter.

I would really like to force a conversation on this subject as I've been seeing this "that source is a blogger and is there by invalid" being bandied about by overzealous editors.BcRIPster (talk) 18:31, 14 May 2008 (UTC)

Certainly. You're looking for WP:SPS. Jakew (talk) 18:36, 14 May 2008 (UTC)

Ok, I just read that again and it says "blogs ... and similar sources are largely not acceptable." with this added annotation that goes on to say that blogs with an editorial review are ok. Ok, this is the most facetious qualifier when in the current media environment. Mainstream media is just as suspect of running invalid data these days with their careless parroting of talking points. I just really can't see that this is a valid distinction any more.BcRIPster (talk) 18:54, 14 May 2008 (UTC)

There is nothing inherent about the blog format that precludes it from being a reliable source. Most blogs aren't, though. -Chunky Rice (talk) 19:04, 14 May 2008 (UTC)

(edit conflict) Well, in order to produce a quality encyclopaedia, we need to insist on good sources: high quality sources with a reputation for getting the facts right. If, as you say, mainstream media are increasingly "suspect of running invalid data" (and I'm inclined to agree with you), then the answer is to reduce our reliance on mainstream media. Jakew (talk) 19:10, 14 May 2008 (UTC)
The problem is that anyone can go and start a blog and put anything they want in it. For this reason, Blogs will always be presumed unreliable, unless there is reason to believe otherwise. The existence of editorial review is a fairly objective criterion for this purpose. Someguy1221 (talk) 19:45, 14 May 2008 (UTC)
Granted but people are using this document and WP:SPS comments about blogs as a blunt weapon to discredit any source that happens to also be a blog without any other justification. This concerns me as the line between who is a journalist blurs more and more.BcRIPster (talk) 01:45, 15 May 2008 (UTC)
The problem isn't the blurring of who constitutes a "journalist", but rather, on who does the fact checking. While a blogger could produce a quality piece equal to that of the New York Times, the latter has a longstanding reputation for being highly accurate in its coverage. The blogger simply doesn't. And lacking that established reputation for factual accuracy, using such sources casts doubt on the accuracy of Wikipedia. The only blogs I would consider reliable for Wikipedia's purposes would be those directly associated with a reliable publication, since the blogger would presumably be held to the same standards as a traditional journalist with that publication. Resolute 03:02, 15 May 2008 (UTC)
That might have been an argument in the past, but is proven not to be so true anymore. Look at all of the newspapers in the last few years who have fired dozens of reports for blatant plagiarism or printing false information that was supposedly "fact checked". Or how about on TV with Fox, CNN and other news networks. There is a site Media Matters that does nothing but documents just the false and or poorly reference science and political based falsehoods which rank in the legion. So would it be valid to cite a news report from the supposedly "fact checked" fox news over a blogger who has proven themselves a reliable resource on the a given subject? In my experience you have to take all sources with a grain of salt at this time in history, and weigh the average between the differences to understand the truth on most topics. The crux of my issue is don't tell me that bloggers are less credible than other sources when those other sources are proven time and time again to have poor credibility short of someone gives them a paycheck.BcRIPster (talk) 03:50, 15 May 2008 (UTC)
Except that, as you noted, reporters caught plagarizing or deliberately passing false information are fired. A blogger who does the same continues blogging. It's that accountability that leaves the traditional media as generally more reliable, even today. Resolute 14:02, 15 May 2008 (UTC)
The firings were only the stories we head about... and a blogger who does the same looses traffic and reputation. Maybe we need to have a weighting systems for considering bloggers based on something like their Alexa/Google ranking, plus... some other value?BcRIPster (talk) 14:12, 15 May 2008 (UTC)
There's no reason to believe that a popular blog is a reliable blog. Indeed, a blogger may be popular because he/she frequently makes controversial claims.
Another aspect to consider is that the mainstream media is generally higher-profile than blogs, and media publications comment on each other (and publish retractions when they err). So if an erroneous claim is made by one media publication, it's usually fairly easy to find other publications that highlight the error. Indeed, if the error is serious, it may itself become newsworthy. This isn't necessarily true of a blog. Jakew (talk) 14:19, 15 May 2008 (UTC)
That may apply to a limited degree to the print media to the best of our ability to guess, but that self correction is clearly not happening in TV/Video based media. It's only "fairly easy" if you know to fact check what you hear on the news. Hmmm.... I also think that if something is notable in a realm of the blogsphere but hasn't been acknowledge by the mainstream media, that is shouldn't be discounted just because of that. As I said before there needs to be someway to weigh bloggers content value. Maybe this is a call for someone to setup up a sentry service to rank blogger reliability. Hmmmm again....BcRIPster (talk) 20:29, 15 May 2008 (UTC)
The "no blogs" rule does have some exceptions... certain specific blogs are considered reliable based upon the reputation of their authors, who are notable experts in a given field. Other specific blogs are considered reliable based upon have achieved a notable status for their journalism (ie won major journalism awards etc.)... Most blogs, however, do not meet these requirements - having neither won a journalism award, nor authored by a recongnized expert. These should not be considered reliable. In other words... at the moment, a specific blog might be acceptable... but we have to judge that on a case by case, blog by blog (and even statement by statement) basis. Blueboar (talk) 21:16, 15 May 2008 (UTC)
And that IMHO is totally reasonable. My issue is with people brandishing the Reliable page as a blanket excuse to say "all blogs are unacceptable" because of this page. So I'm trying to figure out a reasonable way to counter that and discover some type of methodology that allows for grey space here. I believe this grey space is clearly acknowledged and accepted by some people and that is very promising in my mind, but how does one translate that into something that can be presented so that the people who want to view the world as black&white can be buffered. I'm still stewing this idea of setting up a reliability/ranking site. I just don't know if I have the time to manage it :P BcRIPster (talk) 01:33, 16 May 2008 (UTC)
As a hypothetical question, suppose that you were editing blog and wanted to add a neutral, verifiable and OR-free section on the reliability of blogs, making use of sources with no question as to their reliability. What sources might you cite? Jakew (talk) 23:18, 16 May 2008 (UTC)

Oooo! Ok, case in point. I'm working on an article that was marked AfD to see if it can be saved, and the article is about a virtual world based subject. One of the voters for deletion only said this "With the notable exception of Reuters, "virtual world journalists" = bloggers" !!! Argh! There are dozens of editorial based/commercial publications in Second Life that publish their content externally as a blog, yet this person is throwing out this blanket statement and their vote is going to be considered based on it. This is the kind of stuff that is what brought me to this discussion to begin with. Should I approach the voter through their talk page and try to educate them, or what?BcRIPster (talk) 19:15, 16 May 2008 (UTC)

[edit] Heads up

Massive drama over at WP:RS/N#Boxofficeindia.com. Input needed.

On a related point, where's the discussion where we decided to forgo any mention of one criterion of a sources's reliability for us being extensively quoted elsewhere? --Relata refero (disp.) 19:06, 14 May 2008 (UTC)

[edit] The related point - quotation/citation as a criterion for reliability

I wrote the following post for the BOI drama, which Relata resolved. But BOI brought up this important related point Relata refers to. I think this criterion, which many editors support, should be explicitly reincorporated or clarified on this page:

  • Reply to Mperel's and Sarvagnya's important questions

Sarvagnya asks "which policy or guideline?" says that "The fact that this is used [by reliable sources] is a direct evidence of its reliability." Mperel asks "is a source made reliable because other reliable sources regularly cite it"?

As Relata refero points out, this idea has a long history here at Wikipedia. The phrase "Find out what other people say about your sources" and the suggestion "to cross-check with an independent source" appear in Beland's original version of the page and Radiant!'s first guideline version

But the idea that "cit[ation] by other reliable sources ... is indeed the best possible sign (and our original standard!) that it is reliable" (Relata r.) is still there, hiding in plain sight, implicit in our basic guideline, our lex generalis:

Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy.

The rest of the WP:RS guideline and parts of WP:V is lex specialis explicating this sentence, but as WP:RS wisely states, not exhaustively, as some editors seem to argue. The dictionary meaning of "reliable" - able to be relied on (by whom?) is clearly relevant. (And the source at hand, BOI seems prima facie reliable and making unexceptional claims - Would we be having this discussion if it contradicted known reliable sources?) That accepted source A in fact does rely on source B is evidence of nonzero weight that B is reliable in the ordinary sense, and I and many others hold, according to Wikipedia rules. Look at "with a reputation for fact-checking and accuracy". How do we evaluate reputation? Our practice is not to only use sources for which we could find another source explicitly saying this is a great source with great fact-checking and accuracy - and how do we evaluate "the another source" then?

No, the action of citing and using a source speaks louder (and more frequently) and directly about "reputation" than the rarer, but implied, words "this source is good (accurate, fact-checking), we've checked it out." And citations are easy to mechanically google and find, objective and practical.

As Sarvagnya says "a source doesn't become reliable by 'proclamation' or from editors' assurances or 'judgement' that they are reliable." But he is wrong when the he states that "this exercise" of determining a source's reputation "starts by telling us precisely who runs a site, a newspaper, a journal etc." That is simply not what the key word "reputation" means. Reputation doesn't mean "biographical" "credentials" information about a person or institution, which is a useful but indirect means of establishing reputation. Reputation means what other people or institutions think about him, her or it. Were Michael Scheuer's books unreliable sources when he was still anonymous?

Indeed, reputation of a journal, person or text is often based on mechanical computation of citation indices based on the raw fact of citation alone, and not only here at Wikipedia, where it is often used as a criterion of notability. We should not be reinventing the wheel when we think about source reliability. Looking at "who cites it" is second nature and best practice in source evaluation in many fields, in academia, journalism, law (where there's even a word for it - Shepardizing), to many editors and in the real world in general.

This idea is entirely in line with how other core policies and practices work at wikipedia like NOR - we should refrain as far as possible from substituting our own judgment for that of published sources. Avoiding it would mean we put our "objective" and changing wiki-criteria above published reliable sources' opinions. That isn't the wikipedia way. Another parallel example is notability. Recently at AfD, two of the most acute and respected editors (who argue here pro -BOI) argued to delete, or only weakly to keep Sara Roy - she's only a Research Associate at Harvard (biographical, credentials, objective info) - why have an article on her? But it turned out that she is citably well known as the world's #1 expert in her field. "Reputation" data does and should trump "objective" data which can be misleading, and citation clearly can speak for reputation.

However, it might be a good idea to make this more explicit and less reliant on talmudic/jesuitical disputation.

So my proposal is a line in WP:RS like

That an accepted, high-quality RS uses a source is evidence for the source's reliability for use in the same manner as the accepted RS The "in the same manner" and "high-quality" parts are there to satisfy Nishkid64's and Sarvagnya's "RS-hood by association" concerns.

And maybe we should also have a sentence somewhere, or an agreement, to the effect that we are not trying to reinvent the wheel, but just to adapt the best and standard academic, journalistic, legal, real world practices in source evaluation, which certainly includes this idea, to the wonderful world of Wikipedia.John Z (talk) 06:06, 18 May 2008 (UTC)

I've reverted this. It is clearly not the case in a wide range of situations, and perhaps not the case at all. We try to look for reliable sources. Most primary sources - newspapers, scholarly articles, etc. - consider the entire universe of knowledge, experience, and evidence in the purview. This comes up often, for example in articles about controversial political subjects, when people try to include inappropriate political attack material because it is quoted in a major newspaper. Wikidemo (talk) 22:56, 24 May 2008 (UTC)
I'm not sure you've read this in its entirety. Please phrase your objection more clearly. Is it that you feel that the wording that John restored does not adequately distinguish between sources widely quoted because they are extreme and those that are widely quoted because they are reliable? --Relata refero (disp.) 23:06, 24 May 2008 (UTC)
I think the second sentence in my proposal covers Wikidemo's objections, as far as I understand them. I am completely happy with rephrasing and expansion, but the basic idea is very standard in any relevant field I know of. Of course if all good sources A-Z say source xxx is loony tunes, that is the attitude we should take to it at Wikipedia.John Z (talk) 23:38, 24 May 2008 (UTC)
How about this - "How accepted, high-quality reliable sources use a given source provides evidence for its reliability and reputation. For example, citation without comment for facts is evidence of a source's reputation and reliability for similar facts."John Z (talk) 02:41, 25 May 2008 (UTC)
The fact that a non-scholarly source, like a memoir for example, is cited by one scholar (or newspaper, for that matter) does not mean that editors can or should mine the same source for all sorts of other assertions that are not cited by anyone. People may differ in their assessment of sources. For example, if 95% of scholars are of the opinion that the reliability of a particular memoir is too questionable for them to cite it, but one or two scholars have quoted a couple of paragraphs from it, it wouldn't be right for Wikipedia editors to go and mine that memoir for all sorts of assertions that are quoted by no one in the field, and to treat it as a source that has the same standing as the various scholarly accounts. We should reflect the established consensus. Anything else would be a violation of WP:UNDUE. The wording proposed above still leaves a large loophole. Jayen466 10:28, 25 May 2008 (UTC)
I think that a negative "for example" would be very useful, gave up on wording one to go to sleep, though I think the new first sentence carefully understood rules out the 95% vs. 5% undue weight loophole. As Relata emphasizes, we're trying to restore something essential to this guideline that somehow got removed, perhaps because it is so obvious and standard it goes without saying, cf the remarks of many editors at RS/N on the BOI case, and its outcome. Without some such understanding, we are emphatically NOT reflecting the established scholarly or real-world consensus, not even trying to reflect it, but instead coming up with somewhat arbitrary criteria for sources. There's also the problem of politically motivated scholarly controversy - the A-ite scholars might think source X is dandy and quotable, while the B-ite's might say it is crap without pointing out any real flaws, simply because it is from the A-ite POV. In such a case I think use with attribution, not uncommented use for facts could be appropriate. It is very tough, IMHO futile, to formulate rules for all such possibilities, I think a "positive" and a "negative" example should be enough. My strong preference is for brevity.

Howzabout:

"How accepted, high-quality reliable sources use a given source provides evidence, positive or negative, for its reliability and reputation. The more widespread and consistent this use is, the stronger the evidence. For example, widespread citation without comment for facts is evidence of a source's reputation and reliability for similar facts, while widespread doubts about reliability weigh against it. The goal is to reflect established views of sources as far as we can determine them."

Cheers and thanks for the very welcome input,John Z (talk) 23:49, 25 May 2008 (UTC)

I think that's a lot better. It still leaves a potential concern – what if the source says something really contentious on page 235 that no one has quoted. If we include it in the article, we are in effect assuming that it is just coincidence that no one else has quoted it. But we don't know that; it may have been choice, because the reliable sources felt "I am not going to touch that with a barge pole, it's iffy". (I am thinking of BLP issues and the like; say, a biography of rock star X written by a groupie mentions in passing on some page that something embarrassing happened to X's daughter, also a notable individual today [daft example, but I hope you get my drift].) Perhaps we could add a statement to the effect that this type of sourcing isn't good enough for contentious issues? Btw, I didn't follow the BOI controversy, so am just reading this "as is". Cheers, Jayen466 00:10, 26 May 2008 (UTC)
Good point. Though what we want is "the stuff nobody quoted" , otherwise just use the quote. Here is the background: BoxOfficeIndia is a site, perhaps the site for information about Indian cinema. Problem is, it's a little hard to judge by many criteria - Who runs it? - we had a name, but from a blog or something similar; What kind or organization? How do they operate, fact check? etc. So there was a lot of argument involving lots of issues, self-published sources etc. There had been some good cites of its usage for facts by major accepted reliable sources like the Times of London, but perhaps not enough, so argument continued, with most who expressed themselves on the issue of "citing showing reliability" being for the concept. Sarvagnya was the major opponent and he posed intelligent and productive questions, see above. Relata was the party pooper. He swooped down and carpet bombed everyone with citations from the Pope, Mickey Mouse, Albert Einstein and other mythical culture heroes citing or writing in the Journal of Ultimate Truth that BOI was holy writ when it came to Indian films, please pass the curry popcorn, and the debate immediately ended. The information from BOI was stuff like sales figures, not anything that would cause serious contention. Relata and others, I think, stated that they wouldn't use it for contentious BLP material.
I found it hard to come up with good additional rules that weren't overrestrictive besides saying, see BLP, or Usage of any source must comply with other policies, like BLP. What if the bio was widely used for even more seriously negative information about the rock star or his daughter? The nearer what we want to use from the source is to the quoted/cited/vetted material, the more reliable it is - I think the words "for similar facts" can cover that. I think the BLP issues really come up when the source people/organization is not easy to point to, when there is no one in sight to be left holding the bag for us. But then the part in BLP about not using self-published sources covers it. Basically I want the first sentence in there somehow. Part of the difficulty is that this criterion can be, and actually is used in conjunction with other criteria to 'improve' the questionable repute/reliability/mainstreamness of hard-to-evaluate sources like a paper published in an obscure, in - house scholarly journal, or in a journal that has become less "reliable", or a widely disseminated, semi-published old preprint or collection of seminar notes that is cited or well-regarded by "better" sources. For contentiousness, the degree of contention matters, and the degree to which the cited source represents a significant POV.
So to address these concerns, I'll try adding: "If outside citation is the main indicator of reliability, care should be used to adhere to other guidelines and policies, and to not unduly represent contentious or minority claims." to the above. Sorry for taking so much time, real life intrudes. Thanks again!John Z (talk) 08:47, 1 June 2008 (UTC)

[edit] Guideline conflict - Source vs Common

I would like some assistance with an issue. I have recently requested an RM for article Second Battle of Kharkov. The name is derived from a single source by David Irving, an author who has been proven to be unreliable by a Canadian Court of law, and amoral by the Austrian Court of law. While his amorality is not the issue here, his translation of memoirs written during the Second World War in the early 1960s has led to a number of subsequent authors in the field of military history unquestioningly using the name although it is completely lacking in original sources or other works by contemporary authors, German, English or Soviet.

The issue in terms of Wikipedia editing is in essence that although the use of the completely fraudulent historical event name has gained some commonality of usage in English, it is based on nothing that would be accepted in any reference work within the scope of academic research. Can the name continue to be used although it is unsubstantiated by other sources? In other words the current title of the article is OR by the wider academic, and Wikipedia standards, regardless of having gained acceptance in English usage.

There are other issues that complicate the RM, because I have had to create my own title for the event due to multiple commonly-named event occurrences. Not a good solution, but I have not had any proposals for alternatives yet.--mrg3105 (comms) ♠♥♦♣ 22:54, 17 May 2008 (UTC)

As has been pointed out to you several times on the talk page, the "correct translation" or the origin of a particular name is utterly irrelevant as far as choosing a title goes. And there is no conflict of sources here, as no RS is needed to explicate the common name for a thing. If there is a term that is overwhelmingly used in English for an event, and no apparent official english name, then that is the title that will invariably be used. It's up to you, the one seeking to change the name, to demonstrate that "Struggle for Kharkov" is the accepted term amongst historians or otherwise. Challenging the scruples of the term's author or the proper translation from Russian will get you no headway. Else, you're stuck in a content dispute, so follow dispute resolution. But given how things look for your suggestion, you may want to just move on. Someguy1221 (talk) 23:43, 17 May 2008 (UTC)
Actually this goes to a deeper issue of using a common name that is pure invention by a discredited source. I had become common only because less stringent subsequent authors used it without regard or consultation of sources. In fact the reasons for the original invention of the name are because there never was an original name in German and never could have been if they had actually understood and read the sources for the event, even from the German point of view.

The question is therefore - does Wikipedia accept baseless inventions for article titles because they have gained some acceptance by bad writers? Does "anything goes" the convention in Wikipedia?--mrg3105 (comms) ♠♥♦♣ 09:02, 19 May 2008 (UTC)

[edit] The problem with "third party"

"Articles should be based on reliable, third-party, published sources with a reputation for fact-checking and accuracy."

I think we need an explanation of what "third-party" means, in this context. For instance, in an article about a football team, a "third party" is presumably someone who isn't one of us (a Wikipedian) or affiliated with the team, but is someone else uninvolved with either. But what about an article that covers one side of a controversy involving 2 parties? I'm in dispute with an editor who insists that, in an article about Subject A which covers Subject A's attack on Subject B, we can only cite Subject A (the primary source) and Subject C, D etc (those not attacked by Subject A): we cannot cite Subject B's response to Subject A's attack, because Subject B is "not a third party". I think this is ludicrous, and can lead to a whole host of problems, especially when a fringe source attacks mainstream science/history/whatever: according to this interpretation, we cannot cite any rebuttals from the relevant mainstream experts, as they're not "third parties" in the dispute! --Robert Stevens (talk) 10:11, 19 May 2008 (UTC)

Clearly they are both involved parties. Note that there is no prohibition on using involved parties as sources; just on basing whole articles on them. Both the attack and the rebuttal can be cited. There also may be BLP issues involved if the charges are personal though. <eleland/talkedits> 10:17, 19 May 2008 (UTC)
However, we're not counting 1-2-3 in the same fashion in the two scenarios described above (the football team and the A/B controversy) to reach a consistent definition of what a "third party" actually is. Which is rather weird. --Robert Stevens (talk) 10:38, 19 May 2008 (UTC)

[edit] Scholarly sources vs. news sources

Is there not a case to be made for using scholarly sources in preference to news sources, where the former are available? To give an example, September Dossier mentions several British press reports, published in national dailies a few years ago, reporting that Saddam Hussein had the capability to strike British targets with WMD within 45 minutes. Due to ensuing events, everybody now knows that this claim was false. But there is a distinct likelihood that there are newspaper articles lurking in web archives that are similarly inaccurate but, because public interest in the issue waned, have remained uncontradicted in news sources. These are the sources that Wikipedia editors are invited to use. The associated risk of repeating erroneous claims is less present in academic research. Scholarly research is not driven by newsworthiness, and looks at issues and their developments in more detail than all but the very best of journalistic accounts.

So would it not make sense to require that scholarly works be given priority over news sources, and that where academic works and newspaper accounts contradict each other, the scholarly account should be preferred? A historian's study is unlikely to assert that Iraq had the capability to launch WMD attacks against British targets. While it might mention that there was such a claim – because in this case, the claim led to significant controversy – it would not repeat the assertions as statements of fact about Saddam Hussein's military capabilities.

These days, newspaper stories remain archived and accessible for years. Even the best newspapers, such as the New York Times, have had well-publicised scandals concerning the reliability of their reports. I believe we may run a risk of perpetuating errors and "sexed-up news" by having a guideline that assumes that newspaper reports are just as reliable as academic studies. Any thoughts? Jayen466 01:29, 20 May 2008 (UTC)

Completely agree.Momento (talk) 05:17, 20 May 2008 (UTC)
Is this associated with any discussion at talk:Prem Rawat, etc.? In September Dossier example above, the correction became a major political issue so there's no question about the accuracy. Whenever we have conflicting reliable sources we need to try to represent all the significant points of view in their proper context. However there's no question that in the scheme of things scholarly sources are considered the most reliable. ·:· Will Beback ·:· 06:49, 20 May 2008 (UTC)
The underlying issue here is human conflict, whether that be the conflict with Iraq or societal conflict with NRMs. I think it is fair to say that in conflict situations, the media have a tendency to champion populist POVs, sometimes at the expense of factual accuracy. That is not to say that scholarly studies are exempt from bias, but on the whole, scholarly analysis tends to be more sober and neutral. Basically, I would like to see more academic sources used, and a wording in this guideline that encourages editors to seek out and review the academic literature available. As an encyclopedia, scholarly sources should be our preferred sources in those areas where they are available, even though they may be harder to locate than news sources. Instead, in English Wikipedia, we often seem to end up with news sources, primary sources or sources from advocacy groups, bypassing a wealth of academic literature available. Jayen466 17:49, 20 May 2008 (UTC)
Definitely not as a general principal. You have to weigh sources of any sort and not apply strict rules. If your question is how hot the surface of the sun is, or why mountains produce a rain shadow, then sure, a scholarly source is best. If the question is whether social constructs of gender are responsible for pay inequities, or whether one musical genre is more groundbreaking than another, absolutely not. Academic / scholarly sources have a bias, and entire fields go in mistaken directions for decades. Politics is a pretty good example but hardly the most extreme. Dogma within the academy is often very different than prevailing schools of thought in other worlds of though - e.g. business, the law, popular beliefs. If you want to point to scandal there is faked research all the time, political advocacy and propaganda masquerading as scholarship, and paid-for lab results. We simply have to be aware that any source, no matter how esteemed the publisher, could contain false or misleading information. Wikidemo (talk) 18:31, 20 May 2008 (UTC)
We have to go on something though and we are not qualified to say whether an academic field is heading in the wrong direction or not. At least with academic work there is the chance for other academics to dispute it. This happens all the time in the cases you give about pay differentials or musical genres. So we can report the state of academic debate about the topic. News sources are excellent for recent events such as the current US presidential campaign. In areas such as international politics it is often the case that news sources and academic work tend to align with each other, i.e. reporters refer to books by academics and vice versa. Itsmejudith (talk) 19:25, 20 May 2008 (UTC)
I believe one aspect of this is that academic sources, once they become available, have the benefit of hindsight. There is obviously a time lag involved between the time when events are covered in news media, and the publication of historical studies analysing the events in question. We have New York Times archives going back 30 years and more; many of these articles cover topics that have since been the subject of scholarly study and debate. I think that no one would dispute that academic analyses are generally more thorough and better informed than media accounts. They may also include detail that did not make it into the papers. Recreating the history of a topic simply from newspaper articles, and selecting these to "tell a story", in my view amounts to a kind of OR if academic analyses of the same events are available. For example, no one would dream (I hope) of writing a history of the Second World War based on contemporaneous American, German, Russian und British newspaper accounts, without accessing scholarly analyses (perhaps they would if large amounts of these old articles were easily available online, and boy, might we have problems then). But when it comes to more recent events, we seem perfectly happy to access newspaper archives and ignore recent academic discourse.
To give a random example, take the article on the Srebrenica massacre. The most widely cited book on this topic, according to google scholar, is Rohde's Endgame: The Betrayal and Fall of Srebrenica, Europe's Worst Massacre Since World War II. Another scholarly study with a fair smattering of citations in google scholar is Accounting for Genocide: How Many Were Killed in Srebrenica?, from the European Journal of Population/Revue européenne de Démographie, by Helge Brunborg, Torkild Hovde Lyngstad and Henrik Urdal. Now, our article at present has over 100 citations, including a fair number of news articles, and various primary sources – court records, UN reports, etc. But while the book by Rohde is mentioned in the literature, not one of the more than 100 citations refers to it, or the EJoP article. In fact, not one of the academic sources listed under "Further reading" appears to be cited in the article.
A similar situation obtains in the article on the New Age – until a few days ago, the most widely cited academic studies covering the topic, by Paul Heelas and Wouter J. Hanegraaf, had not a single citation in the article. They were not even mentioned in the literature. But we had (and still have) citations from advocacy groups like the Vatican, "Jews for Morality", skeptics' groups, etc., as well as citations of primary literature by New Age authors.
Yes, newspaper articles and advocacy groups are easier to find on the web than scholarly analyses, unless someone undergoes the trouble and financial investment to register with an online library like questia.com (which still only gives access to a proportion of the material available), but it's precisely because newspaper articles are more easy to find that this guideline should impress upon editors the need to seek out and read academic studies. Not least because one might learn something in the process, as well as write a better article. Jayen466 12:39, 21 May 2008 (UTC)

I cannot agree with this. It seems that you are here advocating a slippery slope that would eventually exclude newspaper accounts. Where we have both, both should be cited. We have a fairly decent community understanding for the credibility of certain newspapers and what should or should not be read into them. In a case where there are two "scholarly" articles, and 30 newspaper accounts of the same situation, they all should be summarized in the article. Preference *may* be given to the scholarly accounts, by the editors present there, but that does not imply that the newspaper accounts should be excluded.Wjhonson (talk) 17:29, 21 May 2008 (UTC)

I would not advocate excluding newspaper sources. There are thousands of topics where news sources is all we have. But with all due respect, if we are on a slippery slope, it seems to me we are on the other end of it, with leading academic studies ignored or relegated to "Literature" sections, and entire articles based on news media, advocacy sources, and primary sources selected by editors identifying with their respective viewpoints. Jayen466 22:34, 21 May 2008 (UTC)

By the way Jayen466, we do give space for "minority views" as spoken *by* the minority themselves. That is, we can quote from the Vatican because they have a viewpoint, and that viewpoint is of interest to our readers. The viewpoint of a minority view on a topic should, in my opinion, always be presented enquoted and cited, to show clearly that it is their viewpoint. Wjhonson (talk) 17:31, 21 May 2008 (UTC)

We do give space for minority views spoken by the minority themselves – indeed we do, but subject to fairly sensible restrictions spelt out in WP:SPS as well as WP:PSTS. And don't forget that a good scholarly study will cover and present the minority view, especially if its subject is the minority in question.
To give an example, half of the Scientology article is currently cited to promotional Scientology websites. It rightly carries a marker for inappropriate use of primary sources. This despite the fact that J. Gordon Melton, who writes the Encyclopedia Britannica entry on Scientology, published a fairly sympathetic book on Scientology 8 years ago which covers Scientology thought in a way that Scientologists could hardly complain about. Yet the book is hardly cited (the only three citations it has are the ones I introduced the other day). Primary sources should be our last resort, not our first. In my experience, a collection of news, advocacy or primary sources often makes for a rather disjointed narrative, while the use of scholarly analyses produces coherence by analysing the relationships between these sources – something that we as Wikipedians cannot do, since it is not part of our brief. Jayen466 22:34, 21 May 2008 (UTC)
While scholarly sources are among the best, they are not perfect either. There are scholars who have conflicts of interest with their topics that would not be tolerated in a good newspaper. Many scholars have fringe ideas. Furthermore, it's not always clear what qualifies as a "scholarly" source. Papers in peer-reviewed journals or books published by university presses are presumably scholarly/academic, but what about a book written by a profesor published by a popular press? In short, there are pitfalls with every kind of source and we shouldn't make a blanket policy that says one type of source is always better than another. OTOH, if someone wants to add an encouragement to editors to make extra efforts to find better sources then I don't think that would be controversial. ·:· Will Beback ·:· 17:55, 21 May 2008 (UTC)

I agree with Jayden that academic and historical sources, once they become available, have the benefit of hindsight and an overview of a larger time period. I can recall all the scare stories that appeared in popular media - newspapers and magazines - in the late 1970s and in the 1980s about "brainwashing" cults in the U.S. that were "luring the youth". Media frenzy spurred on the "deprogrammers" to forcibly abduct individuals from new religious organizations. Secondary sources would now reflect this in proper sociological context, while articles from that time would be skewed.

Wikidemo correctly pointed out that there is "faked research all the time, political advocacy and propaganda masquerading as scholarship". In addition, the Internet has websites from various advocacy groups that push anti-Mormon, anti-Catholic, anti-Jewish, etc. agendas.

As Wjhonson reminded us, there can be a slippery slope that could eventually exclude newspaper accounts. The bottom line is that we should weigh the value and the importance of various newspaper accounts if they contribute relevant information (and may even be part of the story of the article), but citing historical or scholarly research should be encouraged - if it is available. Arion 3x3 (talk) 21:35, 21 May 2008 (UTC)

Hello! Jayen466 drew my attention to this debate. My twopennyworth is as follows; first, "slippery slope" arguments are an informal fallacy of argument - I put this aside, since I do not believe anybody wants to or is going to ban press sourcing outright. Second, the value of a news article depends upon the writer, the length, the general reliability of the organ in question, the stated quality of the sources. Some articles are written by specialists, some are cobbled together out of press archives. Not many give the kind of overview of a subject that can suffice to provide the structure of a technical wiki article, though many may add detail.
A historian is trained in source criticism; a journalist is not. The use of second-hand material from archives is so general that the popular press becomes largely self-referential. It's therefore necessary for the editor to provide source-criticism and to do that s/he will have to refer to good quality scholarship. Apart from anything else, press articles are not easily-verifiable sources.
An example from Jayen's wiki world would be Osho - known from thousands of press articles as "the Bhagwan". It can more or less be stated as an axiom that any article that uses this formulation is constructed second-hand from previous reports since first-hand experience would inform the reporter very speedily that "THE Bhagwan" does not occur anywhere in general usage or any primary source. It is not to be found in any serious secondary treatment either. It is simple laziness - the lower bar of truth in the popular press.

So, newspaper articles are fine as content, but rarely suitable or reliable enough for the structure and analysis of an article without the back-up of a good book. I'd certainly say that' if an editor pushes press PsOV and appears to avoid or discount scholarly analysis (as in the New Age article) that editor is on very shakey ground. Redheylin (talk) 22:19, 21 May 2008 (UTC)

[edit] Proposed rewording (changes marked in bold)

[edit] Scholarship

Many Wikipedia articles rely upon source material created by scientists, scholars, and researchers. This is usually considered the most reliable type of source, although some material may be outdated by more recent research, or controversial in the sense that there are alternative theories. A review of the existing scholarly literature should be the first step in starting work on an article. Wikipedia articles should strive to cover all major and significant-minority scholarly interpretations on topics for which scholarly sources exist, and all major and significant-minority views that have been published in other reliable sources, as appropriate.

  • Material that has been vetted by the scholarly community is regarded as reliable; this means published in peer-reviewed sources, and reviewed and judged acceptable scholarship by the academic journals.
  • Scholarly source material may have the benefit of hindsight that is lacking in historical news accounts; it may provide background analysis and help to balance and structure an article.
  • Items that are recommended in scholarly bibliographies are welcomed.
  • Items that are signed are preferable to unsigned articles.
  • The scholarly credentials of a source can be established by verifying the degree to which the source has entered mainstream academic discourse, for example by checking the number of scholarly citations it has received in google scholar or other citation indexes.
  • In science, single studies are usually considered tentative evidence that can change in the light of further scientific research. How reliable a single study is considered depends on the field, with studies relating to very complex and not entirely-understood fields, such as medicine, being less definitive. If single studies in such fields are used, care should be taken to respect their limits, and not to give undue weight to their results. Meta-analyses and systematic reviews, which combine the results of multiple studies, are preferred (where they exist).
  • Peer-reviewed scientific journals differ in their standards. Some court controversy, and some have even been created for the specific purpose of promoting fringe theories that depart significantly from the mainstream views in their field. Many of these have been created or sponsored by advocacy groups. Such journals are not reliable sources for anything beyond the views of the minority positions they are associated with.

[edit] News organizations

Further information: Wikipedia:Verifiability and Wikipedia:Biographies of living persons

Material from mainstream news organizations is welcomed, particularly the high-quality end of the market, such as the The Washington Post, The Times of London, and The Associated Press. However, great care must be taken to distinguish news reporting from opinion pieces. Opinion pieces are only reliable for statements as to the opinion of their authors, not for statements of fact. When citing opinion pieces from newspapers and magazines, in-text attribution should be given. When adding contentious biographical material about living persons that relies upon news organizations, only material from high-quality news organizations should be used.


Comments? Jayen466 23:23, 21 May 2008 (UTC)

  • SUGGEST "analysis and context necessary to the balance and structure of an article". Redheylin (talk) 00:47, 22 May 2008 (UTC)
    • Incorporated in wording above. Jayen466 00:56, 22 May 2008 (UTC)

I would recommend removing the bit about using opinion pieces. Their use at all is highly dubious and they are commonly rejected. They should not be used at all if their use is contentious and giving any encouragement to use them is highly inadvisable. Vassyana (talk) 04:37, 22 May 2008 (UTC)

Agreed. Implemented in proposal above. Jayen466 10:27, 22 May 2008 (UTC)
To address the point made by Will Beback above, I propose adding another bullet point in the section on scholarly sources:
  • The scholarly credentials of a source can be established by verifying the degree to which the source has entered mainstream academic discourse, for example by checking the number of scholarly citations it has received in google scholar or other citation indexes.
(Now added above.) Jayen466 12:12, 22 May 2008 (UTC)
I disagree about removing the caveat on opinion pieces. I understand that this is a grey area... but we do need this explained. Many op-ed pieces are written by notable experts in their field, and as such they should be considered reliable for the opinion of that expert. Because the author is notable... their opinion is notable. Notable opinions should be attributed, but they are reliable. Blueboar (talk) 12:32, 22 May 2008 (UTC)
The caveat whose deletion we are discussing really only says that statements from opinion pieces should be attributed to their author. This clearly makes good sense. The question is whether it needs to be stated here, in such a prominent place in a key guideline, where it could be seen as encouraging the use of opinion pieces in articles. Perhaps the passage could be made into a subsection within Wikipedia:Reliable_sources#Reliability_in_specific_contexts, along with a caveat that the use of an opinion piece in an encyclopedia article must be justified by exceptionally good reasons (notability of the author, amount of comment generated, etc.). Jayen466 12:56, 22 May 2008 (UTC)
How about something like this....
  • Material from mainstream news organizations is welcomed, particularly the high-quality end of the market, such as the The Washington Post, The Times of London, and The Associated Press. However, great care must be taken to distinguish news reporting from opinion pieces. Opinion pieces are only reliable for statements as to the opinion of their authors, not for statements of fact. When citing opinion pieces from newspapers and magazines, in-text attribution should be given. When adding contentious biographical material about living persons that relies upon news organizations, only material from high-quality news organizations should be used.
This would make it clear that editors may discuss a notable opinion that is stated in an op-ed piece... but only as a statement of opinion and not as a statement of fact. Blueboar (talk) 14:53, 22 May 2008 (UTC)
Sounds good to me, and incorporated above. What do you think, Vassyana? Others? Jayen466 15:24, 22 May 2008 (UTC)
I feel more concerned about a matter that perhaps belongs more in "POV" than here - that it is important to distinguish two sorts of reference, those that furnish content and those that supply form. It seems to me that the less the actual form follows sources - whether press or book - that give an overview of the entire subject of the article, the more the editor is reliant on personal synthesis. Therefore, if necessary, an editor ought to prefer to follow a single comprehensive account, and such an account cannot be said to have been given UNDUE WEIGHT as single source among many unless another similarly authoritative comprehensive overview can be produced, at which point editors should co-operate in achieving a synopsis of both. In other words, a heap of mice do not make a lion. A hundred casual references do not outweigh a single general academic treatise or reliable popular overview, since only the latter provide a verifiable source for the OVERALL STRUCTURE of the article, thus providing a rationale for the weight given to each aspect of the subject. Perhaps this is less important in, say, a bio, which follows a common-sense structure, but I think it gets more important the wider (and vaguer, and more controversial) a subject is. Redheylin (talk) 00:15, 23 May 2008 (UTC)
Am in absolute agreement with you. I have tried both approaches in the past, and the result of the approach you outline is infinitely superior, producing a more readable article. Jayen466 00:32, 23 May 2008 (UTC)
I agree with you agreeing. Beauty is Truth, Truth Beauty. Redheylin (talk) 04:53, 23 May 2008 (UTC)
Many well-written opinion pieces are mixtures of fact and opinion. Where they are printed in reliable publications and clearly delineate facts from opinions, the facts in those pieces are reliable. Occasionally you get the opposite - news pieces, even in reliable publications, that mix opinions in with their facts or that contain analysis or reprinted opinions of others. That's more common in some sections (e.g. travel, fashion, food and dining) than it is in others. You often have to look at who is writing the piece and in what capacity. Some columnists are in between editorials and news. Some prominent reporters write occasional opinion pieces. Some papers have guest op ed spaces where you cannot trust any fact at all because the writers are biased, not professionals, and there is no checking or editing of facts. Let me make up a hypothetical example that I hope is uncontroversial written by, say, the head restaurant critic of the New York Times.
"Five years ago this March, Belgian-born chef Thomas Le Gateau left the three-star Michelin restaurant Le Bon Bon for the "Bohemian Grill Club", the new Starwood venture on Hyde Street. Although trying hard to pander to relatively less sophisticated American tastes, he never seemed quite at home in his new venture and, after a reported falling out with the investors barely three months later, left abruptly in disgrace."
I hope it's obvious that the first sentence is (mostly) reliable and the second sentence is (mostly) not. Plus you have to have a little judgment - for instance it is good to question the reliability about any accounts of who really owns a private business, or why an executive might leave. Hope that helps. Wikidemo (talk) 00:52, 23 May 2008 (UTC)
I agree that accademic sources tend to be far MORE reliable than news reports... but that does not mean that news reports are unreliable. Sure, some news sources are more reliable than others... and care always has to be taken to distinguish fact from opinion... but that does not change the basic reliability of news sources. Even opinion pieces can be reliable under certain circumstances (especially for statements as to the author's opinion). The key to this section is to say that news media sources can be reliable. We always need good judgment as to whether a given news article actually is a reliable source for a given statement in a wikipedia article. That is actually true of all sources. Blueboar (talk) 12:48, 23 May 2008 (UTC)

I've dropped in the changes to the scholarship section, as per the above discussion. Looking through the above, I think we are also agreed that the wording for news sources proposed above is at any rate an improvement on what we have, so I'll drop this in as well. Jayen466 13:06, 23 May 2008 (UTC)

This new sentence has some problems: "A review of the existing scholarly literature should be the first step in starting work on an article." Many people do not have a chance to do "a review of the existing scholarly literature" over some topic which is a lot of work and requires good university library resources. If the sentence means that you can't start an article without doing such review, it would be too difficult to write a short new article. If the sentence means that you can't work on an article without doing such review, there would be very few people working with articles at all. Best regards Rhanyeia♥♫ 07:23, 24 May 2008 (UTC)
Note that our present wording says, Wikipedia articles should strive to cover all major and significant-minority scholarly interpretations on topics for which scholarly sources exist. How can an editor live up to this requirement if they don't even take the trouble to check what scholarly material is available, and instead just start hitting newspaper archives, start inserting their favourite or most hated passages out of the primary sources they have read, etc.? Which, unfortunately, is what has happened in a great many articles. As I pointed out before, if you don't have access to a university library (I don't myself), a subscription to questia ($4 a month or so) goes some way giving you access to the most prominent and most widely cited scholarly sources. I contend that ignoring scholarly analysis is, at least for those topics that are subject to scholarly analysis, entirely incompatible with the idea of writing an encyclopedia. I'll reinstate the sentence; let's see how consensus develops. This debate is not finished yet. Jayen466 17:50, 24 May 2008 (UTC)
Questia may be useful sometimes but it's probably not very comprehensive, it's only about humanities and social sciences, and I also looked for some articles and books (of quite usual topic) which I'm reading now and I couldn't find them. Best regards Rhanyeia♥♫ 15:35, 26 May 2008 (UTC)

[edit] Once again!...

...RS is coming up with all sorts of crap that gets nary a mention at V. If someone would like to make significant changes in sourcing policy, attempt it here. In fact, I think we should cut this entire thing down to the words used at Wikipedia:Verifiability#Reliable sources. We can use a template to ensure consistency. God, RS is a joke. Marskell (talk) 17:31, 23 May 2008 (UTC)

Well, WP:V refers the reader here for a discussion of reliable sources: "For a guideline discussing the reliability of particular types of sources, see Wikipedia:Reliable sources (WP:RS)." Nothing said here is inconsistent with WP:V, and I believe we can't just do away with scholarly sources in the presentation here. Jayen466 17:39, 23 May 2008 (UTC)
Jayen, I don't doubt your sincerity. But this nonsense has been going on for two years: because RS is a separate guideline people feel free do with it as they please. It continually produces novel policy interpretations, just like this scholarship section (a variant of which was removed months ago). What V actually says: "For a guideline discussing the reliability of particular types of sources, see Wikipedia:Reliable sources (WP:RS). Because policies take precedence over guidelines, in the case of an inconsistency between this page and that one, this page has priority, and WP:RS should be updated accordingly." Where this page differs from V, the difference should be cut. Marskell (talk) 17:45, 23 May 2008 (UTC)

Apart from somewhat surrealist procedural discussions, was there actually bad guidance? Or did anyone even contend so much? --Francis Schonken (talk) 17:49, 23 May 2008 (UTC)

I don't think this guideline differs from or contradicts WP:V. It just discusses the topic in more depth. We can't delete all mention of scholarly sources in this guideline (there is too little scholarship reflected in WP as it is). The scholarship section has been here for several months at least. Some additions were made as per the discussions above. Which parts of the wording do you feel are out of sync with WP:V? Jayen466 17:53, 23 May 2008 (UTC)
Most of the advice was so obvious it didn't need mentioning ("Items that are recommended in scholarly bibliographies are welcomed.") But some of it is a clear departure from V: "Peer-reviewed scientific journals differ in their standards. Some court controversy, and some have even been created for the specific purpose of promoting fringe theories that depart significantly from the mainstream views in their field." V does not deprecate peer-reviewed science journals. Instructions of this sort can be positively harmful in that they can allow POV pushers to question even quality sourcing.
More broadly, the larger (surreal) point remains: for years now, we have allowed this crappy guideline to sit here without consistent wordng, without conformity to V, without a defined purpose. And everytime it departs from V it creates a problem. Marskell (talk) 18:00, 23 May 2008 (UTC)
For reference, WP:V says

In general, the most reliable sources are peer-reviewed journals and books published in university presses; university-level textbooks; magazines, journals, and books published by respected publishing houses; and mainstream newspapers. As a rule of thumb, the greater the degree of scrutiny involved in checking facts, analyzing legal issues, and scrutinizing the evidence and arguments of a particular work, the more reliable it is.

Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas where they are available, such as history, medicine and science. Material from reliable non-academic sources may also be used in these areas, particularly if they are respected mainstream publications. The appropriateness of any source always depends on the context. Where there is disagreement between sources, their views should be clearly attributed in the text.

Arguably, the previous wording was out of sync with WP:V because it did not make clear that "Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas where they are available". Instead, we said "source material created by scientists, scholars, and researchers ... is usually considered reliable". That is quite a difference.
I appreciate that you may have procedural objections; if you feel this should have been discussed at the Village Pump or elsewhere, or have other concerns, please let me know. Cheers, Jayen466 18:03, 23 May 2008 (UTC)
As for deprecating peer-reviewed journals, I wasn't here when this was inserted. But I am aware that fans of ancient astronaut theories also have "peer-reviewed journals", as do some other pseudosciences and fringe sciences. I assume that the insertion was made by an editor who had those sorts of journals in mind. Jayen466 18:05, 23 May 2008 (UTC)
I'm not sure if your last post is an argument or a joke.
Anyway, I wrote part of the V wording. I know what it says. Why do we need a divergent recapitulation here? We can template the words over easily enough. Marskell (talk) 18:12, 23 May 2008 (UTC)
Suggest we insert mainstream in the first bullet point: "this means published in mainstream peer-reviewed sources". Then we don't need the paragraph that sounds like it is deprecating peer-reviewed journals (which I am sure was not the intent). Jayen466 18:14, 23 May 2008 (UTC)
Suggest we take the wording from here, put it in a template and post it to both V and RS. Then I suggest we cut everything else from this page and start a discussion at V about what the policy needs and what is extraneous. Marskell (talk) 18:18, 23 May 2008 (UTC)
Are you essentially suggesting getting rid of this guideline? Jayen466 18:26, 23 May 2008 (UTC)
Yes. There's nothing radical about it—people have been suggesting getting rid of RS forever (and ever and ever!). Given previous discussions, it's impossible to get rid of the page as a search target—WP:RS must exist as an independant target (not as a redirect to V) or people will start eating their fingers. That being the case, I suggest we bind the page as closely as possible to V through templates and use WT:V as the place to discuss sourcing advice. Marskell (talk) 18:39, 23 May 2008 (UTC)
Could I ask you to post a link here on this talk page if and when you start this discussion? I'd like to follow it and, perhaps, contribute.
I am open-minded and can see potential advantages in either approach – either having just one policy which covers the entire area of sourcing, or having this page as a guideline that explains the relative merits of various types of sources in more detail than WP:V does. And I understand your point about this page potentially taking on "a life of its own". Jayen466 18:49, 23 May 2008 (UTC)
Fair enough, Jayen. I'll post to your talk in a moment. Marskell (talk) 19:01, 23 May 2008 (UTC)
Jayen, please see also Wikipedia:Perennial proposals#Define reliable sources --Francis Schonken (talk) 19:55, 23 May 2008 (UTC)
Tx. Jayen466 22:03, 23 May 2008 (UTC)
It would be useful to see the history. As a point of process, forking is an inevitable result of redundant guidelines. If we want to set any quotation from RS in stone we're going to have to transclude it here or else warn people againste editing it here. On the substance of things, I don't have any preference as to the exact wording, but we should be careful to point out that academic/scholarly sources are definitely not always the most reliable. This comes up all the time, say in politics, where a professor or think tank researcher publishers a controversial thesis, and someone wants to coatrack lots of POV material into an article on theory that it's an especially reliable source. I know qualifiers like "normally" and "usually" should take care of it, and exhortations that context is important and common sense apply, but over-reliance on academic sources can be a problem around here. Wikidemo (talk) 19:50, 23 May 2008 (UTC)

[edit] Hindsight section

I removed the following, because it's either saying nothing, or it's not clear what it means:

"Scholarly source material may have the benefit of hindsight that is lacking in historical news accounts. It may provide background analysis and help to balance and structure an article."

This seems to say only that current scholarly material might be more up to date than old newspaper accounts, which is close to a tautology. SlimVirgin talk|edits 07:47, 24 May 2008 (UTC)

I wondered about that when it appeared but I passed it over. My guess is that "hindsight" should be "insight". -- Boracay Bill (talk) 07:54, 24 May 2008 (UTC)
Yes, could be, though I think that might leave us with the same problem, in that the basic point seems to be that current sources might be better informed than older ones. SlimVirgin talk|edits 08:17, 24 May 2008 (UTC)
See the discussion above under #Scholarly sources vs. news sources for the thinking behind it. The idea is to discourage reliance on newspaper archives alone, and to encourage the use of scholarly analyses putting things into context, where such analyses are available. Jayen466 11:45, 24 May 2008 (UTC)
SlimVirgin, could you explain your objection a bit more? From my understanding, what you view as as a problem, I view as a virtue. I believe encouraging the use of up-to-date sources in some form is a good thing, since we should be presenting the current state of scholarship, as much as possible. Vassyana (talk) 12:09, 24 May 2008 (UTC)
Vassyana, I agree with you. My objection to the sentence was simply that it was almost devoid of meaning. It is basically saying only that more modern sources may have a better overview than older ones. It is trivially true that they may be more informed, but not necessarily true that they always will be. Everything depends on the source and the context.
Jayen, the issue of newspapers versus scholarly sources is a separate one from the issue of modern versus older sources. The policy (WP:V) doesn't prioritize scholarly sources over news accounts, again because everything depends on context. SlimVirgin talk|edits 21:25, 24 May 2008 (UTC)
The relevant wording in WP:V policy is:

Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas where they are available, such as history, medicine and science. Material from reliable non-academic sources may also be used in these areas, particularly if they are respected mainstream publications. The appropriateness of any source always depends on the context. Where there is disagreement between sources, their views should be clearly attributed in the text.

This clearly says that scholarly sources are "usually the most reliable sources in areas where they are available". It says "non-academic sources may also be used in these areas", with the caveat "particularly if they are respected mainstream publications". This does favour academic sources, not to the exclusion of everything else (context), but it does favour them. Jayen466 21:38, 24 May 2008 (UTC)
I was thinking of editing that section out myself. I use to spend long hours digging through newspaper archives. If you want to find out what was actually said at a particular time or place by a particular individual; or wish to know where a particular individual was at a particular time, newspaper archives are often the only sources available unless the person you are researching is especially famous. Old speeches tend to sit as unpublished papers, with the only reliable published source being part the local reporter at the scene recorded. Scholarly sources give breath; newspapers reports of the time give details. For WP purposes we don’t normally require details at that level, and we normally treat such things as primary sources anyway. My point being that the statement while on the surface is generally true, there are two or more sides to everything, and we need at least a more rounded statement. (full disclaimer: I do not now nor have I ever worked for a newspaper/news organization.) Brimba (talk) 13:39, 24 May 2008 (UTC)
Makes perfect sense to me. Thanks! Vassyana (talk) 18:16, 24 May 2008 (UTC)
I understand that a newspaper account may help place a person in a specific locale at a specific point in time, etc., but I think there is nothing in the addition proposed above that is in conflict with that use of newspaper sources. It says, a scholarly source may add the benefit of hindsight that is lacking in a newspaper source (e.g. that Saddam Hussein did not really have the capability to attack British targets within 45 minutes, as was widely reported at one point), that it may provide background analysis, and that it may help to add structure and balance to an article. Where is the contradiction? Jayen466 18:46, 24 May 2008 (UTC)
Mm, I think I understand now what Brimba was trying to get at. Is the point that we should also indicate that newspaper sources may provide/correct detail that may be lacking/inaccurate in some scholarly analysis? I could see the point of that. There's two sides to this: on the one hand, it may beg the question why that detail has not been thought important in scholarly analysis, and why WP should differ from scholarly analysis in according importance to it – there is a whiff of OR about that. The other side is, if some scholar mentions that Martin Luther King was in Washington on a given date, and a New York Times article reports that in fact he was leading a march in Alabama, I'd go with the newspaper source rather than the scholar. It goes both ways. Jayen466 19:09, 24 May 2008 (UTC)

[edit] Google Scholar

I decided to give Google scholar a test drive. I entered in three controversial figures to see how they ranked: Daniele Ganser gets 54 hits, Jared Taylor got 194 hits, and Lyndon LaRouche 735 hits. This leaves me questioning its value in estabilishing reliability, which is what this page is about. Brimba (talk) 14:29, 24 May 2008 (UTC)

Wished it were that easy to test "reliability". In reality you didn't test any of the kind, because then we could make WP:RS a redirect to Wikipedia:Search engine test. You did a flawed test on notability, that's all. --Francis Schonken (talk) 15:35, 24 May 2008 (UTC)
The idea is that you enter the title of the book, journal article, etc. as well as the author, and then look at how often this particular work has been cited. It is true that this may conflate notability and reliability to some extent. Even so, a high number of citations can be taken as an indication that the book or article has at least been taken seriously enough for a large number of people to have commented upon it. I agree the system is not perfect; there are scholarly works that I know have been cited that don't show up in google scholar at all, for whatever reason. I still think it is better to have a flawed tool than none. Jayen466 17:57, 24 May 2008 (UTC)
I agree 100% with Jayen. For academic purposes, a high number of citations in peer-reviewed journals suggests that the work (author) is (rightly or wrongly) taken seriously by scholars. It's one of many tools that editors should use when trying to determine the notability of some really questionable subjects. However, a plain vanilla google search is an entirely separate matter. It irritates me to watch some well-meaning Wikipedia editors mistakenly conflate "Google" with -- to name a few -- JSTOR, LexisNexis, Factiva, WorldCat, and Google Scholar thereby throwing the baby out with the bathwater. The former includes all kinds of blogs and other unreliable personal websites without editorial oversight and reputation for fact checking. The latter--generally speaking--do not. (There are exceptions, of course.) J Readings (talk) 18:12, 24 May 2008 (UTC)

I'd work on a basic description of the testing method you have in mind (with the necessary ifs and buts) either here or at Wikipedia:Search engine test (which has a lot of info, most of it reservations against, but not so much info regarding reliability testing). I'm a bit doubtful whether a significant role for this type of testing is possible, especially if you're going by counting - certainly you need to check the source you're testing is actually used as a reference, not some random delusive comment about it (otherwise you'd only be testing notability).

To give an example of such building block for an actual testing method: Brimba clearly thought he could simply run the names of a few persons by Google Scholar and that's it. Jayen gave a first important indication: this actually has only a chance to produce something useful, when limited to "publications" (name of publication + name of author). I call this an important qualification while in Wikipedia guidance it is usual to indicate as well publications, as also *authors* (independent of their publications), as *publishers* (independent of the quoted publication) to be "sources": only the first of these types of sources can be thus tested. You'd need to inscribe that limitation in the testing method.

My experience with search engine tests is primarily in connection with Naming Conventions (WP:NC). Even if the use of search engines is certainly easier there (notability of a certain format of a name vs. testing of reliability), many results were rejected, for various reasons. Some people will even under no circumstances trust a result acquired via search engine of whatever brand, type or build. --Francis Schonken (talk) 18:41, 24 May 2008 (UTC)

Google Scholar reports as "hits" mere mentions as well as citations. Lyndon LaRouche is mentioned in many scholarly papers so he gets many hits. It doesn't mean he's cited frequently. ·:· Will Beback ·:· 09:24, 26 May 2008 (UTC)
Its true that "citations" are warped by those papers that are extensively quoted in order to be debunked, or as the sole major represented of a discredited or outdated marginal view. I'm not sure how to set up a method of handling that. --Relata refero (disp.) 12:15, 26 May 2008 (UTC)

[edit] Most reliable

Would whoever keeps adding that scholarly articles are the "most reliable" sources, please note that this guideline must be consistent with the policy, which does not prioritize scholarly sources. The policy says:

"In general, the most reliable sources are peer-reviewed journals and books published in university presses; university-level textbooks; magazines, journals, and books published by respected publishing houses; and mainstream newspapers. As a rule of thumb, the greater the degree of scrutiny involved in checking facts, analyzing legal issues, and scrutinizing the evidence and arguments of a particular work, the more reliable it is."

It's important not to prioritize any particular type of source out of context. There are times when mainstream newspapers expose issues that researchers have not addressed properly, and there are other times when researchers give a more accurate account that newspapers can. Editorial judgment is needed to evaluate sources in context. It can't be done in a guideline. SlimVirgin talk|edits 21:20, 24 May 2008 (UTC)

I wonder when WP:V started de-prioritising academic sources. That's an absurdity, and part of an unconscionable dumbing-down and politicisation of the project.
In general, your remarks here are puzzlingly off-base. Can you give me an example of this sort of "issue"? --Relata refero (disp.) 22:25, 24 May 2008 (UTC)
Also, I believe you may have extracted the incorrect section of WP:V. The wars on the section beginning "Academic and peer-reviewed publications are highly valued and usually the most reliable sources in areas where they are available, such as history, medicine and science." are now coming back to me. It appears that section has a robust consensus, and the "may also be used" language was specifically chosen to indicate priority. Given that, I'm restoring the "generally most reliable" as more in keeping with WP:V. -Relata refero (disp.) 22:28, 24 May 2008 (UTC)
I've reverted to SlimVirgin's version. There is a lot of discussion on the page about this. WP:V does not prioritize scholarly sources in all cases and it would be a mistake (and a bias) to do so. There are many sources of truth in the world; academia is only one. There are many different fields and types of academics and scholarship; many of them have their agendas. Wikidemo (talk) 23:00, 24 May 2008 (UTC)
Bias? How?
WP:V clearly prioritises academic sources in areas where they are available. This has been extensively discussed on WT:V, now that I've looked. I suggest you re-open the discussion there if you have concerns. Alternatively, I am open to a complaint that "generally more reliable" is not a useful paraphrase of the WP:V wording. The rest is immaterial. --Relata refero (disp.) 23:03, 24 May 2008 (UTC)
You ask, rhetorically, how academic sources might be biased? I think that goes without saying, but if you must know, bias ia a matter of human nature. Need I recount the institutional conditions in academia that lead to sloppy, mistaken, and biased research? WP:V goes quite far enough in the ways it does favor academia. I would not support this guideline going farther than that. Wikidemo (talk) 23:11, 24 May 2008 (UTC)
If there is no systemic bias, it is, of course, a completely irrelevant concern.
If you feel this goes "farther" than WP:V understandably does in favoring academic sources, please suggest alternative wording. Not specifying that WP:V does in fact favor academic research in the areas where it is available is misleading. --Relata refero (disp.) 23:16, 24 May 2008 (UTC)
I've imported the exact WP:V policy wording for the clause in question. Can we all agree that this is safest, given that WP:V is policy and this guideline should be aligned with it? Jayen466 23:28, 24 May 2008 (UTC)
Sure. WP:V itself, complete with examples, is a good "alternative wording". As an aside, though, imagine if the examples were "music criticism, ethnic studies, philosophy, and liberation theology movements" rather than "medicine, science, and history." People might not be so ready to give as much deference to academia vis-a-vis news, law, and other sources of knowledge. The "critical reaction" section of our film articles, for example, is full of citations to professional critics, not film studies departments. You'll find the same in fields like politics, music, popular culture, etc. Quoting WP:V gets back to the issue raised months ago, why do we have RS residing both at WP:V and here? Responding to Relata refero, I don't see what the systematic or non-systematic nature of bias has to do with the reliability of a source. Bias means, among other things, it is more likely to be ill considered or wrong. Wikidemo (talk) 23:40, 24 May 2008 (UTC)
Unless you claim that academic sources systematically demonstrate "bias" more often than other sources, merely pointing out that bias is human is irrelevant to a hierarchy of reliability.
About the other point, I certainly do not think that we should de-emphasise scholarly sources on "music criticism, ethnic studies, philosophy, and liberation theology movements". Where those are available, they should be emphasised. Phil Sandifer was rightly shouted down recently, both on the list and on WT:V, when he thought that that was a mistake. There's a robust consensus, as I said. --Relata refero (disp.) 23:49, 24 May 2008 (UTC)
You're free to advocate for Scholarpedia. I imagine we would deride most popular music there as simplistic and uninspired, discuss social problems in terms of oppression of the majority culture, remove all reference to causation as a natural phenomenon because it has no sound basis in metaphysics, and mention that Jesus will save the oppressed in all the articles about dictatorships. This is Wikipedia, however. The practice is to find the best reliable sources, not the academic source in every case. I see no consensus for a hierarchy of reliability in favor of the ivory tower, but I am aware of some misconceptions on the point. Wikidemo (talk) 00:07, 25 May 2008 (UTC)
Consensus exists. If you, for what appear to be highly idiosyncratic reasons, are displeased with it, that does not mean that you will not have to live with it. --Relata refero (disp.) 00:30, 25 May 2008 (UTC)
No such consensus exists. You seem to be making this up. I'm not sure what you're getting at, even. As I illustate with the examples, deference to academic sources only happens here in certain fields and contexts, notwithstanding the language of WP:V. Certainly not in music criticism, etc. There is not even any debate over the issue because as a matter of editorial judgment people know what a reliable source is. There are a few areas in which there is contention, however. For example, people sometimes try to railroad controversial material into articles about current events or politics because it appeared in an academic book or is supported by someone with academic credentials. Wikidemo (talk) 00:51, 25 May 2008 (UTC)
Wikidemo, just as an illustration, here is a university press-published book on Heavy Metal, "Running with the Devil", written by a professor of musicology. I believe it is very, very far from asserting that Heavy Metal is "simplistic and uninspired." Along with other scholarly works, it is widely cited in our article on Heavy Metal, which has FA status. Jayen466 00:54, 25 May 2008 (UTC)
So there is (perhaps) a reliable source on heavy metal from within academia. I'm not arguing that academic sources are unreliable, just that we cannot establish a priority system by which academic sources are presumed in every discipline to be more reliable than non-academic sources. Here are some random counterexamples I just dug up. this paper is a wonderful, interesting, provocative article in the Yale Law Journal (law journals are peer-reviewed and scholarly, and have a fact-checking standard head and shoulders above journals in many other fields - but they often strive to stake out a position, not elucidate truth). Here is the abstract from a (peer reviewed, scholarly) paper that states that we live in a "patriarchal and heterosexist social landscape that is inundated with mainstream pornographic material and where very specific socially constructed female sexual performances are promoted." I tend to agree, but I would not say that is a reliable source that "we" (whoever that is) are in a patriarchal society. It is customary in some disciplines (critical race theory, for instance, and politics) for people involved in the events they write about to illustrate their papers with personal anecdotes and recollections. Those anecdotes are not fact checked at all, generally.Wikidemo (talk) 01:17, 25 May 2008 (UTC)
I think Realata Refero is absolutely right. We are talking only about areas where academic sources are available, including natural and social sciences. Yes, absolutely, the scientific (academic) sources must be prioritized, simply because scientific research is more reliable than non-science.Biophys (talk) 03:00, 25 May 2008 (UTC)
So you're proposing this only applies to science and not, say, critical race theory or feminist studies? Why?Wikidemo (talk) 03:27, 25 May 2008 (UTC)
I think this applies to any area where scientific methods can be used. It does not apply to non-scientific subjects like theology, astrology, news reports, cinematography, video games and other things like that.Biophys (talk) 04:52, 25 May 2008 (UTC)
Biophys, I'm afraid your list of non-scientific subjects is based on a typical fallacy and unfamiliarity with the respective bodies of scholarly literature. For example, there are plenty of scholars writing on religious topics, as well as sociologists, psychologists and historians who study and research pseudoscientific phenomena such as astrology or ancient astronaut theories (for an example of the latter, see "Ancient Astronaut" Narrations: A Popular Discourse on Our Religious Past). When covering Pope Benedict XVI's historical stance on theological issues, or when writing an article on Raëlism, we should (and do) cite scholars who have studied these fields. Heading for the LA Times archive is not going to produce an encyclopedic article here. Jayen466 09:52, 25 May 2008 (UTC)
Wikidemo, if something is published in the Yale Law Journal, is well cited and widely discussed in the field (and I have no idea whether that is the case for the examples you cite), then it is a priority source for the topics it relates to. I think the problems you describe are more related to WP:UNDUE and WP:NPOV, in the sense that academic views may be "controversial in the sense that there are alternative theories". No one says it is okay for editors to use a single scholarly article to make a sweeping claim in the way you describe, without checking the article's standing and without covering alternative published views. Jayen466 10:18, 25 May 2008 (UTC)
That argument, I'm afraid, boils down to "scholarly sources are the most reliable, except when they aren't." I don't think it's a workable standard. I would not say the problem with my examples was weight. My cited examples do probably represent the majority opinion within their respective fields. I think within the right departments on campus the dogma is that we are in a patriarchal society, a generalization that is no more sweeping than other summary claims we make in articles. The claim that Law and economics offers a rich, ideologically neutral, analytical framework probably appeals to most tenured law professors when they are on record, even if many are privately uneasy that in practice it is associated with mainstream conservatism - a point that seems to be in the paper if one can get a fix on it. Nor is it POV - POV says we should cover all sides, but I would say we should cover neither. Academic positions on whether we live in a patriarchal society, or how conservative a field is, are simply not the final word because they do not have that quality of reliability. They do not lend verifiability to claims in articles based on them. In short, the information is not terribly encyclopedic. You do have a point about respecting scholarly works only within the topics they relate to, but that is not the issue here. Wikidemo (talk) 14:13, 25 May 2008 (UTC)

(deindent) Whatever your personal problems with the methods and metrics of modern scholarship, Wikipedia isn't the relevant place to play out your objections. ("Academic positions on whether we live in a patriarchal society, or how conservative a field is, are simply not the final word because they do not have that quality of reliability" - peh. Who else should we ask? Rush Limbaugh?) Enough. --15:22, 25 May 2008 (UTC)

That's really quite obnoxious. You're making personal attacks against me for my patient efforts to explain to you how things work around here. I suggest you step back and give this page a breather if you're going to start being uncivil to back up your odd agenda to reinvent Wikipedia as a literature review. Wikidemo (talk) 18:07, 25 May 2008 (UTC)
I'm sorry, but your patient efforts would be more believable if you didn't randomly accuse academia of "bias" without any backup, and pick random abstracts to mock. Wikipedia is a tertiary source, so it is a literature review, in a way. If you are uncomfortable with that concept, and the methods and metrics of modern scholarship, may I suggest that this is not the place for you? Thank you for your explanation, by the way, but I suggest you not try "explaining" it to anyone else...:) You're actually saying that "Academic sources do not have that quality of reliability.." - incredible. --Relata refero (disp.) 18:35, 25 May 2008 (UTC)
No, you may not tell me to leave the project for disagreeing with you. You've made your point that academic knowledge is categorically better than all other kinds, even to the point of being "incredible" that one might challenge the reliability of a paper coming form the Eastern Michigan sociology department on whether "women" (quotes in original) have any use for pornography made in our "patriarchal and heterosexist social landscape". That's not how it works here. You're welcome to your opinion if you can express it within Wikipedia's behavioral guidelines, e.g. WP:CIVIL and WP:NPA. Wikidemo (talk) 19:03, 25 May 2008 (UTC)
Umm, not with me, but with our basic principles of reliability. You're free to disagree with me at any time, though less free to randomly accuse me of personal attacks. Again, if you don't like the way that scholarship operates, Wikipedia is not the place to change that. We reflect the world outside. --Relata refero (disp.) 12:10, 26 May 2008 (UTC)

Comment. It has been my consistent understanding that academic publishers generally provide the most reliable sources and that this is generally accepted in the Wikipedia community. It seems appropriate for a guideline to express what is generally accurate. Examples to the contrary presented here seem only to point to examples where the presence of academic sources is a moot concern or wouldn't be addressed the general judgment of reliability, but rather by our standards on undue weight and fringe claims. Vassyana (talk) 19:18, 25 May 2008 (UTC)

[edit] "Cause of notability"

I removed from the extremist section that any material sourced to such a group (in an article about themselves) must be relevant to their "cause of notability." I take it this means it must be related to the reason they are notable, but this is not correct. We can source any material to an extremist group in an article about themselves, bearing in mind BLP and common sense. SlimVirgin talk|edits 21:32, 24 May 2008 (UTC)

It may have been poorly worded, but I though that was a cousin of the notability-relation requirement of WP:SELFPUB. Vassyana (talk) 19:26, 25 May 2008 (UTC)
I think it's a question of common sense, and I wouldn't know how to find words that would cover every eventuality. Basically, a self-published source can be used in an article about him to say anything he wants to say about himself, notable or not (e.g. his date of birth, childhood experiences), but clearly there are limits. If he wants us to give a running commentary on his life on WP just because he does it on his own website (and several SPSs have expected us to do this in the past), then we'd obviously draw the line. SlimVirgin talk|edits 19:54, 28 May 2008 (UTC)

[edit] Reliability and Form

I understand the views of Brimba and other re. the tautology of "hindsight", but this word was chosen because the sentence was written from the POV of history. The question is of reliable overview as against reliable detail. I was saying, sources that give a reliable overview have to be preferred when structuring the article so as to avoid "undue weight" and "personal synthesis". In the case where only one source gives a full overview, until such another source comes along, this does not constitute undue weight in itself.

Otherwise, if we were in 1700, one might write an article on Gravity using Galileo and Newton and be told; "you are unfairly stressing these two, whereas here are 500 sermons and 500 broadsides that say that they are in error. What are two against a thousand? I demand this article be structured according to Thomas Aquinas". And I would say; produce the sermon that gives a reasoned overview, because these are the only sources that can be considered as overall structural sources, otherwise all you get is "500 churchmen railed and 500 balladeers scoffed" at the end.

So it's the objective overview that matters, and this could PREcede the non-exhaustive comments, though in historical studies it is likely to trail after the news, hence "hindsight". Furthermore, when a subject HAS been given some later highly-critical overview, the argument should be stated fairly and fully before it is contradicted (apart from a brief trailer in the lede, maybe). Redheylin (talk) 00:17, 25 May 2008 (UTC)

Take this for example;

You're free to advocate for Scholarpedia. I imagine we would deride most popular music there as simplistic and uninspired,

Under the criterion I mentioned above this could not happen, since comments about "most popular music" would be downgraded. Unless the scholar in question has taken the time to give a detailed criticism, in which case that scholar should indeed be heard. Not that there are not positive academic treatments of popular music....

discuss social problems in terms of oppression of the majority culture,

AHA! The editor thinks scholars have an automatic left-wing bias. Well, there's a strong correlation between liberal values and intelligence, that's so..... but given that it IS intelligence that brings this about, is it "bias"?

remove all reference to causation as a natural phenomenon because it has no sound basis in metaphysics,

Yes that's right, causation is not a natural phenomenon, but it is a description that is accepted as valid by most scholars, as can be readily confirmed.

and mention that Jesus will save the oppressed in all the articles about dictatorships.

sorry?

This is Wikipedia, however. The practice is to find the best reliable sources, not the academic source in every case. I see no consensus for a hierarchy of reliability in favor of the ivory tower, but I am aware of some misconceptions on the point.

I think the idea is that scholars have to be able to defend their statements before their peers. This automatically generates any objection anyone can think of - and all that material is available for your neutral account.

Redheylin (talk) 17:47, 25 May 2008 (UTC)

I appreciate the effort. It may be useful to distill into a guideline, essay, or set of principles when peer-reviewed academic publications are at their most reliable as sources for encyclopedic claims versus when they are not. That's certainly a fertile subject. I suspect you'll find the problem a little larger than it might at first seem, which is probably why when we pass the boundaries of that which is certain and easy to describe clearly we rely on discretion and emerging consensus of dedicated editors, rather than a simple hierarchy of cardinal rules. Wikidemo (talk) 18:21, 25 May 2008 (UTC)
There are times when non-academic sources are more comprehensive. Popular music is a good example: Lewisohn's "The Beatles Complete Recording Sessions" is non-academic but still recognised as authoritative. But 99 times out of a hundred the most recognised authority will be academic, I think.

Redheylin (talk) 03:56, 26 May 2008 (UTC)

I quite agree, and that appears to be the general approach across the 'pedia. --Relata refero (disp.) 12:00, 26 May 2008 (UTC)
Note Lewisohn's book is widely cited in the academic literature as well, including in articles in the peer-reviewed journals Notes, Popular Music, The Online Journal of the Society for Music Theory, the Journal of Musicology, the Musical Quarterly, Popular Music and Society, and the Journal of Popular Music Studies, none of which are likely to deride popular music as "simplistic and uninspired". --Relata refero (disp.) 12:07, 26 May 2008 (UTC)




[edit] Using search links as references

I remember reading somewhere that a search link should not be used as a reference, either to "prove" something from results or lack thereof or simply as a shortcut to multiple sources. But I can't find this policy or guideline now. Does anyone know where I can find this rule or discussion of the unreliability of using a search link as a source? Wednesday Next (talk) 22:22, 27 May 2008 (UTC)

WP:ELNO, currently item 9: Links to the results pages of search engines, Search aggregators, or RSS feeds. -- Boracay Bill (talk) 04:05, 28 May 2008 (UTC)
Then there exists a wholesale breaking of the rules on Wikipedia relating to this, and to other matters mentioned in the link provided above. Big Time. When do we start deleting links found on Wikipedia to personal websites, especially personal fringe websites? They exist everywhere.Wfgh66 (talk) 22:05, 28 May 2008 (UTC)
Thank you. I knew I'd seen it somewhere. Wednesday Next (talk) 16:23, 28 May 2008 (UTC)

[edit] Wikipedia as a source

It seems to be accepted that Wikipedia cannot be used as a source. Here's one example: Wikipedia:Reliable sources/Noticeboard#Using other Wikipedia aritcles as a source. However, I can't find this rule stated anywhere. Shouldn't this rule be stated on the project page somewhere?--brewcrewer (yada, yada) 04:25, 29 May 2008 (UTC)

What you are looking for is Wikipedia:Verifiability#Self-published sources (online and paper), which is linked from Wikipedia:Reliable sources#Self-published sources. Vassyana (talk) 05:19, 29 May 2008 (UTC)
Okay, but if I may suggest it be more explicit. I'm seeing it used more frequently these days, and this is something that most editors might not realize. --brewcrewer (yada, yada) 05:40, 29 May 2008 (UTC)
The flat statement in an official English Wikipedia policy that (quoting) "Articles and posts on Wikipedia may not be used as sources" seems pretty explicit to me. -- Boracay Bill (talk) 05:55, 29 May 2008 (UTC)
Oh, okay. I was looking at the previous line "wikis, blogs, forum postings, and similar sources are largely not acceptable." But the line that you pointed out probably suffices for "explicitness". :-) --brewcrewer (yada, yada) 06:05, 29 May 2008 (UTC)

[edit] An organization as a source of information about itself or its ideas

On another talk page, Hrafn wrote:

Can you present a WP:RS for this viewpoint (reliability would entail that i) the claim wasn't self-serving (ii) was made by somebody with some expertise in the matter who (iii) does not have a track record of dishonesty)? As it stands we have a wide range of WP:RSs stating that it is creationism, and none that it isn't -- so per WP:DUE, this is the viewpoint we present. HrafnTalkStalk 03:30, 30 May 2008 (UTC)

I'm worried that he might be interpreting R.S. policy as preventing the Discovery Institute or one of its members from being used as a source for what the D.I. believes or teaches. Now I know that intelligent design is a highly contentious subject. I also am well aware that it enjoys virtually no support in the mainstream of science. I am the original author of Wikipedia's ID article, and I've followed its, er, "evolution" rather closely over the years.

I don't want us to make the mistake of misrepresenting the amount of support for ID. It has virtually none, and that's what our articles say (as they should). But there is one question I'd like to ask.

ID opponents say that ID is Creationism, and courts agree with them. Would it be consistent with WP:RS to put in one short quotation (and an external link to its source), in which D.I. member Stephen Meyer (a) denies the consensus of the scientific and legal community that ID is creationism, and (b) briefly explains what he thinks the difference is?

In a video interview, Stephen Meyer said,

  • Intelligent Design is an inference from biological data, not a deduction from religious authority. [1]

I don't want to change the article from anti-ID to pro-ID. I just want to add a minority view which is relevant, because it is the view of the article's primary subject, i.e., the Discovery Institute and its intelligent design campaigns; see Discovery Institute intelligent design campaigns. --Uncle Ed (talk) 17:02, 30 May 2008 (UTC)

It's the view of an unreliable source, which might be presented as his own view. However, there's no evidence of the significance of this cryptic statement, and indeed if it's only a statement on a video then the DI themselves don't seem to set great store by it. There's good evidence that ID proponents deny that it's creationism, they've done so in court and these statements under oath have been discussed by a reliable secondary source. This statement by Meyer does not look relevant. . . dave souza, talk 19:08, 30 May 2008 (UTC)
I would also point out that the article clearly indicates that the DI denies that ID is creationism -- to the point of launching vituperative attacks on fellow conservatives who state the majority viewpoint that it is creationism. HrafnTalkStalk 19:12, 30 May 2008 (UTC)

[edit] Censorship of Significant Information References based using Wiki Guidelines

I had been editing article on Effexor mentioned an online petition of users and their families that numbers 19,000 individuals from U.S. Canada and elsewhere. My reference to this was constantly deleted as was the link to the petition page on the basis of reliablity or as being "disreputable". I stopped the edit on a ruling which I never saw as I was ill for some time wiht a major illness. when I returned I posted comments on this and had some editors accuse me of "rants". I have since seen other editors mention in the discussion that the petition should be posted and at my last comment about censorship by some editors using the guidelines as the "doctrine" I have been blocked. I suggest that the guidelines be reviewed to enable inclusion of reference to major information such as online petitions which by their sheer numbers have validity of the experience of the writers. The reference to the petitions is not claimed to be scientific evidence but is the experience of individuals which deserves to be mentioned as it is clearly worthy of note for others who may be thinking of using this drug. The controversy on the anti-depresants continues, but my point here is that in a democratic society, there must be the possibility of mentioning information. The scientific community is not the only source of information. The peitition is verifiable by its existence for what it is.. a petition. There wsa never a claim that it was anything more than that. Freedom of expression of POV so long as it is balanced and presented truthfully and without distortion should be allowed. More care is needed to ensure that the views of certain editors that are subjective are not defended by a mask of "doctrine" in the guidelines. I therefore ask for a review of the guidelines to consider public opinion, petitions, and surveys of opinion or polls that are open and withoutprejudice should be allowed where appropriate. Szimonsays (talk) 06:43, 31 May 2008 (UTC)

If an online petition is relevant and/or notable, it will be reported by a secondary source, in which case we can cite it. If such petition has not been reported anywhere, it is not up to Wikipedia to be the first one to publish information about it. This has nothing to do with censorship, rather, it has to do with the core principles of the project (See WP:V and WP:NOT#OR) ≈ jossi ≈ (talk) 15:50, 2 June 2008 (UTC)

[edit] Usage by other sources

This text is convoluted and hard to understand as written:

How accepted, high-quality reliable sources use a given source provides evidence, positive or negative, for its reliability and reputation. The more widespread and consistent this use is, the stronger the evidence. For example, widespread citation without comment for facts is evidence of a source's reputation and reliability for similar facts, while widespread doubts about reliability weigh against it. If outside citation is the main indicator of reliability, particular care should be taken to adhere to other guidelines and policies, and to not unduly represent contentious or minority claims. The goal is to reflect established views of sources as far as we can determine them.

Can someone convert this to brillant prose? ≈ jossi ≈ (talk) 15:46, 2 June 2008 (UTC)

I will, if you can convert it into a brilliant thought. (Hint) it's clarity that lets the brilliance through. Redheylin (talk) 01:59, 3 June 2008 (UTC)
It seems to say that almost any source is allowed as long as the opposing editor can not find a high level of marital that debunks it. That’s about as counter to policy as you can get. I am not really sure what it is saying, but it needs to at least be moved to the talk page until there is a consensus as to the function it plays in this guideline, if any. Brimba (talk) 13:51, 3 June 2008 (UTC)
As the main criminal who (re)inflicted this, I am perhaps worst suited to judge or amend. Therefore I was staying away from it and waiting for comment. The main points are in the first and last sentence, and are absolutely standard practice in evaluating sources here and anywhere. Discussion above (See Heads up) led to qualifications and explanations in the middle sentences. As explained above, the idea was present in this guideline from its very first version, and is essential to it. The idea is that "almost any source" is allowed - if, and this is a big if - one can find that the source is widely cited or used or extolled, by e.g. peer-reviewed articles. As Relata refero felicitously expressed it yesterday at RS/N, this idea that "[being] widely quoted in reliable sources" means "reliable", is the Zeroth law of reliable sources. I'll fix up the language a bit later today.John Z (talk) 16:34, 3 June 2008 (UTC)

[edit] Suspected source errors

Reliable sources do commit errors. I proposed the following new section to address the common problem that editors, from their own knowledge and belief and original research, sometimes believe they have pinpointed a source error which has not been demonstrated by any other reliable sources. This often reaches clear consensus among editors, and they know they are not to publish the OR, but they do not believe that the statements (though verified to be the statements of a reliable source) are in fact true. When this happens the editors should be guided toward finessing the results of their OR in an acceptable way:

Often a consensus of editors, while performing source-based research, suspects a reliable source's claim to be a simple error, but no other reliable source demonstrates it plainly to be an error. An apparent dilemma arises, because the claim has been verified to have been published by a reliable source, but editors do not believe it to be accurate. Several solutions to the apparent tension are possible. Editors may choose to present the source's claim as is, deferring their own suspicions to the source's authority; or they may simply omit the claim, thus declining to comment on the source's reliability. They may also provide side-by-side cited statements about the silence of any other closely related reliable sources as to the suspected error, if the silence is sufficiently significant in evidencing error and if the editors do not attempt to force any conclusions on the reader.

For background examples see WT:V#"Verifiability INSTEAD OF truth" or "verifiability IN ADDITION TO truth"?, Wikipedia:Articles for deletion/SS Kiche Maru, Moneybomb#Single-day fundraising comparisons. We should not force the editors to publish a verified statement if they believe it will create undue weight in favor of what they believe to be an error. I believe the methods I proposed are a good selection of ways to handle the issue. JJB 14:22, 4 June 2008 (UTC)

I have a problem with this... how do we determine whether a source actually contains an "error"?... the only way is to compare it to what other sources say. If we have no other sources that contradict it, I think we have to take the source at face value and not allow OR to overturn it. I am also concerned that this would allow POV editors to delete a controvercial but sourced statement because they believe that "the source is wrong". We need to set a very high bar for declairing something to be in error... and OR is not the way to achieve this. Blueboar (talk) 14:56, 4 June 2008 (UTC)

I appreciate it, I agree with the high bar and the POV risks-- but it should not be an impassible bar, because reliable sources can be wrong. I'm trying to say that, beyond the option of taking the source at face value, there is also the option of not including the statement, and the option of citing the silence of other sources that might be expected to comment. See also Talk:SS Kiche Maru. This would not be applicable at all to controversial cases, because by definition controversies can find RS on both sides. If it's a fringe view that has no RS, then a consensus of editors will not be likely to arise. This is only applicable to those cases where, say, a reliable newspaper contains a number error, and a fan website mentions and is very familiar with the error and its correction, but nothing can be found beyond the unreliable and nonauthoritative fan site. Or the case mentioned before where an atlas publishes a geographical detail about a road, someone goes there and can say for certain the atlas is wrong, but there is no RS correction to cite. The idea is that when an error is obvious to a consensus of editors, then we should not force them to include the statement just because it's been "verified". How can I best say that? JJB 16:05, 4 June 2008 (UTC)

I don't think we should say it. At a minimum, something like this would need to be approved at WP:NOR before it gets placed in a guideline, but if it were raised there I would express my feeling is that we shouldn't include it. In those rare situations where there is a very consensus that the source has made an error (such as the examples you give), editors can always invoke WP:IAR to deal with it. I am very reluctant to spell out exceptions in the guideline, because I just see too many ways that they could be abused. Blueboar (talk) 16:25, 4 June 2008 (UTC)

[edit] Thoughts on a (potential) reference to a Voice of America broadcast published on YouTube?

I've been in contact with an editor of the Alan Hovhaness page. The situation as I perceive it, including guesswork, is that the editor is an honest person, contributing in good faith, who could potentially make valuable contributions... but is all but clueless as to understanding our verifiability and reliable source policies. I want to make a suggestion to him, but I first want to be sure that I'm sufficiently clued in.

The editor is a pianist of sufficient note as to have received a few mentions in The New York Times. He is associated with the Alan Hovhaness Research Centre. Based simply on guesswork and the look of their website, I have the uncomfortable feeling that this organization may be essentially the labor-of-love personal website of one or two enthusiasts. I love such websites and find them invaluable, and believe that in some cases they are completely suitable as external links, but obviously not as sources.

At the moment, he is trying to insert material into the article based on what a former neighbor of Hovhaness told him that Hovhaness said... or, as he characterizes it, "Hovhaness' own words." And at the moment he is interpreting Wikipedian objections as aspersions on his honesty, and at least wondering whether our explanations are just pretexts.

Now, Googling on "hovhaness centre" turns up, among other things, this YouTube posting. I don't understand Armenian but I'm 98% sure this is a broadcast by the Voice of America, and it appears to be an interview and feature piece about the Hovhaness Centre.

What I would like to explain to him is that such an item probably would be accepted as a reliable source, because the Voice of America qualifies as a news source, and (very important) because the appearance on VOA means that the material had to undergo independent editorial supervision. If the VOA broadcasts it, it is testimony to the material having at least minimal importance, and that it is moderately accurate as far as it goes. (And I personally would discount the likelihood of someone posting a homemade, counterfeit VOA program to YouTube... at least, on such a relatively minor topic).

Of course, the usual caveats about phrasing apply. Such a posting could support the statement "In a Voice of America interview, X reported that Alan Hovhaness said Y," but could not be used to support the unqualified statement "Y".

I don't want him to feel that "reliable sources" sets up an impossibly high bar, by showing that it's a bar that the Hovhaness Centre has already managed to jump.

I don't want him to feel that he can get his original research into Wikipedia just by getting the Hovhaness Centre to "publish" it... unless there's more to the Hovhaness Centre than I think. What he needs to do is to get his material into a VOA interview, or a newspaper story. The newspaper story wouldn't need to be a news item, it could be a feature piece, and it wouldn't have to be The New York Times... the Somerville News ("Somerville's Most Widely Read Newspaper!") would do.

Am I reasonably on base? Dpbsmith (talk) 14:56, 5 June 2008 (UTC)

Is there an Armenian WikiProject whose members might be able to help? -- SEWilco (talk) 18:45, 6 June 2008 (UTC)

[edit] Query

Another user has used this source as evidence that Nicole Ritchie "self-identifies as being black". Per WP:BLP, I would like to see a better source for this potentially controversial information. Perhaps I am being too strict. What do others think? --John (talk) 21:36, 7 June 2008 (UTC)

I am the user and I want to point out that this is an interview from E! Online. The logo may not be visible because it's from the Internet Archive but the copyright is visible at the bottom. Also, the writer "Kristin" is Kristin Dos Santos, a noted entertainment reporter. MrBlondNYC (talk) 01:01, 8 June 2008 (UTC)

[edit] citing other Encyclopedias

I found this interesting thread: Wikipedia talk:Citing sources/Archive 5#Citing other Encyclopedias.

Shouldn't this issue be covered in this guideline?

The Transhumanist    19:32, 8 June 2008 (UTC)

[edit] Unreliable until proven reliable

The onus on proving reliability should be on the user that adds the material, or claims a source is reliable, right? I've seen many users say "prove that the sources I'm adding is not reliable". But that is generally a difficult thing to do. For example, how do you "prove" someone doesn't have credentials in history? It's easy to show that someone does have credentials, by simply pointing them out (e.g. "XYZ was a professor of American history at the university of ABC").Bless sins (talk) 20:41, 8 June 2008 (UTC)

It's not always that easy... but in general I agree. Blueboar (talk) 21:42, 8 June 2008 (UTC)
I'd have to say, the idea that a source needs to be proven unreliable is rediculous and I'd sincerely hope it doesn't have to be explicated in policy. Anyone can start a website or small publication and claim to strive for fact-checking and accuracy, claim to exercise editorial review, and not actually do anything of the sort. So it should be obvious to anyone familiar with the spirit of verifiability that we presume a random new source to be unreliable whenever it makes a doubtable claim. And as for people with "credentials," it is similarly obvious that this is entirely insufficient, given the existence of cranks who have degrees from or even tenure at respected universities. The only real question is what constitutes proof of reliability. I think a good guideline is A1) Written by an expert who has previously published in a reliable source, or A2) exercises editorial/peer review, and B) that it is notable or at least cited by multiple sources of undoubted reliability. The absence of evidence of unreliability should also be required. Should that be made clearer in the guideline? Someguy1221 (talk) 22:32, 8 June 2008 (UTC)