I'd like to ask people to review this part of the policy, which I've often seen misused to keep out POV that individual editors dislike:

Questionable sources are those with a poor reputation for fact-checking. Such sources include websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions. Questionable sources should only be used in articles about themselves.

The point of this provision was to prevent edits such as the white-supremacist website, Stormfront, being used as a source in an article about ancient Egyptians. This is unfortunately a real example, and it was a regular editor who did it, not an anon passing by or a single-issue account. Clearly, where edits like these are being made, there's a need for a provision of some kind in this policy.

However, I've most often see the provision used to keep out POV that editors disapprove of. There are two separable problems with the provision:

• First, the word "extremist" is being extended to cover activist groups that some editors simply dislike. It has even been invoked to try to keep certain political research organizations from being used as sources, even about events that they themselves are directly involved in, on the grounds that the article in question is not actually about them.

• Secondly, regarding groups that really are "widely acknowledged as extremist," it is being used to stop them from being used as sources — about events they are directly involved in — in summary-style summaries in other articles. For example, someone recently tried to stop the Animal Liberation Front from being used as a source about a raid the ALF conducted, on the grounds that the article about the raid was being summarized (summary style) in another article, and that the ALF could only be used as a source in an article about the raid itself. I think when the provision is being used to stop a group or person from being allowed to say "This is what I or we did" (when the issue is genuinely a notable one), then it needs to be reviewed.

Another editor recently tried to add to RS that, even when extremist sources are quoted by reliable sources, they still can't be cited unless the article is about the extremist source. [1] [2] This would mean that Hitler could not be cited by scholarly sources except in the article about Hitler.

Is there a way we can keep the spirit of this provision, but without having the baby constantly being thrown out with the bathwater? In my view, it should be sufficient to stress that what's important is that sources be used "appropriately," (which the policy does say), but clearly it's not enough.

I'd normally suggest alternative wording myself, but I have an interest in this because I often use animal rights groups as sources on issues they themselves are involved in, so I'd like to throw this open for discussion rather than suggesting a change myself. SlimVirgin ^{(talk)(contribs)} 06:22, 18 February 2008 (UTC)

Hmm. As you're involved in a dispute on this very point right now, I'm hesitant to support any change. Plus, this is old, stable, and widely-quoted wording. The only additions I might suggest are self-characterizations or examples of a particular phenomenon. "The ALF describes itself as an organization dedicated to animal liberation" could be acceptably sourced to them on an article not directly about them. On the second point, you might quote a racist organization on racism, as an example for readers. But no specific factual claims should be sourced to such organizations. If the ALF claims to have found a monkey hanging upside down in lab, you can't report that here unless a reliable source has done so. (And PETA isn't one, incidentally.) Marskell (talk) 15:39, 18 February 2008 (UTC)

Marskell, your assumption of bad faith is really very unpleasant. I wrote above: "I'd normally suggest alternative wording myself, but I have an interest in this because I often use animal rights groups as sources on issues they themselves are involved in ..."

So what is the point of your comment that "As you're involved in a dispute on this very point right now, I'm hesitant to support any change"? You're saying that the policy must not change, even a change for the better, in case I win a dispute? That's an extraordinary thing to say. Please apologize if that's what you meant, and if it wasn't, please say what you did mean. SlimVirgin ^{(talk)(contribs)} 18:54, 18 February 2008 (UTC)

I just meant that I'm hesitant to engage with something I don't see as a pressing problem when it's obviously the result of an on-going content dispute. I didn't say policy must not change—I offered two examples that I've previously thought of. Marskell (talk) 19:37, 18 February 2008 (UTC)

It's not the result of an ongoing content dispute. The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them (an edit made by SandyGeorgia and one other editor, which may explain your presence here). Did you read what I wrote above? There have been problems with the way this provision has been interpreted almost since day one. SlimVirgin ^{(talk)(contribs)} 19:46, 18 February 2008 (UTC)

Re. "The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them", I think it isn't there. I think it was in there in total maybe a few hours. I think I was the last one to remove it [3] - talk page showed no consensus for this change, see Wikipedia talk:Reliable sources#Extremist sources. SandyGeorgia's involvement in the issue was minimal: she applied a cpedit to rokus' phrasing [4] That's all, SandyGeorgia did neither protest its removal, nor supported rokus' plea on talk after the phrase's removal.

SlimVirgin, I haven't still quite forgotten your personal attack on SandyGeorgia here, hitting gravely on others in the process. Yes, it was the late hours of a disheartening ArbCom, which I hold as an attenuating circumstance in your favour. I'm only explaining why I gave some time to explaining (1) that the phrase is no longer (and even wasn't very long) in the WP:RS guideline, and (2) it's a bit frivolous to name SandyGeorgia as a complice in something she had barely anything to do with. --Francis Schonken (talk) 22:46, 18 February 2008 (UTC)

What on earth is it with people who are forced to focus on personal comments? If you have something to say here about the extremism provision, please do. Otherwise please take the comments elsewhere. SlimVirgin ^{(talk)(contribs)} 22:52, 18 February 2008 (UTC)

What was unclear about "see Wikipedia talk:Reliable sources#Extremist sources"? --Francis Schonken (talk) 22:56, 18 February 2008 (UTC)

Marskell, In the past I've also had problems with SlimVirgin selectively editing Wikipedia's policies and then immediately turning around and using those very same edits as a cudgel in her ongoing wikipedia disputes. I strongly agree that people in the midst of a dispute should not be editing the policy pages, particularly when such edits are specifically contrived to advanced one's own personal agenda. For the sake of objectivity, this practice must be stopped. SlimVirgin, I call upon you to play fair and play by the rules. --Ryan Utt (talk) 19:39, 18 February 2008 (UTC)

Ryan Utt has spent much of his time at Wikipedia trying to post actionable libel about someone, which frequently has to be deleted, which is why he's had problems with me. Is this Stalk SlimVirgin Day or something? Do I get an award? SlimVirgin ^{(talk)(contribs)} 19:43, 18 February 2008 (UTC)

My presence here is a result of my watching the page; I'm often present here. (As for RS, I'm still with you that we should get rid of the thing; I didn't notice Sandy's edit.) And I also have no idea about any problem with Ryan Utt.

Anyway, it's very hard to disentangle other disputes from this discussion at the moment, so just ignore my signature and think of the following idea. I've mentally toyed with a section on V that could partly address what you brought up: Phrasal attribution, as I call it, (or In-sentence attribution). Basically, we could loosen the rules a little bit on questionable sources if we demanded that they are flagged in-sentence at every iteration ("X has claimed..."). Although it seems a weird example, think of holy books. We would never write "The world was created in seven days" and source it to Genesis. But we're free to write "According to the Book of Genesis, the world was created in seven days". And we can reasonably do so on pages that aren't specific to the Bible. Not that in-sentence attribution would allow for unlimited use of questionable sources—we would have to word things carefully. But it might be a way into a solution. Marskell (talk) 20:24, 18 February 2008 (UTC)

I'd say that in-text attribution is pretty much a given for anything remotely contentious. The question is how far can this be taken? Do we have a working definition (even a rough one) of "extremist"? Then once we have it, what exactly should the limitations on extremist sources be? SlimVirgin ^{(talk)(contribs)} 22:55, 18 February 2008 (UTC)

A source with a politically extreme POV that's still known as reliable by related professionals, how do we feel about that? ClaudeReigns (talk) 20:32, 18 February 2008 (UTC)

That should be fine, but it would depend on the editors and how much they disliked the POV. Do you have a specific example? SlimVirgin ^{(talk)(contribs)} 20:50, 18 February 2008 (UTC)

Some stuff on the American right comes to mind. The American Enterprise Institute. Even editorials in the WSJ. Still think in-sentence attribution is appropriate to such cases. Marskell (talk) 21:03, 18 February 2008 (UTC)

That's not what's meant by "extremist." SlimVirgin ^{(talk)(contribs)} 21:11, 18 February 2008 (UTC)

What's "extremist" depends on where you stand, and where you stand depends, in part, on where you sit. Some people would say that being published in the Wall Street Journal is proof positive of being not "extremist", but solidly within the mainstream. The government in Washington regards the regime in Tehran as extremist, and the government in Tehran may well think the same of the regime in Washington. --FOo (talk) 21:07, 18 February 2008 (UTC)

Yes of course the Wall Street Journal is mainstream. The problem with this provision is being demonstrated in this discussion. By extremist sources, we really did mean things like Stormfront, which was given as an example when the provision was written. At some point that example was removed, and so now people are calling anything they strongly dislike "extremist." SlimVirgin ^{(talk)(contribs)} 21:11, 18 February 2008 (UTC)

I've had occasion to mention before that the intent of the writer of policy and how it is eventually used are not usually the same thing here. This shouldn't come as a surprise either, since its usually the same in the real world.... Relata refero (talk) 09:34, 19 February 2008 (UTC)

I wasn't suggesting the WSJ is an extremist source. I was suggesting that many view it as "A source with a politically extreme POV that's still known as reliable by related professionals." I was just replying to Claude. Marskell (talk) 21:17, 18 February 2008 (UTC)

That bit about Stormfront is hilarious. I was thinking about quoting Zahi Hawass in an article about the plight of the white American male. Kidding! No, the source I was thinking of was the Executive Intelligence Review. Totally POV, but has some reputation for reliability. ClaudeReigns (talk) 06:34, 19 February 2008 (UTC)

Clarification: The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them (an edit made by SandyGeorgia and one other editor, which may explain your presence here). Please stop these kinds of accusations. Regarding my edit at RS, I altered the wording inserted by the previous editor simply because something I thought was unintelligible English was inserted at RS and popped up on my watchlist, and I attempted to make sense of what the editor was trying to say. I didn't take a position on the edit one way or another. SandyGeorgia (Talk) 21:28, 18 February 2008 (UTC)

Sandy, honestly, we don't need this across multiple threads. I'm going to click the red X on my browser now. As I said to Slim, we can't all realistically ignore each other because we all edit and watch policy, but, today at least, we should stop talking. I posted suggestions above that I mean earnestly, but will bring them up again later. Marskell (talk) 21:34, 18 February 2008 (UTC)

(edit conflict) It's not an accusation, just a statement of fact. Someone made a very controversial edit to the guideline, but rather than remove it, you fixed the writing of it, so I assumed you must agree with it, at least in part. Otherwise it's hard to see why you would have left it in. SlimVirgin ^{(talk)(contribs)} 21:37, 18 February 2008 (UTC)

Your mistake was assuming anything. Discussion should continue elsewhere. SandyGeorgia (Talk) 21:47, 18 February 2008 (UTC)

Anyway, I explained the WP:RS incident, from what I knew about it above. --Francis Schonken (talk) 22:48, 18 February 2008 (UTC)

An accurate summary; I took no position on that edit except that something unintelligible had been added to a guideline, which would confuse anyone reading it until it could be resolved. It was later removed by others, so there was no reason for me to continue commenting. SandyGeorgia (Talk) 22:52, 18 February 2008 (UTC)

I have often thought that this entire section should be scrapped, because there's nothing here that careful use of WP:NPOV won't take care of, but it seems to be widely used to keep articles mainstream, and most things that serve that purpose should be retained. Relata refero (talk) 09:34, 19 February 2008 (UTC)

I don't think we can scrap it because it does keep a lot of nonsense out. But I think we need to consider what it means and how to advise that it be applied consistently. SlimVirgin ^{(talk)(contribs)}

Oh, and SV, you could have warned me that you'd changed your mind so I could keep an eye on this page. Not fair! Relata refero (talk) 09:47, 19 February 2008 (UTC)

Yes, and if you see what I wrote there, our advice to use sources "appropriately" really should be enough. But increasingly I'm seeing people stick rigidly to the letter of this policy, and not the spirit, so it means we have to make sure we understand the implications of what we write. SlimVirgin ^{(talk)(contribs)} 13:42, 19 February 2008 (UTC)

Arbitrary break

The text SlimVirgin has highlighted used to say

Questionable sources are those with a poor reputation for fact-checking. Questionable sources should only be used in articles about themselves. (See below.) Articles about such sources should not repeat any contentious claims the source has made about third parties, unless those claims have also been published by reliable sources.

The example of extremist websites was added as a port from WP:ATT. The added text is similar to that which has remained fairly stable since ATT was created in Oct 2006 by SlimVirgin.

Is the ALF "widely acknowledge to be extremist"? Some people think so (see Animal Liberation Front#Listing as a terrorist threat). Balaclavas—check. Destruction of property—check. Disregard for the law—check. Self-published video propaganda—check. Mainstream front organisation—check. But I guess we'll have to agree to disagree on the extremist aspect. Extremism is only one of several possible characteristics of a questionable source. The examples given aren't intended to be complete, but one other example is sources that are "promotional in nature". An activist campaign group typically produces material that is highly promotional (effectively an advert, but probably closer to a Party Political Broadcast without the regulatory framework). As such, it is a questionable source.

In a recent edit SlimVirgin almost cited this section when saying "The ALF may be used as a source in articles about its own actions, per WP:V." Their "actions" is a slight alteration of "about themselves" which is harmess when considered a subset of "about themselves" but not when their "actions" is extended to involve third parties.

The section of Animal testing that recently provoked this debate declares itself to be a summary section of Britches (monkey). But that article is not about the ALF—it is about a monkey, the claims made about its treatment, and commentary on this. WP:V makes it clear that questionable sources may not be used in claims about "third parties". The monkey and the lab where is was found is a third party. In addition, such sources may only be used if the claim "is not contentious". Apparently, the university disagrees with some aspects of the claims and video footage (see SV's edit above). Finally, such sources may only be used when "there is no reasonable doubt as to who authored it". The ALF video is anonymous and likely to remain so. Frankly, I prefer not to get my reliable information from an anonymous bod in a balaclava on some internet video.

So whether or not you think WP:SS extends the domain of "articles about themselves" to include summary sections, this particular questionable source is being misused. I see no reason to change WP:V in order to make this misuse acceptable. Colin°^Talk 14:02, 19 February 2008 (UTC)

Colin, I don't think the "extremist" part was moved from ATT to here. I think it was moved from RS, though it may have gone via ATT at some point.

There are two issues here. The first is when groups widely acknowledged to be extremist may be used as sources, and you're right that the ALF is such a group. Certainly, they may be used as sources about themselves in articles about them, and obviously that includes their actions, because they don't exist without their actions.

One question is whether that extends to summary-style summaries of articles about them in other articles.

But that's just one issue. The more pressing issue is what we mean by "widely acknowledged as extremist." For at least a couple of years, I've seen this provision invoked many times to keep out sources with a strong POV that editors don't like. For example, it has been used to try to keep out material from Political Research Associates, the Southern Poverty Law Center, and the Anti-defamation League. But having a strong POV obviously shouldn't be equated with "widely acknowledged as extremist." So the question is how we pin down what we mean by the phrase. SlimVirgin ^{(talk)(contribs)} 14:13, 19 February 2008 (UTC)

This is where the "extremist" provision was added to this policy, three months ago. My memory is that it was originally in RS, but was removed, but I'll need to check the history. SlimVirgin ^{(talk)(contribs)} 14:16, 19 February 2008 (UTC)

I added it to RS in August 2005. It said: "However, that a source has strong views is not necessarily a reason not to use it, although editors should avoid using political groups with widely acknowledged extremist views, like Stormfront or the Socialist Workers' Party. Groups like these may be used as primary sources only i.e. as sources about themselves, and even then with caution and sparingly. [5]

So the point was that extremist sources may only be used as primary sources. SlimVirgin ^{(talk)(contribs)} 14:25, 19 February 2008 (UTC)

Let's not muddy the water with "primary sources", of which everyone seems to have their own definition. We're dealing with "questionable sources" here, of which extremist sites/organisations is only one example. BTW: your diff on WP:V is the same one as mine, where the author's edit summary is "Added line ported from WP:ATT". Oh, and I think you mean 2005, not 1985!!

LOL!! Sorry. SlimVirgin ^{(talk)(contribs)} 15:04, 19 February 2008 (UTC)

Regardless of its origins, it has stood the test of time.

Nothing here is black and white. There's degrees of extremism and degrees of questionableness. At some point we draw the line and say "you can use that to say 'XYZ claims to be' or 'XYZ claims to have done' but no more". Even something as innocuous as a Head & Shoulders advert couldn't be used to back up a statement that dandruff was easy to remove.

You haven't responded to my point that "their actions" don't extend to claims they make about third parties (the lab and the animals). This is important as it frames the extent to which this policy statement applies. In this regard, I'm citing the section "Self-published and questionable sources in articles about themselves" which is an extension of the paragraph under discussion.

Their actions have to extend to the labs, obviously. The ALF is an activist group. They are their actions. And obviously if they are doing something, they are doing it to a someone or in a somewhere. The point is that they are used to tell us what they did there, not what other people did. SlimVirgin ^{(talk)(contribs)} 15:06, 19 February 2008 (UTC)

I don't think it is "obvious" and it is important. You say the sources are not being used to tell us "what other people did." But they are. They are being used to tell us what scientists did to a monkey in the name of research. The monkey, its condition and the experiment are neither "the ALF" nor "their actions". Colin°^Talk 15:29, 19 February 2008 (UTC)

I think we can be flexible on a "good faith" summary section. One problem is that it is too easy to slap a "main article" template over a paragraph or section. In general, I don't view WP:SS as anything other than a "how to decide what goes where" guideline; it certainly doesn't allow for violations of policy (such as the not infrequent claim that such sections are exempt from the need for citations).

I'm less concerned with establishing a bullet proof definition of "extremist". The closer an organisation or source gets to extremism, fundamentalism or out-and-out propaganda/promotion, the less its utterances should be regarded as sound. If a subject is notable enough to be discussed here then good reliable people will have already discussed it and we can cite them. Having to cite stuff "on the edge" is really a sign that ones argument is either fringe or ignored, and either way, WP doesn't need it. Colin°^Talk 15:01, 19 February 2008 (UTC)

I think you're mixing up different issues. The ALF is only used as a source on issues that mainstream sources have written about. But you keep coming back to only that example, and the rhetoric is getting in the way of the argument.

I come back to this issue since it is what (by your admission) provoked the discussion, and there remain unanswered disagreements over policy interpretation that make it a useful example. But we can move on if you like. Colin°^Talk 15:29, 19 February 2008 (UTC)

The point is that, as it stands, the provision could be interpreted to mean that Adolf Hitler cannot be cited in any article that isn't about him. So in Antisemitism, we cannot, according to this policy, tell the reader what Adolf Hitler said in Mein Kampf about Jews. So we do need to clarify what we mean without diluting the spirit of the provision. SlimVirgin ^{(talk)(contribs)} 15:10, 19 February 2008 (UTC)

I agree that the "in articles about themselves" restriction could get in the way of reasonable use. At this moment, I can't think of an alternative. My suggestion: we start a fresh new talk section dealing with the issue of "restricting questionable sources to articles about themselves". It can then be discussed without the distraction of definitions of extremism, or any current disputes. Colin°^Talk 15:29, 19 February 2008 (UTC)

Good idea. SlimVirgin ^{(talk)(contribs)} 15:41, 19 February 2008 (UTC)

I am, unfortunately, just passing through and I am not going to have time to follow this discussion properly. However what I have read so far of this discussion brings one particular case to mind that some of you might find instructive. I would like to share it with you in case anyone finds it useful or wants to discuss it:

David Irving is not only an extremist; he is also known to be a very dishonest historian. Nevertheless there is one aspect of his work that has been praised by respectable historians, namely his grasp of German troop movements during WWII. To what extent might it be appropriate to cite someone like David Irving in a Wikipedia article? ireneshusband (talk) 04:00, 20 February 2008 (UTC)

Troop movements, of course. ClaudeReigns (talk) 07:49, 20 February 2008 (UTC)

Only if there are other sources that confirm it. He is so untrustworthy, I think he requires more verification. OrangeMarlin ^{Talk• Contributions} 17:50, 22 February 2008 (UTC)

The point I would like to make is on articles like global warming where articles which question the evidence are immediately labelled as "extremist" and "unreliable" by a vociferous group who maintain a stanchly pro-warming POV and effectively block out any balance. Now I say this as someone who at the time was a believer in global warming who tried to put a single link to a related article on oil running out. Unfortunately the result of this one-sided dominance is that the article on global warming is laughable and seriously undermines the credibility of wikipedia as a source. From my experience on that article, trying to be a neutral judge between the two sides, I am quite convinced that groups of people gang up and use both the rules of wikipedia and their admin status to ensure very POV articles. One huge problem is that one side can present largely coincidental "evidence" which suggests something, and then block any article which questions whether the evidence is coincidental saying it is merely "opinion". Thus huge number of reports of possibly entirely coincidental events can be quoted by one side, whilst those articles expressing proper scientific sceptism are blocked. Bugsy (talk) 10:44, 20 February 2008 (UTC)

Suggestion

We might consider consigning all discussion of what is a good/bad source that happens in this policy to WP:RS. WP:V should defer to WP:RS, WP:NPOV, WP:FRINGE, etc. in determining what qualifies as a good source. WP:V should only be about being able to verify points for inclusion. A fact can be unverifiable for a lot of reasons, sourcing being only one issue. ScienceApologist (talk) 19:18, 20 February 2008 (UTC)

WP:DIS implies that WP:VER covers misrepresenting the contents of a source WP:DIS/Definition of disruptive editing and editors/ :"Cannot satisfy Wikipedia:Verifiability; fails to cite sources, cites unencyclopedic sources, misrepresents reliable sources, or manufactures original research."

However I can find no direct statement on WP:VER adressing "misrepresention of sources,".

This may be why one experienced editor advised me, "In the talk page link you give above, you repeatedly say things like "This discusion is about accurately reflecting the content of sources." This is a weak argument, because it's not directly supported by policy and because one individual doesn't get to dictate what a whole discussion is about."

At present policy is seen to be blank on accurately representing the content of a source. And editors are using this blank space to disrupt and destroy WP. (What confidence can you have in an article in which sources may or may not have been accurately represented? And how can you work collaboratively in such an environment?)

My attempts to understand how to proceed in a real life case of sources being repeatedly misrepresented can be seen at [Wikipedia:Editor_assistance: Editor repeatedly misrepresents content of source on BLP] and my attempts to understand why this is being classified as "a content dispute" by some can be seen on my talk page. I find it disturbing that a basic like "accurately representing a source" needs to be included in policy. However that seems to be the case. SmithBlue (talk) 05:00, 22 February 2008 (UTC)

See WP:NOR#Using sources. Relata refero (talk) 12:08, 23 February 2008 (UTC)

Various wordings

<div class="metadata divbox divbox-red" title="After edit by Jack-A-Roe (talk · contribs)" >

Because this is the English Wikipedia, for the convenience of our readers, English-language sources should be used in preference to foreign-language sources, assuming the availability of an English-language source of equal quality, so that readers can easily verify that the source material has been used correctly.

Because translations are subject to error, whether performed by a Wikipedia editor or a professional translator, readers should have the opportunity to check the original material for themselves. Therefore, when the original material is in a language other than English:

• Translations published by a reliable source are generally preferred to editors performing their own translations.
• Where editors use their own English translation of a non-English source as a quote in an article, there should be clear citation, including a direct quotation of the relevant portion of the foreign-language original, so that readers can check what the original source said and the accuracy of the translation. The direct quotation can appear in the footnote citation to avoid cluttering the article text.
• Where editors use a paraphrase of a non-English reference in an article, or use a non-English source to support a general statement of fact, a direct quotation of the relevant source text should appear in the footnote citation.

<div class="metadata divbox divbox-blue" title="Proposal by CaliforniaAliBaba (talk · contribs)" >

Because this is the English Wikipedia, for the convenience of our readers, editors should use English-language sources in preference to sources in other languages, assuming the availability of an English-language source of sufficient quality, so that readers can easily verify that the source material has been used correctly.

Where editors make assertions of fact based on non-English sources, they must ensure that readers can verify for themselves the content of the original source, the reliability of its author/publisher, and its agreement with those assertions.

Translations of direct quotes from non-English sources given in the body of an article should be accompanied by a footnote giving the original text. Translations published by reliable sources are preferred to translations performed by Wikipedians.

The following is a separate issue outside of the above proposal: cab (talk) 11:24, 19 February 2008 (UTC)

• Though it is not common practise on Wikipedia to support every paraphrase or general statement of fact with a direct quote from the source, other editors may request such a quote using the {{check}} tag; in that case, the editor who originally added the citation should provide the direct quote, to aid verification by users who do not have access to the source in question.

Non-English sources discussion

I don't see any discussion or consensus for this edit [6] which changed the wording of the section in question per above.

I fail to see why we should be flooding Wikipedia with masses of unfree text on tens of thousands of articles on non-Anglosphere authors, movies, musicians, political parties, etc. as would be required by this way of doing things, nor do I understand what benefit it brings; if you can read the language of the non-English source, you can certainly find the quote which supports a certain sentence (except in contentious cases), and if you can't read the language of the source, then a quote doesn't help you anyway. cab (talk) 04:25, 18 February 2008 (UTC)

I'd suggest the following above wording instead. Comments? cab (talk) 04:33, 18 February 2008 (UTC) (modified cab (talk) 04:13, 19 February 2008 (UTC)) (refactored cab (talk) 09:08, 19 February 2008 (UTC))

if you can't read the language of the source, then a quote doesn't help you anyway Well, on the other hand, it could help the person reading it to ask a trusted source whether the translation is true to the original text, or at least find out with the help of an online translator. Not many English-speaking Wikipedians will necessarily have access to non-English language sources. If #2 asks editors to provide an original quotation in a footnote, I don't see why not do it for #3 as well. The English citation for the latter would be a paraphrase after all, not a word-for-word translation. — Zerida ☥ 05:46, 18 February 2008 (UTC)

Many Wikipedians won't have access to just about any kind of source you can think of, which has nothing to do with the language, but the location of publication, the cost of obtaining it, etc. But we don't require that quotes be given in those cases. Don't see what's so special about non-English sources that requires us to pollute articles with a lot of unfree text to support non-contentious statements, especially where the original source is online in the first place, but also in general when we're talking about a recently published book or periodical you could get through inter-library loan. cab (talk) 01:34, 19 February 2008 (UTC)

At least in the first case they don't have much of an excuse, but it's not uncommon for editors to request quotations on the talk page for especially inaccessible sources. The ^{[verification needed]} tag is for this. Of course, if the non-English source is available online, then a link is all that's needed in the footnote. I personally do that quite frequently. If not accessible online, I do think it's a nice addition to have a small quote for the translation or the paraphrased one. The majority of cases will probably end up without a quotation, but I feel that we need a specification that requires editors to supply a non-English quotation if requested. — Zerida ☥ 03:43, 19 February 2008 (UTC)

Well sure, but as you say, we already have a way for people to request a direct quotation in the cases where it's desired. I changed my suggested text to emphasise that some editors might ask for a direct quotation to support a paraphrase, but I don't see that it needs to be made pre-emptively mandatory in the majority of cases where no one actually asked for it. cab (talk) 04:13, 19 February 2008 (UTC)

I don't think I can agree that "it is normally not necessary". If WP:V is devised to make information on Wiki verifiable, I'd say we need more stipulation to ensure that translated text and paraphrases of translated text are indeed verifiable. Editors for example should be able to remove any unverified paraphrases of translated text after a request for a quotation has been made and not answered. — Zerida ☥ 04:35, 19 February 2008 (UTC)

I changed this to "It is not common practise on Wikipedia", which I think can hardly be contested. (Count how many times the "quote" parameter ever gets used on all the citation templates.) Translated text is verifiable by an uninvolved person in the relevent babel category. I still don't see that the issue of quotation is specific to non-English sources; all these argument apply equally to the idea that we need more stipulation in general to ensure material is verifiable even when sources disappear. It's not just non-English websites which are increasingly opting out of web archiving and whatnot. cab (talk) 05:08, 19 February 2008 (UTC)

Don't forget to account for how hard it can be to find a replacement source for a non-English site when a weblink goes dead if you don't have some of the original wording. Just a thought. SandyGeorgia (Talk) 04:00, 19 February 2008 (UTC)

Internet Archive will still cache older copies of non-English web sites, but you're right, and another reason to have those quotes handy. — Zerida ☥ 04:12, 19 February 2008 (UTC)

More and more sources (particularly news sources) are opting out of the internet archive with robots that prevent archival. When I have to search for something in Spanish, even though I speak Spanish, it's very hard for me to find text if I don't have original wording. SandyGeorgia (Talk) 04:15, 19 February 2008 (UTC)

Why is this any different from the case of English websites going dead? Bikasuishin (talk) 10:32, 19 February 2008 (UTC)

Oppose because of systemic bias. --Kaypoh (talk) 08:23, 19 February 2008 (UTC)

Er, we're trying to come up with a policy, not do a straw poll yet. Do you have any suggestions about how the section in question should be worded? cab (talk) 08:58, 19 February 2008 (UTC)

If you discourage and make it more difficult to use sources in other languages, you are making the problem of systemic bias worse, because there are many good sources in other languages, about things in other countries. So I oppose this. --Kaypoh (talk) 10:10, 19 February 2008 (UTC)

I concur. We certainly don't need to make it harder for editors to contribute to articles about non-anglophone cultures or to bring a non-anglophone perspective to existing articles. I don't see any problem with the original wording. Bikasuishin (talk) 10:28, 19 February 2008 (UTC)

The existing language is quite restrictive about how sources in other languages should be used. I proposed loosening it. You oppose that too? Do you have an alternative proposal? Simply saying "oppose", "oppose" means we don't get anywhere and it remains the way it is. cab (talk) 10:27, 19 February 2008 (UTC)

I assume Kaypoh meant "revert" rather than "oppose", and I happen to agree. It's not necessary to specifically point at the {{check}} tag for non-English sources. Bikasuishin (talk) 10:31, 19 February 2008 (UTC)

Okay; but personally, I found the original (green) wording just as bad because it kept going on and on about "translation". 99% of the time we're referencing non-English sources, it's not to give a direct quote, it's to support a normal sentence like "so and so went to ABC School and graduated in 1999". So when some people mistakenly believe non-English sources aren't allowed at all, I point them to that policy section, and then they think it means that we have to translate every non-English source we use. cab (talk) 11:09, 19 February 2008 (UTC)

Fair enough. I like your amended version better. Bikasuishin (talk) 11:57, 19 February 2008 (UTC)

I'm inclined to agree with cab that there's no fundamental difference between english & other sources. Other things being equal, we prefer more accessible to less accessible sources. This includes both language & rarity.

Another point here. Some foreign languages use quite exotic characters, which some people may have difficulty inputting on the computers they use. Peter jackson (talk) 11:29, 19 February 2008 (UTC)

I prefer cab's version. It is more feasible, I think, even still tedious. Practically the proposed (and now argued) version seems to make it impossible any translation from other language Wikipedias, I'm afraid (already the content of the article was paraphrased, without citation of the original, so the requirement cannot be fulfilled). --Aphaia (talk) 00:59, 24 February 2008 (UTC)

I have made the change described above [7]. Please give some alternative language if you have any objections. Thanks, cab (talk) 03:00, 28 February 2008 (UTC)

The whole point of the revisions I am trying to make is to avoid mixing two issues: normal inline citation of non-English sources, and giving translations of direct quotes as part of the article body (a tiny minority of all uses). I can quote real examples of people demanding translations and direct quotations from sources to support regular, uncontroversial article content if you don't believe this is confusing. cab (talk) 08:19, 29 February 2008 (UTC)

I still maintain that this shouldn't be viewed as a problem. If you can access a source that others who don't speak the language cannot, editors should be able to verify the material through a quotation, at the very least on the talk page when requested. This allows editors to remove unverified material, particularly when it involves muddier issues like BLP. The question of what constitutes "controversial" can vary in these cases. I do object to the last modification you made, which is a significant and sudden alteration from the initial proposal that had received consensus. The last version suggests that editors are not really allowed to translate material from non-English sources, which is not policy, and that they can basically ignore requests for verification if they simply include a footnote (unless, I guess, it's on the internet). Frankly, I think the policy is better off not saying anything about non-English sources than leaving it open to these kinds of interpretations. However, WP:V assumes a degree of reliability that would not necessarily apply to sources from different countries, and often don't, so we do need something that covers non-English sources and still make them verifiable. — Zerida ☥ 23:28, 29 February 2008 (UTC)

I agree with Zerida et alli. To me, this is all a matter of "style" or "citation design" and ongoing discussion- a key point. In books on history, one has to provide their sources and that is the way it should be done academically. Many times, direct translation of the original text(s) are not given as it would make for one ugly, confusing textbook. Usually, one cannot directly discuss the citation with the author, but only try to write an opposing viewpoint to deaf ears. If the guideline is there and a time where questionable content arises, the rule can then be summoned up to solve the matter. If the resource seems good or a reasonable attempt at a resource and some citation is provided and there is no debate, then let it be (I believe this happens naturally on the Wiki). Once the resource is cited, it is usually up to the reader to track it down- at their own time and "expense". If the reader does this, finds that the article has misquoted or misled, then it can be debated. If a reader has opposing content in English, then the foreign language and English versions can somehow be meshed or we can have a famous edit war (which is interesting, although lame). The intent of the guidelines should be just that- guidelines. That seems to work best because on Wikipedia we can edit, change, author, expand and fact debate in "real time". If a resource is called into question, or debated...it is reasonable that a translation of the resource to be given in order to back it up. The wording in the original article is rather vague and it does seem that everything has to be translated- maybe that could be cleaned up. Maybe make it "if facts are debated" or something similar. —Preceding unsigned comment added by Mekugi (talk • contribs) 10:39, 9 March 2008

In a recent blurb in Fox News it was noted that a company called Fastcase has inaugurated the Public Library of Law ( www.plol.org ), which is a free online legal research site. Moreover, Harvard University will soon begin posting on the internet, free of charge, research and articles produced by faculty. Harvard Office of Scholarly Communication to handle this project and Harvard expects the technical work to set this up to be completed by April 1st . David Shankbone 15:53, 20 February 2008 (UTC)

Not impressed. It will take decades for such databases to catch up with the level of sophistication and depth of private databases such as LexisNexis and Westlaw. Every law school student soon grows accustomed to the convenience of accessing the Lexis ALLCASES file and in less than 20 seconds, running a single text query across every American case ever published, from the early 1700s to yesterday evening. Plus there are services like Shepard's Citators and KeyCite which the open-source databases simply can't match yet. --Coolcaesar (talk) 04:46, 25 February 2008 (UTC)

As a legal database it is not much to write home about... but as a research database for use on Wikipeida it is. There has been a lot of debate as to whether Lexus/Nexis and Westlaw are to be considered reliable sources (due to the fact that you need to pay to access them). There will be no doubts about this site. Blueboar (talk) 05:03, 25 February 2008 (UTC)

Of course they were RSs--inconvenient though they were--but it is naturally preferable to use a free source when it is available, and we should at some point start tryingto change the existing links. There are printed sources also, & most large public libraries have them for the US and their state. shepherds is still useful in print if you dont need last weeks.

As for Harvard, accumulating significant content will take years and years--and from a single university. Alternative approaches, like pubmedcentral, make everything in the subject available, though sometimes after a time delay. DGG (talk) 20:23, 26 February 2008 (UTC)

The current policy says:

self-published books, newsletters, personal websites, open wikis, blogs, forum postings, and similar sources are largely not acceptable.

Does that include archives of mailing lists?

Recently i have been involved in several notability and verifiability discussions about constructed languages. If you are curious, see:

• Wikipedia:Articles for deletion/Kēlen (2nd nomination) (delete)
• Wikipedia:Articles for deletion/Slovio (2nd nomination) (delete)
• Wikipedia:Articles for deletion/Lingua Franca Nova (no consensus)

A repeating argument in these discussions is that constructed languages is a modern art form, and that it exists mostly on mailing lists and therefore mailing lists are acceptable sources for establishing notability and verifiability.

I claim that mailing list archives are not very different from blogs, forums and open wikis, at least from the technical point of view. A mailing list can be setup for free and content can be submitted by essentially anyone; usually such mailing list aren't even moderated as Wikipedia is. Also the count of users of the mailing list is often used to demonstrate notability of a subject; i claim that this is not enough, because the mere count of users can be exaggerated by the use of socks, spam bots, etc. (I am not saying that proponents of conlanging do that on purpose. Spam bots don't care which mailing list they spam, and socks can be created by trolls that don't represent the serious people involved with the language.)

This topic is related not just to constructed languages. A mailing list can be a source of information about a free software project. That's the case with Perl 6 and a number of related articles. This is a topic in which i have a lot of interest; it is notable, because it has also been covered by many other mainstream publication, but the Perl 6 mailing lists are used as sources for the information in the article; doesn't that contradict the policy here?

I don't think that cold, hard rules can be established about this, but in any case mailing lists should be mentioned as a problematic source. --Amir E. Aharoni (talk) 16:22, 19 February 2008 (UTC)

Unless they can be independantly verified... I would not consider them reliable. Blueboar (talk) 16:52, 19 February 2008 (UTC)

I agree in general, but this is a problem that keeps cropping up, namely where most of a certain topic is discussed in media we find unacceptable e.g. mailing lists, personal websites, or Usenet. I think at some point we're going to have to grasp the nettle, because what's happening is that people are ignoring this policy and just getting on with writing articles. The policy does need to be descriptive as well as prescriptive. SlimVirgin ^{(talk)(contribs)} 18:27, 19 February 2008 (UTC)

(edit conflict)I don't want to say this, but it is my personal opinion that certain mailing lists, like certain moderated usent groups, are reliable sources for certain issues. This is a long-standing problem in SF-related articles. Relata refero (talk) 18:31, 19 February 2008 (UTC)

I think it's a long-standing problem in several areas, particularly pop culture, but also in Scientology, where a lot of the criticism is or was on Usenet. When we were writing ATT, we tried to add an exception to the normal rules for certain subjects, specifically for those that tend only to be discussed in sources we would otherwise regard as unreliable. I tried to introduce it here once too, but was opposed. I do think it's an area we need to address. SlimVirgin ^{(talk)(contribs)} 18:45, 19 February 2008 (UTC)

I just looked up the discussion in case it's helpful here. It's at Wikipedia talk:Attribution/popculture. The version of the exception clause that I preferred was something like:

Some articles on popular culture and fiction rely on self-published materials on bulletin boards, blogs, Usenet, and fansites, because few other sources exist for them. In such cases, the material used must have been posted by named, or well-known pseudonymous, individuals with a known expertise in the area, although not necessarily professional expertise. Anonymous posts should never be used. If in doubt about how to use a source in this area, consult the relevant WikiProjects for advice. Subject areas well-covered by reliable sources, such as science, medicine, history, politics, and current affairs, are excluded from this provision.

Every time I've tried to raise this, people have squealed for fear of opening the floodgates, and I can understand that, but I still think we need something like it. SlimVirgin ^{(talk)(contribs)} 18:54, 19 February 2008 (UTC)

Why don't you start by writing up a something as an essay that we can slowly start using as a reference, to show people how it might apply? Relata refero (talk) 19:01, 19 February 2008 (UTC)

I'm not the right person to do that, because I don't edit in these areas myself, so I really don't understand the extent of what's needed. The ones who do edit those subject areas tend simply to ignore this policy, as far as I can tell. SlimVirgin ^{(talk)(contribs)} 19:53, 19 February 2008 (UTC)

Well, I have the link to the ATT discussion now, so I can link to the arguments there if something comes up, and we can see what develops. Relata refero (talk) 19:57, 19 February 2008 (UTC)

This FAC provides the starting point for crafting wording for an exception. It was decided in this case that the most reliable (and professional) sources in that industry were bloggers, and that was the main medium in which the most knowledgeable experts published. This example allows discussion of professional sources, different than, say, comic books or Pokemon or whatever. SandyGeorgia (Talk) 22:03, 19 February 2008 (UTC)

I saw a lot of pop culture lists when I regularly reviewed FLCs. It is perfectly possible to write a FL without relying on self published material and I'm not aware of any FLs that did so. By allowing such material, you might as well throw away WP:V for those articles IMO. It is ridiculously easy to create a blog or geocities web site--no harder than creating a Wikipedia account. Some of these sites look really professional, until you read the About page and discover it is just a couple of students mucking about in their evenings (and who probably also edit Wikipedia). Pop culture wikiprojects don't tend to be staffed by sages and may claim XYZ is has "known expertise in this area" when in fact XYZ hasn't left high school. Our sources should be written by folk who's day job is affected by their reputation for writing reliable and professional material. Such self-published material is also highly likely to be copied off Wikipedia in the first place. That's not to say there aren't a few WP:IAR cases, as Sandy suggests. But I'm not convinced we need to seriously weaken policy in order to accommodate this stuff. Remember that "pop culture and fiction" are about 90% of Wikipedia. Colin°^Talk 23:34, 19 February 2008 (UTC)

That might mean that about 90 percent of Wikipedia is having to ignore this policy. Phil Sandifer has made some good points about this. He specializes in comic studies, and much of the source material is apparently in fanzines and the like. People have made the same point about Scientology articles -- that most of the criticism, including the amateur expert criticism, is on Usenet or self-published sites. Other examples given during earlier discussions are stamp collecting and things like quilt making. I think so long as you write into any exception that it doesn't apply to science, history, politics etc, it'd liberate the areas that need it without affecting the ones that attract the cranks. SlimVirgin ^{(talk)(contribs)} 23:44, 19 February 2008 (UTC)

constructed languages is a modern art form, and that it exists mostly on mailing lists and therefore mailing lists are acceptable sources for establishing notability and verifiability. - What concerns me about this statement is that it comes perilously close to original research - that is, using mailing lists as primary sources, or, if you will, as examples to prove a particular argument. But we have no idea if any particular mailing list is widely read, or the extent to which a particular posting is truly representative. That's quite different than saying "Oh, widely recognized expert X said ABC on this mailing list". In the latter case, this is close to what we allow now - if a blog contains a posting by an expert on a given topic, then it's an acceptable source. -- John Broughton (♫♫) 02:47, 20 February 2008 (UTC)

This issue is perennially raised and I agree it's not going away anytime soon. However, I strongly disagree with any liberalization of sourcing requirements. There are a number of common poor practices that are not enshrined in policy (and in fact, discouraged or prohibited by policy), so I am unconvinced of arguments stating policy should conform to the more liberal sourcing practices based on the descriptive approach to policy building. (I just don't see the need for looser requirements and am not at all convinced it's a tolerable practice, let alone a good practice that should be documented in policy.) I have yet to be presented with an instance where such sources are actually needed, outside of topics such as those solely notable for being seen as notable in Usenet (which probably don't belong here). In the examples of both constructed languages and SEO above, there are plenty of standard reliable references available on such topics. For example, SEO is very widely published about in both mass market books and tech periodicals. From my perspective, the argument for greater permissibility in forum/mailing list/self-published circumstances is usually grounded in an argument that there would be no sources otherwise (in which case, we shouldn't have an article on it) or that the online sources are easier to access (an argument of which I possess a dim view). Another common argument is that proper sources don't exist (often raised for programming languages such as PHP and pop culture such as television episodes), but as I mentioned above, I have yet to encounter a situation where normal reliable sources are lacking. (Tackling my examples, programming languages have numerous books and periodicals discussing them, and even individual television episodes are covered by a wide variety of entertainment periodicals and critics.) Just some thoughts. *hands out grains of salt* Vassyana (talk) 04:17, 20 February 2008 (UTC)

I know in my part of librarianship the main information sources have long been informal mailing lists, with the formal peer reviewed publications not really being taken very seriously, except by the library school faculty who are required to publish there. I think this is the case in many technical subjects. But this does amount to a certain abandonment of the concept of peer-review and editorial control --the user is expected to evaluate the material without it. As this spreads in the world we will have to revise how we do things, or we will progressively be relying on out of date material. DGG (talk) 06:39, 20 February 2008 (UTC)

I don't necessarily believe that a trend away from print into new media will necessitate much alteration to the verifiability policy. Open journals and internet library access are on the rise. There's no noticeable trend away from the publishing=academic cred model in the public, commercial or (more importantly) academic sectors. Self-published sources from established experts (in blogs, university department websites, etc) are already held to be generally acceptable [on wiki]. At worst, we may eventually need to soften/remove the caution about using such materials from well-regarded experts. Vassyana (talk) 06:52, 20 February 2008 (UTC)

You're almost certainly mistaken. If you hop over and ask the people at Making light or the chaps at the Google features working group what should be used as reliable sources, they aren't going to agree with you at all. I don't think rigidity helps us here.

And you're also absolutely wrong that SPS from established sources are held to be "generally acceptable". I don't know which academic areas you're talking about, but its certainly not true of the ones I know well. Relata refero (talk) 07:30, 20 February 2008 (UTC)

I only meant that SPS from established experts are generally acceptable on wiki. (I'm adding a bracket note to the post to prevent further confusion.) Vassyana (talk) 17:16, 20 February 2008 (UTC)

Why wouldn't a post in a mailing list from a recognised authority on a subject be acceptable? I belong to a cartography mailing list, for instance, and although I'm no experts some of the posters are extremely well known experts in the field with loads of credentials. Shouldn't that be acceptable?--Dougweller (talk) 10:58, 20 February 2008 (UTC)

I'm not prepared or disposed to argue this point, but I have been involved in a few mailing-list and usenet newsgroup discussions where doubt emerged (in some cases not quickly) about whether persons involved in the discussions were impostors. -- Boracay Bill (talk) 11:43, 20 February 2008 (UTC)

Doug, yes, we can use self-published posts — on mailing lists or elsewhere — from recognized authorities, so long as there's no serious reason to doubt who wrote it. See WP:SPS: "Self-published material may, in some circumstances, be acceptable when produced by an established expert on the topic of the article whose work in the relevant field has previously been published by reliable third-party publications." SlimVirgin ^{(talk)(contribs)} 18:22, 20 February 2008 (UTC)

Self-published sources are acceptable to source non-contentious information ( the rest of the policy after "largely not acceptable" goes on to list exceptions to that; the Perl mailing lists you mentioned up top are a good example of when this is acceptable), but they do not establish notability. Squidfryerchef (talk) 22:34, 20 February 2008 (UTC)

I continue to have serious reservations about using mailing lists as reliable sources. To start carving out exceptions to a general rule is a very bad idea... Slippery slope and all that... I do understand that in some specific situations and articles they might be reliable and desirable... but that is why we have WP:IAR. I just don't see the need to modify the policy to account for about a few, relatively rare cases. Blueboar (talk) 22:42, 20 February 2008 (UTC)

There is no slippery slope, those kinds of sources have been used since the beginning of WP, and only in the past couple of years have they come under fire, and nobody's proposing changing the policy. We also don't "have" WP:IAR, which is usually taken as an editor "crying uncle" at this point. The more important idea is recognizing that the "Exceptional claims require exceptional sources" rule has a flip side. Squidfryerchef (talk) 23:10, 20 February 2008 (UTC)

I would think (as a trivial example) that something like this archived perl FAQ post would not be seriously challenged. Yet in a similar discussion surrounding the notability and reliability of Haze Gray and Underway, the home of the sci.military.naval FAQ, was recently contested. See Talk:Haze gray and underway. This is troubling in no small part because it is a rich repository of World War I naval photographs.LeadSongDog (talk) 23:12, 20 February 2008 (UTC)

Since policy is descriptive rather than prescriptive, nothing is ever prohibited "by policy". The fact is, people ARE using mailing lists, etc. as sources for these articles. Since, on Wikipedia, "policy" merely reflects what's already happening, it's a no-brainer that the "policy" page needs to be modified to reflect that. Kurt Weber (Go Colts!) 22:54, 24 February 2008 (UTC)

Wait, are you suggesting that policy be modified to make permissible whatever people happen to be doing at the moment? In that case, policies would be utterly useless...Someguy1221 (talk) 23:04, 24 February 2008 (UTC)

FWIW, Kmweber self-describes as Libertarian. That said, the above would seem to be necessarily his position ;/) LeadSongDog (talk) 01:02, 25 February 2008 (UTC)

What you and a lot of others don't get is that that's precisely how "policy" is supposed to work on Wikipedia, as anyone who's been around long enough can tell you. The choice of the word "policy" for these was unfortunate, as people tend to become confused (as you have) and think it means what it does everywhere else. But it doesn't. On Wikipedia, "policy" is anything but prescriptive; it merely describes what people are already generally doing. That's all it does. And yes, I'm a Libertarian, but really, theories of government do not apply to the operations of a private organization such as Wikipedia. You'll be surprised how strict I am with my band kids. Kurt Weber (Go Colts!) 03:03, 25 February 2008 (UTC)

I wholeheartedly concur with the positions carefully expressed by Amir E. Aharoni, Blueboar, SlimVirgin, and LeadSongDog, and oppose the positions expressed by Dougweller and Kurt Weber. Kurt, your first recorded edit was apparently 7 May 2004, while mine was 22 August 2004, so we are both relatively experienced editors here. You should be well aware that most of the Arbitration Committee disagrees with your position and has interpreted policies as binding on numerous occasions to ban or restrict editors who were essentially submitting garbage to the encyclopedia (and creating more of a mess for everyone else to clean up). This goes back to the whole problem expressed by the famous New Yorker cartoon, "On the Internet, nobody knows you're a dog."--Coolcaesar (talk) 04:41, 25 February 2008 (UTC)

I agree with the current policy, and think that mailing lists are excluded by that policy as they are self-published. For example the Dino Mailing List contains many posts from serious authorities like Tom Holtz and Ken Carpenter, but also many from arm-chair paleontologists; some posts from arm-chair paleontologists can only be described as fanciful; and even a few of the posts from serious authorities are musings or running an idea up the flagpole (a legimate part of the process of forming hypotheses). In addition mailing lists (including Dino Mailing List) tend to attract people of similar views, and can therefore be very hostile about differing views. Dino Mailing List is used as a "sandpit" by some serious authorities, but I wouldn't use it as a source in a Wikipedia article. For the same reasons I would exclude any other mailing list, although I'm happy to use mailing lists as pointers to reliable sources such as new peer-reviewed articles. Philcha (talk) 12:06, 28 February 2008 (UTC)

What is wrong with using it to quote Ken Carpenter, for instance? And on the lists I have in mind, there is no question as to the identity of the well known posters, that is just not a problem.--Doug Weller (talk) 13:27, 2 March 2008 (UTC)

And since I've just been talking to Peter T Daniels on Usenet about his Wikipedia entry, that reminds me -- he's a world class expert on writing systems, why can't I quote something he says on UseNet or on a mailing list? So long as they can be accessed on the Internet, that is.--Doug Weller (talk) 17:00, 2 March 2008 (UTC)

I suppose one can always invoke WP:IGNORE in special cases, although that might mean having to defend the edit afterwards. There are of course some cases that blur the lines. A newspaper of record usually has on its website a way to comment on published articles. While the paper is the publisher, there is little if any editorial control of the comments made and the author attributions are essentially unverifiable. Comments in a blog may trigger real-world discussions that make it into dino-media. I rather like the idea of using pageranking as a less subjective measure, but only for statements that have no scandal factor, which still distorts pagerank.LeadSongDog (talk) 17:30, 1 March 2008 (UTC)

Say for instance I am editing a page about a video game, is it considered legitimate to cite the game itself as a source? In particular, the page about cultural differences between eastern and western video games. I would like to cite specific games where different plot elements are used, but wasn't sure if they were considered a primary source or not one at all.

Thanks. Fllmtlchcb (talk) 00:20, 26 February 2008 (UTC)

Games should only be cited in articles about themselves, for the purpose of providing only basic description about themselves. Citing several games to compare cultural differences would constitute original research, and you should instead stick to what secondary sources say on the subject. Someguy1221 (talk) 01:06, 26 February 2008 (UTC)

I understand saying a certain game is an example of an idea or concept without a citation is OR. I am wondering if it is considered OR to say that a particular game specifically contains a certain plot element would be considered OR, because the game as a reference source can verify any claims made about it. An example of this would be saying that a main character dies before the climax in Final Fantasy 7 (it's 10 years old, so I figure that's a well known spoiler). Fllmtlchcb (talk) 05:23, 26 February 2008 (UTC)

I think you could find a secondary source on that. The problem with directly citing primary sources is that it erroneously suggests that your own interpretation of them is the truth. We cite primary sources when we quote them, but not for interpretations of fictional events. For that we need a secondary source, like a review or a published article.

Given the popularity of FF7 it shouldn't be too hard to find some sort of coverage of that plot element -- either a review, or possibly even an academic publication. --FOo (talk) 05:59, 26 February 2008 (UTC)

FF7 would be simple to find a person's review of to cover the content of the game, which makes it a bad example in hindsight. Rarer, more obscure games like Kuon for the Playstation 2 would not be as easy to find sources for, but greatly shows the different cultural themes in video games made by eastern developers. Unless I write a review of the game myself and quote that as a secondary source (which is self-serving and unreliable), then there is no way for me to find information pertaining to the game at hand. It would be easy to apply the 'ignore all rules' rule, but I don't think it would do Wikipedia good to actually do that. In fact, Kuon doesn't even have a page on wikipedia because of its obscurity. Fllmtlchcb (talk) 06:15, 26 February 2008 (UTC)

This is really a WP:NOR question... As long as you are stick to a purely discriptive comment about the plot line of the game, you can cite the game itself... this is similar to how we treat any creative work, such as TV shows, Books, Movies etc. The key is to avoid any analitical comments or statements of conclusion based upon the game play. So... while you could say, for example, that "Character A dies before reaching the end of the game", but you could not discuss the impact that that death has on the game play. For that you would need a secondary source. Blueboar (talk) 15:08, 26 February 2008 (UTC)

Thank you. This is the answer that I was looking for. Fllmtlchcb (talk) 18:00, 26 February 2008 (UTC)

If it's in the manual, then it's not OR and you can cite the game as its own reference. Ditto anything that's explicitly said in game screens (e.g. upgrade X costs Y and has effects Z). IMO the same applies to anything that can be directly observed from the game screens, e.g. "it takes 10 screens to set up a game of Space Empires III". After that it starts to become less clear. I think that elementary deductions are OK, e.g. if upgrade A is a prerequsite for upgrade B then the total cost of getting to B is the cost of A + the cost of B; but I've met people who disagree. IMO without a citation "a main character dies before the climax in Final Fantasy 7" would be OR, because to prove it you have to disassemble the game and then enumerate all possible paths through it. Philcha (talk) 00:11, 1 March 2008 (UTC)

This too is a WP:NOR issue, but basic arithmetic is clearly exempt, as anyone can easily replicate it. The key is that the inputs to the calculations must be identified. We do this all the time in unit conversions, even using templates for it, eg. The Indy 500 is a 500-mile (805 km) race.LeadSongDog (talk) 17:01, 1 March 2008 (UTC)

I understand that we should not use the game forums as a source, however... what if it is all we have? Case in point, Puzzle Pirates. The administrators and game creators regularly add information to the forums regarding what is coming out and what is new. We also have "PoEmail" updates which inform the community of news, however those things are not held in a repository and there is no "news" section of their website where we can cite information. This is a problem when trying to cite information because the game is not that widely known so there are not many (if any) legit third party articles abound. I'm going to post a notice about this converstaion on the YPP talk page so that others can come over and see what is said. Queerbubbles | Leave me Some Love 15:05, 3 March 2008 (UTC)

If the game is "not that widely known" to the point that the only place that mentions news about it is the game forum... I have to wonder whether it is notable enough for inclusion in Wikipedia. In any case... the solution is to wait... until reliable third party sources (such as industry magazines, review websites, etc.) report on the information. Blueboar (talk) 15:30, 3 March 2008 (UTC)

It has approx. 3 million people playing, however that is spread out over the entire world... so in a sense, its not that widely known. Its not so much that the only news is in the forum, as thats the place where the producers decided to put it. Also, its been around for quite a few years now, people just havent picked it up to discuss the newest additions. Queerbubbles | Leave me Some Love 15:33, 3 March 2008 (UTC)

We still have to wait until a reliable source picks up on the information. You might consider contacting the producers, and explain to them that if they want the latest information about their game to be mentioned in Wikipedia (which they probably would), they will have to release the information in a different way. It is not up to us to bend our rules so that new information about a game can be mentioned. Blueboar (talk) 16:04, 3 March 2008 (UTC)

Oh I completely agree, I just wanted input. Queerbubbles | Leave me Some Love 16:37, 3 March 2008 (UTC)

There's a discussion going on, where a secondary makes a claim about something said in a primary source. The primary source is readily available, and it seems obvious that the statement attributed to the primary by the secondary is simply not in the primary. Does WP:V require that the seemingly faulty claim stay simply because it is verifiable, or does the faulty claim make the secondary source an unreliable source? Murderbike (talk) 18:30, 3 March 2008 (UTC)

Any rule can be ignored if enforcing it would hurt Wikipedia (in this case, by allowing obviously false information to remain in an article). I've removed referenced material for exactly that reason in the past. Someguy1221 (talk) 19:29, 3 March 2008 (UTC)

Question... Why not cut out the secondary source all together and quote what the primary source says? Remember, primary sources CAN be used, they just have to be used with caution (see WP:PSTS. Blueboar (talk) 19:50, 3 March 2008 (UTC)

WP:V never requires that material stay in an article. Editors are expected to use judgment about the accuracy of sources in deciding whether to use a particular source. If a source is simply erroneous, then of course we shouldn't use it as if it was correct. More often, it's a matter of interpretation, in which case the article can attribute the claims specifically, rather than saying they are fact. "According to Smith (2000), Jones (1991) argues ..." — Carl (CBM · talk) 04:02, 6 March 2008 (UTC)

I've come across a range of bogus cites of material where the real source is secondary, but the ultimate source is being cited as if it had actually been read. In a situation I'm looking at now, it appears that this ultimate source doesn't actually exist. One can find this situation arising in a lot of urban legend material. I've added a short passage to indicate that the actual source must be verifiable too. I'm headed over to WP:RS and WP:CITE to address this issue as well, if it isn't already handled there. Mangoe (talk) 13:25, 2 March 2008 (UTC)

I've added link to WP:SAYWHEREYOUGOTIT. I'm keeping the principle here because we need "source" defined right up front.Mangoe (talk) 13:36, 2 March 2008 (UTC)

What do you mean by "verifiable" here, Mangoe? It's standard practice to cite "Smith, John. Book. London 2000 cited in Jones, Sue. Book etc. Are you saying something additional is needed? (Also, that's not what's meant by a secondary source: that just means a source not directly involved in the issue s/he's writing about.)SlimVirgin ^{(talk)(contribs)} 18:25, 2 March 2008 (UTC)

In this case, we have a REDFLAG claim cited to what looks like a reliable source (a website). That website repeats and cites something from another original source (a newspaper article). But upon checking it has been discovered that the original source does not actually exist. Blueboar (talk) 20:02, 2 March 2008 (UTC)

Thanks, Blueboar. I didn't realize this was about a particular issue. Yes, that happens, but I don't think we can legislate for it here. It could happen with any source, not just one cited by another one. That is, an editor could add a reference to a book that doesn't exist, but we can't really ask Wikipedians to go to their local library to check that every book they see cited is a real one. SlimVirgin ^{(talk)(contribs)} 20:12, 2 March 2008 (UTC)

Actually, it's relatively easy to check whether the book exists because practically all books and periodicals ever published (at least in English) are on WorldCat. The really hard verification problem is whether the particular book, chapter, section and/or page cited actually contains the assertion for which it was cited. That is very time-consuming and difficult. --Coolcaesar (talk) 21:04, 2 March 2008 (UTC)

It may be very easy to check, CC, but realistically how often is it done? It really only happens when people have reason to be suspicious, or the material is contentious. I'm guessing that most source material placed between ref tags is never looked at. SlimVirgin ^{(talk)(contribs)} 21:14, 2 March 2008 (UTC)

Well, I check it, if the claim seems dubious enough and there's some hope of hunting down the source. But I'm compulsive that way.

(undent) The case in question was a classic urban legend setup: someone made a dubious claim about the death of George Washington, justifying it with two citations of a Denver Register, dates and all. Since the claim is contradicted by eyewitness accounts, I started chasing it. It's easy enough to find the same material on a number of Catholic blogs and sites, and by and large they repeat the same reference. However, googling for the DR produced nothing other than a reference to a Denver Catholic Register, and further investigation has shown that this wasn't published on the cited dates, and was unlikely to have contained such articles. I'm still pursuing this, but it seems to whole thing was mafe up out of whole cloth.

Ignoring the point I've had to fight for a few times that for a source to be "reliable" in any given case, it has to be right, the other problem is that none of the findable sources for this tale would count as reliable, though surely one of them is the real source. That's really the point I want to add: that the actual source referred to must be reliable. That's the point of WP:SAYWHEREYOUGOTIT, but I think if the present policy is supposed to instruct, it needs to say that. I think the current location in WP:CITE is too buried. And I've run into this problem a lot, though this is the first time in a while I've had what appears to be a completely fictional source (though I see it all the time on Snopes). Mangoe (talk) 00:17, 3 March 2008 (UTC)

The argument that for a source to be "reliable" in any given case, it has to be right flies in the face of the lead sentence of this project page ("The threshold for inclusion in Wikipedia is verifiability, not truth"), and would place WP in the position of acting as an arbiter of truth and rightness vs. untruth and wrongness. Take, for example, the conflict between the claims made in the Agunaldo vs. Worcester sources cited here. -- Boracay Bill (talk) 01:17, 3 March 2008 (UTC)

I think what Mangoe means is that the source has to be "accurate" (as opposed to "right"). "Truth" is subjective... "Accuracy" is not. Blueboar (talk) 01:55, 3 March 2008 (UTC)

Not that it has to be right, but that it cannot be wrong. A source that is demonstrably false is demonstrably not reliable. "Verifiability, not truth" means that truth alone isn't good enough, not that truth is irrelevant. Also, I would point out that in the example you cite, the facts are the reports from the time, not the answer to the question posed in the cited section; and there is little question as to whether the immediate secondary sources are reporting the primary sources accurately.

I've had two occasions where someone has argued that a mistake had to be repeated in Wikipedia simply because it came from a "reliable" source (one from a news report, another from the son of the subject of the article). Oddly enough, it took examination of school yearbooks in both cases to shut down the argument. But for instance it is implausible to claim that a person attended a day school three hours away from where they lived at the time simply because a reporter made a very obvious mistake. To support verifiability, the results of verification have to be accepted. Mangoe (talk) 02:01, 3 March 2008 (UTC)

I see what Mangoe means. I can think of a few cases of this nature. An author writes something that makes an exceptional claim. To back the claim, the author cites a statement made by an authoritative figure of some sort. Something sounds fishy, and you check the source, only to discover that either a) there is no record of that person making that claim, b) the person did state something concerning the subject –but there is a wide gulf between what they said and what the author claims they said, c) the source used is already widely regarded with suspicion, and the author who cited it failed to make that clear. I like to check sources, and have seen it a number of times.

When dealing with exceptional claims we use to require that they be “supported by multiple reliable sources” as one way to protect against this; unfortunately that requirement has gone by the wayside.

It would seem logical that we include a way to deal with those cases where on paper the source looks entirely reliable, and yet the simplest check shows a clear reason to think otherwise. Having said that I see a lot of problems with the proposed wording, although the underlying principle I am in agreement with. Brimba (talk) 03:04, 3 March 2008 (UTC)

The issue in the case that set this off is that the cited source wasn't used at all; some other source was, and the real source contained a citation. I discovered it because the cited source in this instance doesn't appear to exist, but there seems to be a general problem, particularly with a lot of fringey/questionable material, with editors' real sources being different from those they cite. An editor must cite the real source, so that we can verify the real source. The "what is false is not reliable" is a different though related issue. Mangoe (talk) 04:16, 3 March 2008 (UTC)

Are you saying that, in addition to WP:SAYWHEREYOUGOTIT... you would like SAYWHERE '''THEY''' GOTIT, and both have to be reliable? I am not sure if this is possible. Blueboar (talk) 04:38, 3 March 2008 (UTC)

No, it's not that complex. I just want to get the verifiability rationale for WP:SAYWHEREYOUGOTIT into this policy. We need to be able to verify where the editor got it-- indeed, if that source is reliable, then we shouldn't ordinarily need to go further. The problem I see with WP:SAYWHEREYOUGOTIT is that it is obscure, as there are less than ten links to it. In reality the problem is all over the place. Mangoe (talk) 04:58, 3 March 2008 (UTC)

The fact is that there are no reliable sources. All sources are liable to include mistakes (murphy's law?). A statement from something officially classified as RS has to be accepted as "fact" until another RS is found contradicting it. Then, in some cases, it may be possible to dismiss 1 as out of date or non-specialist. Otherwise, the "facts" suddenly become "POVs". Peter jackson (talk) 11:58, 7 March 2008 (UTC)

I couldn't find this in archives, so decided to bring this up here.

Verifiability is a core policy of Wikipedia and defines its authoritativeness in terms of a reference source. Ultimately an FA article is one which is utterly and completely verifiable 100%. This means that no fact or a collection of statements based on a fact can be refuted or their verifiability questioned.

How can a verifiability grading be established?

Looking at the scope of human knowledge there are several general observations one can make about verifiability of various sources of information available in print before Internet (bI) expressed in classes:

• 12 A published work with no references of sources - fiction
• 11 A journalistic work attributed to "in the opinion of" - someone's opinion
• 10 A published work with a few biographic sources, but no in-text references - speculative research
• 9 A journalistic work quoting "an authority on" - single source opinion
• 8 A published work with a bibliography and a few references - original research loosely based on fact
• 7 A journalistic work quoting "an institution which" - sponsored original research
• 6 A published work with a bibliography and substantial number of chapter references - research seeking to advance subject understanding
• 5 A journalistic work citing "a commonly known fact that" - a known, but incompletely understood research
• 4 A published work with a bibliography and footnotes attached to sentences - research evaluating current best-known subject understanding
• 3 A journalistic work citing "latest research" - results of current state of best subject understanding
• 2 A published work with a bibliography, footnotes attached to sentences, and definition of terminology used in the work - definitive subject-area reference work
• 1 An identical result of independent research by at least three different authorities on the subject - fact

If Wikipedia is a reference work, then articles' rated grade would define for its users the level of trust they can put in it.--mrg3105 (comms) ♠♥♦♣ 03:57, 6 March 2008 (UTC)

This seems to be slanted towards academic practices in the natural sciences in the 60 years. Other areas of scholarship, such as history, make much less use of footnotes or any other means of inline citation. "The Declaration of Independence was signed in 1776" is a statement of fact, but it is not the "result of independent research by at least three different authorities on the subject". The term footnotes is imprecise, because many well-cited works use endnotes, it has been the publishing industries preference for decades. Newton's Philosophiæ Naturalis Principia Mathematica lacks references of sources, so it is, of course, fiction. Dsmdgold (talk) 16:19, 6 March 2008 (UTC)

The intent is to encourage provision of verifiable and readily available sources for article referencing. These are likely to be more available if published since the Second World War.

The Declaration of Independence is a primary source which is also a manuscript. Newton's Philosophiæ Naturalis Principia Mathematica was considered fiction in his time (To contemporaries who found the idea of attractions across empty space unintelligible, he conceded that they might prove to be caused by the impacts of unseen particles; the action at a distance was considered by Newton's contemporaries to be a disguised revival of the medieval occult features, and the existence of a perfectly void space (i.e. of the vacuum) aroused a vivid repulsion), and not readily accepted as I understand it. In contemporary terms it would be considered original research, and probably unpublishable in its existing form. Academia has higher standards now.

The choice of using either footnotes or endnotes is usually left to the author/s or the editor. --mrg3105 (comms) ♠♥♦♣ 02:31, 7 March 2008 (UTC)

Sources published since WWII are more likely to be available than others. That, of course, does not mean that, outside of the natural sciences, they are the best available. Newton did not write his works as fiction, nor was it received as such. Gulliver's Travels was received as fiction. There is a substantial difference between the conclusion that a scientific work is incorrect and the conclusion that it is a work of fiction. Fiction has a meaning that entirely different from "not true". The choice between footnotes is usually not left to the author, it is a choice of the publisher. The term footnote means a note at the bottom of the page. It should not be used unles that is what is meant.

All of that said, you seem to have missed my main point, which is that you are basing your separation of fact from fiction on the basis of academic standards within the natural sciences. Other disciplines have other standards. Historians, as a rule, do not annotate every sentence like biologists do. "The Declaration of Independence was signed in 1776" is a statement of historical fact, that is not the result of multiple independent authorities. The actual nature of the Declaration is irrelevant. I could have equally said that Osbern de Arches was landholder listed in the Domesday Book in Yorkshire, but that is a historical fact that is a little more obscure. The point is that historical facts can often only be verified by consulting a single source. It however doesn't make them less true than say for example the ABO blood typing system. Dsmdgold (talk) 03:25, 8 March 2008 (UTC)

"non science" does not equal "fiction". It an be a variety of different modes of discourse. Philsophy is not fiction; "Walden" is not fiction. personal experiences , autobiography, speculative discussions of literature--none of them are fiction. They are attempts to explain aspects of the world, and their authority is in their convincingness and the response of the reader, based upon the effect of the prose, and the trust in the experience and knowledge of the author. Newton and Euclid do not give sources, but their work is mathematics, the very antithesis of fiction--the reader is not expected to check their references, he is expected to follow their arguments and confirm it personally. thereare intermediate forms. Defoe's Journal of the Plague Year is imaginative journalistic reconstruction--it was taken in its day as direct reporting, and given an authority it did in fact not have. Pepys' journals are not fiction--his description of plague and of fire is direct journalism, and can be read both as primary historical sources, or asa masterful story without knowledge of the actual references. the nature of hman knowledge is not simple or uniform,and sources must be evaluated with intelligence and with knowledge. Historical events typically do need more than one source for scholarly work--the exact date for the signing of the Declaration is a nontrivial problem,and much historical work has gone into the exact identification and verification of Domesday. What is needed for the purpose of an encyclopedia depends on the degree to which it is controverted, and the determination of that requires general knowledge and freedom from bias. There is no formula. DGG (talk) 07:47, 8 March 2008 (UTC)

The key point about using any sources, but particularly what we call "questionable" ones, is that they have to be used appropriately. While many editors understand this instinctively, others stick rigidly to whatever is written in the policy. The way it is written at the moment, we would not be able to cite Adolf Hitler's views in Mein Kampf in any article not about Adolf Hitler.

Broadly speaking, a "questionable source" is one with a poor reputation for fact-checking. Questionable sources may be used as primary sources about themselves, events they are directly involved in, and things they have done. They may not be used as secondary sources. That is, they may not be used to discuss issues they have no direct knowledge of.

The question is whether we restrict this use to articles about themselves. This is the point that seems to be leading to absurdity. On the other hand, I'm also not sure we want to remove it entirely, because otherwise we would have to rely on the idea of using them "appropriately," which not everyone "gets." So the question is how to find and word a common sense position that stops questionable sources being overused or used indiscriminately. SlimVirgin ^{(talk)(contribs)} 15:41, 19 February 2008 (UTC)

The point I wanted to make earlier, but held off on while waiting for other input, was that "appropriately" just isn't a useful concept here. We might not as well have a statement if its undercut by "appropriately".

That being said, I disagree that restricting them to articles about themselves is necessarily absurd. Relata refero (talk) 18:29, 19 February 2008 (UTC)

I'm going to pull out something from above as a caution. Re the ALF: "The point is that they are used to tell us what they did there." But we can't trust them to tell us what they did there. If they wanted to lie about, exaggerate, or sensationalize their own actions, of course they'd have no problem doing so. It seems reasonable to suggest further allowances for self-referential claims, but those aren't necessarily more trustworthy then claims about third parties. (Again, self-characterizations should be allowed.)

But there may be some room to move here. Also per previous "I'd say that in-text attribution is pretty much a given for anything remotely contentious." Do you mean that the people who "get" it do it, or that everybody "gets" it? I don't think everybody takes it for a given, and it could be a useful starting point. "Questionable sources may sometimes be used outside of articles about themselves, where they are clearly attributed and described in-text." Think of that as the topic sentence of a paragraph and it might go somewhere.

Third, I understand Colin's reservations about mentioning primary sources, but this is one area that probably does deserve a sentence. We should be free in an article about anti-semitism to write "For example, in Mein Kampf...", as a primary source example, without considering it a policy violation. Marskell (talk) 18:58, 19 February 2008 (UTC)

Yes, I meant that people who "get" it will know when to use in-text attribution, and indeed when to use a source at all. And I agree that it'd be useful to mention the primary source issue. It's precisely for areas like this that the distinction is useful. SlimVirgin ^{(talk)(contribs)} 19:12, 19 February 2008 (UTC)

Just a reply to your last point: yes, as long as that particular passage is pointed out as antisemitic by a 2eme source. If its a bit where AH is talking about socialists instead, then I can imagine an editor with a particular POV wishing to say its actually about antisemitism, not anti-socialism... OK, not an ideal example, but I hope you see my point.

About the others, I'll hold back for a bit. I'm deeply conflicted about this section. Relata refero (talk) 19:05, 19 February 2008 (UTC)

I didn't quite follow your point about AH. My point is simply that editors writing an article about, say, the history of antisemitism need to be allowed to cite AH as an example, and to cite his work directly so long as there's no OR. Ditto with socialists if he's regarded as relevant there, though I suspect he wouldn't be. SlimVirgin ^{(talk)(contribs)} 19:12, 19 February 2008 (UTC)

Yes, that's a good point. If we want to use a primary source as an example of a particular attitude, we need to show that the work is considered an example of that attitude. Marskell (talk) 19:13, 19 February 2008 (UTC)

Mock-up

OK, what about something like the following:

"Questionable sources may sometimes be used outside of articles about themselves, where they are clearly attributed and described in-text. A questionable work may be used as a primary source example of an attitude or viewpoint, if reliable sources have also used it as such. Where an extremist group or individual has made claims about their activities, self-characterizations, or statements of intent, they may be used in articles related to them if the questionable nature of the source is made clear in-text."[Footnote with examples]

I could live with something like this. Marskell (talk) 19:38, 19 February 2008 (UTC)

I don't like the last part, because it would involve an NPOV violation to say in effect: "This is what X says, but, lol, they tend to be a bunch of liars." And I'd want to tweak some of the rest:

If questionable sources are used in articles not about themselves, material sourced to them must be clearly attributed in the text. For example, a questionable work may be used as a primary-source example of an attitude or viewpoint, if reliable secondary sources have also used it as such. Where a widely acknowledged extremist group or individual is used as a source in articles not directly about themselves or their activities, the material must be clearly relevant to the issue at hand, and the source's relevance within that area must have been acknowledged by a reliable secondary source.

I think we should sit on this for a bit, and look around to see what its implications would be. SlimVirgin ^{(talk)(contribs)} 19:50, 19 February 2008 (UTC)

Ya. I would quibble with your quibbles, so extra opinions would be good. I don't think the versions are that far apart. Marskell (talk) 19:58, 19 February 2008 (UTC)

Using questionable sources as examples of attitudes would let people shoehorn rhetoric, argumentation, and objections into controversial articles. Imagine the phrase "xxxx criticized animal rights group yyyyyy as hypocritical and lacking credibility because they spent $1 million promoting an animal sanctuary on land near their rich suburban neighborhood even though the money was raised from a company that zzzzzz." I think we should make clear that simply sourcing something as an attitude or opinion does not make that attitude or opinion any more neutral or relevant to the article than introducing it as a fact. For a viewpoint to be mentionable in an article it has to be encyclopedic in the sense that the fact of people having that viewpoint is relevant to the notability of the subject. The fact that people have 9/11 conspiracy theories, for example, may be true and sourceable, and it may even be independently notable, but it is not germane to an article about Israel, even if people think Israel did it. Merely establishing that there was a controversy or difference of opinion, even a less extreme one, doesn't make any given opinion on the subject relevant. That's especially true in politics where there is a controversy and opposing viewpoint on every matter. Wikidemo (talk) 20:03, 19 February 2008 (UTC)

That's true enough. We'd also need a precisely put sentence demanding relevancy. Marskell (talk) 20:25, 19 February 2008 (UTC)

Perhaps we should stick to questionable sources only being used as sources about themselves, their actions, and events they're directly involved in. It's the issue of restricting them only to pages about themselves that leads to the Hitler problem. But my advice is that we sit on this for some time, a few weeks even, and start looking at articles with this in mind, so we can spot benefits and disadvantages of changing the wording before we do it. SlimVirgin ^{(talk)(contribs)} 20:30, 19 February 2008 (UTC)

Unless I'm missing something in the proposal, I don't agree with it (I'm not clear how the three sentences work together). It would allow biased sources to push (even more) POV on Chavez-related articles. "The New York Times said that VenezuelaAnalysis.com reported X" (in an article not about VenAnalysis.com) is very different from "VenAnalysis reported X"; the former is OK, the latter is not. As long as in independent, secondary, reliable source says something about a biased or extreme source, we can cite it with the proper attribution (in this example, The New York Times), but only if it has been mentioned by independent secondary reliable sources. SandyGeorgia (Talk) 20:35, 19 February 2008 (UTC)

About a biased source? No, the policy doesn't say that. Just about all sources have a bias. A questionable source is more than just biased; the issue is lack of editorial oversight, lack of fact-checking etc. SlimVirgin ^{(talk)(contribs)} 20:48, 19 February 2008 (UTC)

At the moment, I think you could cite a secondary source that discusses the influence of Mein Kampf regarding anti-semitism. I'm not convinced a change of policy is required. Also, this could be misused by editors attempting to introduce their own views into articles. Addhoc (talk) 21:41, 19 February 2008 (UTC)

I completely agree with everything SlimVirgin has been saying. Questionable sources can and should be used in articles about themselves, and in some cases, about the writers of the questionable source, but never as an authority on the subject the questionable source speaks about. A few examples include:

• The use of sources that are clearly racist in articles pertaining to that particular race or to races in general. Example: Mein Kampf.
• The use of sources detailing conspiracy theories in articles relating to history and politics. Example: Loose Change (2007 film).
• The use of sources detailing unscientific explanations for natural phenomena in articles about those phenomena. Example: the Discovery Institute).
• The use of paranormal research in science or psychology related articles. Example: Dean Radin.
• The use of denialist sources in articles about the things which they deny. Examples: the Institute for Historical Review; ExxonMobile-funded "research" on global warming.
• Hoaxes and fabrications. Example: The Protocols of the Elders of Zion. --GHcool (talk) 21:52, 19 February 2008 (UTC)

EDIT: Its been called to my attention[8] that I did not make myself clear enough in my previous post. I agree with SlimVirgin when she says, "[E]ditors writing an article about, say, the history of antisemitism need to be allowed to cite AH as an example, and to cite his work directly so long as there's no OR." I would extend this logic to citing [Dean Radin]] in Wikipedia articles about ESP. On the other hand, I am opposed to citing Hitler in articles related to Judaism or Radin in articles related to legitimate science. --GHcool (talk) 22:48, 19 February 2008 (UTC)

Agreed. This is where the issue of "appropriateness" comes in. It's very hard to legislate for it, because it boils down to common sense. SlimVirgin ^{(talk)(contribs)} 23:25, 19 February 2008 (UTC)

Perhaps we can write a list of "red flags" that editors should think twice about before using in articles that cover a broad topic. Among these red flags could include racism, pseudo-science, etc. --GHcool (talk) 23:41, 19 February 2008 (UTC)

I take exception to the efforts to tamper with the policy in this way. If something about Mein Kampf is notably anti-semitic, then a third party has probably noted it in some way. [9] Is there some reason SlimVirgin is itching to quote Hitler in multiple articles? I'm confused. ClaudeReigns (talk) 07:07, 20 February 2008 (UTC)

• no way should we "soften" the present policy. There can, of course, always be debate on what "questionable" means in a given context. This is a matter of WP:UNDUE. But, if a valid viewpoint is being removed on grounds that it is being attributed to questionable sources, it can bloody well be asked of those who wish to see it represented to provide good sources. If there are no good sources, the "viewpoint" has no place in the article. If there are any, cite them directly. Really, what is this about? The phrasing is alright as it stands. See any case on WP:FTN to see why we should not relax it. dab (𒁳) 09:40, 20 February 2008 (UTC)

I think "the questionable nature of the source is made clear" is a recipe for edit wars and WP:Weasel violations. Can we not instead allow "clear direct quotations" ? It isn't ideal because you might want to paraphrase Hilter if he was too verbose but extending "articles about themselves" to "articles related to them where they are used to provide verbatim quotations" seems a possible mid ground? --BozMo talk 11:30, 20 February 2008 (UTC)

• Oppose. I don't think we should change the policy. Questionable sources are exactly that and we simply shouldn't use them when possible, and certainly not outside of articles about them. As someone said above, if something from a questionable source is notable enough to be worth putting in Wikipedia, then it probably has been said in a third party source. It seems unlikely to me there are many situations wherein something would be worth noting from a questionable source in an article which is not about the aforementioned questionable source when it hasn't been commented on externally. Titanium Dragon (talk) 11:54, 20 February 2008 (UTC)

I agree with everything Slim Virgin has written, but am also loathe to make any change in wording that could be construed as weakening this provision. Now - I am just thinking off the top of my head and this suggestion might make things worse! but - would it help any to say something like, questionable sources may be quoted to illustrate a point of view, but never to further an argument? Whether this particular suggestion is constructive or not, I agree that this is a serious enough matter that we should take our time and mull over any change before acting on it. Slrubenstein | Talk 17:47, 20 February 2008 (UTC)

Rather than restricting to "articles about themselves" the restriction should be to "statements about themselves". I see no logical reason for saying that a source is reliable in one article and not in another--what matters is the relation between the statement and the source. I think the reason behind this dispute is that the current version has been used to enforce WP:UNDUE, rather than WP:V, which is like using a hammer as a screwdriver. :) For example, if it is decided that it is appropriate due weight to mention an opinion by a group of extremists in an article not about themselves, then the best source to verify that opinion are the extremists themselves. To decide whether the weight is appropriate or not, a secondary source is probably needed. But the question is entirely about due weight, not about verifiability. --Itub (talk) 10:53, 21 February 2008 (UTC)

Yes, good point about UNDUE; I think you're right that we've been getting it mixed up with V. I also agree with Slrubenstein's point that questionable sources could be used to illustrate a point (and it would have to be a point that reliable sources have confirmed as a notable point), but not to further an argument — although I wonder if we'd ever be able to write that up clearly enough to prevent endless questions about what the difference is. SlimVirgin ^{(talk)(contribs)} 11:23, 21 February 2008 (UTC)

My suggestion may ultimately just be a shorthand way of saying "in a way that is NOR compliant" and if people were to agree that my phrasing would in some way help things, perhaps it would help to add something like "see NOR for more specific guidance on this point." Also, just for the record, at most I would see my proposal as a necessary but not sufficient ground for inclusion of such material, i.e. we should clarify other requirements. Slrubenstein | Talk 15:42, 21 February 2008 (UTC)

"Illustrate a point of view, but never to further an argument." I know what you're trying to say, but we wouldn't be able to separate the two and enforce it. I illustrate points to further my arguments, after all. Marskell (talk) 17:21, 21 February 2008 (UTC)

I agree that "statements about themselves" would be a far more reasonable restriction here. Restricting sources based on the article seems too rigid in some circumstances, and perhaps too broad in others. So is there any objection to this proposed minor change? Any downside to it? PSWG1920 (talk) 21:48, 10 March 2008 (UTC)

Take a second look at what started this thread. "... provision used to keep out POV that editors disapprove of." All you need do is read the talk history of almost any paranormal article to see how this is a real problem which usually involves the same group of editors. In fact one editor seems to have recently read SlimVirgin's quote: "Questionable sources are those with a poor reputation for fact-checking. Such sources include websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions. Questionable sources should only be used in articles about themselves."

Ab editor in the EVP article wrote: "AAEVP is not quite good enough to source this. They make up all kinds of things at their website. What would be best is if we found someone who didn't believe in EVP reporting on the classification scheme (per WP:FRINGE#Independent sources). Barring that, if we could find one of the people mentioned in our article (like Raudive, for example) who used the classification scheme, at least that would be more authoritative than some website that Tom Butler made up one day. ScienceApologist (talk) 16:39, 23 February 2008 (UTC) [10] This editor is hard-over anti anything paranormal and is not above distorting the truth, as in this statement for which he has zero proof.

I am personally uncomfortable with changing the rule on fringe because I have "fringe" problems in my field and I would not want their statements used as evidence unless the statements are empirically supported. I expect the same treatment. The problem comes when a group of editors work together to dominate an article and exclude sources that might harm their point. In EVP, an online skeptical dictionary is given authority while the long-lived British Society for Psychical research is rejected as unreliable and fringe.

In-text qualifiers are useful, but they are too often used as innuendo to cast doubt rather than explain. A previous arbitration concluded that qualification in the introduction of such points as the unproven nature of the subject is sufficient.

In Electronic Voice Phenomena, [11], and in many emerging fields of study, virtually all of the literature on the subject is original research or published by specialized journals with limited circulation. For instance, the AA-EVP [12] averages around 1,700 unique visitors a day based on our server count and our quarterly NewsJournal is delivered to some 500 people and since it is also emailed, we think read by nearly 2000. That is a pittance compared to a mainstream journal, so of course, it is fringe. And of course, it publishes the original research of other researchers. Yet, there is hardly any source of accurate information about the state of the art other than what can be found in the publications of the small groups around the world. In my opinion, the rules leave Wikipedia to the skeptics, and the only remedy is a rule change or deletion of fringe articles because they cannot be accurately reported. Tom Butler (talk) 21:57, 23 February 2008 (UTC)

"The problem comes when a group of editors work together to dominate an article and exclude sources that might harm their point."

I also don't want to see an increase in crank views being included, but crank and fringe don't necessarily mean the same thing. The difficult has been how to word a policy that welcomes all reasonable, considered viewpoints, including minority viewpoints, and excludes only the nutty ones. I think we have failed with this policy, because I see it too often being misused, as you say, to exclude sources that people simply don't like. But I'm not confident that I could find the words to strengthen the inclusion of minority viewpoints without opening the floodgates to the cranks. SlimVirgin ^{(talk)(contribs)} 22:10, 23 February 2008 (UTC)

I support SlimVirgin in this discussion. If a RS secondary source refers to an extremist source it should be possible in some cases to accept that the RS has done fact-checking and verification. MaxPont (talk) 18:17, 25 February 2008 (UTC)

I can think of only 3 cases in which unreliable sources should be cited:

• to support statements in any article about the views of the author of the source
• where the author of the source is making an admission against themself
• to support purely descriptive statements about the source itself

I can't think of anything else that can be trusted. Peter jackson (talk) 11:51, 7 March 2008 (UTC)

"Should" indicates a preferred course of action, but is not necessarily a requirement. You don't need such sources to say that water is wet or George Washington was a President of the United States. Changing these in such a manner will invariably lead to clubbing new users. "Should" is the most accurate term. Furthermore, there are plenty of reasons not to use third party information or those that aren't known for fact checking (i.e. a notable website). I'm not saying it isn't preferred, but you don't need to delete everything that isn't explicitly cited. — BQZip01 — ^talk 04:59, 8 March 2008 (UTC)

The policy doesn't require that everything has to be individually cited - far from it. Most things only need to be "verifiable" in theory. Only contentious material, quotes, and material likely to be challenged are required to be individually cited. — Carl (CBM · talk) 17:03, 10 March 2008 (UTC)

Even beyond the question of IAR and everything having exceptions, "must" is a harsh word to use in any policy because it begs the question "what happens if I don't do it?" There is actually very little that one must do on Wikipedia. There is a lot of stuff one should do if one wants an article to be a good one, or one's edits not to be reverted. For example a sentence must always begin with a capital letter, other than the occasional proper noun like eBay. But so what? "Should" seems adequate for most content issues. "Must," on the other hand, suggests that it's a behavioral violation if you don't do it. Like, one must not vandalize articles. Wikidemo (talk) 20:16, 10 March 2008 (UTC)

I agree that should is better than must. SlimVirgin ^{(talk)(contribs)} 01:21, 11 March 2008 (UTC)

Hi i just want to know what can be done if some body has published an article about Qur'an and in that the claims are not as per text of AL-Qur'an? like Britannica has many references but that are not written in text of Qur'an. should it not be such that any verse of Qur'an quoted for any claim, must contain that claim and not such as that because some body has published an article and the claims are not shown in the referenced ayats of Qu'an. Here in wiki pedea is also same thing is hapening that people are writing about Qur'an but not as per text of qur'an which is not reliable in that sence that a person is writing for quran, quoting qurani verses which donot have in the text of that referenced verse of Qur'an which means that article would be about Qur'an and not As per Qur'an. there is need of such article where the text of Qur'an should be the reliable souce to quote quran. thanks Farrukh38 (talk) 20:20, 9 March 2008 (UTC)

Given that you mention Britanica, which is usually highly reliable, I am going to guess that this is a case of disagreement over a given translation. If so, we have to remember that the Qur'an is in Arabic, but we write in English. There is always going to be some question as to whether any given English translation is perfectly accurate or not. We have to make judgement calls as to whether the translator can be relied on to do a good job. A high quality source like Britanica probably does. That said, if the translation is contentious, attribute the passage to the source (as in: "According to the Encyclopedia Britanica, verse 12 is translated as follows: <EB's translation>")... and then contrast it with another reliable translation (as in: "However, Prof. Joe Blow of the Arabic Language Dept. at Harvard University translates the verse as: <Blow's translation>" ). Blueboar (talk) 17:01, 10 March 2008 (UTC)

I suspect I know the answer to this one, but I'm going to ask anyway. First the specifics. Our article on Jeanne Tripplehorn states "After graduating from high school, Tripplehorn became a D.J. at Tulsa radio station KMOD, using the air name "Jeanie Summers"." I know this to be false, because I went to school (or at least went to the same school at the same time) with Ms. Tripplehorn, and know that she worked at the Radio Station while she was in high school, not merely after. I can document this, from my high school yearbook. Therein lies the problem. In my mind, yearbooks probably should be considered self-published sources, which would put them off limits for information on third parties, especially in a BLP. An additional complication is that yearbooks are hard to come by, unless you went to that school that year. The Tulsa library does not have a copy of the relevant year, and I can't think of any other institution that would, except perhaps the state or local historical societies. I am assuming that unless a better source can be found that this is one (minor) piece of misinformation that will have to go uncorrected. Dsmdgold (talk) 15:39, 10 March 2008 (UTC)

I think this should count as a reliable source in this instance. Yearbooks are "official" publications of the school that issues them, and thus are not "self-published" in the way you are thinking. Chances are, the school has a copy of the yearbook in question in their archives... and anyone wishing to verify the information could contact the school. Blueboar (talk) 16:45, 10 March 2008 (UTC)

Something I've noticed lately is the widescale use of unreliable sources as references. For example, Urban Dictionary. In the last few days I've removed urbandictionary.com references from about 100 articles, using the Special pages External links search to find them. (I'm not pulling them from the External links sections of articles, only the ones which are actually being cited in the body of the article as a reference) Having pulled all of these I can find, I've now stumbled upon the fact that rotten.com's library is being used as a reference on a large number of articles.

So my question is, is it pretty much agreed upon that going through and mass-removing all these inappropriate sources is a good idea? Urban dictionary is, of course, not remotely reliable; anyone can type anything as an urban dictionary entry and then have their friends all vote thumbs up on it (or use proxies and clear their cookies and vote thumbs up on it 200 times themselves). Rotten.com's library articles are written by some unknown person, no sources are cited, the library entries have no reputation for accuracy or fact checking, we haven't the slightest idea if anything in their articles is true. Here's an example of one of their pages I just removed from one of our articles: http://www.rotten.com/library/bio/crime/spree-killers/woo-bum-kon/

And, while I know it would be optimal for me to replace all these unreliable sources with reliable ones myself, this would be something of a mammoth undertaking which I don't have the time or inclination to engage in. I figure we're better off with no reference at all for some statement in an article rather than a completely unreliable reference.--Xyzzyplugh (talk) 22:05, 10 March 2008 (UTC)

No arguments here... delete away. Blueboar (talk) 01:00, 11 March 2008 (UTC)

Digital single-lens reflex cameras have a "long standing" categorization of professional, prosumer, and consumer. However, which cameras classify as which is primary hearsay and ambiguous. Several of the templates ({{Canon DSLR cameras}}) have a history of people changing these categories around and without any sources.

So here's the question. Which trumps which: reviews or manufacturers? If a manufacturer says "compact professional" but reviewers say "prosumer" then which is it? Does the manufacturer trump reviewers? My opinion is that manufacturers do because it's their product ("We made this: it's compact professional") while the reviewers are more-or-less a self-published source ("I say it's prosumer because of X").

In an article this could be hashed all with references, etc. But in a template there is no room for discussion so it's one of the following: none, one, or both. There are no articles hashing out the various categories that the user could be pointed to.

So which trumps: manufacturers or reviewers? Do the reviewers have "more say" simply because there are more of them? Cburnett (talk) 22:27, 11 March 2008 (UTC)

I would say manufacturers should be considered as primary sources for statements they make about their own products. If a review is published by a reliable source, I would say the review should be considered as a secondary source. Generally, reliable secondary sources are preferred to primary sources because they tend to be less conflicted, although the latter can be used with care in articles about the subject (the product) as described in WP:PSTS. However, if there are only self-published reviews and no third-party published ones, I would recommend relying on the manufacturers' statements. - Neparis (talk) 00:22, 12 March 2008 (UTC)

How would you classify engadget.com and digitalcamerareview.com? Cburnett (talk) 00:26, 12 March 2008 (UTC)

I am not familiar with either of them, but at first glance both Engadget.com and digitalcamerareview.com appear to be third-party published secondary sources. - Neparis (talk) 13:58, 12 March 2008 (UTC)

This is a lot like the ongoing genre difficulties in several (probably quite a lot of) articles on bands and musicians... there, the general practice is to list all genres that are well-attested in reliable sources. In both cases, there is a problem that not even experts will agree on one unambiguously, because there is no empirical definition of the different categories. If there are multiple reliable sources (and in this case I'd include the manufacturer) that say different things, then say all of them, with the relevant citations. SamBC(talk) 14:08, 12 March 2008 (UTC)

For music I could understand because there are many more genres and I think pretty much everyone can understand that they differ. However, putting "Compact Professional/Prosumer" just begs for no one to understand (in particular anyone from either camp) and for persistent and constant removal of "the other one". Cburnett (talk) 23:09, 12 March 2008 (UTC)

In the above discussion, Citing Product as a Reference, it's established that when dealing with creative works (books, video games, movies, etc), it's entirely acceptable to use the work as its own factual reference. It seems to me that this really should codified in policy, instead of just leaving it to the realm of common sense. Something along the lines of "Creative works may be used as their own first-person source in strictly factual matters."

On a related note, it would be helpful to have some policy or guidance regarding plot summaries (assuming there isn't already and I've missed it). By a strict interpretation of the rules, it seems that most of them should be packed solid with references. Clayhalliwell (talk) 00:33, 12 March 2008 (UTC)

I wouldn't mind something like this, although it's actually dealt with in the NOR policy, here. But if people think it's needed here too, we could add something brief and link to NOR. SlimVirgin ^{(talk)(contribs)} 00:55, 12 March 2008 (UTC)

I'd certainly appreciate it, as I'm sure others would as well ^{[citation needed]}. Even though this guidance is indeed present in the NOR policy, it makes sense to put information where people will look for it. It never would have occured to me to look in an article forbidding original research for guidance on acceptable sources. IMHO, the entire Sources section in NOR really doesn't belong in that article, and should logically be moved here. A well-formed article should describe what its subject is, not what it isn't. Clayhalliwell (talk) 15:59, 12 March 2008 (UTC)

This seems like a stupid question, but I've spent some time looking with no luck. Is there a policy on citing sources in foreign languages? I mean, it seems like that kind of defeats the verifiability requirement on english language wiki pages, but is there a policy for this or anything specific that I can point to when discussing the issue on talk pages?

Thanks, Tom

Mmyotis (talk) 01:25, 12 March 2008 (UTC)

Wikipedia:Verifiability#Non-English_sources--mrg3105 (comms) ♠♥♦♣ 02:59, 12 March 2008 (UTC)

Suggest adding the following paragraph in a section called "What verification is" before the "Burden of proof" section.

In Wikipedia, every challenged or controversial claim needs to be verified. Verification is done by adding citations to sources in the article, either in footnotes after each statement or in a section at the end. The statement that's being verified needs to be clearly apparent in the source cited.

Note: It seems to me we shouldn't go right to "Burden of proof" and other heavy, legalistic-sounding concepts without a simple explanation of what it is we're talking about. Best, --Shirahadasha (talk) 03:29, 13 March 2008 (UTC)

The wording was changed from assuming the existence of English sources of equal quality to assuming the existence of English sources of sufficient quality on Feb. 28. I am very uneasy with the change. Scholarly work, and other sources of interest to Wikipedia, are published in many languages. This is a normal state of affairs. In my area of mathematics, for example, there is hardly a researcher who cannot read both English and French.

1. Some sources more clearly justify the assertions in the text than others. These sources should usually be preferred. Translations can always be requested in cases of uncertainty.
2. Sources also provide additional contextual information, and naturally some sources do this better than others.
3. There is no reason for policies or guidelines to substitute their abstract judgment for that of an individual editor, likely knowledgeable about their field, in weighing issues of quality vs. accessibility on a case-by-case basis, so long as they are reminded to consider accessibility as a criterion. In many cases, the difference in benefit to the reader will be clear, even though an English source might be "sufficient" in the sense of justifying a particular statement.
4. There is just something a bit unseemly about the idea of going out of our way to exclude foreign-language sources. We cannot insulate readers from the fact that the world - and science - functions in many languages, and they should not have an expectation that sources should be provided in their language.

I recognize that there will be cases where some readers prefer a source in English, even of lesser quality. There is nothing preventing an additional source in English from being added if one is available in such cases, and perhaps something should be added to WP:V about an additional source. But this is no reason to deprive people of the best source. Joeldl (talk) 05:46, 12 March 2008 (UTC)

I reverted the change back to "sufficient" English sources based the necessity to have a level of transparency in the information provided, which I consider a major point of this policy. As this is the English Wikipedia, with a stated goal of provided a free, easily accessible encyclopedia to the English speakers of the world, we should be writing for the monolingual reader as much as possible. While I am 100% on board with your suggestion that additional sources be provided in foreign languages, I think the basic criteria for inclusion of a source for the purposes of verification be that it is reliable, illustrates the point in question, and is in English wherever possible.

To give a concrete example of potential problems that using the best source regardless of language might bring about, I speak and read Japanese as a second language (ja-4). If I was writing an article about a certain temple, I might wish to reference its homepage--after all, it is reasonable to think that a page about the history of a temple written by that temple would be the best source available. I have no problems reading what is written and can tell you very easily why 2002 was an important year. However, the vast majority of the people we are writing this encyclopedia for could not, and so I think it is appropriate to provide them with an English source that, while not necessarily the best source available, is sufficient in that it explains what happened at Todai-ji in 2002, thereby fulfilling the twin goals of verification and information transparency. --jonny-mt 07:18, 12 March 2008 (UTC)

How obscure would a language have to be before the source became (practically) unverifiable? I've been trying to improve some World War I related articles, but in many cases the sources cited, particularly in the Balkans, are in just one side's language. It's just not credible to expect balanced assessment of the POV in a (in Serbian) or (in Turkish) language reference. It would imply that to make the assessment of POV a reader would need to read that language, the former enemy's language, and (in English) without having a WP:COI. This tiny pool of neutral polyglots is then further reduced by needing them to be interested enough to edit the article. This leads inevitably to the acceptance of propagandistic sources as functionally equivalent to reliable ones. Clearly this is not acceptable. See Živojin Mišić for instance. I'd suggest that as a minimum a published translation to a lingua franca (say one of the big five or ten second languages) must be available in order to have a source treated as both verifiable and reliable.LeadSongDog (talk) 14:21, 12 March 2008 (UTC)

The last non-English citation I did was yesterday when I realised that a track by Mannfred Mann's Earth Band was based on a Schubert Impromptu. I decided that this would count as a "surprising" fact which might be challenged, googled for sources [13] and decided that the one that stood out as reliable was from a German public service radio company. Unlike certain other sites they do not identify exactly which impromptu was used, but I decided that that was not the "surprising" element of the fact and included that detail at The Roaring Silence even though my WP:RS did not give it. In this case, I did not find an English language site that asserted the identity of the music, but even if the French language wiki which did identify the impromptu had been in English, I would not have used it as wikis are not WP:RS. If the German radio station had given full details of the work and it was an English WP:RS that gave only partial info I might have cited both thus giving an English source for people to check the main "surprising" element of the fact and a mroe detailedd foreign source; however, I think I would have just used the English because the way that key names change between English and other languages is likely to confuse a lot of readers. The process relies on a degree of judgement but here I would be opting for a "sufficient" quality source not an equa quality one.--Peter cohen (talk) 18:18, 12 March 2008 (UTC)

Given that both of the original works discussed are easily obtainable, and that anyone interested can compare them for themselves without special training or equipment I wouldn't be too fussed over WP:RS on the comparison. Just listen for yourself and see if the wiki is correct in this instance (i.e. vet the questionable secondary source) by comparing Schubert's "Impromptu No. 3 in G flat Major opus 90 D899" to Manfred Mann's "Questions". Is it WP:OR? No -- it's duplication of prior research. LeadSongDog (talk) 20:59, 12 March 2008 (UTC)

LeadSongDog actually raises a good point, perhaps unwittingly: restriction to English sources introduces bias, since the English-language community, like any other, has its own biases. I hadn't thought of that, since I was so focused on science, but I think it's a strong argument against priveleging English sources excessively. In fact, it's in cases like that that Serbian and Turkish sources become important. It is erroneous to assume that we cannot use Serbian sources on issues like that, for the same reason it would be erroneous to refuse to use English-language sources in covering the Iraq war, for example.

I'm not persuaded by jonny-mt's argument, since I think verifiability does not mean verifiability by a monolingual reader. It is good when that can be achieved, the same way it is good when we can provide an online source, but it is just one consideration. I insist that sources of lesser quality must be the ones to be considered additional.

As for LeadSongDog's point about widely understood languages, perhaps that needs to be accommodated in a case-by-case weighing of accessibility vs. quality to be performed by individual editors.Joeldl (talk) 01:37, 13 March 2008 (UTC)

There is no problem with non-English sources as long as they are translated by an experienced and impartial translator. An extreme example of the unacceptable is the translation of a Hebrew Bible by a known neo-Nazi, even one with a PhD in Hebrew.--mrg3105 (comms) ♠♥♦♣ 02:13, 13 March 2008 (UTC)

The point I was trying to make was not that verifiability can only be considered from the view of the monolingual reader, but rather a takeoff on one of your points--it is good when the monolingual reader can verify something, and given that the only language shared by everyone who reads the English Wikipedia is English, that good should be provided whenever possible. Putting aside LeadSongDog's interesting point for the moment, presenting a foreign-language source to a monolingual reader (which is the majority of our readers) has no benefit, and so it should be avoided whenever possible. As for the question of whether the English source or the foreign-language source would be the "additional" source...well, that's just semantics. --jonny-mt 12:02, 13 March 2008 (UTC)

One might even argue that a published quality translation is the secondary source that establishes the notability of the primary source. Of course both should then be cited. But I would dispute jonny-mt's assertion that there is no benefit in presenting the foreign-language source to the monolingual reader. For one thing, it helps raise awareness that there is a relevant literature in other languages. It provides keywords that assist in creating inter-language links that in turn help improve articles in both languages. It also helps wp:build the web. The reader might even be inspired to attempt reading a new language. Of course we could always ask the wizards to code up user skins that allow them to hide content that their Babel settins say they can't read, but I don't really want to encourage that. LeadSongDog (talk) 17:54, 13 March 2008 (UTC)

It is my understanding that when citing a newspaper article, it is the published newspaper article that is being cited, not the url of the online version of that article. And that therefore, if the newspapers website no longer hosts the article, moves it to a different url, or restricts access to pay-only, the citation is still valid. Is my understanding correct? Dlabtot (talk) 18:13, 12 March 2008 (UTC)

My opinion is that it depends upon what newspaper and why it is being used.WP:RS says "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. When citing opinion pieces in newspapers and magazines, in-text attribution should be used if the material is contentious." If it was being used, for instance, to report an archaeological discovery, I'd argue it was only a valid citation until reports in journals, etc are available (knowing how often news sources get it wrong, or someone just looks at the headlines and they have little to do with the article). And what is there is nothing stored on the web?

An interesting example is not a newspaper but the BBC News on line. They announced the discovery of a 7500 year old city in India, and it was immediately all over the web. A few hours later someone noticed that they'd confused another story about a 7500 year old city in Egypt, and the city in India was only 2500 years old. Too late though, it was already being cited all over.--Doug Weller (talk) 18:51, 12 March 2008 (UTC)

Well, that's off topic, actually. If a reliable source gets something wrong, they'll issue a retraction or correction (in which case we should keep in line with it) or another reliable source will point it out (in which case we fairly report all sides if there is a genuine dispute, or we just ignore one source entirely if we decide their version is insignificant). But here, we are considering the case of a reliable source removing an apparently correct article from web access. In this case, Dlabtot, you are entirely correct; the citation is to the article itself and not to the website it was hosted on (although you might find it in a web archive). The only advice I would offer is to avoid using any offline-only sources where a content dispute is ongoing or likely (especially in BLP issues); Injecting a source that most editors can't read themselves can be problematic if there is reason to doubt the material within. Someguy1221 (talk) 19:03, 12 March 2008 (UTC)

To answer the question asked... yes, a news story that appears in both the online edition and the print edition of a newspaper is still valid, even though the online edition has been moved to archives. This is for several reasons... a) the print version is still easily verifiable by going to your public library. b) The on-line version is still verifiable by going to the archives (the fact that you may have to pay for it does not change whether it is verifiable). In fact, this last point holds true for statements made purely in an on-line edition of a paper.Blueboar (talk) 00:30, 13 March 2008 (UTC)

In general Wikipedia doesn't discriminate against print sources, print sources are perfectly valid, so a source doesn't become unreliable simply by moving between media. Best, --Shirahadasha (talk) 03:33, 13 March 2008 (UTC)

Hi,

I've been involved in a long running dispute on the article Pontic Greek Genocide. I've argued that based on the lack of any published material, prominent adherents and mainstream coverage, the view fails on the above Wiki policies and the article needs retitling and rewriting.

The rationale being used to keep the article in its current state is that sources have been found which refer generally to a genocide of Greeks in the Ottoman Empire. However, I've argued that a collection of sentences from sources not direclty linked to this event does not consitute a wide body of academic work which would establish this view as the majority. I have noted for example that a search on Google Books returns zero books related to this event. As WP:V states, "Exceptional claims require exceptional sources", one could at the very least expect one book to be published on such an event if it indeed is a major view held by academia.

The other major argument being made to keep the article is a recent resolution passed by the International Association of Genocide Scholars which recognises a genocide of Greeks and Assyrians. I've argued that although the resolution presents a notable aspect to this article, the fact it is a resolution and not based on a body of scholarly work still does not make this an exceptional source. Indeed, prominent members of the IAGS have stated as much by stating "The current resolution strikes me is an oversimplified statement that does not have the support of major scholars in our organization" (Peter Balakian) and "There is almost no single scholarly work done on the treatment of Greeks during the First World War." (Taner Akcam). So we have evidence here that not even this resolution represent a unanimous decision among major scholars.

Excuse the long comment, but I wanted to make the situation clear and would welcome some outside views how this dispute should proceed. Thanks, --A.Garnet (talk) 09:23, 14 March 2008 (UTC)

To clarify, I dont intend to use this page as place to carry this dispute, only for editors knowledgeable on Wiki policy to comment whether I and others are right to dispute the article based on the above. --A.Garnet (talk) 09:56, 14 March 2008 (UTC)

I did a Google search on "exceptional claims require exceptional sources" on the archives and I got nothing, just the hit here on WP:V. Racism and genocide are, very sadly, endemic rather than exceptional in history. But you know more about this subject than I do ... perhaps you can give sources that describe the people and time in such a way that those things really would seem exceptional? If you're saying that Turks have not been described accurately by European writers over the last few centuries, I'm certainly with you there. - Dan Dank55 (talk) 13:48, 14 March 2008 (UTC)

I just realized that my comment could be considered offensive, but I don't think it is, in context. The First World War was a time of intense nationalism and xenophobia, and it wouldn't be an "exceptional" claim that the Turks acted no differently from anyone else at the time. - Dan Dank55 (talk) 21:42, 14 March 2008 (UTC)

Is it true that a television program can't be verified if you have to pay to view it, available both online and via satellite (DirecTV) television? While it may not be so easily verified, the citation is wholly verifiable in my opinion. -- Dougie WII (talk) 21:06, 14 March 2008 (UTC)

One page you want is m:Video policy, but be aware that people who review links for appropriateness definitely don't like a video that would take an hour of their time to watch instead of a few seconds, even when it meets requirements. You'll be a lot better off if you can find an online transcript. Also note from WP:RS, "Material from mainstream news organizations is welcomed", but then, they will generally have transcripts available. I take it you're not talking about something produced by a mainstream news organization. - Dan Dank55 (talk) 22:00, 14 March 2008 (UTC)

Thanks for the tips, but I'm not talking about news, I'm talking about a fictional drama series used in an article about that series and characters in that series. -- Dougie WII (talk) 22:54, 14 March 2008 (UTC)

For verification (ie citation), having to pay is a non-issue. We have long held that pay-to-view websites are verifiable... I do not see video as being any different. Blueboar (talk) 23:39, 14 March 2008 (UTC)

What is exactly mend by "third party sources"? I would say any source other then Wikipedia and it's ediors. Here a slightly different definition is used: "...we require reliable third party sources ... because whenever you have a sentence referring to the primary source that is not a quote, that sentence conveys your personal belief in what the primary source says. To solve this dilemma, we have included WP:V..." By my definition a "primary source" may very well be a "third party source": Kant's work is a primary source when aditing Immanuel Kant but it is also a "third party source" since he is not an editor of Wikipedia (it is evident his books are not Wikipedia itself). In NOR the definitions of Primary, secondary, and tertiary sources are explicitly given. Using two different words ("third party" or "tertiary") suggests two different meanings. My debetor seems unaware of a distinction. If no distincion exists then why use two different words? If it does exist, please make the distinction explicit. Any clarification is welcome. (My spelling might be quite off - apologies for that) ErikvdL (talk) 10:29, 15 March 2008 (UTC)

(ec) "Third-party" is used in the normal sense of being independent from the subject. Since this is a point of confusion that comes up periodically, I went ahead and added a parenthetical comment to the policy to clarify this point.[14] There may be a better way to express this in the policy, as a parenthetical comment can create a clumsiness in phrasing. If someone has a better way to express this, I'd encourage them to fix it. Vassyana (talk) 10:44, 15 March 2008 (UTC)

Hi ErikvdL, I think secondary and tertiary sources are usually considered third party. Addhoc (talk) 10:41, 15 March 2008 (UTC)

We use the term "third party" to signal a degree of independence from the subject of the article, though we can't be too precise about what we mean by "independence," because it'll vary enormously, and boils down to editorial judgment. For example, we have an article about a band started by X, and we're looking at the sources to decide whether they're third-party. X has written about his band, and the only other person who's written about it is X's wife on her blog, or in a fanzine published by her brother. At that point, we would say that we need an independent third party to write about the band too, before we can agree to have an article about it.

However, if X's wife is a journalist, and instead of publishing her story on her blog, she had it published in the New York Times, we might decide that the NYT is a third-party that decided X's band was worth writing about. Other editors might disagree with this, arguing that the wife being the author overrides the independence of the NYT, and that an even more independent source is required before we can properly call it a third party.

This issue is separable from the issue of primary, secondary, and tertiary sources. A court transcript is a primary source of information on X, where X is a defendant, but it is nevertheless a third-party source, because it's independent of X. We would still not want to base an article on it, but for different reasons.

I don't think we should include this kind of explanation in the policy itself, but it might be a good idea to add something to a footnote. SlimVirgin ^{(talk)(contribs)} 11:56, 15 March 2008 (UTC)

I like SlimVirgin's idea of a footnote; it's not reasonable at all to expect everyone to know what we mean by third-party. It's a concept with much subtlety that is defined by practice on Wikipedia and not by the dictionaries. Anyone who doesn't know our nuanced meaning still won't know it with the addition of "in relation to the subject", and since so many people rely on their knowledge of policy and get uncomfortable when the language changes when they weren't looking, I'm in favor of putting it back the way it was (without "in relation to the subject") and adding a concise footnote. Also, "third-party in relation to the subject sources" is really more of a German than English construction, even when you add the parentheses for clarity. - Dan Dank55 (talk) 14:32, 15 March 2008 (UTC)

I really don't think there's much nuance or subtlety to the meaning, but otherwise I don't disagree. Vassyana (talk) 16:06, 15 March 2008 (UTC)

←The follow examples come to mind, but I am still learning so tell me if I have misjudged it. There's a somewhat notorious student organization at a university in your state; the question is whether they've got the notability to support a Wikipedia article. There's a man or woman in your state who wrote several articles about the organization in a couple of respected newspapers, and there were letters to the editor in response that didn't deny the claims. None of the other possible sources, such as the school newspaper, are considered reliable sources. Are the newspapers, and the writer, considered third parties if they are situated 200 miles away? 20 miles away? In the same town? If they rely on much of their income from people affiliated with the school? A little income? Does it make a difference if the editor is a new account, or a widely trusted Wikipedia editor? I see potential for subtlety, given that the phrase "third-party" is bearing the entire weight of representing the concept of "objective, disinterested" in the phrase "If no reliable, third-party (in relation to the subject) sources can be found for an article topic..." - Dan Dank55 (talk) 01:19, 16 March 2008 (UTC)

George reverted my edit of the order of the comma, here's the part of MoS that covers it:

Inside or outside Punctuation marks are placed inside the quotation marks only if the sense of the punctuation is part of the quotation (this system is referred to as logical quotation). Correct: Arthur said that the situation is "deplorable".

• (When a sentence fragment is quoted, the period is outside.)

Correct: Arthur said, "The situation is deplorable."

• (The period is part of the quoted text.)

- Dan Dank55 (talk) 00:17, 16 March 2008 (UTC)

P.S. Yes, I learned the opposite order in school, too. But the conversations at WT:MoS were clear on the point. Also see section 6.10 of the online Chicago Manual of Style. - Dan Dank55 (talk) 01:02, 16 March 2008 (UTC)

A number of editors are claiming that an entire book can be used to reference a specific claim, based on the statement in WP:CITE that "Page numbers are not required when a citation accompanies a general description of a book or article, or when a book or article, as a whole, is being used to exemplify a particular point of view." Repeated requests for a page number or quotation from the book are met with repetitive assertions that they are not required. Can one indeed ignore a request for a page number and quotation to back up a specific claim? Does WP:V not matter? Jayjg ^(talk) 01:06, 11 March 2008 (UTC)

All quotations need an inline citation, which would include a page number, as does any statement that's challenged. Book citations are only okay when providing a broad outline of a position. SlimVirgin ^{(talk)(contribs)} 01:19, 11 March 2008 (UTC)

Absolutely. I think the problem is in the section for WP:CITE; in the cases in question, we have bibliographic information, not citations per se. Perhaps rewording would clarify matters. Could we be graced with an example or two? Mangoe (talk) 01:21, 11 March 2008 (UTC)

No, editors can not just ignore page number and quote requests, but sometimes an entire book does indeed support a statement. Generally, the kinds of statements that are going to be supported by an entire book (or chapter of a book) will be very broad. The more specific the claim, the more narrow and exact the citation will probably need to be. We would need to know the specific statement and book being cited to know whether the statement you are talking about can be cited to an entire book, or whether it needs more detail. Blueboar (talk) 01:41, 11 March 2008 (UTC)

The specific claim being made is

"Some critics who use the analogy extend it to include Arab citizens of Israel, describing their citizenship status as 'second-class'."

and the citation used is

Davis, Uri. Israel: An Apartheid State. 1987.

I've stated that, given that the title of the book is not Arab citizens of Israel are second-class citizens, they need to provide an actual page number and quotation, but they are adamant that they do not. Jayjg ^(talk) 01:50, 11 March 2008 (UTC)

While I tend to be on the side of requiring clear citations, including page numbers, this particular dispute seems a bit "pointy" on both sides. On one hand, no one should be able to seriously dispute that a book obviously accusing Israel of apartheid against Arabs generally indicates (in its viewpoint) that Arabs are second-class citizens of Israel. On the other, it should be relatively simple to cite such a claim to the introduction or conclusion of such a book, or to cite a reliable third-party source that says the book makes such claims. So really, those disputing the use of the book and those refusing to provide a full citation are being a bit silly in this case. Vassyana (talk) 02:42, 11 March 2008 (UTC)

You may be unfamiliar with the topic, Vassyana. There are, broadly, two claims that might be made about Israel's treatment of Arabs and the apartheid analogy; first, that they treat Arabs in the occupied territories as if under a system of apartheid, and secondly that they do the same to Arab citizens of Israel. The edit Jay is referring to uses Davis as a source, implying that he claims the latter, as well as the former. Jay is asking for a citation to support that. SlimVirgin ^{(talk)(contribs)} 03:02, 11 March 2008 (UTC)

Ah, that makes sense. It's not my particular topic and what limited material I have read about such claims did not distinguish between occupied territories and citizens in Israel, except by degree. Thanks for the clarification! Vassyana (talk) 03:10, 11 March 2008 (UTC)

In this case, I concur with SV. The statement we are verifying is that an author uses a particular phrase as a discription for Arab citizens of Israel. That needs a specific citation to a place where he uses it. If he uses it in multiple places throughout the book, then either a "first occurance" ref or a multiple page ref (with "etc." to indicate repetition) is needed. Blueboar (talk) 03:17, 11 March 2008 (UTC)

I've not edited that topic page but when I saw this discussion, I got curious because I've seen disputes on other topics resulting from citing a whole book as a source. It's especially hard when the book is obscure. Wikipedia policy aside, I looked on Google - they don't have any excerpts of the book, and Amazon only sells used copies. But I located a newspaper article in Le Monde diplomatique, Israel: an apartheid state?, that cites the book in footnote 9. I'm not sure if the article answers your question or not, but it looks like it could be useful. --Jack-A-Roe (talk) 03:23, 11 March 2008 (UTC)

Thanks for finding it, Jack. SlimVirgin ^{(talk)(contribs)} 08:56, 11 March 2008 (UTC)

It does help; the article uses the reference to refer specifically to the treatment of Palestinians in the territories, not Israeli Arabs. Jayjg ^(talk) 23:28, 16 March 2008 (UTC)

I've shown on the article talk page that Uri Davis does, in fact, apply the apartheid analogy to Palestinian citizens of Israel. I don't know what he says in that particular book, however, that is his personal view on the subject. <eleland/talkedits> 00:12, 17 March 2008 (UTC)

The use of the term "apartheid analogy" in the above comment is misleading as the article is about analogy to south Africa while Davis specifically claim that there is no analogy to south Africa:

"The legal regulation of apartheid in Israel is structured in terms that are very different from the structures of legal apartheid in the Republic of South Africa"

In any case this is a side issue to the fact that so far Eland could not provide a citation to the "second class" mention in the lead in this way: "Some critics who use the analogy extend it to include Arab citizens of Israel, describing their citizenship status as "second-class".[2][3][unreliable source?][4][unreliable source?][5][cite this quote][page # needed]" Zeq (talk) 03:58, 17 March 2008 (UTC)

Exercising your usual standard of care I see; I cited that information specifically to Seeking Mandela more than a week ago. [15] <eleland/talkedits> 09:13, 17 March 2008 (UTC)

"Clearly, the effort to create in Palestine a Jewish majority ex nihilo, could not but further entail the further dispossession and expulsion of the majority of the native population, and the legislation of the remaining non-Jewish, largely Palestinian Arab, population under Israeli rule into the status of second-class citizens", Uri Davis, Israel:An Apartheid State, p 21. RolandR (talk) 10:25, 17 March 2008 (UTC)

As has been pointed out countless times on the article's talk page, the book as a whole is being cited in the lead to demonstrate that "some critics" do indeed use the term apartheid to describe the status of Arab citizens of Israel. There is a lengthy quote from the book later in the article, cited to the relevant page. But it is superfluous to cite a specific page in the lead, since this is the theme of the whole densely-argued and extensively documented book. The first sentence of the back cover blurb reads "In this book, Dr Uri Davis examines those official structures of the Israeli state that define the national status and citizenship rights of its population".

The LMD article does not actually to cite the book as referring only to the 1967-occupied territories, since immediately following the footnote to Davis' book the article continues "The applicability of the South Africa model to Israeli-Palestinian relations is problematic. The first issue is the geographical delineation of Israeli "apartheid": does it cover all of Israel or only the WBGS?". But, since the article does not give a page number, it is difficult to see in what way they believe the book refers to areas beyond the Green Line; certainly in the book itself, these areas are only referred to tangentially, since they are not here Davis's subject.

The suggestion that Uri Davis does not consider that Palestinian citizens have a "second-class" status, and that he does not believe and consistently argue that Israel is an apartheid society within its pre-1967 borders is almost laughable. He has devoted his entire life and career to establishing and documenting this analysis, and it is hard to accept the good faith of those who seem determined to twist his arguments into their diametrical opposite. RolandR (talk) 00:45, 17 March 2008 (UTC)

Parsing the quote, clearly that one citation alone is not sufficient to support all the assertions in the quote:

"Some critics who use the analogy extend it to include Arab citizens of Israel, describing their citizenship status as 'second-class'."

Breaking it down, the assertion says

1. multiple critics (not "a critic" -- do they have names?)
2. who [presently] use the analogy (onus to show that it is used)
3. extend [the analogy] to include ACOI
4. describing (This is an assertion even if not itself controversial)
5. [the ACOI's] citizenship status
6. as 'second-class'. (The onus is to show the critics each use the quoted phrase)

On its face the citation fails to support the use of the present tense in the assertion. The cite is for 1987, over two decades ago. Even assuming the assertion to be entirely correct, it is not supported. Please do the homework on the article before bringing disputes to a more general forum such as this. Some rewrite can keep the sense of the statement while rendering it supported.LeadSongDog (talk) 13:35, 17 March 2008 (UTC)

1. The article does not say "multiple critics" but "some critics". We have cited Electronic Intifada and Uri Davis; this should be enough, since we are not trying here to produce here a definitive list. But we could also add, for instance, Mazin Qumsiyeh [16], Saree Makdisi [17], Bruce Dixon [18], Ilan Pappé [19] and very many more.
2. This is absurd; when does the validity of a citation expire? In any case, the analogy is in regular use, and has been used for decades now. To give one example, Uri Davis stated in a 2004 interview about his autobiography Crossing the Border "If the apartheid distinction in South Africa was between white and non-white, the apartheid distinction in Israel is between Jew and non-Jew". It is ridiculous, and insulting to everyone's intelligence, to pretend that, since he wrote a book in 1987, he can no longer be assumed to have the same views in 2008. And in any case, we are dealing here with verifiability, and the assertion has indeed been verified.
3. I have provided above a further quote from the Davis book, explicitly meeting each of the other points here. He discusses the citizenship status of Palestinian citizens of Israel, asserting this to be "second-class". That is the theme of pages 13-55 of the book; do you want me to scan the lot and put it in as an extended footnote? RolandR (talk) 15:45, 17 March 2008 (UTC)

1. In normal English usage "some" people is taken to mean at least three, but in any case it's an unnecessarily vague word, adding little semantic value.
2. I didn't say the citation expires. Of course it doesn't. But it describes his position at that point in time, not later. If people never changed their minds they'd never write more than one thing. (This would be very bad news for the publishing industry!) As you now have found a subsequent ref, do make the more precise observation you use above, with the Davis quotation. Showing that he maintained the position in 1987 and 2004 certainly is a very strong indicator that he still does, but why not say something along the lines of "The critic Uri Davis has argued that..." and cite one or two pages where he does so? I doubt there's a need for the full pp.13-55 in order to understand what he is saying.

Please try to assume good faith, most especially when discussing controversial subjects. It's all too easy to drive off neutrals, leaving only polarized viewpoints. Most of us here are simply trying to build a good encyclopedia.LeadSongDog (talk) 16:50, 17 March 2008 (UTC)

I'm sorry if I did not assume good faith; I assure that this was in no way directed at you, but resulted from my frustration at the constant run around we have been given over this citation (which, it should be noted, was to establish that some critics hold this view, not to document the view itself). Anyway, in the light of this discussion, I have introduced much of this material to the article -- where it was promptly reverted on the grounds that it had not been discussed, and then re-reverted.RolandR (talk) 20:13, 17 March 2008 (UTC)

Roland, the problem is that it is you who has been giving others "the constant run around... over this citation." All that was required was a page number and representative quote, not this stonewalling. Jayjg ^(talk) 01:36, 19 March 2008 (UTC)

Jay is using an understantment here. Roland have been involved in NPA violations and editwarring to protect his use of the source. As the parsing above shows there is no real proof offered in the source to justify the way it is used in the article lead. To top it off there is now use of Activism websites such as "From Occupied Plaestine" (a blog) and a web page called "Boycott Israel". This is an encyclopedia and some minimal standards should be adheered too. Zeq (talk) 05:03, 19 March 2008 (UTC)

Hi, today I heard someone say that verifiability (not truth) is the threshold for inclusion, and thus false information (if backed by reliable sources) should not be removed if no reliable sources refute the false information. What should we do? --Kjoonlee 22:04, 13 March 2008 (UTC)

If a reliable source makes a claim that you believe is false, you can ignore that part if you can provide a very convincing reason to disbelieve the source. But if this is part of an actual controversy, and there are multiple reliable sources taking contradictory positions, then it is our responsibility to present all significant opinions. We don't throw one out because we like another one better. Could you also give more context on the relevant issue? Someguy1221 (talk) 22:24, 13 March 2008 (UTC)

If you believe the information to be "untrue", attribute it to the person who who makes the claim. However, if multiple reliable sources are stating something as fact, and there are no reliable sources that refute the the statement, then we need to consider that it is likely that the claim reflects a majority view. Blueboar (talk) 22:41, 13 March 2008 (UTC)

To start with.... point IE8 at http://www.webstandards.org/files/acid2/test.html and you get the Acid2 smiley. Point IE8 at http://webstandards.org/files/acid2/test.html and you don't get the Acid2 smiley. The IE team is aware of it ( http://blogs.msdn.com/ie/archive/2008/03/05/why-isn-t-ie8-passing-acid2.aspx ) but it's been reported many times that IE8 "passes" Acid2. No reliable sources have yet been found which say specifically that IE8 fails Acid2 (except for maybe the blog post by the MSDN team). Anyway... if a loophole exists which allows people to add false information to Wikipedia, IMHO it needs to be plugged right away. What can we do to stop that sort of thing? --Kjoonlee 22:58, 13 March 2008 (UTC)

As with what I said, a consensus of editors can refuse to obey a policy if applying it would lead to a "bad" result. If you can't reason it out with the editors already involved, consider following dispute resolution. Someguy1221 (talk) 03:05, 14 March 2008 (UTC)

Isn't that a case of primary sources becoming relevant in the absence of secondary sources? Leushenko (talk) 03:32, 16 March 2008 (UTC)

If a reliable source is saying something you can prove to be false, it strikes me that calls our notion of reliable into question.LeadSongDog (talk) 18:03, 16 March 2008 (UTC)

Saying that something is a "reliable source" gets the information in the door; it doesn't make the information right, that's what debate is for. It isn't news that some news sources are often used as reliable-enough sources to establish notability, but are notoriously inaccurate. Indeed, there's a profitable industry in the U.S. (on the left, the right and up the middle) devoted to laughing at them. If you're saying "and therefore we need to chuck the whole idea of reliability", not so much. - Dan Dank55 (talk) 18:58, 16 March 2008 (UTC)

Is there a centralised place to discuss specific breaches with established members? Talk pages seem to not get any attention. --Nathanael Bar-Aur L. (talk) 15:43, 18 March 2008 (UTC)

There is a reliable sources noticeboard. You could try there. SlimVirgin ^{(talk)(contribs)} 18:25, 18 March 2008 (UTC)

Thank you, I missed those noticeboards. I think it may be more of a original research noticeboard issue. --Nathanael Bar-Aur L. (talk) 20:07, 18 March 2008 (UTC)

Looks like SlimVirgin little of the change "Wikipedia may not be used as [a] source" to "Wikipedia may not be cited for anything other than uncontentious information about Wikipedia itself." I'm not sure why this should be so. On some questions like the USA Congressional staff edits to Wikipedia or the Seigenthaler incident WP itself kind of needs to be cited, just as self-published and questionable sources in other fields need to be cited. <eleland/talkedits> 10:10, 21 March 2008 (UTC)

To provide some context for ongoing discussions, I have merged recent threads dealing with WP:SELFPUB, what it actually means, and whether or how it should be changed (also, this shows that I am not the first person in recent weeks to suggest significantly rewriting this policy!) PSWG1920 (talk) 12:56, 21 March 2008 (UTC)

Sources that are "widely acknowledged as extremist"

I'd like to ask people to review this part of the policy, which I've often seen misused to keep out POV that individual editors dislike:

Questionable sources are those with a poor reputation for fact-checking. Such sources include websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions. Questionable sources should only be used in articles about themselves.

The point of this provision was to prevent edits such as the white-supremacist website, Stormfront, being used as a source in an article about ancient Egyptians. This is unfortunately a real example, and it was a regular editor who did it, not an anon passing by or a single-issue account. Clearly, where edits like these are being made, there's a need for a provision of some kind in this policy.

However, I've most often see the provision used to keep out POV that editors disapprove of. There are two separable problems with the provision:

• First, the word "extremist" is being extended to cover activist groups that some editors simply dislike. It has even been invoked to try to keep certain political research organizations from being used as sources, even about events that they themselves are directly involved in, on the grounds that the article in question is not actually about them.

• Secondly, regarding groups that really are "widely acknowledged as extremist," it is being used to stop them from being used as sources — about events they are directly involved in — in summary-style summaries in other articles. For example, someone recently tried to stop the Animal Liberation Front from being used as a source about a raid the ALF conducted, on the grounds that the article about the raid was being summarized (summary style) in another article, and that the ALF could only be used as a source in an article about the raid itself. I think when the provision is being used to stop a group or person from being allowed to say "This is what I or we did" (when the issue is genuinely a notable one), then it needs to be reviewed.

Another editor recently tried to add to RS that, even when extremist sources are quoted by reliable sources, they still can't be cited unless the article is about the extremist source. [20] [21] This would mean that Hitler could not be cited by scholarly sources except in the article about Hitler.

Is there a way we can keep the spirit of this provision, but without having the baby constantly being thrown out with the bathwater? In my view, it should be sufficient to stress that what's important is that sources be used "appropriately," (which the policy does say), but clearly it's not enough.

I'd normally suggest alternative wording myself, but I have an interest in this because I often use animal rights groups as sources on issues they themselves are involved in, so I'd like to throw this open for discussion rather than suggesting a change myself. SlimVirgin ^{(talk)(contribs)} 06:22, 18 February 2008 (UTC)

Hmm. As you're involved in a dispute on this very point right now, I'm hesitant to support any change. Plus, this is old, stable, and widely-quoted wording. The only additions I might suggest are self-characterizations or examples of a particular phenomenon. "The ALF describes itself as an organization dedicated to animal liberation" could be acceptably sourced to them on an article not directly about them. On the second point, you might quote a racist organization on racism, as an example for readers. But no specific factual claims should be sourced to such organizations. If the ALF claims to have found a monkey hanging upside down in lab, you can't report that here unless a reliable source has done so. (And PETA isn't one, incidentally.) Marskell (talk) 15:39, 18 February 2008 (UTC)

Marskell, your assumption of bad faith is really very unpleasant. I wrote above: "I'd normally suggest alternative wording myself, but I have an interest in this because I often use animal rights groups as sources on issues they themselves are involved in ..."

So what is the point of your comment that "As you're involved in a dispute on this very point right now, I'm hesitant to support any change"? You're saying that the policy must not change, even a change for the better, in case I win a dispute? That's an extraordinary thing to say. Please apologize if that's what you meant, and if it wasn't, please say what you did mean. SlimVirgin ^{(talk)(contribs)} 18:54, 18 February 2008 (UTC)

I just meant that I'm hesitant to engage with something I don't see as a pressing problem when it's obviously the result of an on-going content dispute. I didn't say policy must not change—I offered two examples that I've previously thought of. Marskell (talk) 19:37, 18 February 2008 (UTC)

It's not the result of an ongoing content dispute. The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them (an edit made by SandyGeorgia and one other editor, which may explain your presence here). Did you read what I wrote above? There have been problems with the way this provision has been interpreted almost since day one. SlimVirgin ^{(talk)(contribs)} 19:46, 18 February 2008 (UTC)

Re. "The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them", I think it isn't there. I think it was in there in total maybe a few hours. I think I was the last one to remove it [22] - talk page showed no consensus for this change, see Wikipedia talk:Reliable sources#Extremist sources. SandyGeorgia's involvement in the issue was minimal: she applied a cpedit to rokus' phrasing [23] That's all, SandyGeorgia did neither protest its removal, nor supported rokus' plea on talk after the phrase's removal.

SlimVirgin, I haven't still quite forgotten your personal attack on SandyGeorgia here, hitting gravely on others in the process. Yes, it was the late hours of a disheartening ArbCom, which I hold as an attenuating circumstance in your favour. I'm only explaining why I gave some time to explaining (1) that the phrase is no longer (and even wasn't very long) in the WP:RS guideline, and (2) it's a bit frivolous to name SandyGeorgia as a complice in something she had barely anything to do with. --Francis Schonken (talk) 22:46, 18 February 2008 (UTC)

What on earth is it with people who are forced to focus on personal comments? If you have something to say here about the extremism provision, please do. Otherwise please take the comments elsewhere. SlimVirgin ^{(talk)(contribs)} 22:52, 18 February 2008 (UTC)

What was unclear about "see Wikipedia talk:Reliable sources#Extremist sources"? --Francis Schonken (talk) 22:56, 18 February 2008 (UTC)

Marskell, In the past I've also had problems with SlimVirgin selectively editing Wikipedia's policies and then immediately turning around and using those very same edits as a cudgel in her ongoing wikipedia disputes. I strongly agree that people in the midst of a dispute should not be editing the policy pages, particularly when such edits are specifically contrived to advanced one's own personal agenda. For the sake of objectivity, this practice must be stopped. SlimVirgin, I call upon you to play fair and play by the rules. --Ryan Utt (talk) 19:39, 18 February 2008 (UTC)

Ryan Utt has spent much of his time at Wikipedia trying to post actionable libel about someone, which frequently has to be deleted, which is why he's had problems with me. Is this Stalk SlimVirgin Day or something? Do I get an award? SlimVirgin ^{(talk)(contribs)} 19:43, 18 February 2008 (UTC)

My presence here is a result of my watching the page; I'm often present here. (As for RS, I'm still with you that we should get rid of the thing; I didn't notice Sandy's edit.) And I also have no idea about any problem with Ryan Utt.

Anyway, it's very hard to disentangle other disputes from this discussion at the moment, so just ignore my signature and think of the following idea. I've mentally toyed with a section on V that could partly address what you brought up: Phrasal attribution, as I call it, (or In-sentence attribution). Basically, we could loosen the rules a little bit on questionable sources if we demanded that they are flagged in-sentence at every iteration ("X has claimed..."). Although it seems a weird example, think of holy books. We would never write "The world was created in seven days" and source it to Genesis. But we're free to write "According to the Book of Genesis, the world was created in seven days". And we can reasonably do so on pages that aren't specific to the Bible. Not that in-sentence attribution would allow for unlimited use of questionable sources—we would have to word things carefully. But it might be a way into a solution. Marskell (talk) 20:24, 18 February 2008 (UTC)

I'd say that in-text attribution is pretty much a given for anything remotely contentious. The question is how far can this be taken? Do we have a working definition (even a rough one) of "extremist"? Then once we have it, what exactly should the limitations on extremist sources be? SlimVirgin ^{(talk)(contribs)} 22:55, 18 February 2008 (UTC)

A source with a politically extreme POV that's still known as reliable by related professionals, how do we feel about that? ClaudeReigns (talk) 20:32, 18 February 2008 (UTC)

That should be fine, but it would depend on the editors and how much they disliked the POV. Do you have a specific example? SlimVirgin ^{(talk)(contribs)} 20:50, 18 February 2008 (UTC)

Some stuff on the American right comes to mind. The American Enterprise Institute. Even editorials in the WSJ. Still think in-sentence attribution is appropriate to such cases. Marskell (talk) 21:03, 18 February 2008 (UTC)

That's not what's meant by "extremist." SlimVirgin ^{(talk)(contribs)} 21:11, 18 February 2008 (UTC)

What's "extremist" depends on where you stand, and where you stand depends, in part, on where you sit. Some people would say that being published in the Wall Street Journal is proof positive of being not "extremist", but solidly within the mainstream. The government in Washington regards the regime in Tehran as extremist, and the government in Tehran may well think the same of the regime in Washington. --FOo (talk) 21:07, 18 February 2008 (UTC)

Yes of course the Wall Street Journal is mainstream. The problem with this provision is being demonstrated in this discussion. By extremist sources, we really did mean things like Stormfront, which was given as an example when the provision was written. At some point that example was removed, and so now people are calling anything they strongly dislike "extremist." SlimVirgin ^{(talk)(contribs)} 21:11, 18 February 2008 (UTC)

I've had occasion to mention before that the intent of the writer of policy and how it is eventually used are not usually the same thing here. This shouldn't come as a surprise either, since its usually the same in the real world.... Relata refero (talk) 09:34, 19 February 2008 (UTC)

I wasn't suggesting the WSJ is an extremist source. I was suggesting that many view it as "A source with a politically extreme POV that's still known as reliable by related professionals." I was just replying to Claude. Marskell (talk) 21:17, 18 February 2008 (UTC)

That bit about Stormfront is hilarious. I was thinking about quoting Zahi Hawass in an article about the plight of the white American male. Kidding! No, the source I was thinking of was the Executive Intelligence Review. Totally POV, but has some reputation for reliability. ClaudeReigns (talk) 06:34, 19 February 2008 (UTC)

Clarification: The thing that really triggered it for me was the edit to RS that even if secondary sources cite an extremist source, we still can't use them (an edit made by SandyGeorgia and one other editor, which may explain your presence here). Please stop these kinds of accusations. Regarding my edit at RS, I altered the wording inserted by the previous editor simply because something I thought was unintelligible English was inserted at RS and popped up on my watchlist, and I attempted to make sense of what the editor was trying to say. I didn't take a position on the edit one way or another. SandyGeorgia (Talk) 21:28, 18 February 2008 (UTC)

Sandy, honestly, we don't need this across multiple threads. I'm going to click the red X on my browser now. As I said to Slim, we can't all realistically ignore each other because we all edit and watch policy, but, today at least, we should stop talking. I posted suggestions above that I mean earnestly, but will bring them up again later. Marskell (talk) 21:34, 18 February 2008 (UTC)

(edit conflict) It's not an accusation, just a statement of fact. Someone made a very controversial edit to the guideline, but rather than remove it, you fixed the writing of it, so I assumed you must agree with it, at least in part. Otherwise it's hard to see why you would have left it in. SlimVirgin ^{(talk)(contribs)} 21:37, 18 February 2008 (UTC)

Your mistake was assuming anything. Discussion should continue elsewhere. SandyGeorgia (Talk) 21:47, 18 February 2008 (UTC)

Anyway, I explained the WP:RS incident, from what I knew about it above. --Francis Schonken (talk) 22:48, 18 February 2008 (UTC)

An accurate summary; I took no position on that edit except that something unintelligible had been added to a guideline, which would confuse anyone reading it until it could be resolved. It was later removed by others, so there was no reason for me to continue commenting. SandyGeorgia (Talk) 22:52, 18 February 2008 (UTC)

I have often thought that this entire section should be scrapped, because there's nothing here that careful use of WP:NPOV won't take care of, but it seems to be widely used to keep articles mainstream, and most things that serve that purpose should be retained. Relata refero (talk) 09:34, 19 February 2008 (UTC)

I don't think we can scrap it because it does keep a lot of nonsense out. But I think we need to consider what it means and how to advise that it be applied consistently. SlimVirgin ^{(talk)(contribs)}

Oh, and SV, you could have warned me that you'd changed your mind so I could keep an eye on this page. Not fair! Relata refero (talk) 09:47, 19 February 2008 (UTC)

Yes, and if you see what I wrote there, our advice to use sources "appropriately" really should be enough. But increasingly I'm seeing people stick rigidly to the letter of this policy, and not the spirit, so it means we have to make sure we understand the implications of what we write. SlimVirgin ^{(talk)(contribs)} 13:42, 19 February 2008 (UTC)

Arbitrary break

The text SlimVirgin has highlighted used to say

Questionable sources are those with a poor reputation for fact-checking. Questionable sources should only be used in articles about themselves. (See below.) Articles about such sources should not repeat any contentious claims the source has made about third parties, unless those claims have also been published by reliable sources.

The example of extremist websites was added as a port from WP:ATT. The added text is similar to that which has remained fairly stable since ATT was created in Oct 2006 by SlimVirgin.

Is the ALF "widely acknowledge to be extremist"? Some people think so (see Animal Liberation Front#Listing as a terrorist threat). Balaclavas—check. Destruction of property—check. Disregard for the law—check. Self-published video propaganda—check. Mainstream front organisation—check. But I guess we'll have to agree to disagree on the extremist aspect. Extremism is only one of several possible characteristics of a questionable source. The examples given aren't intended to be complete, but one other example is sources that are "promotional in nature". An activist campaign group typically produces material that is highly promotional (effectively an advert, but probably closer to a Party Political Broadcast without the regulatory framework). As such, it is a questionable source.

In a recent edit SlimVirgin almost cited this section when saying "The ALF may be used as a source in articles about its own actions, per WP:V." Their "actions" is a slight alteration of "about themselves" which is harmess when considered a subset of "about themselves" but not when their "actions" is extended to involve third parties.

The section of Animal testing that recently provoked this debate declares itself to be a summary section of Britches (monkey). But that article is not about the ALF—it is about a monkey, the claims made about its treatment, and commentary on this. WP:V makes it clear that questionable sources may not be used in claims about "third parties". The monkey and the lab where is was found is a third party. In addition, such sources may only be used if the claim "is not contentious". Apparently, the university disagrees with some aspects of the claims and video footage (see SV's edit above). Finally, such sources may only be used when "there is no reasonable doubt as to who authored it". The ALF video is anonymous and likely to remain so. Frankly, I prefer not to get my reliable information from an anonymous bod in a balaclava on some internet video.

So whether or not you think WP:SS extends the domain of "articles about themselves" to include summary sections, this particular questionable source is being misused. I see no reason to change WP:V in order to make this misuse acceptable. Colin°^Talk 14:02, 19 February 2008 (UTC)

Colin, I don't think the "extremist" part was moved from ATT to here. I think it was moved from RS, though it may have gone via ATT at some point.

There are two issues here. The first is when groups widely acknowledged to be extremist may be used as sources, and you're right that the ALF is such a group. Certainly, they may be used as sources about themselves in articles about them, and obviously that includes their actions, because they don't exist without their actions.

One question is whether that extends to summary-style summaries of articles about them in other articles.

But that's just one issue. The more pressing issue is what we mean by "widely acknowledged as extremist." For at least a couple of years, I've seen this provision invoked many times to keep out sources with a strong POV that editors don't like. For example, it has been used to try to keep out material from Political Research Associates, the Southern Poverty Law Center, and the Anti-defamation League. But having a strong POV obviously shouldn't be equated with "widely acknowledged as extremist." So the question is how we pin down what we mean by the phrase. SlimVirgin ^{(talk)(contribs)} 14:13, 19 February 2008 (UTC)

This is where the "extremist" provision was added to this policy, three months ago. My memory is that it was originally in RS, but was removed, but I'll need to check the history. SlimVirgin ^{(talk)(contribs)} 14:16, 19 February 2008 (UTC)

I added it to RS in August 2005. It said: "However, that a source has strong views is not necessarily a reason not to use it, although editors should avoid using political groups with widely acknowledged extremist views, like Stormfront or the Socialist Workers' Party. Groups like these may be used as primary sources only i.e. as sources about themselves, and even then with caution and sparingly. [24]

So the point was that extremist sources may only be used as primary sources. SlimVirgin ^{(talk)(contribs)} 14:25, 19 February 2008 (UTC)

Let's not muddy the water with "primary sources", of which everyone seems to have their own definition. We're dealing with "questionable sources" here, of which extremist sites/organisations is only one example. BTW: your diff on WP:V is the same one as mine, where the author's edit summary is "Added line ported from WP:ATT". Oh, and I think you mean 2005, not 1985!!

LOL!! Sorry. SlimVirgin ^{(talk)(contribs)} 15:04, 19 February 2008 (UTC)

Regardless of its origins, it has stood the test of time.

Nothing here is black and white. There's degrees of extremism and degrees of questionableness. At some point we draw the line and say "you can use that to say 'XYZ claims to be' or 'XYZ claims to have done' but no more". Even something as innocuous as a Head & Shoulders advert couldn't be used to back up a statement that dandruff was easy to remove.

You haven't responded to my point that "their actions" don't extend to claims they make about third parties (the lab and the animals). This is important as it frames the extent to which this policy statement applies. In this regard, I'm citing the section "Self-published and questionable sources in articles about themselves" which is an extension of the paragraph under discussion.

Their actions have to extend to the labs, obviously. The ALF is an activist group. They are their actions. And obviously if they are doing something, they are doing it to a someone or in a somewhere. The point is that they are used to tell us what they did there, not what other people did. SlimVirgin ^{(talk)(contribs)} 15:06, 19 February 2008 (UTC)

I don't think it is "obvious" and it is important. You say the sources are not being used to tell us "what other people did." But they are. They are being used to tell us what scientists did to a monkey in the name of research. The monkey, its condition and the experiment are neither "the ALF" nor "their actions". Colin°^Talk 15:29, 19 February 2008 (UTC)

I think we can be flexible on a "good faith" summary section. One problem is that it is too easy to slap a "main article" template over a paragraph or section. In general, I don't view WP:SS as anything other than a "how to decide what goes where" guideline; it certainly doesn't allow for violations of policy (such as the not infrequent claim that such sections are exempt from the need for citations).

I'm less concerned with establishing a bullet proof definition of "extremist". The closer an organisation or source gets to extremism, fundamentalism or out-and-out propaganda/promotion, the less its utterances should be regarded as sound. If a subject is notable enough to be discussed here then good reliable people will have already discussed it and we can cite them. Having to cite stuff "on the edge" is really a sign that ones argument is either fringe or ignored, and either way, WP doesn't need it. Colin°^Talk 15:01, 19 February 2008 (UTC)

I think you're mixing up different issues. The ALF is only used as a source on issues that mainstream sources have written about. But you keep coming back to only that example, and the rhetoric is getting in the way of the argument.

I come back to this issue since it is what (by your admission) provoked the discussion, and there remain unanswered disagreements over policy interpretation that make it a useful example. But we can move on if you like. Colin°^Talk 15:29, 19 February 2008 (UTC)

The point is that, as it stands, the provision could be interpreted to mean that Adolf Hitler cannot be cited in any article that isn't about him. So in Antisemitism, we cannot, according to this policy, tell the reader what Adolf Hitler said in Mein Kampf about Jews. So we do need to clarify what we mean without diluting the spirit of the provision. SlimVirgin ^{(talk)(contribs)} 15:10, 19 February 2008 (UTC)

I agree that the "in articles about themselves" restriction could get in the way of reasonable use. At this moment, I can't think of an alternative. My suggestion: we start a fresh new talk section dealing with the issue of "restricting questionable sources to articles about themselves". It can then be discussed without the distraction of definitions of extremism, or any current disputes. Colin°^Talk 15:29, 19 February 2008 (UTC)

Good idea. SlimVirgin ^{(talk)(contribs)} 15:41, 19 February 2008 (UTC)

I am, unfortunately, just passing through and I am not going to have time to follow this discussion properly. However what I have read so far of this discussion brings one particular case to mind that some of you might find instructive. I would like to share it with you in case anyone finds it useful or wants to discuss it:

David Irving is not only an extremist; he is also known to be a very dishonest historian. Nevertheless there is one aspect of his work that has been praised by respectable historians, namely his grasp of German troop movements during WWII. To what extent might it be appropriate to cite someone like David Irving in a Wikipedia article? ireneshusband (talk) 04:00, 20 February 2008 (UTC)

Troop movements, of course. ClaudeReigns (talk) 07:49, 20 February 2008 (UTC)

Only if there are other sources that confirm it. He is so untrustworthy, I think he requires more verification. OrangeMarlin ^{Talk• Contributions} 17:50, 22 February 2008 (UTC)

The point I would like to make is on articles like global warming where articles which question the evidence are immediately labelled as "extremist" and "unreliable" by a vociferous group who maintain a stanchly pro-warming POV and effectively block out any balance. Now I say this as someone who at the time was a believer in global warming who tried to put a single link to a related article on oil running out. Unfortunately the result of this one-sided dominance is that the article on global warming is laughable and seriously undermines the credibility of wikipedia as a source. From my experience on that article, trying to be a neutral judge between the two sides, I am quite convinced that groups of people gang up and use both the rules of wikipedia and their admin status to ensure very POV articles. One huge problem is that one side can present largely coincidental "evidence" which suggests something, and then block any article which questions whether the evidence is coincidental saying it is merely "opinion". Thus huge number of reports of possibly entirely coincidental events can be quoted by one side, whilst those articles expressing proper scientific sceptism are blocked. Bugsy (talk) 10:44, 20 February 2008 (UTC)

Suggestion

We might consider consigning all discussion of what is a good/bad source that happens in this policy to WP:RS. WP:V should defer to WP:RS, WP:NPOV, WP:FRINGE, etc. in determining what qualifies as a good source. WP:V should only be about being able to verify points for inclusion. A fact can be unverifiable for a lot of reasons, sourcing being only one issue. ScienceApologist (talk) 19:18, 20 February 2008 (UTC)

Restricting questionable sources to articles about themselves

The key point about using any sources, but particularly what we call "questionable" ones, is that they have to be used appropriately. While many editors understand this instinctively, others stick rigidly to whatever is written in the policy. The way it is written at the moment, we would not be able to cite Adolf Hitler's views in Mein Kampf in any article not about Adolf Hitler.

Broadly speaking, a "questionable source" is one with a poor reputation for fact-checking. Questionable sources may be used as primary sources about themselves, events they are directly involved in, and things they have done. They may not be used as secondary sources. That is, they may not be used to discuss issues they have no direct knowledge of.

The question is whether we restrict this use to articles about themselves. This is the point that seems to be leading to absurdity. On the other hand, I'm also not sure we want to remove it entirely, because otherwise we would have to rely on the idea of using them "appropriately," which not everyone "gets." So the question is how to find and word a common sense position that stops questionable sources being overused or used indiscriminately. SlimVirgin ^{(talk)(contribs)} 15:41, 19 February 2008 (UTC)

The point I wanted to make earlier, but held off on while waiting for other input, was that "appropriately" just isn't a useful concept here. We might not as well have a statement if its undercut by "appropriately".

That being said, I disagree that restricting them to articles about themselves is necessarily absurd. Relata refero (talk) 18:29, 19 February 2008 (UTC)

I'm going to pull out something from above as a caution. Re the ALF: "The point is that they are used to tell us what they did there." But we can't trust them to tell us what they did there. If they wanted to lie about, exaggerate, or sensationalize their own actions, of course they'd have no problem doing so. It seems reasonable to suggest further allowances for self-referential claims, but those aren't necessarily more trustworthy then claims about third parties. (Again, self-characterizations should be allowed.)

But there may be some room to move here. Also per previous "I'd say that in-text attribution is pretty much a given for anything remotely contentious." Do you mean that the people who "get" it do it, or that everybody "gets" it? I don't think everybody takes it for a given, and it could be a useful starting point. "Questionable sources may sometimes be used outside of articles about themselves, where they are clearly attributed and described in-text." Think of that as the topic sentence of a paragraph and it might go somewhere.

Third, I understand Colin's reservations about mentioning primary sources, but this is one area that probably does deserve a sentence. We should be free in an article about anti-semitism to write "For example, in Mein Kampf...", as a primary source example, without considering it a policy violation. Marskell (talk) 18:58, 19 February 2008 (UTC)

Yes, I meant that people who "get" it will know when to use in-text attribution, and indeed when to use a source at all. And I agree that it'd be useful to mention the primary source issue. It's precisely for areas like this that the distinction is useful. SlimVirgin ^{(talk)(contribs)} 19:12, 19 February 2008 (UTC)

Just a reply to your last point: yes, as long as that particular passage is pointed out as antisemitic by a 2eme source. If its a bit where AH is talking about socialists instead, then I can imagine an editor with a particular POV wishing to say its actually about antisemitism, not anti-socialism... OK, not an ideal example, but I hope you see my point.

About the others, I'll hold back for a bit. I'm deeply conflicted about this section. Relata refero (talk) 19:05, 19 February 2008 (UTC)

I didn't quite follow your point about AH. My point is simply that editors writing an article about, say, the history of antisemitism need to be allowed to cite AH as an example, and to cite his work directly so long as there's no OR. Ditto with socialists if he's regarded as relevant there, though I suspect he wouldn't be. SlimVirgin ^{(talk)(contribs)} 19:12, 19 February 2008 (UTC)

Yes, that's a good point. If we want to use a primary source as an example of a particular attitude, we need to show that the work is considered an example of that attitude. Marskell (talk) 19:13, 19 February 2008 (UTC)

Mock-up

OK, what about something like the following:

"Questionable sources may sometimes be used outside of articles about themselves, where they are clearly attributed and described in-text. A questionable work may be used as a primary source example of an attitude or viewpoint, if reliable sources have also used it as such. Where an extremist group or individual has made claims about their activities, self-characterizations, or statements of intent, they may be used in articles related to them if the questionable nature of the source is made clear in-text."[Footnote with examples]

I could live with something like this. Marskell (talk) 19:38, 19 February 2008 (UTC)

I don't like the last part, because it would involve an NPOV violation to say in effect: "This is what X says, but, lol, they tend to be a bunch of liars." And I'd want to tweak some of the rest:

If questionable sources are used in articles not about themselves, material sourced to them must be clearly attributed in the text. For example, a questionable work may be used as a primary-source example of an attitude or viewpoint, if reliable secondary sources have also used it as such. Where a widely acknowledged extremist group or individual is used as a source in articles not directly about themselves or their activities, the material must be clearly relevant to the issue at hand, and the source's relevance within that area must have been acknowledged by a reliable secondary source.

I think we should sit on this for a bit, and look around to see what its implications would be. SlimVirgin ^{(talk)(contribs)} 19:50, 19 February 2008 (UTC)

Ya. I would quibble with your quibbles, so extra opinions would be good. I don't think the versions are that far apart. Marskell (talk) 19:58, 19 February 2008 (UTC)

Using questionable sources as examples of attitudes would let people shoehorn rhetoric, argumentation, and objections into controversial articles. Imagine the phrase "xxxx criticized animal rights group yyyyyy as hypocritical and lacking credibility because they spent $1 million promoting an animal sanctuary on land near their rich suburban neighborhood even though the money was raised from a company that zzzzzz." I think we should make clear that simply sourcing something as an attitude or opinion does not make that attitude or opinion any more neutral or relevant to the article than introducing it as a fact. For a viewpoint to be mentionable in an article it has to be encyclopedic in the sense that the fact of people having that viewpoint is relevant to the notability of the subject. The fact that people have 9/11 conspiracy theories, for example, may be true and sourceable, and it may even be independently notable, but it is not germane to an article about Israel, even if people think Israel did it. Merely establishing that there was a controversy or difference of opinion, even a less extreme one, doesn't make any given opinion on the subject relevant. That's especially true in politics where there is a controversy and opposing viewpoint on every matter. Wikidemo (talk) 20:03, 19 February 2008 (UTC)

That's true enough. We'd also need a precisely put sentence demanding relevancy. Marskell (talk) 20:25, 19 February 2008 (UTC)

Perhaps we should stick to questionable sources only being used as sources about themselves, their actions, and events they're directly involved in. It's the issue of restricting them only to pages about themselves that leads to the Hitler problem. But my advice is that we sit on this for some time, a few weeks even, and start looking at articles with this in mind, so we can spot benefits and disadvantages of changing the wording before we do it. SlimVirgin ^{(talk)(contribs)} 20:30, 19 February 2008 (UTC)

Unless I'm missing something in the proposal, I don't agree with it (I'm not clear how the three sentences work together). It would allow biased sources to push (even more) POV on Chavez-related articles. "The New York Times said that VenezuelaAnalysis.com reported X" (in an article not about VenAnalysis.com) is very different from "VenAnalysis reported X"; the former is OK, the latter is not. As long as in independent, secondary, reliable source says something about a biased or extreme source, we can cite it with the proper attribution (in this example, The New York Times), but only if it has been mentioned by independent secondary reliable sources. SandyGeorgia (Talk) 20:35, 19 February 2008 (UTC)

About a biased source? No, the policy doesn't say that. Just about all sources have a bias. A questionable source is more than just biased; the issue is lack of editorial oversight, lack of fact-checking etc. SlimVirgin ^{(talk)(contribs)} 20:48, 19 February 2008 (UTC)

At the moment, I think you could cite a secondary source that discusses the influence of Mein Kampf regarding anti-semitism. I'm not convinced a change of policy is required. Also, this could be misused by editors attempting to introduce their own views into articles. Addhoc (talk) 21:41, 19 February 2008 (UTC)

I completely agree with everything SlimVirgin has been saying. Questionable sources can and should be used in articles about themselves, and in some cases, about the writers of the questionable source, but never as an authority on the subject the questionable source speaks about. A few examples include:

• The use of sources that are clearly racist in articles pertaining to that particular race or to races in general. Example: Mein Kampf.
• The use of sources detailing conspiracy theories in articles relating to history and politics. Example: Loose Change (2007 film).
• The use of sources detailing unscientific explanations for natural phenomena in articles about those phenomena. Example: the Discovery Institute).
• The use of paranormal research in science or psychology related articles. Example: Dean Radin.
• The use of denialist sources in articles about the things which they deny. Examples: the Institute for Historical Review; ExxonMobile-funded "research" on global warming.
• Hoaxes and fabrications. Example: The Protocols of the Elders of Zion. --GHcool (talk) 21:52, 19 February 2008 (UTC)

EDIT: Its been called to my attention[25] that I did not make myself clear enough in my previous post. I agree with SlimVirgin when she says, "[E]ditors writing an article about, say, the history of antisemitism need to be allowed to cite AH as an example, and to cite his work directly so long as there's no OR." I would extend this logic to citing [Dean Radin]] in Wikipedia articles about ESP. On the other hand, I am opposed to citing Hitler in articles related to Judaism or Radin in articles related to legitimate science. --GHcool (talk) 22:48, 19 February 2008 (UTC)

Agreed. This is where the issue of "appropriateness" comes in. It's very hard to legislate for it, because it boils down to common sense. SlimVirgin ^{(talk)(contribs)} 23:25, 19 February 2008 (UTC)

Perhaps we can write a list of "red flags" that editors should think twice about before using in articles that cover a broad topic. Among these red flags could include racism, pseudo-science, etc. --GHcool (talk) 23:41, 19 February 2008 (UTC)

I take exception to the efforts to tamper with the policy in this way. If something about Mein Kampf is notably anti-semitic, then a third party has probably noted it in some way. [26] Is there some reason SlimVirgin is itching to quote Hitler in multiple articles? I'm confused. ClaudeReigns (talk) 07:07, 20 February 2008 (UTC)

• no way should we "soften" the present policy. There can, of course, always be debate on what "questionable" means in a given context. This is a matter of WP:UNDUE. But, if a valid viewpoint is being removed on grounds that it is being attributed to questionable sources, it can bloody well be asked of those who wish to see it represented to provide good sources. If there are no good sources, the "viewpoint" has no place in the article. If there are any, cite them directly. Really, what is this about? The phrasing is alright as it stands. See any case on WP:FTN to see why we should not relax it. dab (𒁳) 09:40, 20 February 2008 (UTC)

I think "the questionable nature of the source is made clear" is a recipe for edit wars and WP:Weasel violations. Can we not instead allow "clear direct quotations" ? It isn't ideal because you might want to paraphrase Hilter if he was too verbose but extending "articles about themselves" to "articles related to them where they are used to provide verbatim quotations" seems a possible mid ground? --BozMo talk 11:30, 20 February 2008 (UTC)

• Oppose. I don't think we should change the policy. Questionable sources are exactly that and we simply shouldn't use them when possible, and certainly not outside of articles about them. As someone said above, if something from a questionable source is notable enough to be worth putting in Wikipedia, then it probably has been said in a third party source. It seems unlikely to me there are many situations wherein something would be worth noting from a questionable source in an article which is not about the aforementioned questionable source when it hasn't been commented on externally. Titanium Dragon (talk) 11:54, 20 February 2008 (UTC)

I agree with everything Slim Virgin has written, but am also loathe to make any change in wording that could be construed as weakening this provision. Now - I am just thinking off the top of my head and this suggestion might make things worse! but - would it help any to say something like, questionable sources may be quoted to illustrate a point of view, but never to further an argument? Whether this particular suggestion is constructive or not, I agree that this is a serious enough matter that we should take our time and mull over any change before acting on it. Slrubenstein | Talk 17:47, 20 February 2008 (UTC)

Rather than restricting to "articles about themselves" the restriction should be to "statements about themselves". I see no logical reason for saying that a source is reliable in one article and not in another--what matters is the relation between the statement and the source. I think the reason behind this dispute is that the current version has been used to enforce WP:UNDUE, rather than WP:V, which is like using a hammer as a screwdriver. :) For example, if it is decided that it is appropriate due weight to mention an opinion by a group of extremists in an article not about themselves, then the best source to verify that opinion are the extremists themselves. To decide whether the weight is appropriate or not, a secondary source is probably needed. But the question is entirely about due weight, not about verifiability. --Itub (talk) 10:53, 21 February 2008 (UTC)

Yes, good point about UNDUE; I think you're right that we've been getting it mixed up with V. I also agree with Slrubenstein's point that questionable sources could be used to illustrate a point (and it would have to be a point that reliable sources have confirmed as a notable point), but not to further an argument — although I wonder if we'd ever be able to write that up clearly enough to prevent endless questions about what the difference is. SlimVirgin ^{(talk)(contribs)} 11:23, 21 February 2008 (UTC)

My suggestion may ultimately just be a shorthand way of saying "in a way that is NOR compliant" and if people were to agree that my phrasing would in some way help things, perhaps it would help to add something like "see NOR for more specific guidance on this point." Also, just for the record, at most I would see my proposal as a necessary but not sufficient ground for inclusion of such material, i.e. we should clarify other requirements. Slrubenstein | Talk 15:42, 21 February 2008 (UTC)

"Illustrate a point of view, but never to further an argument." I know what you're trying to say, but we wouldn't be able to separate the two and enforce it. I illustrate points to further my arguments, after all. Marskell (talk) 17:21, 21 February 2008 (UTC)

I agree that "statements about themselves" would be a far more reasonable restriction here. Restricting sources based on the article seems too rigid in some circumstances, and perhaps too broad in others. So is there any objection to this proposed minor change? Any downside to it? PSWG1920 (talk) 21:48, 10 March 2008 (UTC)

Take a second look at what started this thread. "... provision used to keep out POV that editors disapprove of." All you need do is read the talk history of almost any paranormal article to see how this is a real problem which usually involves the same group of editors. In fact one editor seems to have recently read SlimVirgin's quote: "Questionable sources are those with a poor reputation for fact-checking. Such sources include websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions. Questionable sources should only be used in articles about themselves."

Ab editor in the EVP article wrote: "AAEVP is not quite good enough to source this. They make up all kinds of things at their website. What would be best is if we found someone who didn't believe in EVP reporting on the classification scheme (per WP:FRINGE#Independent sources). Barring that, if we could find one of the people mentioned in our article (like Raudive, for example) who used the classification scheme, at least that would be more authoritative than some website that Tom Butler made up one day. ScienceApologist (talk) 16:39, 23 February 2008 (UTC) [27] This editor is hard-over anti anything paranormal and is not above distorting the truth, as in this statement for which he has zero proof.

I am personally uncomfortable with changing the rule on fringe because I have "fringe" problems in my field and I would not want their statements used as evidence unless the statements are empirically supported. I expect the same treatment. The problem comes when a group of editors work together to dominate an article and exclude sources that might harm their point. In EVP, an online skeptical dictionary is given authority while the long-lived British Society for Psychical research is rejected as unreliable and fringe.

In-text qualifiers are useful, but they are too often used as innuendo to cast doubt rather than explain. A previous arbitration concluded that qualification in the introduction of such points as the unproven nature of the subject is sufficient.

In Electronic Voice Phenomena, [28], and in many emerging fields of study, virtually all of the literature on the subject is original research or published by specialized journals with limited circulation. For instance, the AA-EVP [29] averages around 1,700 unique visitors a day based on our server count and our quarterly NewsJournal is delivered to some 500 people and since it is also emailed, we think read by nearly 2000. That is a pittance compared to a mainstream journal, so of course, it is fringe. And of course, it publishes the original research of other researchers. Yet, there is hardly any source of accurate information about the state of the art other than what can be found in the publications of the small groups around the world. In my opinion, the rules leave Wikipedia to the skeptics, and the only remedy is a rule change or deletion of fringe articles because they cannot be accurately reported. Tom Butler (talk) 21:57, 23 February 2008 (UTC)

"The problem comes when a group of editors work together to dominate an article and exclude sources that might harm their point."

I also don't want to see an increase in crank views being included, but crank and fringe don't necessarily mean the same thing. The difficult has been how to word a policy that welcomes all reasonable, considered viewpoints, including minority viewpoints, and excludes only the nutty ones. I think we have failed with this policy, because I see it too often being misused, as you say, to exclude sources that people simply don't like. But I'm not confident that I could find the words to strengthen the inclusion of minority viewpoints without opening the floodgates to the cranks. SlimVirgin ^{(talk)(contribs)} 22:10, 23 February 2008 (UTC)

I support SlimVirgin in this discussion. If a RS secondary source refers to an extremist source it should be possible in some cases to accept that the RS has done fact-checking and verification. MaxPont (talk) 18:17, 25 February 2008 (UTC)

I can think of only 3 cases in which unreliable sources should be cited:

• to support statements in any article about the views of the author of the source
• where the author of the source is making an admission against themself
• to support purely descriptive statements about the source itself

I can't think of anything else that can be trusted. Peter jackson (talk) 11:51, 7 March 2008 (UTC)

Self-published and questionable sources in articles about themselves - Policy unclear

The written policy here is rather unclear:

Material from self-published and questionable sources may be used as sources in articles about themselves, so long as:

• it is relevant to their notability;
• it is not contentious;
• it is not unduly self-serving;

...

What is it, exactly, which cannot be contentious or unduly self-serving? For example, in the case of a self-published web page being cited in an appropriate article, what if the statements being referred to in wikipedia are not contentious or overly self-serving, but other statements on the cited page are such? Is that then a valid source or not? Or to take it even further, what if the main page of the domain is clearly contentious or self-serving, even if the cited page isn't? In other words, is it just what is referred to which cannot violate the above guidelines, or is it the entire page/article which is cited, or even the entire publication/domain? My best guess is the first, but no matter what, this policy really needs to be clarified, because all three of those interpretations seem possible currently. PSWG1920 (talk) 06:03, 5 March 2008 (UTC)

It's referring to contentious material being used in WP, not just contentious material elsewhere in the same source. Having said that, I agree that we need to fix this section, because we use self-published sources for contentious material all the time. We should say instead that, where a self-published source is used to support contentious material, editors should clearly attribute the view using in-text attribution. SlimVirgin ^{(talk)(contribs)} 23:13, 5 March 2008 (UTC)

I've always read that to say "if the addition of the source is not contentious", ie that such additions must have consensus amongst the editors involved. Tim Vickers (talk) 18:55, 6 March 2008 (UTC)

"It" refers back to the material. The intent of it (I wrote the first version of it, as I recall) was that self-published sources should only be used to talk about issues they're directly involved in — where they are primary sources and have direct knowledge and involvement — and shouldn't be used in contentious areas in which they're not directly involved and might be contradicted by other sources.

Given that SPS's are used all the time for contentious edits about themselves, we really should update it to clarify that, if they're primary sources and the issue is one that secondary sources have written about (and specifically where secondary sources have written about the involvement of that primary source), then it's fine to use them as sources in articles about themselves and their activities. And, as you know, I would want to add "in sections of articles summarizing their activities" too, but that's a separate issue.

The intent of these provisions was to stop e.g. a white-supremacist group turning up at every opportunity to use itself as a source on certain ethnicities. It was never intended to stop well-known people or groups discussing their own activities, within reason. SlimVirgin ^{(talk)(contribs)} 20:31, 6 March 2008 (UTC)

So changing it to "The claim supported is not contentious", would be clearer, ie you could cite a neo-nazi group to say (in the article about that group) that they hate homosexuals, foreigners, gypsies, jews, and black people - since these people usually do that wouldn't be contentious. But you could not cite the group to support a claim about contentious topics in that article - such as the number of black people who might be fraudulently claiming unemployment benefit or the number of illegal immigrants in a country. Tim Vickers (talk) 20:46, 6 March 2008 (UTC)

Well, now hold on... if the "Bigots for a White America" were to claim that all black people fraudulently claim unemployment, and that claim is a major factor in their particluar brand of extremism, the fact that they make this claim could well be something that makes them a notable organization... what seperates them from all the other white-supremacist groups. If so, I think it is perfectly acceptable to discuss this claim in the article about the group, and to cite the website where they make this claim. The trick is to clearly attribute the claim to group, and make it clear that it is a claim... ie that it is their opinion on the issue. Blueboar (talk) 21:12, 6 March 2008 (UTC)

However, if BWM, were a notable group of bigots, and had clearly made that statement, the fact that they had made this statement would not be a contentious claim. Tim Vickers (talk) 21:52, 6 March 2008 (UTC)

In any article about the neo-Nazi group and its actions, you could say "This is what they say. This is why they say they say it." I think you could also say, "The group alleges that Country X's immigration controls are so lax that X number of illegal immigrants have entered, and the effect of this is Y." But you'd have to exercise careful editorial judgment at this point so that the article informs us about the group and its views without turning the page into nothing but a platform for it. The safest thing is to try to stick to expressing views of the group that secondary sources have identified as notable views that the group holds i.e. where secondary sources have said "Group A believes X." But you wouldn't actually have to cite those secondary sources.

This is all a matter of editorial judgment though. It's too detailed to outline in a policy, and editors may disagree about it. SlimVirgin ^{(talk)(contribs)} 21:02, 6 March 2008 (UTC)

So would it be clearer in this case to say:

Material from self-published and questionable sources may be used as sources in articles about themselves, so long as the claim supported:

• is relevant to their notability;
• is not contentious;
• is not unduly self-serving;

...

That would seem to fit with with you're saying above. The problem seems to be that the person commenting above wasn't sure what "it" referred to in the section they quoted. Tim Vickers (talk) 21:06, 6 March 2008 (UTC)

I would need more time to think about the wording, because changing it could inadvertently open loopholes. SlimVirgin ^{(talk)(contribs)} 21:09, 6 March 2008 (UTC)

• SlimVirgin wrote: "... notable views that the group holds i.e. where secondary sources have said "Group A believes X." But you wouldn't actually have to cite those secondary sources."

  I've seen complaints raised at WP:GAC if an article contains claims similar to "Group A believes X" that are not accompanied by an inline citation to an affirmative secondary source. - Neparis (talk) 00:47, 7 March 2008 (UTC)

Do you mean that Group A (the primary source) is cited for its own beliefs, but people are saying a secondary source is needed too? SlimVirgin ^{(talk)(contribs)} 00:57, 7 March 2008 (UTC)

More generally than that — complaints both when there is already an inline citation to a primary source, and also, quite rightly, when there is no source cited. Primary sources are in my view increasingly being deprecated in Wikipedia. In some cases, this is because FRINGE, UNDUE, and REDFLAG are being assigned greater weight by many than in the past. - Neparis (talk) 02:32, 8 March 2008 (UTC)

That might be sensible if you need a great deal of expertise to interpret the material, such as religious writings, I don't think you could cite the bible to show what protestants believe about "graven images", however where there is a simple statement of belief that should be OK. Tim Vickers (talk) 01:07, 7 March 2008 (UTC)

Actually, I have seen this issue raised in a GA review. Where Group A's official website (the primary source) was cited for a statement of its own beliefs in the article about Group A, and the GA review had a problem with that citation and asked for secondary sources. Blueboar (talk) 04:08, 7 March 2008 (UTC)

Another unclear point is the last restriction: "the article is not based primarily on such sources." What does primarily mean here? Does it have to do with raw measurements of the text or the number of references? Or does it mean that the article's existence should not be dependent on self-published/promotional sources? Is it okay if material from appropriate "questionable" sources effectively doubles or triples the length of the article, as long as third-party sources form the basis for the existence of the article in wikipedia? PSWG1920 (talk) 22:32, 10 March 2008 (UTC)

That's a matter of editorial judgment and it would depend on context, length, quality of edits, notability. The point is that the article should not be based entirely on such sources, or almost entirely. But there's no magic cut-off point. SlimVirgin ^{(talk)(contribs)} 03:07, 11 March 2008 (UTC)

Well, if it's a judgment call, then would it not be better placed in a guideline than an official policy? Also, is it relevant whether or not the article would still exist on wikipedia with only third-party sources? PSWG1920 (talk) 03:15, 11 March 2008 (UTC)

Nobody's suggesting a numerical cut-off. We don't suggest a number when it comes to primary sources in PSTS, but we indicate that great caution is required. We need to do the same thing here, to indicate that extensive quoting from SPSes should, if challenged, be backed up by citations indicating those views are notable and encyclopaedic. For an example of something that would benefit from this, see Sita Ram Goel. Relata refero (talk) 08:25, 11 March 2008 (UTC)

PSWG, lots of advice in policies boils down to judgment calls except in the most obvious of cases. Even if we wanted to, we couldn't legislate for every eventuality, and I don't think we would want to, even if we could. We have to allow people the space to make some decisions based on their familiarity with the topic and with the article. SlimVirgin ^{(talk)(contribs)} 08:59, 11 March 2008 (UTC)

But it seems to me that "the article is not based primarily on such sources" is vague enough that it really doesn't belong in an official policy. Couldn't that kind of thing be left to a guideline such as WP:RS, which emphasizes that common sense and the "occasional exception" apply? In fact, from what I have read on this discussion page, it seems like the entire section is fluid enough that most of it could be moved to WP:RS, leaving only essential restrictions in the official policy. PSWG1920 (talk) 09:31, 11 March 2008 (UTC)

That would imply that it's okay to create articles based entirely on self-published sources, but it isn't. SlimVirgin ^{(talk)(contribs)} 10:00, 11 March 2008 (UTC)

Sorry, I have to qualify that. There are, I believe, articles in some areas of pop culture that are based on self-published sources, because those are the only sources available, but on the whole it's not encouraged. SlimVirgin ^{(talk)(contribs)} 10:01, 11 March 2008 (UTC)

Then "primarily" could be changed to "entirely" in the policy. PSWG1920 (talk) 10:05, 11 March 2008 (UTC)

I really don't think it would be helpful to pin people down. Articles shouldn't be based entirely or primarily on self-published sources, but sometimes the quality of those sources might be very high, or the subject unusual and very interesting, so that an exception could be made, based on editorial judgment at the time. All policies are subject to common sense. SlimVirgin ^{(talk)(contribs)} 10:08, 11 March 2008 (UTC)

That's why I keep suggesting that this be moved to WP:RS, in which it is made clear that common sense and occasional exceptions apply. PSWG1920 (talk) 10:11, 11 March 2008 (UTC)

But it's made clear here too that sourcing is a matter of editorial judgment. There's no reason to move that provision out of the policy just because it's not nailed down to within an inch of its life. SlimVirgin ^{(talk)(contribs)} 10:52, 11 March 2008 (UTC)

I don't really see how that is made clear in the Verifiability policy. At least, it is stated more prominently and unmistakably on the guideline pages. PSWG1920 (talk) 11:19, 11 March 2008 (UTC)

I just noticed that you left a note on my talk page about this a few days ago. Sorry not to have responded. You were asking about these edits. If these websites were published by Bates or any official Bates organization, they can be used according to this policy, because the article's about him. If published by others, it's trickier, and it would depend on the quality of the sources. I certainly wouldn't rule them out though. SlimVirgin ^{(talk)(contribs)} 11:28, 11 March 2008 (UTC)

Some of the sources which were deleted are websites of Bates method teachers, which were labeled by said editor as advertising. I am more concerned, however, with the edits he seems intent on making than those he has already made. He has repeatedly stated in the discussion page that most of the article should be deleted due to lack of support by third-party sources. As I noted in the original comment on your talk page, my main reason for letting you know was that, since you're an admin and talked about rewriting the policy, observing what is transpiring there provides some insight into how this and other policies and guidelines are being interpreted. PSWG1920 (talk) 12:21, 11 March 2008 (UTC)

What does "themselves" mean?

"Material from self-published and questionable sources may be used as sources in articles about themselves," What does themselves mean in this context? Please say in just a few words, because I just can't wade through all the verbiage above. In simple-minded curiosity, your friend, GeorgeLouis (talk) 15:56, 15 March 2008 (UTC)

It means pretty much what it says. Material from subjects may be used in articles about those subjects. For example, if a professor with a wiki article has a self-published website detailing their personal opinions about a particular theory, it could possibly be used as a source for the scholar's opinion in the article about the professor. Vassyana (talk) 16:08, 15 March 2008 (UTC)

"Themselves" might be the right word here. The language doesn't seem clear to me either, and thanks for pointing it out, but I'll leave it alone while other people are working on the section (above). Corporations, animals, computers, and robots can do something "themselves", but you can't say "the rocks fell on the floor themselves". Source material is always the product of something to which the words "itself" or "themselves" could be applied, such as persons, corporations, and computers. - Dan Dank55 (talk) 16:30, 15 March 2008 (UTC)

Hello. Thank you for the consideration, but that could certainly be part of the discussion on how the policy should be rewritten. Although it may be that just changing the wording around "themselves" will make its meaning more clear. I moved the section on that to the bottom (as it was previously a subsection), to show that it is general-purpose and not focused on only one aspect of WP:SELFPUB. PSWG1920 (talk) 09:09, 16 March 2008 (UTC)

Let's be clear though, because WP:SELFPUB is so often misapplied in an attempt to include self published material that it actually excludes. It lists a very high bar that must be passed in order to qualify for inclusion. For example, a very high percentage of self published material is unduly self-serving. Dlabtot (talk) 17:59, 15 March 2008 (UTC)

Potential rewrite of WP:SELFPUB

Based on what I have read in above sections, and having seen how this policy has been perhaps wrongly interpreted elsewhere, I have come up with a potential re-write of WP:SELFPUB:

Self-published and questionable sources may be used to reference relevant information about themselves, including what viewpoints they present, with these conditions:

1. Potentially controversial claims made only by such sources should not be presented in wikipedia as generally accepted fact; it is especially important to make clear the dubious nature of such claims if they regard third parties or any living persons.
2. Such sources should not be used to establish the notability of specific topics they deal with (i.e. to justify discussion of such in wikipedia), unless the sources in question are themselves the main subject of the article.

I think the wording of these conditions would help solve a few problems discussed above. Saying that questionable sources can be used to "reference relevant information about themselves" would seem to cover #1 and #5 of the current restrictions, and also would protect sources from being excluded solely by article. From the discussion here, I get the sense that #2, #3, and #4 are intended to prevent such claims and ideas from being presented in wikipedia as true or valid, rather than exclude any mention of such. I'm not sure what the intent of #6 is. Does this exclude web sites which don't reveal the author's full name? As I previously explained, #7 as is really seems too vague to be a policy. PSWG1920 (talk) 04:28, 16 March 2008 (UTC)

There's a problem with allowing self-published and questionable sources to be used outside articles about themselves and their activities. I agree it should be changed so that they're not restricted to their own articles, but I'm not clear what a good alternative wording would be, because we don't want to go to the other extreme and have people try to use them all over the place. SlimVirgin ^{(talk)(contribs)} 16:14, 17 March 2008 (UTC)

I thought that saying that such sources could be used to "reference relevant information about themselves", in conjunction with the condition that in general such sources should not be used to establish notability of any specific topic (that would include subtopics of articles) might be sufficiently tight. Perhaps it needs to be made clear in what specific situations and articles self-published and questionable sources are not acceptable? PSWG1920 (talk) 21:10, 17 March 2008 (UTC)

Relevant in what sense? And what kinds of articles are we allowed to use those sources in, that is the question. SlimVirgin ^{(talk)(contribs)} 21:23, 17 March 2008 (UTC)

My thinking is that spelling out what "kinds of articles" questionable sources may be used in is too general here. By "relevant" I meant relevant to what is being discussed; the statements themselves would have to comply with other policies. PSWG1920 (talk) 21:33, 17 March 2008 (UTC)

We had an example recently of Stormfront being used as a source in an article about ancient Egyptians, arguing that they weren't really Egyptian or something (I forget the exact claim). Any change to this section has to make sure we don't encourage the use of questionable sources in this way. Currently, Stormfront would only be allowed to be used as a source in an article about itself. SlimVirgin ^{(talk)(contribs)} 21:39, 17 March 2008 (UTC)

Perhaps greater restrictions should be placed on extremist sources specifically, or sources promoting bigotry? PSWG1920 (talk) 21:49, 17 March 2008 (UTC)

The problem is that there are lots of questionable sources spouting nonsense that aren't extremist or bigoted. SlimVirgin ^{(talk)(contribs)} 22:02, 17 March 2008 (UTC)

How about adding to the proposed second condition:

Such sources should not be used to establish the notability of specific topics they deal with or claims they make (i.e. to justify discussion of such in wikipedia), unless the sources in question are themselves the main subject of the article.

I suppose this would possibly allow Stormfront to be used as a source in an article not specifically about itself, but only if there was also a reliable third-party source which dealt with their specific claims which were being discussed. PSWG1920 (talk) 22:12, 17 March 2008 (UTC)

Yes, that would work, but would it make itself redundant? That is, if we had a source other than Stormfront, we would never want or need to use Stormfront, except in an article devoted to it or its activities. SlimVirgin ^{(talk)(contribs)} 18:22, 18 March 2008 (UTC)

As is noted below, that problem is probably already covered by WP:UNDUE, and in any event is of limited relevance to WP:SELFPUB, since really, any discussion of Stormfront's views, even via a third-party source, would presumably not be acceptable outside a very narrow range of articles. PSWG1920 (talk) 18:46, 18 March 2008 (UTC)

Do you mean redundant to the first condition or to the sentence itself? PSWG1920 (talk) 18:51, 18 March 2008 (UTC)

I think more than notability the issue is that self-published sources require third party supporting sources in order to establish due weight. Stormfront's opinion on the ancient Egyptians shouldn't be included in an article about ancient Egyptians because it has absolutely no standing in terms of being a respected significant opinion - not because it isn't notable. Absurd claims can get a lot of press but we shouldn't be including them as though they were respected minority opinions as we would with something that meets WP:UNDUE. -- SiobhanHansa 14:54, 18 March 2008 (UTC)

The problem is with the word "respected." At what point does an organization become so extremist or ridiculous that their views, even about their own activities, and even where those views are relevant, become unacceptable? The policy as it stands does not allowed Mein Kampf to be cited as an example of Hitler's views, except in articles about Hitler. Clearly, that's absurd — we might want to cite them in articles about antisemitism, for example — so we do need some rewording of the policy to remove the "in articles about themselves" restriction. But how do we do that without going to the other extreme and allowing them everywhere? I think that's what PSWG is grappling with. SlimVirgin ^{(talk)(contribs)} 18:22, 18 March 2008 (UTC)

Good point. However, I would say that aspect is already covered by WP:UNDUE. I assume that Stormfront's opinion would not merit mention in an article about ancient Egyptians even if Stormfront itself wasn't cited (i.e. if the information came only through a reliable third-party source which addressed Stormfront.) Likewise, any mention of the modern flat Earth viewpoint doesn't belong in the Earth article, even if it's only via third-party sources and not the Flat Earth Society directly. Cited sources don't seem to be the primary issue in those cases, whereas WP:SELFPUB is about sources that can be cited. Thus it doesn't seem that cases like the aforementioned need to be addressed here. PSWG1920 (talk) 15:48, 18 March 2008 (UTC)

Then I think I fail to understand the proposed addition that such sources should not be used to justify the notability of specific topics they deal with (maybe the Stormfront example mislead me). Could you elaborate? -- SiobhanHansa 17:09, 18 March 2008 (UTC)

SlimVirgin originally brought up the Stormfront example, I was just trying to accommodate that. What I was trying to do with that condition was to get to the intent of current restriction #7, paraphrased, the article should not be based primarily on self-published or questionable sources, which to me seems too vague to be a policy. Also, I felt that saying Such sources should not be used to establish the notability of specific topics they deal with or claims they make would help cover the current restriction #1, that citations from self-published sources must be "relevant to the notability" of the subject. PSWG1920 (talk) 17:29, 18 March 2008 (UTC)

This seems entirely ill-advised. Isn't the whole point of WP:SELFPUB to restrict the use of self-published sources? So that articles on politicians or political organizations, to give a salient example, will not be based on what those politicians or political organizations say about themselves? Surely joesmithforsenate.org should not be considered a reliable source for an encyclopedia article about Joe Smith. If anything, I would tighten the existing standards, not loosen them as is being suggested. Dlabtot (talk) 21:59, 17 March 2008 (UTC)

Self-published sources are allowed to be used in articles about themselves, but the restrictions limit the extent of it. SlimVirgin ^{(talk)(contribs)} 22:02, 17 March 2008 (UTC)

Yeah, that's what I said. The question is, why would we loosen those restrictions? Dlabtot (talk) 23:18, 17 March 2008 (UTC)

PSWG has explained above why s/he wants the change. SlimVirgin ^{(talk)(contribs)} 18:22, 18 March 2008 (UTC)

Perhaps you or PSWG would be so kind as to point out to me, where the resaon for wanting the change was explained? Of if it is easier you could just briefly summarize that reason in your response. Thanks in advance. Dlabtot (talk) 18:58, 18 March 2008 (UTC)

See here and here. PSWG1920 (talk) 19:04, 18 March 2008 (UTC)

Ok, I've read those discussions, but they don't seem to address my question. Let me ask you directly in the hopes that you will provide a direct answer. The question is: Why would we loosen the restrictions in order to allow self-serving, contentious self-published material in articles about themselves? Maybe I'm misunderstanding. The way I read it, your changes would allow the "Joe Smith" WP article to include "Joe Smith is the only candidate with a pro-family agenda" and use the joesmithforsenate website as a source. Is this what you intend, or is it an unintended consequence, or am I just misunderstanding something? Dlabtot (talk) 19:20, 18 March 2008 (UTC)

Well, certainly those statements should not be treated in wikipedia as fact. But, it should be okay to refer to or even quote those statements in an article about Joe Smith, from a Neutral POV. PSWG1920 (talk) 19:25, 18 March 2008 (UTC)

Well, that's my fundamental objection to eviscerating SELFPUB in the manner you propose. If anything, it should be made more restrictive, not less. BTW, you did not respond to my direct question. Could you please attempt to do so? Thanks. Dlabtot (talk) 19:32, 18 March 2008 (UTC)

As for why the restrictions should be loosened, I can only point to the discussions above which I linked to. Note that SlimVirgin was the one who originally brought up the idea of doing such; I later pointed out an aspect of the policy which is currently very unclear, which is why I got involved in this discussion to begin with. PSWG1920 (talk) 19:39, 18 March 2008 (UTC)

I'll just renew my objection to the weakening of the SELFPUB restrictions. Especially in light of the absence of a cogent reason to do so. Dlabtot (talk) 19:51, 18 March 2008 (UTC)

Let me elaborate on your "candidate Joe Smith" example. If such statements are referred to in wikipedia, and there is any specific, concise response from the political opposition, that should be mentioned in the same wikipedia article, so the article itself remains unbiased. That would probably be covered under WP:NPOV. PSWG1920 (talk) 20:30, 18 March 2008 (UTC)

That would have the effect of turning Wikipedia into an online version of Crossfire. I say, keep it encyclopedic; articles should be based on reliable, independent sources. Dlabtot (talk) 20:46, 18 March 2008 (UTC)

Which is why I said a "specific, concise response" (the same should be true of the original statement.) It would not be necessary then to keep presenting counterpoints. PSWG1920 (talk) 21:00, 18 March 2008 (UTC)

I think you may be misunderstanding the policy, Dlabtot. It allows SPSs. It allows politicians' articles to use their websites to say "he says he is the only X candidate." That is not a loosening of policy. That is the current policy. It allows those sources because we want to hear what people have to say about themselves, because we don't allow someone to be maligned in a newspaper, have it repeated on Wikipedia, and have no way of addressing the allegations without going through another newspaper. So we allow people to use their blogs to say, "By the way, the allegation in the New York Times that I am a raping, pillaging, child abuser" is false." What is being discussed here, or at least what I am discussing, is the issue of what we call "questionable" sources, not self-published sources. SlimVirgin ^{(talk)(contribs)} 21:10, 18 March 2008 (UTC)

So, if I understand you correctly, you are saying "I am the only pro-family candidate" is not a self-serving statement? (By the way, this is not a hypothetical, although I've paraphrased, it is a real-life of example from the Alan Keyes article. Which achieved GA status after fixing this and many other SELFPUB problems.) Dlabtot (talk) 21:30, 18 March 2008 (UTC)

It is not unduly self-serving, and it's fine to say in an article about X that he says he is the only pro-family candidate, although undoubtedly there will be other sources who say otherwise. There was no need to fix that in order to get GA status, so long as it was written properly. SlimVirgin ^{(talk)(contribs)} 21:50, 18 March 2008 (UTC)

"as long as it's written properly"... umm, yeah. You might want actually look at the article back when it was primarily based on his campaign website before you make a judgement on whether it was written properly and whether this proposed rewrite of WP:SELFPUB would have helped or hurt the editing of that article. Dlabtot (talk) 21:58, 18 March 2008 (UTC)

Perhaps a better example of why we need WP:SELFPUB would be the article Bates Method, where User:PSWG1920 is currently involved in a content dispute over the issue of basing an article primarily on contentious, self-serving, self-published sources. Dlabtot (talk) 21:22, 18 March 2008 (UTC)

Again, so much depends on how the material is presented. SlimVirgin ^{(talk)(contribs)} 21:50, 18 March 2008 (UTC)

Purpose of SELFPUB

I think before any meaningful discussion can be achieved on the wording of the relevant section, we need to agree on what purpose it holds. I think this can be summed up in a series of possible uses of a source, as well as questions to ask about the source. There are four possible uses of a source I can think to distinguish:

1. Never use it
2. Only use it in an article on the author/publisher
3. Can cite it everywhere as an opinion
4. Can cite it everywhere as a fact

There are also five questions I would ask about the source:

1. Is the author a respected expert in the subject area?
2. Do the claims made within fall in line with those made by reliable sources? (in which case it could only be useful for providing greater detail)
3. Do the claims made within conflict with those made by reliable sources? (or at least appear partisan/biased in comparison)
4. Are the claims made within on a specific matter covered by no reliable sources?
5. Do the claims made within pertain to third-parties, possibly living persons?

2, 3, and 4 could really be considered a single question: How do the claims within compare to those made by reliable sources? I think if agreement can be made to what association of uses and answers is desired, which at the very least should be the basis of any arguments here, then we only have to figure out what wording gets across the desired message. The desired use for certain answers should also be blindingly obviously.We could also debate the use of such sources in establishing notability, but that's really tangential. Someguy1221 (talk) 11:01, 19 March 2008 (UTC)

I think the emphasis needs to be on what sources can be used to reference what information. Specifically a clear distinction should be drawn between sources which can only be used to reference claims or philosophies (usually questionable sources), and sources from which [what are treated in wikipedia as] facts can be cited (usually third-party sources.) Further, it needs to be spelled out more clearly what, if anything, would exclude a source from being used in wikipedia at all, and what types of statements, if any, are unacceptable to cite anywhere even as opinions. I think that the current rules prohibiting certain statements from being referenced are perhaps being misinterpreted by some as completely excluding any questionable source which contains such statements. Beyond that, the issue of what information is appropriate for what articles is primarily covered by WP:NPOV. PSWG1920 (talk) 08:21, 20 March 2008 (UTC)

My proposal for rewriting WP:SELFPUB

I think it's fine except the extraordinarily vague qualifier 'unduly' should be removed. The term "self-serving" means "serving one's own selfish interests" - why do we need the "unduly" qualifier? Are there some guidelines for when self published material that serves the selfish interest of the publisher crosses the line into unduly serving the selfish interest of the publisher? Dlabtot (talk) 22:14, 18 March 2008 (UTC)

Let me ask your opinion here (I'd also like to hear from anyone who wants to weigh in.) Should literature promoting Young Earth creationism ever be an acceptable source for wikipedia (and is it ever acceptable according to a strict interpretation of the current Verifiability policy)? What about web sites promoting geocentrism? PSWG1920 (talk) 10:07, 19 March 2008 (UTC)

The policies apply to all topics. I'm not sure what the point of your question is and what it has to do with dropping the word "unduly" from WP:SELFPUB. Dlabtot (talk) 13:04, 19 March 2008 (UTC)

I wasn't really talking about the word "unduly", I am asking a general question to try to get a feel for how you understand the policy regarding sources. PSWG1920 (talk) 13:28, 19 March 2008 (UTC)

Dlabtot I think "unduly" is in there because all self-published websites are self-serving to some degree. They're there to do what their authors want them to do, so of course the information is self-serving. Unduly to me means that the author's need has not been balanced with the need to serve the reader - a page that's more about letting the author promote whatever their thing is than about providing information that a reader could reasonably be expected to seek. I think it's always a judgment call - in part because it depends on context. -- SiobhanHansa 21:56, 19 March 2008 (UTC)

Everybody promotes a point of view, so such a definition of "self-serving" would be too broad. Factors involved would be whether it describes its viewpoint in an informative tone, whether it constantly uses superlatives to describe a person or organization, and similar. Does it provide information or present an exposition of a viewpoint, or is it more in the nature of advertising? This has to do with the way the viewpoint is presented, and has nothing to do with the viewpoint itself, certainly not whether one agrees with the viewpoint or not. Young Earth Creationism, could be done either way, as could essentially any other viewpoint. This has nothing to do with whether editors personally believe the viewpoint true. Best, --Shirahadasha (talk) 04:55, 23 March 2008 (UTC)