Wikipedia talk:Verifiability/Archive 18

Archive 15

Archive 16

Archive 17

Archive 18

Archive 19

Archive 20

→

Archive 25

Recent changes -- necessary?

I don't know that I agree with the necessity of this recent change -- it seems to make the policy needlessly wordy, and discusses the policy itself (self-referencing) instead of just stating the policy. Thoughts? /Blaxthos 08:22, 12 April 2007 (UTC)

I think it is necessary (this is why I fixed it) because of a step forgotten between verifiability (the goal) and need of source (the mean). The rewording allows to be specific : "Assertions that are likely to be challenged should therefore be verifiable, or be removed", which is always true. In the previous version the conditions that led to remouval were hazy, and (in my experience) led to endless disputes. But of course, my wording may not be perfect. Michelet-Me laisser un message 09:05, 12 April 2007 (UTC)

---

Proposed intro

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader should be able to check by himself, without undue effort, whether material added to Wikipedia should be given credit or not. Any material that is challenged casts a doubt on the encyclopedia's quality, and break the consensus on its content. Assertions that are likely to be challenged should therefore be verifiable, or be removed.

The easiest way to verifiability -and the only possible one for quotations- is to provide a reliable source. Wikipedia:Verifiability is one of Wikipedia's core content policies. The others include Wikipedia:No original research and Wikipedia:Neutral point of view. They should not be interpreted in isolation from one another, and editors should try to familiarize themselves with all three. Jointly, these policies determine the type and quality of material that is acceptable in Wikipedia articles.

---

OK, the Wikipedia:Consensus procedure now asks you to discuss why this proposed introduction was reverted. So if you reverted it - what's wrong with it? Michelet-Me laisser un message 17:24, 12 April 2007 (UTC)

What was the issue you were having trouble with on the French Wikipedia that you thought this change might help? SlimVirgin ^(talk) 17:10, 12 April 2007 (UTC)

Extremists that tend to defend POV statements by supressing everything they don't like, on the basis that it has no source. This gives endless discussions on formal arguments, and most of the time the resulting article finds itself in a terrible shape. My problem is to re-focus such discussion on the goal itself (verifiability). Michelet-Me laisser un message 17:15, 12 April 2007 (UTC)

Hello, Michelet. When en: was the size of fr:, we had many conversations like this, and many people believed what you believe. Over time, however, and after some incidents (example) that en: has been through but haven't happened to fr: yet, most en: editors now believe that the danger of good unsourced material being removed is less of a problem than allowing uncited information to remain in articles while people argue about it. fr: will have to find its own level of comfort with uncited material being in articles. I suspect that over time, fr: editors will come to much the same conclusion that en: editors have, as it becomes harder and harder to know every editor even in small fields of interest (fr:'s search engine visibility is already comparable to en:'s). I may be wrong, and fr: may decide to do something very different; and we can all learn from how different projects handle such decisions. That said, fr: is blessed with a number of very thoughtful, very dedicated users, and a conversation about how fr: should deal with editors removing unsourced information really belongs at fr:. Jkelly 18:01, 12 April 2007 (UTC)

It depends on what area one looks at; I know one fairly technical, and highly POV, social science article, where the resident POV-pusher routinely harasses readily sourceable inclusions by newbies which don't suit him. They go away more often than they learn our citation methods. Septentrionalis PMAnderson 00:22, 13 April 2007 (UTC)

"a conversation about how fr: [...] belongs at fr:" Theoretically true, but practically, en: policy is directly translated into fr: without being discussed. So the only way a change in fr: policy may be obtained is by changing it for the whole WP foundation, on en: Michelet-Me laisser un message 18:41, 12 April 2007 (UTC)

Thanks for sharing your respected en: experience, Jkelly ;) but as a matter of fact, I agree with all that you write. Maybe I made myself unclear, I think this is not the point. If you read the proposed version, you will find that there is little difference between what I propose and the practical conclusion that "everything should be sourced". Mainly, the difference is that it allows for internal link (see article XYZ), trivial observations (the sky is blue) and largely consensual material (London is a Capital city). This will not (IMHO) lead to new similar incidents.

My concern is changing the focus between sourcing and verifiability (which is after all the principle's name). ♦ When focusing on sources, articles tend to evolve naturally toward citation collections (wikisource excerpts?), and arguments tend to focus on form: presence or absence of source. ♦ The issue here is to try and change the trend: focus on verifiability, write fluent articles that aim at transmitting knowledge (and not quotations), and argue about verification (credible or not). ♦ The idea is just to invert the focus induced by the present formulation: clearly state that "Assertions that are likely to be challenged should therefore be verifiable, or be removed", and (only) then introduce sources as the natural (and most of the time, only) mean to verifiability.

What do you think? Would that be evil? :( Michelet-Me laisser un message 18:41, 12 April 2007 (UTC)

Michelet, the current wording has served us well. I think "The threshold for inclusion in Wikipedia is verifiability, not truth" (or 'attributabily' in WP:ATT) is a crucial tenet of our content policies and our encyclopedia. I don't think we should change it into more complicated and less clear verbiage. Crum375 18:54, 12 April 2007 (UTC)

So what? The proposed intro says the same thing. Michelet-Me laisser un message 18:59, 12 April 2007 (UTC)

It was hard to know what you were saying, Michelet. For example, the "easiest" way to verify is with a reliable source. What other way is there? And "'Verifiable' in this context means that any reader should be able to check by himself, without undue effort, wether [sic] material added to Wikipedia should be given credit or not" seems meaningless to me. SlimVirgin ^(talk) 19:00, 12 April 2007 (UTC)

♦ Other ways are: explanations, internal links to specialized articles... that's about all. ♦ The point of "undue effort" is simply to weight the cost/efficiency for additional justification. If justification is easy, no citation is required. ♦ Sorry for the typo (I just claim an en-3, see?) - couldn't you try to contribute to a possible meaningfull redaction? Michelet-Me laisser un message 19:41, 12 April 2007 (UTC)

It's really fine as it is. Reliable sources are the only way to verify, in the way we use the term. Not internal links or explanation. If your problem is on the French Wikipedia, why do you want to change the policy here? SlimVirgin ^(talk) 20:48, 12 April 2007 (UTC)

While you are at it, I suggest that the first sentence "The threshold for inclusion in Wikipedia is verifiability, not truth" to be changed, since the actual bar for inclusion is higher than just verifiability. See Wikipedia:Notability (only sufficiently "notable" topics may be included), Wikipedia:What Wikipedia is not (some types of material just are plain not suitable for an encyclopedia, period), Wikipedia:Conflict of interest ("vanity" is not allowed in an encyclopedia) and other official policies and guidelines that all, collectively, set a bar for the type and character of material to be included in the encyclopedia. Not everything verifiable is suitable for inclusion in Wikipedia! So it's part of the threshold but not the whole thing. mike4ty4 00:11, 15 April 2007 (UTC)

A threshold may be necessary but need not be sufficient. SlimVirgin ^(talk) 19:10, 16 April 2007 (UTC)

Is this a reliable source?

Note: this is a continuation of a discussion from Wikipedia talk:Attribution/Archive 14.

Is [1] a reliable source for this:

"The south end was opened up in 2003 and the new plan was put into effect on September 20, 2004."

It's self-published, but the author has been cited by newspapers, including:

New York Times: "They were converted from horse power to an electric conduit system around 1898, said Joseph Brennan, a railroad and subway enthusiast who has created an..."

New York Sun: "There are eight completely abandoned subway stations, a subway historian, Joseph Brennan, said."

Columbia Daily Spectator: "According to Joseph Brennan, Columbia Senior User Services Consultant, who also studies subway history and life, the City Hall station was one of the first..."

I have not been able to find any newspaper articles or MTA press releases describing this change, but multiple sources which may be unreliable confirm the September 20, 2004 date. --NE2 23:13, 22 March 2007 (UTC)

Does anybody dispute the assertion? Septentrionalis PMAnderson 00:29, 23 March 2007 (UTC)

Not as far as I know. --NE2 00:35, 23 March 2007 (UTC)

NE2, he wouldn't be an RS for anything contentious, but if this isn't disputed, you could perhaps use him but attribute the edit to him in the text. Something like "according to Joseph Brennan, a private xxx enthusiast ..." Strictly speaking, only professional self-published sources may be used, so if anyone objected to that edit, it would have to be removed, but the point is that no one is likely to object (reasonably) if the material is completely non-contentious. SlimVirgin ^(talk) 03:08, 23 March 2007 (UTC)

You do know that some people object to such references on principle, right? Don't get me wrong, I don't see anything wrong with using the source in this case, but there's always someone who'll remove such a source because "it isn't a reliable source". --Conti|✉ 03:16, 23 March 2007 (UTC)

Why should there be a lower standard of verifiability for non-contentious material? Aren't there people who come along and read the encyclopedia every once in awhile rather than just editing? Isn't saying a source is good enough until challenged pretty much the same as saying it is not a reliable source?—eric 04:07, 23 March 2007 (UTC)

All sources have limitations and if we only had claims that were without doubt by anyone we would only have a blank page. The point is to be as useful and credible and accurate as we possibly can be so we have all these policies to help us do that. But the policies only work in the context of intelligent people using good sense. The policies can not substitute for actual informed understanding and reasonable judgement. Keep in mind our objective to produce an unbiased free (libre) credible encyclopedia and you won't go far wrong. WAS 4.250 06:10, 23 March 2007 (UTC)

"Completely non-contentious" could cover 80?, 90? percent of Wikipedia content. Would this source be appropriate for a good article? a featured article? or only the obscure articles which few editors are likely to visit? If, as you rightly suggest, an editor familiar with the subject matter uses good sense and decides this source is appropriate to use why should some random editor be able to come along and force it's removal? Or why should an editor working on non-contentious material be able to slack off and apply a lower standard of verifiability? I read SV's statement as: this source is good enough for our readers, but not good enough for any random editor that stops by the article.—eric 08:09, 23 March 2007 (UTC)

I see this person's research is referenced several times in wikipedia. After brief googling, I see that it is possible to write a wikipedia article, in which, using solid references, it may and should be demonstrated that Brennan is a recognized expert in this very narrow topic. After that he will easily fit the provision that self-publishing of an expert are amdissible references in his field of expertise. `'mikka 06:29, 23 March 2007 (UTC)

---

I found another source for the date: an update to "Tracks of the New York City Subway" by Peter Dougherty. This book is self-published, but has been sold at the New York Transit Museum, and is thus "officially" recognized by the NYCTA as being a good book.[2] Is this reliable enough? --NE2 12:31, 14 April 2007 (UTC)

Given that both Doughery and Brennan have been quoted and interviewed by various newspapers, perhaps they count as "amature experts" and thus have a greater level of reliability than the average self-publisher? Blueboar 12:40, 14 April 2007 (UTC)

Probably, though this policy or whatever it is (as well as Wikipedia:reliable sources) says "well-known, professional researcher writing within his or her field of expertise". I thought I had put this on Wikipedia talk:Reliable sources rather than here (Wikipedia talk:Attribution); do you think I should move this there? --NE2 13:10, 14 April 2007 (UTC)

I'd suggest WT:V as the best place for this kind of query at the moment. It involves a question at policy level: the policy as worded precludes the source. I'd say, however, that the policy is wrong in this case, and the authors being acknowledged experts in the subject have adequate status to allow the use. JulesH 10:50, 15 April 2007 (UTC)

Unless anyone objects, I'm going to add this fact to BMT Nassau Street Line and cite both sources, with HTML comments clarifying why the sources are probably reliable. There may actually be a fully reliable source - the NYCTA probably posted fliers about the change, since some platforms are no longer used - but I have not been able to find a photograph (for instance I see no fliers in [3] or [4]). There may also be an issue of verifiability there, unless a copy or photograph of the flier is kept somewhere like a library. --NE2 10:12, 16 April 2007 (UTC)

Truth v Verifiability

Quote: The threshold for inclusion in Wikipedia is verifiability, not truth - the etymology of the word "verifiability" is the latin word for truth, verum, so this statement is a contradiction.

— superbfc [ talk | cont ] — 22:41, 15 April 2007 (UTC)

We know. We tried recently to change it — see Wikipedia:Attribution — but there was resistance. SlimVirgin ^(talk) 19:05, 16 April 2007 (UTC)

The definition of verifiability is:

that any reader should be able to check that material added to Wikipedia has already been published by a reliable source.

this says nothing about truth, and there are sound reasons why many editors prefer that it doesn't; chiefly that cranks and nationalists insists that their PoV is Truth, and that Wikipedia must include it unless we can prove otherwise to their satisfaction. Would adding such a paragraph assuage Superbfc's concerns? Septentrionalis PMAnderson 20:27, 16 April 2007 (UTC)

But that is not what the word "verifiable" means. It doesn't mean "already published" (which is how we use it); it means "confirmed as true," which is precisely what we don't mean. SlimVirgin ^(talk) 20:35, 16 April 2007 (UTC)

It seems to me that within the boundaries of Wikipedia, we mean 'verifiability' as the ability for a typical reader to establish as true that the statement that is included in an article was in fact made by the specified source, and that the source is reliable. In other words, the 'truth' for us is the validation or verification of the attribution, not the essence of the subject itself. It is because of possible confusion about this point that we tried to move to 'attributability', to remove all doubt as to what we mean. Crum375 20:47, 16 April 2007 (UTC)

Whereupon the change got slammed as abandoning "truth". Perhaps another paragraph outlining the argument that attributable statements are likely to be true? Septentrionalis PMAnderson 21:11, 16 April 2007 (UTC)

Because people didn't understand that we don't use verifiability in the right way. We're making fools of ourselves, in fact. SlimVirgin ^(talk) 21:56, 16 April 2007 (UTC)

You are assuming that there exists 'absolute truth', and assuming it does exist, that the bulk of the 'experts' today are closer to it than others. Both are far from established facts, depend on the subject and context, and we can only speculate about them. Our fundamental goal here is not to present absolute truth as to the subject matter — it is simply to report what reliable sources have published about the issue, in a neutral and balanced manner. Any attempt to try to establish the absolute truth is outside our mandate as a tertiary publication and would violate WP:OR and WP:NPOV. Crum375 21:20, 16 April 2007 (UTC)

No, I am not assuming absolute truth; I am getting very tired of being assigned to positions which I have not supported. In this case, I have gone out of my way to avoid supporting any such thing; I merely propose language which may reassure some of those who routinely argue against this policy. Septentrionalis PMAnderson 21:46, 16 April 2007 (UTC)

Why do you say that "attributable statements are likely to be true"? SlimVirgin ^(talk) 21:57, 16 April 2007 (UTC)

Perhaps "more likely" would be closer. For most definitions of truth (Sextus Empiricus always excepted), an attributad statement is more likely to be "true" than a random assortment of words; a statement which is asserted by all relevant reliable sources is at least as likely to be "true" as any other statement on the subject. All relevant reliable sources have been known to be wrong together; but so have other methods of collecting probable statements. Septentrionalis PMAnderson 22:04, 16 April 2007 (UTC)

To be able to say 'more likely' to be true, logically there has to be a concept of absolute truth, otherwise there would be no defined direction to go to. So if we accept that our goal precludes identifying the absolute truth, then we shouldn't talk about approaching it, either. Crum375 22:21, 16 April 2007 (UTC)

Do you have a source for this argument? For that matter, have I said either "approach" or "direction"? Do let me know, so I can correct this slip of the typewriter. In the meantime, how are you going to reassure those editors who believe and have said that attributability endangers Wikipedia's truthfulness? Septentrionalis PMAnderson 22:34, 16 April 2007 (UTC)

An "attributable statement" is simply one for which a reliable source exists. That tells us nothing about its truth value. SlimVirgin ^(talk) 22:27, 16 April 2007 (UTC)

Congratulations. And what are you going to say to the question: why should we care about collecting these meaningless noises?

Neither of these is my question; I'm a Hilbertian. But I see the PoV which caused WP:ATT to fail once is still in full cry. Septentrionalis PMAnderson 22:34, 16 April 2007 (UTC)

It hasn't failed yet. Which POV are you referring to? SlimVirgin ^(talk) 22:37, 16 April 2007 (UTC)

(What is an "Hilbertian", PM?) ≈ jossi ≈ (talk) 22:46, 16 April 2007 (UTC)

That no effort can or should be made to reassure the substantial percentage of editors who think that "verifiability = truth"; although a simple effort to address a purely verbal confusion would, as Askari Mark said, resolve much of this worry. They !voted against once; they'll do so again. A few words to the effect that an attributable WP is likely to be fairly accurate will sway some of them; we can't know how many until we try, and SlimVirgin declines to try. A truly sea-green and incorruptible mind: Attributability One and Indivisible! Septentrionalis PMAnderson 22:57, 16 April 2007 (UTC)

I have almost no idea what you mean. Verifiability does imply truth, as the word is normally used. The way we use it is idiosyncratic, probably because the editors who started the page were unsure of its meaning. This is a common problem in Wikipedia, but we shouldn't have to live with it forever. We should change the name. That's why we wrote WP:ATT. An attributable statement is not more likely to be accurate. I can think of plenty unattributable, accurate statements, and plenty of attributable, false ones. I'm sure you can too. SlimVirgin ^(talk) 23:01, 16 April 2007 (UTC)

There are exceptions; but if attributable statements are not more likely to be accurate on the whole and on average, why do we bother to require them? Why should anybody bother to read us? What good are they to the encyclopedia? Septentrionalis PMAnderson 02:37, 17 April 2007 (UTC)

An encyclopedia such as Wikipedia is not a book of ultimate truths. It is a book accurately representing published mainstream knowledge, properly balanced, neutrally presented and attributed to reliable sources. If a person wants to get a neutral summary of published materials on a given subject, he comes to us. If he wants to get 'the truth', he goes to his local place of worship. Crum375 03:05, 17 April 2007 (UTC)

Precisely; and no one in this discussion has said otherwise. It would be nice, in the middle of these delightful anti-clerical rhapsodies, not to offend those who are dismayed by ATT because they think (as I do not think) that it leads away from mere humble factuality. Septentrionalis PMAnderson 03:51, 17 April 2007 (UTC)

Or more simply, when you've had yes for an answer, can some of us go back to building consensus for it? Septentrionalis PMAnderson 03:53, 17 April 2007 (UTC)

You'll have to forgive my denseness, but I am not following. ATT was not a change in policy - it was merely a packaging issue, so I don't see how the ATT merge would have any bearing on this discussion. Also, I don't understand where we need any fresh consensus - we already have long-standing consensus for the point that Wikipedia is about verifiability or attributability, not [ultimate] truth, as you seem to agree. Crum375 12:42, 17 April 2007 (UTC)

It was widely perceived - and was denied consensus - as being a change in policy. It certainly wasn't intended to be a change; and I don't happen to believe it was. But it is that consensus which needs to be built.

A useful first step would not scaring away editors who are genuinely worried about factuality by suggesting that they want Wikipedia to be a theological exercise. Along the same lines, SV's last comment in this section can be read as suggesting that we don't care about whether what WP says is the case, as long as we have a paper trail. I hope she doesn't mean that, but means that being occasionally wrong is a cost of our policy.

And here I'm not talking, I repeat, about any high-faluting "absolute" truth. (Nor am I talking about a concern I share, although many editors clearly do.) One example that was brought up during the ATT discussion is a particular covered bridge, on which our source is a newspaper article, some years old. Our article says, on that authority, that the bridge is open; apparently, since then, it has been closed and converted into a barn. [Apparently, since then it has been reopened] What to do about this case is not clear: some editors would merely express regret, some would amend to "it was openclosed , as of 19--"; some would move the statement to the talk page, pending present evidence. But whatever we do, saying "it doesn't matter" is divisive and silly. Septentrionalis PMAnderson 16:47, 17 April 2007 (UTC)

(outdent) You said above "For most definitions of truth (Sextus Empiricus always excepted), an attributad (sic) statement is more likely to be "true" than a random assortment of words". That still assumes that there is some ultimate truth, and that we are striving to achieve it. But that is simply wrong - we are not trying to achieve any ultimate truth - that is left for others - we are just trying our best to present the existing human knowledge, neutrally balanced and properly attributed to published reliable sources. As far as the ATT merge issues - I must admit that I have yet to understand what repackaging 3 policies into a single one has to do with the substance of the policies. Crum375 17:48, 17 April 2007 (UTC)

• No it doesn't. It merely assumes that it is possible to assert (in some sense, however loose or non-ultimate, of "true") that "it is true that this bridge has been converted into a barn". Sextus Empiricus, would, I admit, deny the possibility; but I think attempting to write a policy which commits us to epistimological skepticism is in itself a violation of consensus. Septentrionalis PMAnderson 04:00, 18 April 2007 (UTC)
  • Consensus is not applicable when dealing with our core mission. Verifiability in the sense that we seek attribution to reliable published sources and not ultimate truth, is part of our core mission, along with NOR and NPOV. If tomorrow 90% of some forum decide otherwise, that mission would still not change. Consensus is fine for implementation details and auxiliary policies, but not the core mandates. Hence, we are still building a tertiary source, which means we neutrally summarize attributed published material in a balanced fashion. We do not strive to achieve 'truth', except in the sense that our readers should be able to verify as true that our sources in fact said what we say they said. Crum375 04:25, 18 April 2007 (UTC)
    • This is both true and absolutely irrelevant to the point at issue. Both Crum and I agree that WP:NPOV and WP:ATT express our core mandate; therefore it is not in danger. Therefore the change between them is a question of implementation, and does require consensus. Septentrionalis PMAnderson 04:42, 18 April 2007 (UTC)
      • If what I say is "absolutely irrelevant", please explain how - to me it is the essence of this thread. I don't see what you mean about ATT and NPOV and "the change between them" - ATT is merely a repackaging of NOR and V as they are, it does not address or include NPOV per se, and certainly does not contradict it. As far as the message you left me on my Talk page (and please let's keep the discussion here so others can see it), I am not sure what it means. If you are saying that perhaps in some cases we should use OR instead of attribution to reliable published sources, I would disagree, as it would be clearly in violation of our core principles. Crum375 11:54, 18 April 2007 (UTC)
        • • I asked you three questions:
            • Is "This covered bridge has been converted into a barn" an ultimate truth?
            • If it is, how? What definition of "ultimate" are you using?
            • If it isn't, why are you dragging "ultimate truth" into this discussion? I'm only interested in non-ultimate truths.

            That's three questions. It would help if you would answer at least one of them. Septentrionalis PMAnderson 19:04, 18 April 2007 (UTC)

(outdent) Here are my answers. They are repetitive because your questions are repetitive:

1. Any statement we publish, e.g. about the status of a bridge or a barn, that is not attributed to a published reliable source, and is challenged or likely to be challenged by anyone, is WP:OR and would therefore violate our core content policy as an encyclopedia or tertiary publication. 'Ultimate truth' is anything that deviates from strict attribution.
2. If it's OR, it seems to be an attempt to find an 'ultimate truth', which is outside our mandate here. My definition of 'ultimate' is an attempt to find an absolute ('final') truth, which would be OR.
3. I am referring to ultimate truth because any attempt to include a statement in an article which is not properly attributed to a published reliable source is OR, i.e. an attempt to establish an absolute or ultimate truth. All we care about here is attribution, neutrally presented and balanced.

Hopefully this makes my points clearer. Thanks, Crum375 22:21, 18 April 2007 (UTC)

1. • So the covered bridge being converted into a barn is an ultimate truth, because it hasn't been covered by a newspaper. I don't know about anyone else, but that's a novel doctrine to me; it sounds vaguely like the Vienna Circle, but which of them would deny that the bridge is an observable?
   • Is there any other support for this? If not, should we base policy on it? Septentrionalis PMAnderson 21:11, 21 April 2007 (UTC)

Claims of consensus

The following. from WP:RS is a universal rule. If we agree on it, it should be here as policy. If we don't agree on it, it shouldn't be anywhere. Which? Septentrionalis PMAnderson 20:27, 16 April 2007 (UTC)

In particular, claims of consensus must be sourced. The claim that all or most scientists, scholars, or ministers hold a certain view requires a reliable source, which says that most of them hold that view. Sources which simply assert the view are not enough; if that is all the sources say, opinions should be identified as those of particular, named sources.

This is a policy page, so we can't get into detail. We say that quotations and any material that is challenged or likely to be challenged needs a source. That obviously includes "all scientists accept the theory of man-made global warning." SlimVirgin ^(talk) 21:46, 16 April 2007 (UTC)

This is ignoring the problem, which is that people write "all scientists support X" and then source that with one scientist, who does support X, but says nothing about all scientists. Septentrionalis PMAnderson 21:50, 16 April 2007 (UTC)

I agree that it's a menace, but people do all kinds of crazy things on this website. We can't go into that level of detail on a policy page. Why not add it to (or keep it in) RS? That's only a guideline, so this kind of thing is appropriate there. SlimVirgin ^(talk) 21:59, 16 April 2007 (UTC)

Because if it's in a guideline, it will be under a tag which says: this is not written in stone, it has exceptions. The editors we both deplore will claim their case is the exception. If this is a universal rule, it belongs here. Septentrionalis PMAnderson 22:10, 16 April 2007 (UTC)

It is bad practice to write policy based on exceptions. There are hundred other situations in which the common sense and best judgment of editors will be needed. We do not write these into policies. ≈ jossi ≈ (talk) 22:14, 16 April 2007 (UTC)

If it requires judgment and is subject to common sense, we put it in a guideline; that's why we have them. Septentrionalis PMAnderson 22:19, 16 April 2007 (UTC)

If you look at this policy, it's very tight and deals only with principles. We don't give examples. So if we were to add yours, it would stick out like a sore thumb, then other people would want to add theirs etc. It was for this reason that we wanted to develop ATT and have the accompanying FAQ, but you were one of the people who opposed that, so now we're back to the rotten old three-page system: two tight policy pages, and one mess of a guideline. I suggest you leave it out of RS too, so it's not under a "not set in stone" tag, and just use the "challenged or likely to be challenged" umbrella, which is policy. SlimVirgin ^(talk) 22:25, 16 April 2007 (UTC)

A third possibility exists, you know: it is generally true, but there are too many exceptions to have it be policy. This is the main reason for guidelines existing. JulesH 15:28, 17 April 2007 (UTC)

I agree in general; but what exceptions are there to this rule? Septentrionalis PMAnderson 04:16, 18 April 2007 (UTC)

Which rule? SlimVirgin ^(talk) 22:27, 18 April 2007 (UTC)

The one quoted in italics at the top of this section, and called a rule. Septentrionalis PMAnderson 22:49, 18 April 2007 (UTC)

As I said below, I think it needs to be left to individual editors on the page, even though broadly speaking I agree with you. Who is calling it a rule? SlimVirgin ^(talk) 23:04, 18 April 2007 (UTC)

Journalists

Jules, I recall that there was consensus for this to be removed. [5] I've seen it be abused quite a few times, and I suppose the key question is why journalists should be an exception. SlimVirgin ^(talk) 20:18, 16 April 2007 (UTC)

Because a journalist does research on the area he covers as his profession. Septentrionalis PMAnderson 21:09, 16 April 2007 (UTC)

But what if he is writing outside his area of expertise, assuming he has one? SlimVirgin ^(talk) 21:10, 16 April 2007 (UTC)

Then he should be treated like a biologist, writing outside his profession. The solution is to reword to make clear that that limitation applies to both. Septentrionalis PMAnderson 21:12, 16 April 2007 (UTC)

Your edit [6] doesn't help. If he really has researched a particular field, he's covered by the "professional researcher" description. If not, then the question remains: why are journalists exceptions? SlimVirgin ^(talk) 21:16, 16 April 2007 (UTC)

I agree, the exception always seemed puzzling. Why was it originally added? It's been in the text for a long time. Gimmetrow 21:30, 16 April 2007 (UTC)

I don't recall why it was added originally, but I've seen it abused several times, so I think it's time to get rid of it. A journalist who is genuinely known as a researcher in a particular area is covered by our "professional research" exception. Any other journalists should be treated the same as any other non-expert self-published source. SlimVirgin ^(talk) 21:43, 16 April 2007 (UTC)

I also think there should be no special exclusion for any specific profession - it is enough to refer to a 'researcher', and that should cover all professions. Crum375 21:46, 16 April 2007 (UTC)

Very likely because someone argued that a journalist is not a "researcher"; if there is consensus that journalists are, there should be no problem with "professional researchers, including journalists." Septentrionalis PMAnderson 21:48, 16 April 2007 (UTC)

I think a professional journalist can be a good researcher, and often has access to other assistant researchers, so the bottom line is the same. I see no reason to single one profession out. Crum375 21:58, 16 April 2007 (UTC)

Journalists who are paid to research are by definition professional researchers. There's no need to say "including journalists," because it doesn't include all journalists, obviously, so it will lead to confusion. Why not just have one clearly written sentence for once, without confusing clauses added after the fact that then need to be explained somewhere else? SlimVirgin ^(talk) 22:01, 16 April 2007 (UTC)

In short, we don't need to address real problems, revealed by experience (and I recall following this page one of the times "journalists aren't real researchers" came up), because Wikipedians are rational, and will see in a policy everything that is logically implied by it. What Wikipedia have you been editing? It certainly isn't this one. Do what you please; we can pick up the pieces after the collapse. Septentrionalis PMAnderson 22:17, 16 April 2007 (UTC)

The phrase in question appears to originate from a discussion of whether a source is a noted expert in the field and follows appropriate professional standards of ethical investigation, either of journalism or academia. The text on blogs was rewritten 29 September 2005 to state that blogs were not acceptable unless set up by a "well-known person, or a known professional journalist or researcher in a relevant field". The parallel construction suggests that "researcher" here meant academic researcher. Is that where "journalists aren't real researchers" came from? I am also curious how this phrase in policy has been abused. Are people trying to say all bloggers are journalists? Gimmetrow 22:44, 16 April 2007 (UTC)

Yes, that sounds like the occasion I was thinking of; but there may have been a later move to remove "journalist" as well. Septentrionalis PMAnderson 22:57, 16 April 2007 (UTC)

• I would have interpreted "researcher" in this context to mean academic researcher, or soemthign clsoe to one. Some journalists, particualrly investigative journalists, do soemthing close to research, but many do not, at least as i use the word "research". However an experienced professional journalist, writing in the field s/he has been accoustomed to cover, and thereby putting his or her reputation the line to at least some degree, is in a sense a living secondary source. In any event, most of the same reasons for accepting self-published serious content from researchers as reliable also apply to journalists IMO. If we simply leave it to implication that journalists are included within "researchers", we open the door to long arguments that journalists are not covered by thsi provision. No one has indicated any resaon why retaining the long standing wording is a harmful -- at worst it is redundant, and IMO it is far from intuitively obvious that "researcher" includes "journalist". I therefore strongly urge that "researcher or journalist" be restored to the policy, and it seems to me that changing a long established wording on thsi point should have waited until after there was a clear consensus to do so. DES ^(talk) 23:33, 16 April 2007 (UTC)

The problem with "journalist" and "researcher" is that they have no clear definition. The issue I think we're trying to address is that they are recognized as experts in their field. Perhaps we can find a clear definition for expertise, such as having published articles on the topic in major publications within the prior 10 years. Otherwise anyone who evergot a freelance article printed can claim to be journalists for the rest of their lives, and thus automatic reliable sources on any topic. -Will Beback · † · 23:43, 16 April 2007 (UTC)

There has been an agreement to do this, DES. I think it may have been on the ATT talk page at one point. If you think journalists are a special breed of expert, you have to argue why. And they certainly do regard themselves as professional researchers when they're specializing in a particular area. SlimVirgin ^(talk) 23:53, 16 April 2007 (UTC)

If anyone else made such a claim, SV would demand evidence. I don't, because it doesn't matter. WP:Consensus can change. Is there any reason not to say journalist now? Septentrionalis PMAnderson 00:14, 17 April 2007 (UTC)

I wasn't part of the original discussion, AFAIR, but to me it seems discriminatory to single out just one profession out of many. Why not mathematicians? Health scientists? Authors/Biographers? Legal scholars? There is a near-infinite variety, and I see no reason to just mention one profession. Crum375 00:22, 17 April 2007 (UTC)

We can include them when and if somebody objects that "researcher" (or DES's proposal below) doesn't cover them. Septentrionalis PMAnderson 02:34, 17 April 2007 (UTC)

(after ec) I don't recognize discussion on the ATT talk page (which i regard as merely a rejected proposal) as establishing any consensus for this page. I see that some qualification may be needed for both "journalist" and "researcheer", but it seems to me that the potential for abuse is as large in the one case as in the other, and until we can come up with a good qualification text, then i think the previosu wording should stand, and I am incinded to restore it now. Also, IMO the issue is not so much one of expertise as of professional ethics. A professional journalist who spouts off wildly on a blog or other self-published forum can lose vital reputation, or even a job. this imposes restraint in many ways comperable to the standard editorial process (which is, after all, far from infallible). An experience journalist may not be an "expert" in the sense that an academic is, but typically s/he will know a good deal about the field that s/he routinely covers, will understand what are reasoanble and unreasoanble cliams, and will have a reputation at risk. I would favor qualifing "journalist" with "experienced professional" until a better qualification standard gains consensus. DES ^(talk) 00:32, 17 April 2007 (UTC)

If we wanted to change this to "researcher, experienced professional journalist, or recognized expert in the field" I wouldn't object. DES ^(talk) 00:34, 17 April 2007 (UTC)

Works for me. If we need to qualify journalist specifically, "journalist covering the field" sounds rignt. Septentrionalis PMAnderson 02:32, 17 April 2007 (UTC)

"Researcher" is too vague. Any wikipedian can claim to be a "researcher" even if they are unknown, unpaid and unqualified. We don't need to open that door to abuse. It should be clear that we're discussing researchers who have being a researcher as their profession, not hobby. Wjhonson 05:39, 17 April 2007 (UTC)

How about "professional researcher, journalist or other recognised expert in the relevant field"? JulesH 15:18, 17 April 2007 (UTC)

The problem with "researcher" is that it carries a strong connotation of academic, especially scientific research. In fact, at least one dictionary (Princeton WordNet), gives as its only definition of the word "a scientist who devotes himself to doing research". Yes, it does have other meanings, but this is what people are going to think first when they see the text. I'm in favour of changing the wording, as suggested above, to something that clarifies that we mean researcher in the wider sense. JulesH 15:18, 17 April 2007 (UTC)

I'd also agree with Will Beback's suggestion above of limiting the context to somebody who has been published (preferably on multiple occasions) by a notable publication. I don't think time limiting it is useful; we can continue to use sources produced by a trustworthy source long after they've stopped producing new work. In that case, I think "researcher, journalist or other expert whose work has been published in a notable publication in the relevant field" might be the way forward. JulesH 15:25, 17 April 2007 (UTC)

Hmm. "[...] notable professional publication [...]" would be better. Wikipedia is a notable publication, but we wouldn't want every one of our editors to be suddenly classed as experts on any subject whose articles they've edited. JulesH 15:27, 17 April 2007 (UTC)

How about "An experienced professional researcher, journalist, or other expert in the field, who has previously published work in the field, in a publication that would itself be a reliable source." DES ^(talk) 15:36, 17 April 2007 (UTC)

Inserting, with the the fields changed to a relevant field and that field. I think it should probably be shorter, but the contest between brevity and precision is better conducted in place. Septentrionalis PMAnderson 04:13, 18 April 2007 (UTC)

Direct and immediate reversions are not helpful; please tweak either the old wording or the new, and we may meet in the middle. One of these deserves another. Septentrionalis PMAnderson 04:38, 18 April 2007 (UTC)

I went back to "professional researcher, scholarly or non-scholarly." What's important in the sentence is that the person is a professional, and not just someone who has set themselves up as a researcher, and that they've previously had their research in that field published by third-party sources i.e. someone other than themselves. So long as we make that clear, there's no need to start listing all the professions we allow, because we allow them all. A dentist who has been paid in the past to write papers for dentistry journals would be a reliable self-published source. Or an engineer, a journalist, a school teacher, a cleaning lady — so long as someone has paid them in the past to write up and publish their research in a field relevant to the topic of the article, they may self-publish it and be deemed a reliable source in that field by Wikipedia. SlimVirgin ^(talk) 05:27, 18 April 2007 (UTC)

I see that SlimVirgin is engaging in the "slippery slope" fallacy. Nonsense. Just because we say "journalists" does not mean we have to say "hairdressers". Whether journalists are covered has been questioned; that's why several of us want to include them explicitly. When hairdressers become an issue, we can consider whether it is better to include both professions or neither. But I'm not holding my breath until they are. Septentrionalis PMAnderson 05:44, 18 April 2007 (UTC)

I think the point is that we're wondering why you want to mention journalists specifically. We don't get much material from self-published journalists, to the best of my knowledge, so I'm unsure why this is such a big deal. And we wouldn't be allowed to use them unless they had become specialists of some kind in a particular area, which most journalists don't. So we're talking about a very small number of potential sources: journalists who are specialists in a particular area, whose research in that area has previously been published by reliable third-party sources, but who are now publishing their material themselves. Can anyone think of even one example? SlimVirgin ^(talk) 06:05, 18 April 2007 (UTC)

I just dealt with a similar situation, which is why I posted my concern above. An "award winning" freelance journalist/blogger named "Nicholas Stix"/user:70.23.199.239 started adding links to his writings to numerous pages, and fighting when other editors removed them. WP:COIN:70.23.199.239 Then another editor started defending him and re-inserting the links. That other editor insisted that because Stix is a journalist his blogs count as reliable sources.[7] One of the blog main links in question is downright offensive.[8] So I want to make sure that the wording of this policy isn'tt changed to allow "journalists" of this type to be automaticlaly considered suitable sources. -Will Beback · † · 06:28, 18 April 2007 (UTC)

That's exactly the problem. SlimVirgin ^(talk) 21:59, 18 April 2007 (UTC)

Of course, it doesn't say that they're automatically reliable sources: it says they may be reliable sources: clearly this means that other factors need to be taken into consideration -- this would, naturally, include what the source says and the subject area that's covered.

But then gets back to the question of why single out journalists as one profession among all the hundreds of professions we might deem acceptable. SlimVirgin ^(talk) 21:59, 18 April 2007 (UTC)

SV: I see quite a lot of stuff added from journalists' self published sources. For instance, http://www.bbc.co.uk/blogs/theeditors/ is cited on a number of pages, and http://blogs.guardian.co.uk even more. I don't, personally, see anything wrong with using these sources: while they aren't edited prior to publication, they are written by professionals who pay attention to the information they put into them to ensure accuracy. JulesH 07:39, 18 April 2007 (UTC)

Those blogs aren't self-published. They're published and edited by the BBC and the Guardian. They're not really blogs in other words. SlimVirgin ^(talk) 21:59, 18 April 2007 (UTC)

Jules, your point about adding "notable." We already say "reliable," and I'm not sure what notable would mean other than reliable. "These may be acceptable so long as their work has been previously published by reliable third-party publications." SlimVirgin ^(talk) 05:31, 18 April 2007 (UTC)

True. Ignore me, I wasn't reading the entire sentence. JulesH 07:39, 18 April 2007 (UTC)

• Self-published material may be acceptable when produced by a well-known, professional researcher in a relevant field or a well-known professional journalist. These may be acceptable so long as their work has been previously published by reliable third-party publications.

I don't see anything there limiting acceptability based on "what the source says and the subject area that's covered". Jayson Blair is a well-known journalist who's been published in a reliable third-party publication. So apparently, according to this policy, he will be a reliable source for the rest of his life. Separately, how do we define "journalist"? Anyone who ever got a free-lance article published? -Will Beback · † · 00:45, 19 April 2007 (UTC)

Exactly. SlimVirgin ^(talk) 00:48, 19 April 2007 (UTC)

The proposed wording on journalists is:

Self-published material may be acceptable when produced by a an experienced professional researcher, journalist, or other expert in a relevant field, who has previously published work in that field, in a publication that would itself be a reliable source.

This is restricted when applied to journalists; it is restricted with any other class of expert.

The argument for including journalists explicitly is:

1. Everyone in this section appears to be agreed, more or less, on when we include self-published researchers, and that this does include journalists; some of us feel that implicit inclusion is enough. We are discussing only how to convey that agreement to others.
2. But in fact, some editors have argued against the inclusion of journalists which do meet our criteria, because they're not researchers. This would be easier to answer if they were mentioned explicitly.
3. Some editors have also included journalism which is not written by experts, or out of their field. This might well be easier to deal with if we could cite: "journalist ... who has previously published work in that field, in a publication that would itself be a reliable source."
4. I have not seen either of these problems with any other profession. When do, we can deal with it then.
5. "Fairness" has nothing to do with it. This is an encyclopedia, not an equal opportunity commission.

Which of these numbered points is obscure? Septentrionalis PMAnderson 21:39, 21 April 2007 (UTC)

I think researchers (scholarly or non-scholarly) adequately conveys that we mean researchers in any profession or field. As I asked above, can anyone think of a real example of a journalist who has specialized in a particular area, who has had his work in that area published by a reliable, third-party publication, but who now chooses to self-publish his work in that area instead of having it published elsewhere, and whose self-published work we might want to use as a source? I can't think of a single example. Until we know we have some actual examples, I feel there's no point discussing whether we want to make special mention of them. SlimVirgin ^(talk) 22:04, 21 April 2007 (UTC)

Let's see what happens; if no more confusion arises, fine. Septentrionalis PMAnderson 05:25, 27 April 2007 (UTC)

Can we at least add a time limit onto having been published, and a limitation on specialty? Something like 5 years?

• These may be acceptable so long as their work in their specialty has been previously published by reliable third-party publications with five years.

Remember, we consider college newspapers to be reliable sources. We need to put some limits on this journalism exemption. -Will Beback · † · 06:00, 27 April 2007 (UTC)

I don't think it makes sense to use a time limit. If an article is a reliable source today, it will presumably be a reliable source in five years' time, also (although it may be a reliable source to out-of-date information). I do think your concern about journalists who have published in small publications should be addressed though; perhaps 'notable' should be added to 'reliable' to further restrict the acceptable venues of publication? JulesH 08:07, 27 April 2007 (UTC)

Sorry, I'm coming to this discussion late. I have a small amount of experience as a journalist, and I think it would be a mistake to explicitly mention journalists. Yes, some of them, by virtue of their research, eventually become notable experts on a topic. But my feeling is that the majority wouldn't qualify as professional researchers. I have a certain visibility in a very narrow field of technology, and some of my readers do think me an expert. But the truth is, I'm not. I wince at the thought that somehow a statement of mine would be treated as authoritative. I like the current version. TimidGuy 11:58, 27 April 2007 (UTC)

I think the college paper is a non-problem. The wording above would require that the journalist be an expert in the field, and have published work on that field. With the possible exception of blogging about the college itself as an alumnus, college publication is unlikely to pass that standard. But I'm willing to leave things as they are until it becomes a substantive issue again. Septentrionalis PMAnderson 15:55, 27 April 2007 (UTC)

Citing unpublished emails

Wikipedia:Templates for deletion/Log/2007 April 19#Template:Cite email discusses a template to use unpublished emails as reference. So far most !votes are Keep. PrimeHunter 00:06, 20 April 2007 (UTC)

Thanks for letting us know about it. SlimVirgin ^(talk) 00:55, 20 April 2007 (UTC)

Unbelievable. Jayjg ^(talk) 01:00, 20 April 2007 (UTC)

Homage and Shows Referencing Other Shows

If a show makes a reference or homage to another material that is popular, such as one character from one show dressing in the exact same signature outfit of another character or other similar references, how does one verify such a reference? With an image? With the episode where the reference took place? It will be extremely difficult to find a source where animators, directors, and show creators say "yes, we did this shot or dressed this character in this specific way because we wanted to reference that show." It's well known that such types of references and homages exist, so how does one verify them?

For a classic example, see the Lupin in Pop Culture section in Lupin III.

You verify it with the directors, or some other reputable source, saying that it is a homage. -Amarkov moo! 21:31, 22 April 2007 (UTC)

Indeed. A valid source could be, for instance, a magazine review or some other similar article. JulesH 07:50, 23 April 2007 (UTC)

The problem with what you are suggesting is that, as posed above, directors/writers don't always say stuff about homages in published sources as handy as magazines. What about discussion board posts? Online chats? Fan convention appearances? Blogs? There seems to be a systemic bias against these types of sources, although DVD commentary tracks seem to be unobjectionable. Having said that, there is often a long lag time between when a TV episode airs and when the writer/director actually speaks to any specific homage. It can take years before the DVD of a show is released, what do we do in the meantime? I don't think we need to take Wiki-verifiability or the OR policy to such an extreme degree as to discount the common sense of editors who notice and report on the homage/cultural reference/trivia/whatever. Over-reliance on policy (WP:BASH, WP:CREEP) makes us not see the forest for the trees. Wl219 10:41, 23 April 2007 (UTC)

It's an area of policy I've been actively trying to change for some time now, but there is serious resistance to it. Self-published sources are not (generally) considered acceptable, even for this type of claim. That said, there have been numerous discussions that hold that despite the policy, it is acceptable to use a self-published source from an acknowledged expert. This is an application of WP:IAR, which supersedes requirements about types of source when it applies. The canonical example is that of J Michael Straczynski's postings on various USENET groups concerning Babylon 5, which are almost universally agreed to be acceptable. If you have sources that are on a similar level of authority to these, then I'd suggest using them. Otherwise, I'd say the claims should probably be removed as original research. I don't think noticing similarity between the dress styles of two characters in different productions is enough to call it homage; it is by no means certain that this is the cause of the similarity, and therefore it would definitely be original research to call it one without a source to back the claim up.

You don't need the source to be a director or writer for it to be valid, though. Find anyone published in a reliable source making the claim, and you can duplicate it. If the writer isn't completely authoritative, you have merely to attribute it as their opinion, not absolute fact. JulesH 12:22, 23 April 2007 (UTC)

Coming back to the original message for a moment: the intention may be to say "what do I do when I observe a homage in a programme?". The answer is, if you want to incorporate it in the article, find a source. The program is not a source. Screen captures are not a source. If you can't find a source, it is original research, and it should not be incorporated. I think that is pretty much policy. The best way to deal with this is to ask for sources where you suspect someone has only observed, and if none is forthcoming after a while, delete the original research. Alternatively: maybe the programme is a primary source and observations are allowed. There is a whole lot of popular culture in articles which hinges on this point. Notinasnaid 12:36, 23 April 2007 (UTC)

Indeed; the programme itself is a primary source, and direct observations may be used as sourced statements of fact. However, the way they were used in the article in question amounts to synthesis, as described in WP:NOR. A source that comes to the same conclusion is needed. JulesH 16:35, 23 April 2007 (UTC)

The RuneScape article is biased

And when someone complained about the bias on the talk page, the response was that it was because of this policy. --59.189.61.231 11:59, 28 April 2007 (UTC)

Looking at Talk:RuneScape, I don't see what you're talking about. Neither the words "bias", "WP:V" or "verifiability" appear on the page. Could you be more specific about the issue? JulesH 09:04, 2 May 2007 (UTC)

Non-negotiable

When was the part about this policy being non-negotiable removed from the wording?--Crossmr 04:35, 2 May 2007 (UTC)

Reagan-Matlock

Many of the statements in Ronald Reagan provide undescribed works by one Matlock as the source. I assume this is Jack F. Matlock, Jr.. Since he was an appointee of Mr. Reagan's, can he be considered authoritative, especially as some of the assertions for which he is provided as the source are dubious? John FitzGerald 20:09, 2 May 2007 (UTC)

Conflict between Deletion policy and Verifiability policy

I have posted a discussion on the Deletion Policy talk page about a conflict I read between these 2 policies. Please see Wikipedia talk:Deletion policy#Conflict between Deletion policy and Verifiability policy for specifics on it. I would really like to hear the comments on this. Slavlin 18:13, 7 May 2007 (UTC)

A proposed change to the nutshell:Specific claims must be checkable by non-experts

I propose adding the line "Sources must be cited in a way that permits a non-expert to check specific claims",

If this nutshell is accepted, I want to propose corresponding changes to this policy and WP:CITE to implement this requirement.

We've had a significant problem with people adding huge bibliographies to the end of articles, resulting in the reader being so snowed by an avalanche of material that there is no effective way of checking whether a particular claim is actually in any of these sources or whether they represent smoke and mirrors added for puffery purposes. Requiring that claims be individually sourced essentially forces footnote-style sourcing in articles containing a large amount of content and/or very controversial claims. This is extra work, but unless sourcing is done in such a way that specific claims can actually be checked against specific sources, our verification policy can be easily defeated by a very modest amount of academic fraud. As an extreme example, suppose someone added the entire Library of Congress catalog to an article. The article may well be sourced, but providing so formidable a haystack means that there is simply no practical way for a reviewer wishing to check sources to determine whether a particular needle is really there or not. This proposed change clarifies that checkers can add maintenance tags asking to identify the specific source(s) behind specific disputed claims, and that disputed claims which are not individually associated with a specific source should be regarded as unsourced and can be removed. Best, --Shirahadasha 20:06, 10 May 2007 (UTC)

I can agree with this... it seems to fit with the intent of an editor who places a {{fact}} tag in an article. He or she is requesting that a specific claim or statement be cited (using the <ref>citation info</ref> format.) Blueboar 12:35, 11 May 2007 (UTC)

What if you used Wikipedia?

On the article about Leslie (Star Trek), I used information from Wikipedia, and somebody keeps tagging it and deleting as as unverifiable. I've tried to find other sources to back my info up, and he says they are all unreliable, and redeleting my information. Do I need to verify information from another Wikipedia page?Ye Olde Luke 22:11, 10 May 2007 (UTC)

Using Wikipedia as a source in an article in Wikipedia is a "self-reference"... not good encyclopedic practice. Each article should stand on its own, and things said in each article should be verifiable to sources outside of Wikipedia. Also, you should know that Wiki's in general have long been considered un-reliable sources, as the cited article may change completely between when you use it as a reference and when a reader looks to verify your statement. In addition, Wikis have no editorial oversight or fact checking, or peer review. We can not assume that the information in a wiki is accurate. Nor do we know the qualifications or expertese of the editors who wrote the article.

What you can do is provide an internal "see also" link to the appropriate wikipedia article. Blueboar 12:45, 11 May 2007 (UTC)

Wikipedia as an indirect source?

First, I presume that the many pages that apparently do no more than mirror wikipedia are not acceptable sources. Assuming so, is there any commentary addressed to a situation where there is a reasonable belief or a fair suggestion that a third party who otherwise has the markers of reliability has in this instance relied upon Wikipedia or one of its mirrors?

Consider, e.g., the statement made by Dr. Susan Davis Allen in a speech that the University of California system was amongst the 8 originally named Public Ivies.[9] Regardless of whether that statement is true, the speech is not academic research and carries an air of limited investigation ("As far as I have been able to ascertain..."). If, indeed, she prepared this speech by using Google, it is highly likely that she found the many, many references to a prior wikipedia article making just that same claim.

Note, if you will, that this particular statement is dicta in her speech, and even if this speech was subject to peer review, or anything like it, it is very, very unlikely that any critic would address this fact. Obviously, this case could be resolved without resolving this particular dilemma -- by consulting the R. Moll's book itself. However, the example still highlights what could be a real and difficult to otherwise resolve concern. (And it strikes me that the more successful Wikipedia is, the more often this issue could arise.)

There are three initially apparent possibilities:

1. If the third party is anything more than a mere mirror, the data it considered do not render an otherwise reliable third party unreliable, without regard to the specific data considered.

2. Rule #1, except where there is reasonable reason to believe that the third party is just echoing wikipedia's statements.

3. Rule #1, except where the third party states or admits wikipedia was the source.

Cka3n 22:31, 10 May 2007 (UTC)

What happens if the Internet Archive removes a site?

On New York Bus Service, I used a cached version of the old website. The site is now blocked via robots.txt. What do I do? --NE2 03:40, 11 May 2007 (UTC)

Un-sourced statements

I removed[10] a bunch of un-sourced material from Verio, per this discussion, but User:Ronz keeps adding it back without sources except a "fact tag" at the top of the article that has been there for days. I don't want to edit war with him over it. Do I have the right to remove un-sourced material from Wikipedia? -- Stbalbach 01:39, 14 May 2007 (UTC)

Unsourced material should be removed, but note that does not mean that you can constantly revert to keep it removed. You should try the steps in dispute resolution. -Amarkov moo! 03:16, 14 May 2007 (UTC)

WP:V is very clear on this point:

*The obligation to provide a reliable source lies with the editors wishing to include the material, not with those seeking to remove it.

It's not up to me to prove it should be removed. The burden of evidence is for those who want to include it. I don't need an RfC to remove un-sourced material. -- Stbalbach 03:47, 14 May 2007 (UTC)

That doesn't mean that you are allowed to edit war to enforce it. Edit warring is bad whether or not you are right. -Amarkov moo! 04:28, 14 May 2007 (UTC)

Agreed. So what do you suggest? -- Stbalbach 04:38, 14 May 2007 (UTC)

Self-published sources

I have reworded this section without intending to change the meaning substantially. The previous wording assumed that if self-published news was worthy, it would also be reported in third-party media. This is not necessarily the case. If Robert Scoble writes something about technology, I believe it is better to cite his blog than to cite a third party rehash if we are relating his expert opinion on a topic. If we are writing about a technology topic in detail, the NY Times recital of what Scoble said may not contain as much detail as the original source. Jehochman ^☎ / _✔ 05:35, 16 May 2007 (UTC)

I keep running into a problem with Internet related articles. Many of the best trade publications are published online only, and some take the form of blogs. These blogs have editors and fact checking just like a paper trade journal, but Wikipedia seems to have an inherent bias against using anything called a "blog". As soon as I cite something like this, people react negatively without even looking at the source. "Blog! It must be garbage. We can't use that as a source." How do we address this sort of reaction? Jehochman ^☎ / _✔ 06:43, 16 May 2007 (UTC)

Agree. This is unfortunately a form of WP:BIAS that will only get worse in the future. The web documents itself, and is the best source for information on web based subjects. And if anything, web sources are gaining credibility as mainstream sources are losing it. Blogs must be used with caution, but should not be devalued as greatly as they are, especially when they are the primary source for primary source data. For example, the IETF conducts its business online--are we to ignore this data because it is only published on the web? Dhaluza 09:42, 16 May 2007 (UTC)

Agreed. I've been thinking for a while now that a reworking of the handling of self-published sources section is necessary, and I don't think the changes made (and reverted by SV) go far enough. SV: I have to admit I don't understand the comment you made when reverting these changes. Could you explain why you feel they were incorrect in more detail? JulesH 10:11, 16 May 2007 (UTC)

The change said something about not using SPSs if non-SPSs are available, but that strikes me as an unwise thing to add, because it all depends on context. SlimVirgin ^(talk) 17:56, 16 May 2007 (UTC)

Could we state: (1) Blogs may be acceptable sources when they have professional or expert editors who ensure the reliability of published material; and (2) Wikipedia does not prefer one media format over another. I'd like something to point at next time the torch-and-pitchfork gang comes after one of my blog references. There's wicked bias against blogs. Here are a few examples of blogs that should be considered reliable sources:

• A List Apart - has expert, volunteer editors (they now sell advertising, and might get paid)
• Search Engine Land - has paid editorial staff
• Matt Cutts - when he writes about Google and search engines, but not when he writes about his cats

Thank you for considering this. Jehochman ^☎ / _✔ 13:03, 16 May 2007 (UTC)

We've been through the wording of this very carefully, here and at ATT. I don't see the benefit of changing it. What do you feel the problem is with the current wording exactly? If the only thing you're worried about is the professional blog aspect, the blogs you're talking about aren't self-published (the ones run by organizations with editors etc) i.e. they're not really blogs. SlimVirgin ^(talk) 17:56, 16 May 2007 (UTC)

The current wording stigmatizes all blogs. Maybe we could resolve this by changing:

"For that reason, self-published books, personal websites, and blogs are largely not acceptable as sources," to

"For that reason, self-published books, and personal websites and blogs are largely not acceptable as sources. Online magazines and journals published in blog format may be acceptable if they have editors and fact checking."

I think this would reduce bias against online magazines that call themselves "blogs." As you said, these aren't really blogs, but people get confused because of the name. Jehochman ^☎ / _✔ 18:52, 16 May 2007 (UTC)

I do think more emphasis is needed that it is editorial policy, not publication medium, that's relevant here. I've encountered many people who wholeheartedly believe Wikipedia's policy is that web sources are deprecated in favour of print ones, which really shouldn't be the case. If a web source is reliable (e.g., it is written by a well-known expert on the subject at hand and has had independent editorial oversight) then it is as acceptable as any other source that satisfies these criteria (if not actually preferable, for the convenience of our readers). But this obviously isn't clear enough, perhaps because we actually get our hands dirty and talk about all the different types of source that may be self-published. JulesH 19:41, 17 May 2007 (UTC)

break

We have three editors who support modifying the policy, and one editor who doesn't see the need. Can we get more comments please? Jehochman ^☎ / _✔ 18:41, 17 May 2007 (UTC)

What about a policy that instructs editors to cite self-published sources inline, rather than in a footnote? That way the data is presented in the style of a quotation. If it is unarguably true that Blogger So-and-So said something, why can't an article say: "Blogger So-and-So wrote on Month DD, YYYY, that Subject Matter is blah blah blah." That could be an element of a (perhaps temporary) compromise on the issue.

There has been an extended discussion at Talk:Paul_Krugman#Lying_In_Ponds_Website_as_Reference that is highly relevant to the debate here. Jjb 05:41, 20 May 2007 (UTC)

Yes, that's a proper way to cite a well-known blogger, and I've used that method. However, we also have magazines that are published as blogs. Some Wikipedians seem to see a website that uses blog software, and instantly they think, "unreliable." This is the bias I am concerned about. The software used to build a website has no bearing on the reliability of the information presented. Reliability depends upon the editorial policy of a site. To me, Wikipedia seems to be biased towards mainstream media. For a topic like web development, blogs like A List Apart are much more reliable than, for instance, the New York Times. Jehochman ^☎ / _✔ 06:01, 20 May 2007 (UTC)

Somebody else made a modification to this section, so I've gone ahead and made two more small edits: one to clarify, and one to remove redundancy. I believe these reflect the consensus that existed previously. If these are acceptable, I am completely satisfied with the current version. Jehochman ^Talk 08:04, 26 May 2007 (UTC)

comment WP:RS/WP:RELY, Reliable Sources covers this aspect very well. Instead of the addition of "Online magazines and journals published in blog format may be acceptable if they have editors and fact checking." would I suggest that it will say something like ", unless it is a reliable source per WP:RELY.". WP:RS is independent of the medium used for the publication. It's about the actual source, the entity who is providing the information, be it a person or organization.

What also is important is to look at each subject and determine what the typical means for what type of publications are. Einstein writing a scientific publication and Einstein writing a letter to a colleague makes a difference. The letter might contains personal (biased) opinions and speculation, which he would not put into a publication (yet). It is important to find out where serious and well researched information are being published and what is being used for speculation, personal opinion and debate. Some noteworthy information remain speculation and no definite answer would be possible. In those cases should the information still be included in an article, but somehow flagged to make clear to the reader that this is not a proven fact. It might makes sense to separate the resources in the article into references that are used to verify a fact and resources that refer to theories and speculations. --roy<sac> ^Talk! .oOo. 06:37, 30 May 2007 (UTC)

p.s. WP:RELY touches on the subject of context where a publication was made in the section about "Scholarly and non-scholarly sources". It states that it is okay if references are "scholarly and non-scholarly, so long as the sources are reliable". I stated above that I would only partially agree with that and prefer the separation of content and references based on the criteria if it is scholarly or non-scholarly or fact/widely accepted versus theory/speculation. --roy<sac> ^Talk! .oOo. 06:43, 30 May 2007 (UTC)

Sources incorrect?

There are a lot of sources that could be used for verification that are either partially or wholly inacurate.

Does that not compromise the authenticity of any information submitted even if it is verified?

Also, if an article is 'wrong', yet there are no sources to verify the removal of the incorrect articles, what can be done?

Davidtakesthesquare 18:18, 17 May 2007 (UTC)

Note that the policy doesn't require a source in order to remove information, only to add it. That said, I'm not sure how we'd know something is wrong without a source that contradicts it, other than in the most obvious cases. JulesH 19:35, 17 May 2007 (UTC)

For example, with musical genres, I would say that you can identify a band's genre by their sound, although it may not be written in a certain source. So if some bands are attributed to a certain genre that they do not resemble closely, should it not be edited? Davidtakesthesquare 20:11, 17 May 2007 (UTC)

Genre identification is, in many cases, rather subjective. Bands may also belong in more than one genre, although this is somewhat unusual. Ideally, this information would be sourced, although for non-controversial cases I don't see including the information pending a source being located being an issue. If somebody disagrees with a categorisation, I'd suggest removing the classification until a source is located. In the case of a sourced statement of genre that doesn't appear to be correct I'm not sure there's anything to do without finding a source that suggests another one -- it is a highly relevant piece of information that we should always have in such articles. JulesH 12:03, 20 May 2007 (UTC)

If sources disagree about what genre a musical piece or a band is in, WP:NPOV would suggest it's better to include both views rather than delete one based on an editor's own opinion. It's a very common thing for different sources to have different opinions about these things. Best, --Shirahadasha 16:15, 20 May 2007 (UTC)

verifiable source

On Goguryeo controversies and Goguryeo, an editor is repeatedly using several sources that may both violate WP:V, WP:NPOV, and WP:NOR.

He is using sources like

• [[11]]

• [[12]]

all of which are in Chinese and most of the editors cannot read them.

Is using these sources allowed? I believe that these sources are not NPOV and primary sources can be interpreted in many ways and I'd like to remove all the content cited by these sources. However, the Goguryeo article is going through an intensive debate and argument. I do not want to heat up the fighting anymore than it is now, so I am requesting a more knowledgeable editor to see if these sources are indeed allowed to be used. thank you Good friend100 20:29, 22 May 2007 (UTC)

Chinese language isn't an issue as long as there aren't any equivalent sources in English. We don't actually require sources to be neutral, but we do require them to be reliable. Unfortunately, I can't read Chinese so I'm not in a position to assess them, but in general we don't accept wikis as sources (we require sources to have been published by a reliable publisher, while articles on wikis are self-published), so unless the wiki article is a republication of a source (presumably public domain) that has been previously published by a reliable publisher, it shouldn't be used. The first source won't even load for me, but the URL makes it sound like it might be a newspaper, in which case it probably is acceptable (although for an academic topic like this, an academic publication should be strongly preferred, and if an academic publication on the subject disagrees it should probably be used either to rebut the point or to replace it entirely). JulesH 09:53, 23 May 2007 (UTC)

Yes, the first link is some news site I think. The reason why I am asking is because the dispute includes using what sources. One user claims that Chinese historical works and the links from Chinese Wikisource is ok, and I am wondering if this is ok.

Is the chinese wiki (2nd link) acceptable? I'd like to verify completely so that I can use any Korean or other language sites for sources. I don't want to start another war. thank you again. Good friend100 21:02, 23 May 2007 (UTC)

As JulesH mentioned, wikis in general are not good. There is a policy accepting most published works, regardless of the language, though preference is given to academic ones rather than simple news clippings. --Cheers, Komdori 23:14, 23 May 2007 (UTC)

Yes, I understand that already. I am asking again, is the wikisource allowed to be used? Or is it unverifiable and cannot be used as a source? Good friend100 02:53, 24 May 2007 (UTC)

Is Baidu Baike, a Chinese encyclopedia, fair for using as a source? The link is the article for "Goguryeo". I am going to remove all POV sources from Goguryeo and I am wondering if this source violates WP:V since it isn't in english. Good friend100 03:04, 24 May 2007 (UTC)

The source is OK, although an encyclopedia isn't ideal. Again, it isn't a problem that the source is in Chinese, although if an English source says the same thing and is as reputable it should be used instead. The way of treating a POV source isn't normally to remove it, but to make clear that it represents only a single point of view (e.g. by prose attribution). See WP:NPOV#Attributing_and_substantiating_biased_statements for some suggestions there. JulesH 06:20, 24 May 2007 (UTC)

What about sources like [[13]]? Editors think that this source cannot be used because its a Korean POV, I agree with this, but I don't understand why mygoguryeo is POV while Chinese sites and Chinese language articles are ok. Good friend100 20:57, 24 May 2007 (UTC)

Verifiable vs Concensus

If a User has used a verifiable source which in itself appears ok, but goes against all other known verifiable sources that quote a different fact, how can this be reversed? I initially considered this be be an issue over concensus, but according to wikipedia policy/guidelines, I would need to have a verifiable source specifying that there is a claim of concensus. This is in most cases not possible or even plausable. What policy/guideline can I quote that allows multiple individual sources take precidence over one source? Maggott2000 22:37, 23 May 2007 (UTC)

In that particular case, it sounds from what you've said like the best thing would be to note that there's a difference of opinion on the fact. -Amarkov moo! 00:10, 24 May 2007 (UTC)

verifiability, not truth

"The threshold for inclusion in Wikipedia is verifiability, not truth."

For the reasons below, I suggest changing this phrase to: "The threshold for inclusion in Wikipedia is objective verifiability, not subjective truth."

REASONS:

The original phrase is confusing, as it directly opposes two desirable qualities. I have seen it explained as "Wikipedia makes a point of asserting that it does not intend to promote truth.", for example.

However, what I read in the WP:V context is that WP supports verifiability (as objectively verified truth) versus truth (as subjective or intersubjective truth), meaning a stronger (because neutral and equally verifiable by anyone) form of truth than what people usually say as "that is the truth" -- meaning their truth. In particular, it is well-known that two different people living the same moment will have different, albeit still truthful, recollections.

Considering the importance of this policy, I suggest the wording should be changed as above to avoid the confusion. I hope this is useful. Edgerck 23:57, 23 May 2007 (UTC)

Since it is germane, I will repeat here what I said on the above user's talk page. "Verifiable" in Wikipedia terms does not mean, as it means in science, that you can go out and perform an experiment to test a point's validity, or as it means in mathematics, that you can draw up an airtight proof. It means that you can visit the referenced source and check that what is being asserted to be said was actually said. If it meant either of the former, then one could justify edits by saying, "Check for yourself: go perform the experiment" or "Just do the proof, it's simple algebra". But, in the opinion of this user, Wikipedia doesn't work that way, for the simple reason that it cannot work that way. I'll say no more of this here. I have no source to cite, showing I am correct. :-) Robert K S 01:00, 24 May 2007 (UTC)

Thanks, Robert. Let me clarify and sorry for being too concise in the first posting.

Yes, I mean "verifiability" exactly as you say WP does, as almost any journalist does, and as some physicists do. And that is why it makes sense to talk about "objective verifiability" -- for example, a book that A. Crank did self-publish (perhaps with different texts for every copy) is not objectively verifiable. The position of the Sun in the sky is objectively verifiable. A paper in the Physics Review Letters is objectively verifiable (in addition to be peer-reviewed, which is a second level of verifiability albeit intersubjective).

More generally, in our interactions with the world, we can perceive the distinctive views:

• Objective Reality: money, market, companies, governments, law, books by reputable publishers, stars.
• Intersubjective Reality: medical diagnosis, bank-client relationship, majority of e-commerce, hacker organizations, criminal gangs.
• Subjective Reality: truth, thoughts, dreams.

which cannot be harmonized except to some extent. Truth is subjective. No wonder people have a hard time making truth be intersubjective (ie, acceptable by two people in a dialogue), or even attempting the impossible task of making truth objective. Further, objective does not mean immutable. The market is objective (as it would be very difficult for one person to single handedly control it), but changes every minute.

In summary:

• "Objective verifiability" is what WP verifiability is, just (hopefully) making sure that it is not understood as something I or just a few people can verify.
• "Subjective truth" is a pleonasm IMO (because trust MUST be subjective), but is there because many people think that truth is objective.

"The threshold for inclusion in Wikipedia is objective verifiability, not subjective truth."

Hope this is useful. Edgerck 01:41, 24 May 2007 (UTC)

As I point out on your talk page, your examples show that your version of "objective verifiability" is not identical to Wikipedia's "verifiability". I can only hope to assure you that once you understand the distinction and why it is important, you will appreciate supremely Wikipedia's present policy, and the utter, utter unimplementability of the change you propose. Robert K S 01:55, 24 May 2007 (UTC)

I think you just stopped reading a bit too soon, when you cite the Sun example. Please keep reading and you will see that WP objective verifiability is in my list as "A paper in the Physics Review Letters is objectively verifiable". The counter example (which is NOT quite WP objectively verifiable) is in my list as "a book that A. Crank did self-publish".

The Sun example (which you cite in my talk page), as well as the market example, are there but are clearly a non-WP example. The concept of "objective verifiability" > WP.

Hope this is useful. Edgerck 02:09, 24 May 2007 (UTC)

I think the suggested change would be a good clarification of a statement that has a long history of confusing people. Any other comments? JulesH 06:23, 24 May 2007 (UTC)

In short, what I gather from this policy is that a "verifiable" lie will be accepted in Wikipedia, while an "unsourced" fact will not. Right? (In short, I find this policy revolting) AVM 01:40, 1 June 2007 (UTC)

Question

I noticed that sometimes a source is linked to a blog. Now I think it's unreliable source because it could be something that is totally made up. Yet many users' attitude are that all online sources are better than print material (even though some come from reliable sources like government reports). I think this negatively affects the quality of articles. OhanaUnited 10:50, 24 May 2007 (UTC)

Such an attitude is contradicted by our policies and guideline. You are absolutely right, blogs are not reliable sources. It is true that reliable online sources are preferred to reliable offline sources such as books, but any reliable source is better than an unreliable source.—greenrd 12:51, 24 May 2007 (UTC)

I disagree. Blogs can be reliable sources. Our policy states that self-published material by an expert can be used when the expert is writing in their own field, see Matt Cutts. Also, a personal blog could be used as a primary source in some situations. There are e-magazines publishes as blogs, see A List Apart. These may have editors and fact checking, in which case they are reliable sources. There is common misunderstanding that anything called a "blog" must be no good. That's absolutely wrong and biased. Each situation has to be judged on the merits. Jehochman ^☎ / _✔ 12:57, 24 May 2007 (UTC)

With a few exceptions like MSN Messenger development team blog, majority of the blogs are written by those whose specialty are not in the field that they refer the reference to. Most are honest mistake, believing the information they found from Google are correct. But a few are wrong like determining a game's population by number of accounts when there are people who created multiple accounts. When all other means are exhausted, is it better to not cite a source or citing a bad source? OhanaUnited 06:07, 25 May 2007 (UTC)

Unless the information you're trying to include is important to the subject, it would usually be best not to include it rather than cite a bad source. If the information is important (particularly if it is impossible to give balanced coverage of the topic without it) citing a bad source is preferable, IMO. A blog isn't always unreliable, and without knowing the subject at hand it's difficult to tell what is and isn't a reliable source in this respect. If we are talking about people's opinions, for instance, then the blogs of people who hold that opinion are good sources, but some evidence that the opinion is notable would normally be required. JulesH 07:49, 25 May 2007 (UTC)

Cause sometimes people just copy and paste what they found on google and put it on their blogs. They would take every word and think what they find is right. Should we put something to safeguard this situation from happening? OhanaUnited 04:54, 26 May 2007 (UTC)

I don't see how this could happen in any of the circumstances we accept blogs, i.e. for information about the blog's author, for information that the blog's author is an expert on, or for information about the blog's author's opinions (if they're deemed relevant to the topic of the article, which must be done by some other process than ensuring verifiability). I find it highly unlikely that any blog author is going to cut and paste such information into their blog, as it's information they likely know better than a random source they found on google. But in the situation where an expert *does* cut and paste from another source, I think we can usually infer from this that they do trust that source, and the information in the cut & paste section can be considered as reliable as anything the expert wrote themselves (unless they continue to state that they disagree, of course). So, no, I don't think any such safeguard would be useful. Unless you have a specific case in mind? JulesH 09:38, 26 May 2007 (UTC)

I'd like to make a minor change

In the section on self published sources in articles about themselves, I would like to change:

• it does not involve claims about third parties, or about events not directly related to the subject

to:

• it does not involve claims about third parties
• it does not involve events not directly related to the subject

Some editors interpret the "or" to be an "and", this would clear up that misconception. The existing sentence is grammatically incorrect anyway. Does anyone object? - Crockspot 02:38, 26 May 2007 (UTC)

I see no problem changing it, but it's not grammatically incorrect as it is. If you want to split it in two, it would need to be "it does not involve claims about third parties," and "it does not involve claims about events not directly related to the subject." SlimVirgin ^(talk)

The existing "it does not.... or about.." is what I find grammatically incorrect. There should be an "is not" after the "or" for it to be a proper sentence. Anyway, nitpicking. I'll go ahead and make the change and see if it stands. - Crockspot 12:23, 26 May 2007 (UTC)

Suggested subsection: Wikipedia articles as sources

I suggest that a subsection should be added in Sources headed something like Wikipedia articles as sources, clearly stating that Wikipedia articles should not be cited as sources. That point can be taken from the info in this section of WP:NOR, but the point should be explicitly stated here. -- Boracay Bill 05:10, 27 May 2007 (UTC)

English vs. non-English sources

The current wording of that section is in direct contradiction with issues like NPOV and countering systemic bias. a "source of the same quality" isn't necessarily a "source that says the same thing with the same authority".

It's a very real issue e.g. for articles about international relations: English-speaking countries are allies, and the prevailing POV in their media and other sources is different from those in other countries. By preferring sources in English, we're preferring the POV of English-speaking countries. Zocky | picture popups 14:55, 31 May 2007 (UTC)

You're right that there is a problem, but it's not as serious as you make out. Most "notable points of view" are available in English translation, and Wikipedia can even accept editors own translations under certan conditions. Systemic bias will remain if editors fail to write in the NPOV material, but that's not a problem caused by policy.

Far more concerning is cases where references to the English version are taken out, and the foreign language material itself deleted/distorted, as here. PalestineRemembered 09:49, 2 July 2007 (UTC)

Why?

I am confused by this. To many people, Wikipedia is considered a "reliable source". If the Washington Post were to get all of it's information from the New York Times, which got all of it's information from the BBC, then by the time the story got to the Washington Post reader it would be rather skewed. And furthermore, there would be no point in reading the Washington Post because you would get more reliable info from the New York Times or the BBC. Just an example. Obviously all of these newspapers have their own journalists and researchers going out to try and find the facts. This brings us to Wikipedia. If we want to remain a reliable source we have to do more than just say the same thing someone else said. Time was Wikipedia was built by experts who knew their stuff. Now anyone can post what anything they like and rather than getting deleted it just gets a little [citation needed] tag. If they can find an article talking about it, then no one will ever care. We have to do more than see if one other source said it. We have to see if it is generally excepted as true amongst the experts. Plenty of things are verifiable and completely false. Plenty of things are difficult to verify and completely true. Sure more sources would be good and should be a goal for but it SHOULD NOT be the end all or the be all of this site. If it is the site will quickly degenerate into worthlessness. Olleicua 20:12, 31 May 2007 (UTC)

This is the purpose of the guideline for reliable sources. While everything has to be verified with a source, the source itself must be reliable. So, while someone may be able to dig up something that says that martians landed in their backyard 50 years ago, unless it passes the standards set by that page, it can't be included. --13² 20:20, 31 May 2007 (UTC)

In theory you are absolutely correct. Unfortunately, in practice, there are pages on this site that are full of statements that are linked to sources that are not necessarily reliable or otherwise biased and that are ignored. There are also pages full of facts that are well agreed upon by experts that are full of [citation needed] tags added by someone who didn't know what they were talkiong about. Whether a source is reliable or not also happens to be a subjective issue and therefore should not be involved in the Wikipedia process. The fact that someone who some people trust said something does not mean that it should be believed by everyone. Olleicua 20:28, 31 May 2007 (UTC)

Oh no, I totally agree with you. There are definitely flaws, but that comes with the territory. When you allow anyone to edit, it's going to happen. Usually, however, I think this is caused by someone who just does not know any better. I think the best that can be hoped for is that people who understand the guidelines and know what they're doing go through those articles and tag/de-tag them as needed. If a user is seen doing these things, it should be gently brought to their attention on their talk page so they can learn from the mistake and correct themselves in the future. Unfortunately, it's a massive effort that would never end. --13² 20:51, 31 May 2007 (UTC)

I agree with this. "Verifiability not truth" is an Orwellian absurdity.Don Quixote de la Mancha (talk)

SPS about oneself

I've seen instances where a person has a blog or writes an article on an SPS and that article cannot be used as a source, even about the author. For example, if someone wrote a controversial opinion on a blog, why couldn't their Wikipedia page reflect that they made that comment? It seems that the concerns of reliability are less relevant (assuming the person actually wrote the quote in question) if the person is simply speaking about themselves. In short, can something written by a given author in an SPS be used as a source about that author or their opinions? Oren0 22:58, 31 May 2007 (UTC)

WP:V makes clear that this sort of SPS is ok IN AN ARTICLE ABOUT THE AUTHOR, and provided all six of the conditions spelled out in WP:V are met. But it is not ok as a source in any other article (like a larger artcle that covers the author along with other topics, or as part of a larger topic). UnitedStatesian 00:20, 1 June 2007 (UTC)

What is the reasoning behind the first of these criteria, "it is relevant to their notability"? Why can't we use a SPS for basic facts, biographical infromation here. Suppose someone writes on their own web site or blog that they "grew up in Hope, Arkansas," or "my latest album was influenced heavily by "Weird Al" Yankovic's Dare to be Stupid," neither fact makes them notable but it may be relevant to an article about them or their work. -MrFizyx 23:23, 15 June 2007 (UTC)

I read the sentence differently. I read it to mean close to "it is relevant information taken into account how they are notable." Rhanyeia 14:43, 18 June 2007 (UTC)

Oh, perhaps we should edit the policy to make it more explicit. What do others think? -MrFizyx 14:48, 18 June 2007 (UTC)

Verifiability of criticism?

On the article for the show Man vs. Wild, there has been some fighting recently over the "Controversy" section, which was recently removed altogether. Concerning the show, the Discovery Channel forum is packed full of viewers complaining that the show is a fraud, as are numerous other forums unrelated to the Discovery Channel. Example: google search for "Man vs Wild" and "fake"; you'll get ~17,000 hits.

The question is: how can this be reported on the article with citations -- or can it be reported at all? It's not like Discovery is going to call one of their own shows fake, and there are hardly any reviews of the show at all from "reputable" sources. As the article stands, the widespread criticism isn't even mentioned.

I added many references to the Criticisms section before it was deleted, but the whole section was deleted because the deleter didn't like the quality of the references. I didn't create the criticism section, but since it was full of "fact" requests, I filled them in and expanded the section. The types of references included:

• Links to video clips of the episodes and time offsets into the episodes to see the obvious cases of cheating that people have been complaining about on the forums (a raft supposedly lashed with hibiscus actually bound with manilla cord, a "free climb" of a waterfall actually involving a climbing harness, wearing a life jacket when swimming in a river in the Rockies, and so on.)
• Links to many, many different threads of people complaining about the episodes. This includes people who lived in the areas where he visits. Example: in one episode, he pretended to be lost in the sierras -- first climbing down from a mountain, rafting through a river, walking through a meadow, and swimming across a lake to freedom. Yet, a rafting photographer who used to work on the river that Bear went down and knows the whole area well pointed out that at no point was he far away from civilization, there was a road on the side of the river he went down, rafting companies went by it every 15 minutes, and to get to the lake, he had to walk right past several rafting companies.

Is it impossible to point out these criticisms here? Was my choice of references appropriate? I could also have linked to pictures from the host's own photo gallery clearly showing, for example, that the ends of the bamboo poles are saw cut, not burned off as presented in the episode, and the raft is tethered to the crew's boat near the island, not adrift in the ocean as presented in the episode. Would that be any better, or would that count as "original research"?

Let me know what you think the proper solution to a problem like this is. -- Rei 16:24, 31 May 2007 (UTC)

You can only really include this kind of thing if you find a reliable source talking about it. If these criticisms about the show are relevant, I'm sure somebody (a magazine or newspaper reviewer perhaps) must have commented on them. JulesH 19:49, 31 May 2007 (UTC)

New Discovery Channel shows rarely get treatments like that (only the big ones like, say, MythBusters, tend to). I've searched and can't find any non-forum/non-blog reviews that are more in depth than your basic two-paragraph Discovery Channel blurb, and there's only a handful of even those. So... no way to comment on the tens of thousands of annoyed viewers posting all over the place, then, or even to note that they exist? Even though they're pointing out things that you can clearly see with your eyes in the videos (primary sources), plain as day, and commenting based on expertise (such as local residents to the areas he visits)? Also, isn't "many viewers complaining" a primary source to the claim that "many viewers are complaining"? -- Rei 21:33, 31 May 2007 (UTC)

No. In principle, they could all be sockpuppets of a few people. Of course I don't think that is plausible, but the possibility exists.—greenrd 10:06, 1 June 2007 (UTC)

The other question that invalidates this is what does "many" mean? It would usually be interpreted to mean a significant number, but how do we determine what a significant number of viewers is? I think any way of doing this would inevitably be original research. JulesH 10:48, 1 June 2007 (UTC)

Okay. So, to sum up, in the videos (available at YouTube for a reference) and stills (available at Bear's website), you can see with your eyes rope on the raft, him wearing a life jacket in a river, him wearing a climbing harness when 'free climbing' a waterfall, etc... but we can't mention this, right? Not until a newspaper decides to do a non-blurb review of the show (which may never happen) and comment on this fact, correct? -- Rei 18:26, 1 June 2007 (UTC)

We can, as editors, make descriptive comments about an image. "This is a cat", "he is standing by a car", "this is a sunset" and so on. What we cannot do is make interpretive or evaluative comments "This is the largest cat in the universe", "This is the worst car ever made", "He will attempt to climb the waterfall but because he's a sorry-ass loser he's going to fall into the pool." Wjhonson 01:54, 2 June 2007 (UTC)

Question: If the show is isn't discussed in any independent sources, should we include it? It's one one thing to relax our normal verifiability standards for popular entertainment that's basically fiction. But It's another to do so for anything that claims, or is believed, to be factual. Under our regular standards, we couldn't include anything that can't be independently verified. The whole purpose of verifiability is to protect against this sort of thing. If there's no really commentary on such a show at all, perhaps it simply doesn't belong in the encyclopedia. I'm not just talking about the criticism part here, I'm talking about the whole article. This article has no independent sources at all. If it can't be verified, shouldn't we delete? Best, --Shirahadasha 04:05, 3 June 2007 (UTC)

E-mail

I had a question concerning a comicbook and I sent the writer of it an e-mail. He responded to me. I was wondering if there are any ways to cite this e-mail as the information would be relevant to the article I am working on.--CyberGhostface 15:37, 20 June 2007 (UTC)

I don't think the e-mail is a reliable source, and on top of that it looks like you are conducting original research. I STRONGLY encourage you to leave that information out of Wikipedia. UnitedStatesian 18:18, 20 June 2007 (UTC)

Idea for a tag re: reliability

There is an interesting proposal at WP:RS to creating an inline tag for situations where a source is reliable, but hardly the best of sources available. The proposed wording would be something polite and non-confrontational, like [more reliable source requested]. To give you an idea of where such a tag might be used... take an article on an historical topic, containing a statment that cites to a Histroy Channel documentary as a fact source. While the History Channel is not completely unreliable, I hope you would agree that there are probably sources that are more reliable for the same information. However, more community input is needed. (Please join the discussion at Wikipedia Talk:Reliable sources#Inline tag?) Blueboar 22:41, 24 June 2007 (UTC)

Article should discuss the subject of print sources directly

The opening paragraph of this project page is widely interpreted to mean that web sources are better than print sources because they are more easily checked by random readers. This is not the case and needs clarification.

The discussion of self-publication is sloppily written in such a way to allow for harassment of area experts who are sufficiently prominent to also be involved in the publication process of books and journals.

Also, Wikipedia must recognize the fact that many extremely reliable print sources are not readily accessible to random wikipedia readers. Many volumes published by prestigious university presses are printed in editions of only a few hundred copies and may cost hundreds of dollars. They are none-the-less reliable and verifiable sources. The issues arise with newspapers from the pre-web era. --Pleasantville 15:52, 5 June 2007 (UTC)

I agree with Pleasantville's comment concerning the opening paragraph. The phrase "any reader should be able to check" ignores the (unfortunate) fact that many of our readers don't have access to a large research library. If the best sources for a particular topic happen to be found only in a few libraries, we should use them, despite their inaccessibility. By doing this, we get a more accurate encyclopedia, and we also give the public access to information that they couldn't otherwise obtain.

Here is a possible rewording of the second sentence that would solve the problem, I believe. Instead of this:

"Verifiable" in this context means that any reader should be able to check that material added to Wikipedia has already been published by a reliable source.

this:

"Verifiable" in this context means that every addition to Wikipedia must be based on existing material published by a reliable source.

Opus33 21:18, 22 June 2007 (UTC)

How can a reader determine that an assertion has been published by a reliable source if they can't see the source? While academic journals, etc, may cost hundreds of dollars, most reliable ones are collected by university libraries that are accessible (albeit with difficulty) by most readers. If readers can't verify the source text then it's impossible to make sure has been accurately summarized. ·:· Will Beback ·:· 21:33, 22 June 2007 (UTC)

Hello, the answer is that we only need to have enough editors available who can check the reference sources--as I argued above, it's hopeless and counterproductive to insist that all readers be able to check the source. In fact, however, most of the big libraries are at universities, and make these libraries available to their students--thus to a huge population of WP editors. So I don't see any real problem here. Opus33 21:45, 22 June 2007 (UTC)

Verifiability Standard

Is it true that if a reliable source publishes a book and makes a statement in the book that he or she does not validate with any kind of proof, that this book can still be used as a source for that statement? For example, if author Smith says that Jones murdered 20 people yet offers no footnote or proof to that effect, can editor Johnson edit an article saying that Jones murdered 20 people and list author Smith as a source for this claim and that this falls within WP guidelines for Verifiability? Jtpaladin 00:17, 26 June 2007 (UTC)

Hello, This actually relates to the previous topic so I have juxtaposed them. I would say, you have to rely on the quality of the source. So, if Smith submitted his book to (for example) the University of Chicago Press, we can be certain that the editors of that press found qualified reviewers, who know the topic of the book, have read through the whole book (they get paid to do this) and made sure that the evidence for the claim of 20 murders was solid. This would have to have happened before the book qualified for publication. University of Chicago Press has to do this, because otherwise long-term they would lose their credibility as a serious academic publisher. On the other hand, if Smith finds a so-called "vanity press", and pays them $5000 to publish the book, then the testimony of the book is worthless.

More generally, the important criteria are: is there peer review, and does the publisher have strong motivation to adhere to the standards of scholarship? Opus33 00:59, 26 June 2007 (UTC)

Postscript for Jtpaladin: since writing the above, I noticed that you are involved in a big controversy. My advice would be: find a biography of Joe McCarthy, published by an established academic press, that says that alcohol did not play a role in his death. Cite it, and you'll be on very solid ground for purposes of editing and debating your fellow editors. Opus33 03:32, 26 June 2007 (UTC)

Opus33, thank you very much for your comments and taking the time to address the issue. My biggest concern is that knowing how political book publishers are more interested in making money from sensational charges and the wanton manner in which these books are released without footnotes and sources for their charges, this really speaks to the heart of Wikipedia's basic foundation of verifiability.

How we can allow just any author or set of authors to get away with making unsubstantiated claims when we know for a fact that they are unsubstantiated, opens Wikipedia for attack as a repository of fraudulent information. The standard should be, does the author provide sources for whatever charge they are making? In the case in question, you have biographers who are not doctors, who do not have access to medical records, who have never met the person of which they speak, possess no medical evaluation of the person, no authority upon which to speak regarding medical matters, and provide no source for this medical allegation. And this is supposed to be OK with Wikipedia rules?!! It's absurdity of the highest order. The only thing we know about this case it that the person in question died from a particular cause of death. That's it. Why we can't simply quote that fact and then allow for speculation apart from that fact is a question I can't get answered.

In my opinion, the reason for so much conflict in Wikipedia is because of this very loose standard regarding the use of authors who are either lying, merely copying what they heard from another author, or are simply ignorant of the matter.

Let me give another example. Let's say that 100 books have been written about a particular subject. Then, one day, key material information is discovered that make all the 100 books wrong about the subject. So you have one published source saying one thing and 100 published sources saying another. According to Wikipedia guidelines, "If your viewpoint is held by an extremely small minority, then_whether it's true or not, whether you can prove it or not_, it doesn't belong in Wikipedia, except perhaps in some ancilliary article. Wikipedia is not the place for original research."[14] Does this seem right?

I know that under "What is a reliable source?" WP:RS, it states, "The reliability of a source depends on context; what is reliable in one topic may not be in another. A publication by a world-renowned mathematician may not be a reliable source on topics of biology. In general, an article should use the most reliable sources available to its editors." So it appears that if you are not someone with a background in autopsies, medical examinations, medical evaluations, etc. your opinion on the cause of death of a person beyond what is stated in the coroner's report, is inadmissable as a qualified source under this rule. Does my analysis sppear correct to you? In the case in question, "the most reliable sources available to its editors" is the coroner's report since none of the authors have access to the subject's medical records or an opinion from a medical authority who has examined the person in question. Is this a correct line of thinking? I think so.

Of course, this has no bearing on the author's opinion on other issues of the biography of the person being discussed. However, I have seen flat out falsefication of the record of this person by the authors and yet the publishing company, who is principally interested in profits, allows this information to be published without proper citation. So, then the Wiki editor has in his hands, a book from a published source that even though the author is lying, distorting, or simply erring in the record of the subject, it can be used as a source in Wikipedia. A simple way to avoid the abuse associated with such a situation is to require that if an author is being cited as a reference for a topic, that the source that the author is using be the true piece of evidence in making the argument of the author. For example, if author Smith were to say that Mr. Jones spent 3 years in the military, that it be incumbent on author Smith to state the document or source that author Smith is using in order to make the claim that Mr. Jones was in the military for 3 years. Because as I understand the rule now, and correct me if I'm wrong (and please do so by showing me the Wiki rule regarding it), a Wiki editor can simply state that author Smith said that Mr. Jones was in the military for 3 years and that is all that is required for proper ciation in Wikipedia. So if author Smith has the only published source for this information then 100 other authors simply can cite author Smith as a source for the claim that Mr. Jones was in the military for 3 years. Again, does this describe Wikipedia's guidelines? Am I right in this assertion? Please let me know.

In summary, my concern is that as we develop more and more into a partisan world where publishing a book is a great deal more simple than it once was; where books now get published all the time with scant levels of foot-noting; where the claims of conspiracy theorists get published by reputable publishing companies; where fact-checking takes a backseat more and more in favor of releasing a book that will generate great profits for the publisher, the Wiki rule of verifiabilty without requesting access to the facts or sources upon which the claims of the author are made, will take Wikipedia down a slippery slope of irrelevancy. I believe this is because the current rule of verifiability seemingly omits any requirement that the author be required to provide evidence supporting their claim for a particular issue.

The simple solution would be that if author Smith is making a claim, the Wiki editor can use author A as a source for that claim by citing the publication, page number, etc. However, if that editor is challenged on author Smith's claim, that it be incumbent upon the Wiki editor to further cite the precise source upon which the author rests his claim. If the author has no source or the source is just another in a long chain of authors repeating the same thing over and over without reference to a document or other source-material which could be used as simple evidence, then the Wiki claim can be removed. In other words, just because some author in a book makes a claim of something, does not necessarily make that claim a fact. Why we would allow an author to make a claim of anything less than requiring scholarly supported proof is mystifying.

On the other hand, maybe I've missed some guideline that addresses my concern, if so, I would appreciate a link to it.

Again, I appreciate your time and help and would be very thankful if you could further address this matter. And if in fact Wiki rules are as I outlined above with the author Smith example, that perhaps some dialog be initiated to require a stricter guideline regarding the presentation of evidence to support an argument made by an author. Thank you. Jtpaladin 16:05, 26 June 2007 (UTC)

Sources discussion from RS

Copy-paste from RS talk

Someone left a note on my talk page saying there was a discussion here about scholarly/non-scholarly sources, but I don't see it. The point is that we use academic and non-academic sources, obviously, and this page must reflect that. SlimVirgin ^(talk) 21:51, 25 June 2007 (UTC)

Searching on scholarly would have found it, but it's #A bit that I find questionable. above. Septentrionalis PMAnderson 22:02, 25 June 2007 (UTC)

The policy says "Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy. Sources should be appropriate to the claims made: exceptional claims require exceptional sources." No mention of scholarly sources. Any attempt to prioritize them over other mainstream sources flies in the face of V and NPOV. SlimVirgin ^(talk) 07:35, 26 June 2007 (UTC)

If we want a caveat based on subject area it might be "Primacy should be given to scholarly sources, particularly in the sciences. [That doesn't fly in the face of NPOV at all] Historical and arts topics will often have a greater range of reliable material, scholarly and non-scholarly." It needs a caveat if its to go in at all, but singling out Pop culture, as Tim's edit did, is not the way to do it.

But this will take up two lines on V, so I suggest moving it there. While we're at it, let's tighten the wording on Exceptional claims and move it over as well. Marskell 07:49, 26 June 2007 (UTC)

The exceptional claims thing is already in V, as I recall. It would definitely be better to discuss this there, because this is just a guideline, and it can't contradict the policies. SlimVirgin ^(talk) 08:00, 26 June 2007 (UTC)

The exceptional claims is here as a single sentence. I'd suggest unpacking it midway between the length here and at RS, and then removing it from RS. It's a little meandering over there. Marskell 08:25, 26 June 2007 (UTC)

Hello, I can't quite follow Slim Virgin's original comment (the phrase "I don't see it" is ambiguous). For what it's worth, here is a restatement of the main content of the previous two sections:

• Publishing organizations that do a lot of prepublication peer review have special claim to our attention, and should be favored over those who don't.
• The organizations most likely to do peer review are the ones where heads roll when error is commited (example). I believe that the outlets that normally show this kind of commitment to accuracy are prestigious journalistic outlets and academic publishers. In contrast, the compilers of (for example) amateur web pages can ignore accuracy as they please.
• We need to combat the truly alarming view (see Pleasantville's contribution) that only resources available to every single reader can be used as WP reference sources. The correct criterion is to use the best sources, so long as they are accessible to at least some WP editors for checking. This is why we should change the wording of the Wikipedia:Verifiability page, as I suggested in italics above. Your opinion on this proposed change is solicited.

Thanks for listening. Opus33 15:41, 26 June 2007 (UTC)

Sources

Somethingwe need to decide is if the WP:Verifiability policy should define reliable sources, or if this should be done on the WP:Reliable sources page. I think the verifiability policy should state that reliable sources should be used (as it does at the moment) and the specific page on reliable sources define what this term means and give examples of how it is applied. Does this seem reasonable? Tim Vickers 17:04, 26 June 2007 (UTC)

I don't think a definition would be helpful, because what counts as a reliable source depends heavily on the context. What we encourage people to do is use the most appropriate sources, and then we have to leave what that means to their editorial judgment. If they don't have any editorial judgment, a guideline won't help them anyway. Anything that says peer-reviewed sources are always better in a certain context would almost certainly be wrong at least some of the time. SlimVirgin ^(talk) 17:18, 26 June 2007 (UTC)

Almost everything in Wikipedia's policy will be "wrong at least some of the time", which is why we have WP:IAR. But does that mean we should give up on providing guidance? Let's work toward a useful guideline instead of leaving the question wide open to the whims of individual editors. Raymond Arritt 17:29, 26 June 2007 (UTC)

I think we need to give some guidance on what we mean when we use the term "reliable sources" and give some examples of what is reliable and not reliable. Furthermore, I think the best place to do this is on the WP:Reliable sources page. If we give no guidance at all on this subject new editors will have no idea what the WP:Verifiability policy means when it uses this term. Tim Vickers 17:34, 26 June 2007 (UTC)

RS is a discredited guideline (which doesn't actually say anything that isn't in another policy), so anything added to it will be ignored by most good editors anyway. If you want to develop something, so long as it's not detailed, it's best to develop it here, but we can't, as was suggested at RS, prioritize peer-reviewed material over other reliable, mainstream sources. That idea flies in the face of everything Wikipedia stands for. SlimVirgin ^(talk) 17:43, 26 June 2007 (UTC)

I don't agree that sources can't be prioritized. Mainstream is a judgement call, and whether a source is reliable is a judgement call (which is why I think RS should me modified, not verifiability). I'm not sure why WP:RS is discredited, as a merge was tried but failed, so it seems to have some support. And I'm also not sure why saying that material (where available) produced by experts in their respective field, whose work is analyzed by their peers, all with a lot at stake to get it right would not take precedence over the work by just about anyone else ("flies in the face of everything wikipedia stands for?") R. Baley 18:05, 26 June 2007 (UTC)

Reliability is in part, and I hope Slim will agree, a product of the credibility of the publishing organization and the thoroughness of the editorial review process. Peer review and publication in reputable scientific journals offers both extensive editorial oversight and a highly credible venue. For many complex scientific issues, the editorial oversight present in peer review is much more rigourous than for most other mainstream sources, e.g. newspapers and magazines. Certainly we expect the description of things like the greenhouse effect to be more complete and accurate in the the Journal of Climate than what appears in the New York Times, and in a case of conflict I would assume the journal is correct and the Times made an error or simplification. That is the kind of thing I think is intended by saying that peer reviewed works are generally more reliable when discussing scientific conclusions, because we would ordinarily expect them to be more complete and more thoroughly reviewed. Dragons flight 18:40, 26 June 2007 (UTC)

DF, I agree with what you wrote. Everyone sensible knows it is almost always true without adding it to a policy. The danger of adding it is that there will be exceptions, and the stress on scholarly sources will be used as a stick to beat people with. I was asked by e-mail for an example of an article that doesn't prioritize the scholarly view, so I'll paste here what I wrote in reply — For example, at Rudolf Vrba, we give the academic historians' view that the reports in 1944 of gas chambers and mass murder at Auschwitz were disseminated quickly, and that the slight delay in doing so made no difference to the number of people who were killed there. We also present the non-scholarly view that, in fact, there was a substantial delay between the information being known by governments and being distributed to the targeted Jewish communities, and that this delay meant hundreds of thousands were led to their deaths unwittingly when others knew what was going to happen to them. This is a controversial issue, which goes to the heart of the scholar-versus-survivor discourse that plagues all Holocaust historiography. We report both views, without prioritizing either of them, as do other reliable sources, possibly much to the frustration of the academic historians, many of whom see the debate as fundamentally wrong-headed.

The fact is that scholars are sometimes (or even often) wrong, and other publications need to be able to call them on that, without editors here shouting them down with the "not peer-reviewed!" war cry, which I've seen misused too many times. SlimVirgin ^(talk) 18:54, 26 June 2007 (UTC)

Then we need to explicitly state that no primacy is given to peer-reviewed sources, because editors with scientific training will presume otherwise. If that's our policy, let's be up front about it. Raymond Arritt 21:30, 26 June 2007 (UTC)

(edit conflict) Having thought about it, I don't agree that we need to change WP:V to modify the guidelines on WP:RS. WP:V is official policy, so it should be short and sweet so that we don't find ourselves hamstrung too often. WP:RS is refered to on this page, but simply modifying the RS guidelines doesn't contradict Verifiability. I think it is prudent to note that discussion is taking place on RS talk (maybe in a prominent position at the top in an info box?) because obviously what happens on RS affects how WP:V is interpreted/applied).

In short, people here should know about the proposed modifications to the RS guideline, but determining a source's relative reliability (when sources contridict, which one(s) should we use) is fundamentally different than verifying that a source says what we think it says. Modify the RS guideline, not WP:V, they don't contradict each other. R. Baley 18:05, 26 June 2007 (UTC)

The problem is that the way some people want to modify RS (to give priority to peer-reviewed material and, in effect, to introduce scientific point of view through a back door) contradicts V and NPOV, both policies. If people want to change policy, they have to argue for the change on the policy pages. SPOV has been rejected many times, so any attempt to reintroduce it has to take place prominently on a policy page, either here or NPOV, so lots of people see it. SlimVirgin ^(talk) 18:09, 26 June 2007 (UTC)

I'm a little confused—how did the modification violate policies? In the sciences, the view represented by peer reviewed material is a pretty significant POV to discuss in a NPOV presentation. This doesn't mean other views are excluded, nor that unverifiable sources are being used. Gimmetrow 18:39, 26 June 2007 (UTC)

It looks like an attempt to introduce scientific point of view, or scholarly point of view, which as I said above has been rejected many times as a violation of NPOV. SlimVirgin ^(talk) 18:44, 26 June 2007 (UTC)

If peer-reviewed and non-peer reviewed sources are considered equal, then that has to be explicitly stated. Most people with scientific training assume that peer-reviewed sources take precedence, so they (we) need to be set straight. I hope the resulting debate can remain pleasant once this becomes broad knowledge. Raymond Arritt 18:53, 26 June 2007 (UTC)

SlimVirgin, who decided and when that the WP:Reliable sources page is no longer an official guideline? This is marked as a guideline and linked by both Wikipedia:Verifiability and Wikipedia:Citing sources. This is the first I have head of any such decision. Tim Vickers 18:41, 26 June 2007 (UTC)

RS started life as a POV fork of V. It remains a guideline because a small number of people edit warred when others tried to remove the guideline tag, and people couldn't be bothered fighting and gave up on it. There is nothing in it that isn't supported by another policy, so far as I know, so it's, at best, a useless page; at its worst, it contradicted other policies and even itself, and was very badly written, and therefore widely ignored. SlimVirgin ^(talk) 18:44, 26 June 2007 (UTC)

If you do think it is a useless page, you should propose it for deletion. That would be the best way of determining if what you say is really true. Tim Vickers 18:55, 26 June 2007 (UTC)

SV, I don't think you're right that WP:RS is "widely ignored". It's very frequently cited on WP:AFD for instance. If the guideline is deprecated, it should be marked as such (and I for one would welcome such a change). JulesH 19:05, 26 June 2007 (UTC)

I too have frequently see the WP:Reliable sources page cited, it is cited on all the relevant policies and I stopped counting at 10,000 links to this page on Wikipedia. I can't understand how it could be thought that it is no longer applied. Tim Vickers 19:12, 26 June 2007 (UTC)

Something we need to decide is if the WP:Verifiability policy should define reliable sources, or if this should be done on the WP:Reliable sources page?

It's helpful to have a separate page that specifically talks about reliable sources. It's a specific page to point to with WP:RS. I see the page referenced frequently on article talk pages, deletion discussions, and elsewhere. It's anything but a "useless page". And, there is no contradiction. Sources with the best "reputation for fact-checking and accuracy" tend to be peer-reviewed journals and academic publishers, especially for scientific topics and less so for pop culture, biographies, and some other topics. There is no contradiction to say this on the WP:RS page. --Aude (talk) 19:45, 26 June 2007 (UTC)

Should the following be added?

"Wikipedia welcomes material written by scientists, scholars, and researchers, particularly material published by peer-reviewed journals. Editors should be aware that these may be outdated by more recent research, or may be controversial in the sense that there are alternative treatments. Scholarly sources are preferred for the sciences. The arbitration committee has written: "What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such." In other areas, such as history and arts, a greater variety of reliable sources, both scholarly and non-scholarly, may be available. Wikipedia articles should represent all majority and significant-minority treatments of a topic, so long as the sources are reliable."

Seems perfectly agreeable to me, and would solve the current dispute. It meets the concerns re pseudoscience but still allows for reliable non-scholarly sources elsewhere. And its Arbcom wording. Marskell 19:41, 26 June 2007 (UTC)

It helps a little. The Arbcom part is quite good but the stuff at the beginning and the end dilutes it. The second sentence, in particular, directly contradicts the Arbcom wording. Raymond Arritt 19:46, 26 June 2007 (UTC)

In saying "or may be controversial in the sense that there are alternative treatments.", it seems like it says we should give equal treatment to "alternative" theories, such as conspiracy theories and such. WP:NPOV#Undue weight is important here. --Aude (talk) 19:48, 26 June 2007 (UTC)

"...alternative treatment from sources of equal quality"? Marskell 19:50, 26 June 2007 (UTC)

In the sciences there are no "sources of equal quality" compared to peer-reviewed literature. There's peer-reviewed material, then "gray literature" below that, then everything else. A far better solution would be simply to omit the first two sentences, which leaves a concise, consistent guideline instead of something that contradicts itself. The last sentence does not quite contradict the Arbcom wording, but it requires careful reading in light of the Arbcom criteria. Raymond Arritt 19:54, 26 June 2007 (UTC)

Yes, yes, I hadn't meant to suggest something else is equivalent. Simply removing them might be fine. Marskell 20:00, 26 June 2007 (UTC)

How about - "The most reliable material is published by scientists, scholars, and researchers, particularly in peer-reviewed journals and academic presses. These sources are preferred in subjects such as medicine, natural sciences, social sciences, and humanities. In the sciences in particular, the arbitration committee has decided that: "What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such." In other areas, such as current affairs, popular culture and the arts, a greater variety of reliable sources, both scholarly and non-scholarly, may be acceptable." Tim Vickers 20:08, 26 June 2007 (UTC)

I'd prefer the simpler and cleaner version that comes from omitting the first two sentences and the last sentence of what Marskell proposed. (The last sentence is really WP:UNDUE territory, not sourcing.) Then we've got:

Scholarly sources are preferred for the sciences. The arbitration committee has written: "What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such." In other areas, such as history and arts, a greater variety of reliable sources, both scholarly and non-scholarly, may be available.

Clear and direct. Raymond Arritt 20:10, 26 June 2007 (UTC)

I would prefer to merely state that P-R literature, for the sciences, trumps other sources. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such is in danger of misinterpretation as forbidding reliable blogs as sources (do I have any in mind in particular? I wonder). For entire *theories* you need P-R lit; but for explication of details other sources can be useful. The distinction is likely to be lost William M. Connolley 20:25, 26 June 2007 (UTC)

I suggested earlier that, wrt science articles and sourcing, we could add something like this: "Non-peer reviewed may be used where they help to explain ideas, details or jargon, but not to contradict (per undue weight)." (link diff). I think with some tweaking this could work. R. Baley 20:52, 26 June 2007 (UTC)

There is no such thing as a "reliable blog" as a source in science, think what a Nature editor would say if you cited a blog in a paper! Tim Vickers 20:31, 26 June 2007 (UTC)

A very broad division between science and arts / humanities is preferable to listing specific disciplines. And humanities should fall on the other side of the divide. It's perfectly possible to use non-scholarly but reliable sources on such pages. Marskell 20:11, 26 June 2007 (UTC)

Scholarly sources, particularly publications in peer-reviewed journals, are preferred for the sciences. In particular, the arbitration committee has decided that: "What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such." In other areas, such as history and arts, a greater variety of reliable sources, both scholarly and non-scholarly, may be acceptable. - Changes bolded, we need to highlight peer-reviewed journals for the sciences, saying that the ArbCom "has written" makes them sound almost biblical, and the issue is never what is available, the issue is what is acceptable. Tim Vickers 20:21, 26 June 2007 (UTC)

I think we should let the arbcom speak for itself and the first "particularly" is a little difficult—given that we are a tertiary source, a secondary textbook or equivalent might actually be preferable. Marskell 20:35, 26 June 2007 (UTC)

"Scholarly publications are preferred for the sciences. In particular, the arbitration committee has decided that: "What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such." In other areas, such as history and arts, a greater variety of reliable sources, both scholarly and non-scholarly, may be acceptable." Better? Tim Vickers 20:43, 26 June 2007 (UTC)

Fine be me. Marskell 20:46, 26 June 2007 (UTC)

The ArbCom doesn't make policy. Articles should reflect all majority and significant-minority views published by reliable sources within the limitations of V, where "source" can refer to the publisher or the writer. Editors should choose the most appropriate sources given the context, and which sources are appropriate is a matter of good editorial judgment. Clearly, if discussing a scientific theory, scientists who have written about that theory in peer-reviewed journals are going to be better than cleaning ladies who have blogged about it, but that's trivially true and not worth mentioning. However, there may be reliable non-scholarly challenges, and if these have been published by reliable publications, and are significant-minority, or even majority, views, they must be represented, and may not be downgraded in terms of length or tone because not scholarly. That is the very essence of NPOV. SlimVirgin ^(talk) 21:25, 26 June 2007 (UTC)

No, the Arbcom doesn't make policy, but its wording is preferable to the opinion of any given editor in this discussion. It makes precedent. Given that this wording speaks directly to what we are discussing, why should it be excluded? Marskell 21:31, 26 June 2007 (UTC)

We have three core content polices: V, NOR, and NPOV. If you want to change them, you have to argue for the change strongly, not simply state it, or initiate polls, or state that others have said it. You have to argue why it is needed, why it would be an improvement, and show that it wouldn't contradict other policies. You've done none of those things.

As I said above, what you are pushing for is either trivially true and therefore not worth mentioning, or it's a substantive change and arguably a violation of NPOV. SlimVirgin ^(talk) 21:42, 26 June 2007 (UTC)

The rationale for inclusion was that we had been arguing about it, people want some guidance at RS on the subject and this is a better place for it, editors don't agree with you wrt to science sourcing, and the wording of NPOV broadly doesn't either. You have been asked at least twice to show how it does contradict policy and I have yet to see the answer. If nothing is to go in, here or at RS, fine. But an unqualified "scholarly or non-scholarly" should not be included. The "trivially true" argument is nice. Who can argue? I wonder though, if it's so trivially true, why we needed an arb case; why we've had megs of wasted words on topics like Global Warming, Evolution, and Pseudoscience. Given that you don't edit any of these areas, I don't know why you're opposing strengthening the wording for editors who do. Marskell 07:00, 27 June 2007 (UTC)

So NPOV means that every woo-woo idea that a crackpot manages to work into a newspaper is taken just as seriously as material in a National Academy of Sciences report. If that's true, I don't see the point in participating in the project. Raymond Arritt 21:42, 26 June 2007 (UTC)

I agree entirely with SlimVirgin's statements on the need to use reliable sources - we all agree on this and the policy is clear. However, what we are trying to do is give guidance on what "reliable sources" actually are. Not to define in all cases - which would be to expand the policy - instead, add to the guideline page WP:Reliable sources some general guidance on what are usually considered to be reliable sources in various subject areas. Please read the propose addition that was voted on. It does NOT state that the only reliable sources are peer-reviewed and that non-peer-reviewed sources cannot be used. Instead it offers the general guidance on what is preferred in most cases that is appropriate to a guideline page. Tim Vickers 21:36, 26 June 2007 (UTC)

That specific proposed addition had holes, particularly creating an abusable exception for pop culture. Broadly dividing the sciences and arts without wordiness makes more sense. I'm not particularly concerned about cleaning ladies, because they've never been at issue. I'm concerned about wheat-and-chaff on science info (as I'm sure you are, given your mainspace contributions). And I say that not being capable of adding "hard" science—it's the "soft" stuff that I edit that's most worrying (evolutionary descriptions, conjecture on astronomical findings, etc.) Reliable but generic sources are just not good: that an abstract is preferable to the reliable New York Times is a non-question for scientific info—of course we should prioritize the abstract. I followed the Arbcom ruling and it was so relieving. "Scientific theories promulgated outside [textbooks and journals] are not properly verifiable as scientific theories and should not be represented as such." Yes, thank you very much.

But we need to be sensible to the fact that arts and humanities don't work that way. Original documents, memoirs, eye-witness stories, etc. might all be acceptable per policy when published reliably. Marskell 22:04, 26 June 2007 (UTC)

The proposed addition did not make a broad exception for pop culture, instead it just acknowledged that not all areas have scholarly studies - "...alternative reliable sources are used if scholarly publications are not available - which may be the case in topics related to popular culture or current events." We can make this more clear that this is not a blanket exception by saying "...alternative reliable sources are used if scholarly publications are not available - which may sometimes be the case in topics such as popular culture or current events." Tim Vickers 22:14, 26 June 2007 (UTC)

Or we could not refer to pop culture at all. Do we need to? Marskell 22:24, 26 June 2007 (UTC)

It was just an example, but one I thought useful to cover as it is a subject which often comes up at AfD. So you don't think the wording "...alternative reliable sources are used if scholarly publications are not available - which may sometimes be the case in topics such as popular culture or current events." is clear enough? Tim Vickers 22:39, 26 June 2007 (UTC)

Rudolf Vrba

SlimVirgin brings into this the question of Rudolf Vrba. While a fascinating dispute, apparently well covered, I don't see it as having much to do with this text; I would not describe as a difference between scholarly and non-scholarly sources, but as one between primary and secondary sources. I can only judge from the present state of the article, but the points seem to be as follows.

• Vrba asserts now that certain things happened in 1943 and 1944; he asserts that he was an eyewitness to some of them.
• He published this claim in 1963; whether he claimed it privately in 1944 is one of the points at issue.
• Most secondary sources don't believe him (and their reasons could usefully be expanded upon); one (Ruth Linn) does, and has recently published at length.
• The Israeli Supreme Court found his contentions libelous (apparently; the article's treatment of this is oddly phrased).

This is already covered. We prefer secondary to primary sources. We should not give undue weight to lone dissenters, like this Ruth Linn. An article which contravenes existing policy, as this appears to, really should not be cited as an example against new phrasing. Septentrionalis PMAnderson 19:35, 28 June 2007 (UTC)

Edit WP:RS or WP:V ?

This is getting confusing. I would like to discuss what we want to primarily edit (either WP:RS or WP:V) with an eye towards creating a straw poll to clearly lay out choices so a decision can be made (see proposal outlined below). The addition to WP:RS that initially caught my attention was:

"Version 2 - The most reliable material is published by scientists, scholars, and researchers, particularly publications in peer-reviewed journals and academic presses. These sources are preferred in subjects such as medicine, natural sciences, social sciences, and humanities. However, alternative reliable sources are used when scholarly publications are not available - such as in topics related to popular culture or current events."

which received unanimous support among the participating editors (link here). But while this I think embodies the type of edit we want to make, where do we want to put it? (let's argue of the wording later).

I'm ok discussing here, as this forum purportedly has a wider audience. Though it does feel odd to discuss changes to RS on the V talk page. On the other hand, as the verifiability page directly references the RS page, it might have to be tweaked to make sure they don't conflict. However, it is my opinion that creating a hierarchy of Source reliability does not go against NPOV, and this hierarchy is mostly in effect already with respect to science-oriented articles (and as noted above, this idea is endorsed by ArbCom, if that carries weight for any of you).

SlimV and I (and probably others) do agree on one point, I also think it's a bad idea to modify the official policy page with this type of text. I think it was pointed out earlier that there may be exceptions, which is why I think we should primarily modify/edit the WP:RS page (a guideline). The verafiability page seems sufficiantly broad enough to me that adding the text to RS would not create a conflict. R. Baley 20:41, 26 June 2007 (UTC)

Catch 22. Anything that's done on WP:RS will be summarily reverted as inconsistent with WP:V (as has been done numerous times), and yet we're being told no modifications are to be made to the policy page -- even if they're consistent with what Arbcom has said in the past. Raymond Arritt 20:50, 26 June 2007 (UTC)

The cynical side of me suspects that may be the point, I was/am hoping a straw poll could clear things up. R. Baley 21:16, 26 June 2007 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Straw poll proposal

It looks like there are four options:

1. Modify WP:Verifiability (official policy) page -because either WP:RS isn't really relevant, or because it belongs on this page, either way.
2. Modify WP:Reliable Source (guideline) page with no changes to WP:V as none are needed.
3. No modification or edits needed with respect to the type of change quoted above, as it conflicts with NPOV.
4. We have to decide exactly what the text is, before we know where to put it.

This isn't a straw poll at this time. So please don't vote yet. I want to make sure that the above choices are (1) worded well (2) fairly reflect the positions/concerns most everybody has. Whatever we decide, I would like to keep the number of choices to a minimum so as to keep the confusion down (4 may already be too many). Thanks R. Baley 20:41, 26 June 2007 (UTC)

Comments on the above straw poll suggestion

With due respect, that we don't need it for the timebeing. Let's get the section being discussed in. This page has primacy and discussion should proceed from it. Marskell 20:44, 26 June 2007 (UTC)

I just think it's confusing to be discussing not only what to put in, but where. Some people may want the change, but not if it's on WP:V. Some are saying we can't modify WP:RS without discussing here. I think we can modify the RS page, but others have said that to do so violates NPOV, a core policy. That said, if people would rather get the wording right, that's ok by me too, I just thought this might be a better way to go. R. Baley 20:59, 26 June 2007 (UTC)

Sounds a good idea, as SlimVirgin's comments above show, some people appear confused as to if WP:Reliable sources is still being used by the community. I think core policies should be kept as clear and concise as possible, so this proposed addition should go in Reliable sources. Tim Vickers 21:02, 26 June 2007 (UTC)

Or it should go here by the same logic. That we even need RS is an open question. Marskell 21:06, 26 June 2007 (UTC)

I've amended the first choice to reflect this. R. Baley 21:11, 26 June 2007 (UTC)

We don't decide policy with straw polls. This is the kind of thing that used to happen at RS, and it's one of the reasons people don't trust it. SlimVirgin ^(talk) 21:15, 26 June 2007 (UTC)

Surely we can poll to decide what people would like to change and where they would like to change it? I'm not even suggesting yet what the actual words are, but what ever they turn out to be ecxactly, we need to be able focus on exactly what it is we want to do. R. Baley 21:19, 26 June 2007 (UTC)

Kill the straw poll. Policy is not designed to be discussed in polls. Discuss, debate, exchange views, listen to concerns of others and express yours. ≈ jossi ≈ (talk) 21:21, 26 June 2007 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Restart

Agreed about the straw poll, but how do we change a policy like this? I'm all for open discussion, but not to the point that discussion is an end in itself with no possibility of a substantive outcome. Raymond Arritt 21:25, 26 June 2007 (UTC)

The first thing you have to do is make sure you fully understand it and its related policies, NOR and NPOV, and how they hang together. Many of the people commenting here and at RS don't. Then you'll have an idea of the full implications of any change you propose, and you'll be able to anticipate objections, which will speed up the debate. SlimVirgin ^(talk) 21:27, 26 June 2007 (UTC)

See Wikipedia:The role of policies in collaborative anarchy ≈ jossi ≈ (talk) 21:40, 26 June 2007 (UTC)

And you'll see there an explanation of the importance of the concept of attribution, and that all notable views must be included and attributed, in rough proportion to their popularity. This is the most fundamental principle of Wikipedia, and even if you succeeded in changing it here, and I hope you won't, it would make no difference, because NPOV is a Foundation issue. SlimVirgin ^(talk) 21:45, 26 June 2007 (UTC)

We either need to clarify in the guideline page discussing reliable sources what we mean when we say reliable sources, or add such a clarification to this verifiability policy. The proposed addition to the RS page does not contradict in any way the current V policy. Please read it carefully. Tim Vickers 21:51, 26 June 2007 (UTC)

(copied from RS talk) Tim, I see that you hadn't edited NPOV, NOR, V, RS, ATT, or their talk pages, until two days ago, yet you're rushing in to add material that was rejected a long time ago, for good reason, and asking us to go through the arguments for the millionth time, which frankly isn't fair. If you want to make big changes, can I ask you please to spend some time reading the related policies carefully and checking the archives first? SlimVirgin ^(talk) 22:08, 26 June 2007 (UTC)

Popularity? Popularity among whom? Among the general public? e.g. public opinion polls? or rough proportion of coverage to views by reliable/informed sources. If the former, do you realize that nearly half of Americans mistakenly believe that surgery causes cancer? That's what a 2005 poll by the American Cancer Society found. [15] So, do we allow half the space in the relevant articles to talk about this mistaken belief? Among medical professionals, experts, and those informed, surely they have the facts straight. Surely reliable sources, such as peer reviewed journals and those published by academic presses, don't give much weight to the myth. Though, the results of the poll were published in a scholarly journal and should be mentioned somewhere. But, WP:NPOV#Undue weight applies. This is just one example of how public opinion polls diverge from the rough proportion various views are held or discussed by reliable sources. --Aude (talk) 21:55, 26 June 2007 (UTC)

(EC, I already wrote this so. . .) I think if carefully worded, a guideline on sourcing won't contradict NPOV. We don't fill up the Islam page with christianity's take on it in a rough proportion to it's demographic popularity. Similarly, science articles such as global warming would be a disaster if it were edited from the view of a popularity contest (US centric, I know, but even world-wide there is considerable "debate" coming from people with an agenda to push). I am hopeful that I just misunderstand what you're saying.

Also, while the above section was indeed called a "straw poll," I wasn't trying to get official policy content included based on a vote. I was just trying to find out what, if anything, people were wanting to change, and where (if anywhere) they wanted to put that change (as well as the basis of objections clearly described). R. Baley 22:02, 26 June 2007 (UTC)

What SlimVirgin wrote wasn't entirely clear, when she said "Notable views" I'm sure she meant views that have been published in reliable sources. You see, the entire question revolves around what we consider to be reliable sources, which is why this definition is so important. Tim Vickers 21:58, 26 June 2007 (UTC)

I looked again at the WP:V, WP:NOR, and WP:ATT pages. None of these mention "popularity". The only policy page that does is WP:NPOV, which concerns WP:NPOV#Undue weight. There it says "NPOV says that the article should fairly represent all significant viewpoints that have been published by a verifiable source, and should do so in proportion to the prominence of each.", with "verifiable source" pointing back to this page. What is a verifiable source? One that is reliable, as explained in more detail on WP:RS. Reliability comes with good editorial process and review, which is characteristic of scholarly sources, as well as high-quality journalistic sources (e.g. New York Times). For scientific topics, scholarly, peer-reviewed publications have a more rigorous review process and would be considered even more reliable than the New York Times. --Aude (talk) 22:04, 26 June 2007 (UTC)

Well, bizarre in one respect—popularity has nothing to do with anything, certainly not foundation issues. Marskell 22:06, 26 June 2007 (UTC)

Popularity among reliable published sources obviously. Do not nitpick, please. SlimVirgin ^(talk) 22:10, 26 June 2007 (UTC)

Indeed, and in proportion to the prominence of each reliable source - which is where guidance on how to assess the prominence and reliability of each source would be helpful. Tim Vickers 22:16, 26 June 2007 (UTC)

I didn't mean to nitpick. The strength of NPOV in many ways is its anti-popularity. Intelligent design is relatively popular after all, but those two words are (rightly, according to NPOV) not mentioned in the body of Evolution, an FA.

Perhaps another way to put it: if "sources should be appropriate to the claims made" and scholarly sources are appropriate to claims made in the sciences, then the "popularity" of claims is determined by scholarly sources. That seems very simple, and the Arb ruling clearly buttresses it. Marskell 22:18, 26 June 2007 (UTC)

Writing policies

For the editors who've not been involved in writing policies before, but who want to add changes here, the essence of successful policy writing is brevity. Every time you add a word or a sentence, you're opening up misunderstandings or unforeseen consequences, not to mention loopholes for deliberate wikilawyers. Therefore, you have to be a good writer who's in command of his vocabulary, who knows how to make sure every word counts, who understands the wider implications of each change, and who's completely familiar with NOR, V, and NPOV, and the way they work in practice, not just in theory. That means understanding all the things those pages say, and all the things they deliberately don't say, all the reasons for both, and how editors use those additions and omissions for good and for bad.

Anyone trying to whack in a sentence here or there without being totally familiar with the situation is like someone wandering in off the street to bet a pile of money on a horse without knowing the first thing about horse-racing. It's senseless, and it won't lead to anything good. Some reflection would therefore be appreciated. SlimVirgin ^(talk) 23:18, 26 June 2007 (UTC)

1. What does "no fact-checking facilities or editorial oversight." mean? What are "fact-checking facilities"? Do telephones and Google count as facilities that could be used for fact checking? Does anybody object to the removal of the word "facilities"? Is what is meant "fact-checking procedures"? Tim Vickers 00:37, 27 June 2007 (UTC)
2. What examples might there be of a non-scholarly professional researcher? As a scholar is usually defined as a learned person or a specialist in a given area, how is it possible to be a professional researcher and not be a scholar? Tim Vickers 01:14, 27 June 2007 (UTC)

Professional writers and publications have standard processes for fact-checking. That's what we're referring to. A journalist is a non-scholarly professional researcher; most people paid to do research outside a university would count as a non-scholarly researcher. Tim, this really isn't helpful. You seem to be trying to reinvent the wheel. SlimVirgin ^(talk) 01:32, 27 June 2007 (UTC)

Professional writers and journalists are only fact-checked when writing for relaible sources. When writing a blog there may be no fact-checking whatsoever. In those cases we're relying purely on their own integrity and professionalism. ·:· Will Beback ·:· 20:18, 27 June 2007 (UTC)

Journalists go through a rigorous review and editorial process, albeit not quite as extensive as scholarly peer review. For many topics, a New York Times article would be a good, reliable source. Someone outside of a university, but with a think tank, would go through some reputable publisher that has a rigorous review/editorial process. But "professional researcher" is a vague term that I really don't think is helpful. --Aude (talk) 01:38, 27 June 2007 (UTC)

This gentleman calls himself a professional 9/11 researcher. But, has nothing peer reviewed or scholarly to show for it. I think we need to stick with the term "scholarly". --Aude (talk) 01:30, 27 June 2007 (UTC)

He also has nothing published in a reliable source to show for it, so our current exception wouldn't apply to him. JulesH 08:35, 27 June 2007 (UTC)

I fully agree that we need to consider possible unforeseen consequences of modifying a guideline or policy. That's why I'm nervous about the "scholarly and non-scholarly" phrase. I think we're just coming at this from different fields of editing. Yes, on the Holocaust (to take an example from above), non-scholarly sources (e.g. survivors) may have much to say that is of equal or greater value to what scholars have written. However, as an editor who works primarily on medical and scientific articles, I can tell you that opening the door to "non-scholarly" sources will make it even more difficult to enforce WP:WEIGHT, WP:FRINGE, and the other policies try to ensure that such articles are truly summaries of the current state of knowledge. Clearly, "reliable sources" on historical topics, people, places, events etc are not quite the same as "reliable sources" on medical/scientific topics. Yes, an intelligent and scrupulous editor realizes this intuitively, but guidelines are still useful. One solution might be to develop subpages (for instance, WP:MEDRS, which is a proposal under construction to guide the use of "reliable sources" on medically-related articles). I really get the sense, though, that the proposed wording ("scholarly and non-scholarly") is geared toward fixing a problem some editors have encountered, without perhaps a full consideration of its effects on editors in other areas. MastCell ^Talk 02:00, 27 June 2007 (UTC)

Subpages for specific topic areas is not a good idea. Do we want specific guidelines for 9/11 conspiracy theories, and how they should be covered in the main 9/11 articles? What about specific guidelines for global warming? evolution? etc. A general guideline on reliable sources that discusses what would be considered reliable for various topics would be best. To do that on a page entitled "Reliable sources" give extra emphasis and helps reiterate the fact that we want people to use reliable sources. It's a helpful guideline page. --Aude (talk) 02:06, 27 June 2007 (UTC)

I would have to say that if there were to be a special page, it would have to be about all conspiracy theories, not just those related to the events of 9/11. See other articles such as Area 51, Apollo Moon Landing hoax accusations, etc.--MONGO 07:17, 27 June 2007 (UTC)

Also, I've worked on gun violence and criminology topics. There are special interests with strong POVs at work on those pages. High standards for reliable sources has been set on the gun violence pages and understood that it's expected. That has helped maintain stability on those pages, even in the midst of the recent VA Tech shootings, the gun violence pages are still stable. Diminishing and depreciating the WP:RS page or redirecting it won't be helpful. --Aude (talk) 02:09, 27 June 2007 (UTC)

I've replaced "fact-checking facilities" with "fact-checking procedures" as this is closer to the meaning that is intended. I agree that a journalist is a non-scholarly researcher. I think the next sentence covers the concerns expressed above about crank "researchers", so what about just saying "a well-known, professional researcher in a relevant field."? Tim Vickers 02:18, 27 June 2007 (UTC)

Sources by well-known, professional journalists are entirely acceptable. "Non scholarly" still throws in a loophole. Instead of "Well-known, professional researcher (scholarly or non-scholarly) in a relevant field.", how about "Well-known, scholarly researcher or professional journalist in a relevant field." --Aude (talk) 02:29, 27 June 2007 (UTC)

So in an article about physics, a journalist's interpretation is as good as a peer-reviewed article by a physicist? How accurate are the science articles in your newspaper? They must be a lot better than the ones I read... Raymond Arritt 02:33, 27 June 2007 (UTC)

That's why trying to sum up reliable sources here is problematic. Simply saying reliable sources with a link to the guideline page would be best. The guideline page can say something like "The most reliable material for subjects such as medicine, natural sciences, social sciences, and humanities are published by scientists, scholars, and researchers, particularly publications in peer-reviewed journals and academic presses. However, alternative reliable sources are used if scholarly publications are not available - which may sometimes be the case in topics such as popular culture or current events." can be stated, with further examples and discussion covering various topics. --Aude (talk) 02:38, 27 June 2007 (UTC)

Hey, I like that formulation! :) I'm also OK with ""Well-known, professional journalist or an academic researcher in a relevant field." Tim Vickers 02:45, 27 June 2007 (UTC)

I think Aude put it very well. Opus33 03:44, 27 June 2007 (UTC)

Social sciences and humanities should not be included in any formulation. Marskell 07:02, 27 June 2007 (UTC)

I don't think simply removing "(scholarly or non-scholarly)" is an appropriate solution to this problem. Journalists are acceptable sources for many articles. Other kinds of experts may be acceptable for other articles (e.g. professional antique valuers for articles on antiques, lawyers for articles on law, etc.). It has previously been argued that these kinds of sources are accepted as non-scholarly examples of professional researchers. If we are to remove the text, we must have a method of rephrasing that accepts the work of these kinds of experts. The phrasing that User:Marskell recently introduced is a blatant example of systematic bias towards scholarly subjects, and must therefore be rejected. JulesH 08:35, 27 June 2007 (UTC)

Given the potential for misuse (just think of the words "self-published non-scholar" strung together) it needs to be much better incorporated and explained. After the many debates of the last two days, we presently have nothing added wrt to this; for now, that's probably the best compromise. Marskell 09:10, 27 June 2007 (UTC)

That's "self-published non-scholar professional researcher who has been previously published by reliable sources". I don't have a problem with this. I do have a problem with the policy changing so that it isn't acceptable. JulesH 10:36, 27 June 2007 (UTC)

If you're convinced this is a problem, please provide an example of an unacceptable source that is permitted under this definition. Then we can work to ensure that such a source is not allowable. JulesH 10:38, 27 June 2007 (UTC)

To be clear about one thing, the bit is barely two months old. Placing it in, not removing it, is what's changing policy. This idea that we should make no distinction between scholarly and non-scholarly is a development I have seen nothing of before (with middle-of-the-road policy watching). And "self-published non-scholar professional researcher" is something of a contradiction in terms.

Anyhow, was I wrong here? I'm not pulling the pseudoscience argument out of nowhere—these people turn up constantly. My principal argument had been that because we weren't talking about something for which academic scholarship exists, you can't call the man a scholar on the topic. But now we want to deliberately remove that distinction in policy? (See also this lunatic).

And the "relevant field" bit does not cover it—the ufology or astrology editor will argue "no, no in this 'field' this man is very well known, perfectly relevant." Marskell 12:48, 27 June 2007 (UTC)

I have gone back and looked at previous versions of WP:RS, from the fall, before much content of that page was eliminated. One thing it used to say:

"Reliability is a spectrum, and must be considered on a case-by-case basis. Typically peer reviewed publications are considered to be the most reliable, with established professional publications next. Government publications are often reliable, but governments vary widely in their level of reliability, and often have their own interests which will explicitly allow for withholding of information, or motives."

Something like this might be good to say somewhere. It would be good to bring this back on the WP:RS guideline and maybe say something here. --Aude (talk) 12:54, 27 June 2007 (UTC)

This change was made shortly following a number of other changes, specifically one that removed other exceptions that previously existed. It was argued at the time that the people who were covered by the previous exceptions were "non-scholarly researchers" and therefore this phrase was introduced to disambiguate the fact that researcher was being used in an inclusive, rather than exclusive sense. Feel free to look back at the archives.

As to whether you were wrong in the Exopolitics article, without reading more than I have time to I can't be certain but my feeling is, yes, you were. As I see it, the author of the self-published book in question is accepted as an important figure in the history the subject, and therefore anything he has published on the subject should be considered for inclusion. Certainly it should not be eliminated merely on the grounds of self-publication. It doesn't help that we're talking about a fringe subject, but that's an issue of NPOV, not verifiability. My opinion is that the book should be used as a source, but we should be careful to ensure prose attribution is used for its contents (again, as an issue of NPOV).

But, in the end, I'm not suggesting the text should remain as it is: I'm suggesting that we come up with an alternative that strikes a middle ground, which allows for self-published sources written by experts who don't happen to be employed in the position that the majority of people will think of on reading the word "researcher". There are certainly plenty of experts whose opinions are worth including who don't fit this description, and we shouldn't have to ignore this policy to use those experts in an article.

I've thought for a while that this entire section on self-published sources needs a total rethink, and now might be the best time to start it. But I don't think we should be making any changes until we know what the outcome should be. JulesH 14:08, 27 June 2007 (UTC)

If fringe, self-published sources on non-academic topics become includable simply because a fringe community calls the info relevant to the history of their non-academic research...we would have a very serious problem. The P&Gs and, as noted above, Arbcom, do not agree with you.

As for a middle ground, let me invert and ask for a "self-published non-scholar professional researcher" that should be included but wouldn't be without that bracket. I believe survivor stories were mentioned. Marskell 14:27, 27 June 2007 (UTC)

One such source is the use of Teresa Nielsen Hayden's essay on Roger Elwood. While I accept that it may not have been wise to use while the subject was still alive, I can see no argument against it now. Nielsen Hayden is, without doubt, an expert on the workings of science fiction publishing, and her analysis of the influence of Elwood's publications on that field is, as far as I am aware, impossible to replace, because nobody else has done the research to catalogue those effects. JulesH 16:18, 27 June 2007 (UTC)

"Well-known" (as in "well known journalists") is too vague to be a standard. The problem with journalism is that there is no clear criteria for that profession. Have Jayson Blair, Hunter S. Thompson, and Bill White (neo-Nazi) been published in reliable sources? Yes. But I wouldn't use anything any of them wrote as a plain source. As for Nielsen Hayden, she admitted that she had not actually researched the catalogs, but was basing her essay on her recollections[16] and is what we often call "original research". However even the original research opinions of experts are often worth quoting with attribution ("According to longtime SF editor TNH, ..."). But does being an acknowledged expert on SF anthologies make one an expert on animal hoarding? We need to maintain clear language that experts are only experts within their field. ·:· Will Beback ·:· 20:18, 27 June 2007 (UTC)

Was it original research? Of course: that's my primary contention. She's the one who performed the research, she's an expert in the subject, so we should quote her research. I'm all for the use of prose attribution when relying on a self-published source, I think it is usually a good idea (see my suggested changes to the policy below). As for animal hoarding, I would suggest anything in that article that's sourced to her essay on the subject should probably be removed as she isn't an expert on it, but as I recall the article merely contained an external link to her essay, rather than used it as a source. JulesH 08:14, 28 June 2007 (UTC)

Will vs May

I have to agree with SlimVirgin on this one Diff 22:54, 26 June 2007 unreferenced may be removed or it may be referenced. The project Wikipedia:Unreferenced articles is here to begin referencing everything that is unreferenced (there are others also). Though not to long ago (days) it was strongly (and wrongly) attacked for Deleting articles. Jeepday (talk) 23:23, 26 June 2007 (UTC)