Wikipedia talk:No original research/Archive 53

Newbie
So help a newbie out here. Hypothetical situation, cause it's probably very unlikely. Lets say something crazy happens, like maybe the sky turns florescent green for a day, but not a single soul on the face of the earth reports on it. No verifiable sources, but everybody knows it's happened. Does that then mean that Wikipedia cannot have an article on the subject? Extreme situation, I suppose a milder way of asking is can you claim the sky is blue without citing a source.


 * This would never happen: if the sky did indeed look green for a while, it would be widely reported. We should not discuss a hypothetical situation that would never arise.--Gautier lebon (talk) 10:27, 10 July 2010 (UTC)

Second situation... A large number of verified sources publish a piece of misinformation, intentionally, across the board. Everyone knows it is misinformation. No verified source points out it is misinformation, but everybody just knows because it's just something inherently obvious. Must Wikipedia then represent the mistruth as reality in an article if someone decides to create such an article and back it up with their sources? Again, ridiculous scenario, 1984-ish, but I'm trying to get a feel for the reach of the core policies on Wikipedia.

Thanks --HeroofTime55 (talk) 05:24, 10 July 2010 (UTC)


 * How is someone who was asleep for the day to accept that the sky was fluorescent green without reliable sources? That the sky is blue is both common knowledge and easily verifiable. The policies aren't supposed to cover every single case, they just describe common accepted good practice. However I don't think we would be able to include that the sky was green for a day even by invoking WP:Ignore all rules, I'm pretty sure the consensus would be against including it in Wikipedia without some reliable source even if everyone agreed it happened. Dmcq (talk) 06:52, 10 July 2010 (UTC)


 * If the statement is not challenged and not likely to be challenged, a reference would not be necessary. Whether Wikipedia could have an article on the event would depend on whether the event is notable. If the sky turned fluorescent green, there would be sources and it would be notable.
 * The problem with misinformation is probably more likely to be relevant. Religion is an obvious example (presuming that not all religions are right). It might be necessary to achieve consensus (on the Talk page or elsewhere) as to whether a sourced statement is fact or opinion and whether particular positions are given due weight. In such circumstances, the requirements regarding reliable sources might be more stringent, and more discussion might be required. --Boson (talk) 09:05, 10 July 2010 (UTC)


 * Reliable sources often publish misinformation on the 1st of April. Peter jackson (talk) 09:41, 10 July 2010 (UTC)
 * Yes I must unveil the conspiracy to hide the existence of spaghetti trees, details oof which were once released from a reliable source. :) Dmcq (talk) 10:00, 10 July 2010 (UTC)
 * I agree with Boson. If sources (such as newspapers) that are ordinarily considered reliable publish something, then it can be included in Wikipedia.  And it is up to the reader to evaluate whether, in this particular case, the information is not reliable despite its publication by a source.--Gautier lebon (talk) 10:27, 10 July 2010 (UTC)

It's good to keep in mind that while this policy does limit what can be published in an article, the other side of the coin, verifiability, is not a a suicide pact. That is, we don't have to publish something simply because it has been published in a reliable source. If a reliable source published an obvious piece of misinformation, which was never retracted, or refuted by other sources, we can form a consensus to simply not write about it. In the course of discussion on a talk page, when you're trying to decide what sources to use and what sources to ignore, you are absolutely permitted to use original research to evaluate the sources. Someguy1221 (talk) 01:15, 11 July 2010 (UTC)


 * That's the reasonable, common-sense view, but is it actually stated in policy anywhere? Or even guidelines? Peter jackson (talk) 08:52, 12 July 2010 (UTC)


 * I think the sense of it is more apparent than real. How could we determine that something was an obvious piece of misinformation?  Only because we have some source that says it is so.  Otherwise, we are in the position of simply not believing what we read.  WP:V is quite clear that what editors believe is of no consequence for what can be in the encyclopedia.  If, on the other hand, we do have a source that states that something was misinformation, then we have the problem of conflicting reliable sources.  That is covered by WP:NPOV, so again our judgment concerning the truth of the matter is not called for.   RJC  TalkContribs 14:05, 12 July 2010 (UTC)


 * Responding to the question from Peter jackson, see WP:IAR. Wtmitchell (talk) (earlier Boracay Bill) 01:32, 13 July 2010 (UTC)

I was there that day. It was in the 60s at university after finals. We were all drinking and there was great debate about what the actual color was: Variations of green and Chartreuse (color) I think there was a journalism student in our group so I am surprised it never made the papers.--Mike Cline (talk) 14:33, 12 July 2010 (UTC)
 * I suspect that it never made the papers because the Nixon Administration ordered the CIA to suppress it ("Mind control is a terrible thing to waste"). Thankfully, the counter-culture has been keeping a memory of this event alive since then (why do you think the band is called "Green Day"). Blueboar (talk) 15:18, 12 July 2010 (UTC)

Question 1 regarding primary sources
In the article Forced labor of Germans after World War II I had inserted a short summary of a factual event taken from this publication: United States Department of State

Foreign relations of the United States, 1947.

The British Commonwealth; Europe

Volume III

Washington, D.C.:

U.S. Government Printing Office (1972)

The editors of the book follow the following policy

According to editor Nick-D I was wrong to include my sentence, he removed it and elaborated as follows on the article talk-page.

Statement used out of context

I've just removed this from the article: This is taken from a document from the US government asking the French to explain what happened to these POWs. Without the French explanation, it is not usable by itself. Moreover, Wikipedia articles should not be cited to primary sources like this. Nick-D (talk) 07:23, 4 January 2010 (UTC)
 * In discussions between France and the US in early 1947 regarding whether France should begin repatriating its German prisoners it was noted that of the 740,000 handed over by the U.S. to France for forced labor only 450,000 remained; 290,000 had been "stricken off the rolls".

Was Nick-D wrong or right in his interpretation of Wikipedia policy as regards that sentence? Are there any other aspecs worth noting, e.g. cases in which state department publications may or may not be used? Nick-D made his position very clear and elaborated, but now I would like to hear the opinion also of others who have grasped the relevant policies and their application.--Stor stark7 Speak 20:15, 13 July 2010 (UTC)
 * Well, this is clearly a primary source from the US government. Nick-D seems to doubt whether the US knowledge about French bureaucracy ("stricken off the rolls") makes this a simple fact. With these doubts whether this is a simple fact, primary sources are no longer acceptable per se.
 * It is a bit of a borderline case though in my view, and in such cases we should err on the side of safety (i.e. if someone raises the point, accept their worry) Arnoutf (talk) 20:54, 13 July 2010 (UTC)


 * Where did you hear about the US government communication from? That would probably be a much better source for the article. The statement should certainly be qualified better as just being according to the US government. Did the French ever reply? That should be included if they did but if they didn't it doesn't stop the US government query being usable. Dmcq (talk) 21:14, 13 July 2010 (UTC)


 * The primary source document in question is being used without a secondary published source to back it up. The implication is that the 290,000 German POW were dead, in fact they were released. Nick-D and myself made it clear that primary source documents must back up published material in secondary sources. Please allow Nick-D to reply here. --Woogie10w (talk) 21:59, 13 July 2010 (UTC)


 * Stor Stark7 is attempting to misuse these primary source documents to synthesise a case that the US was responsible for the 800,000 deaths of German POW after WW2. The arguments refuting these claims have been covered in the Wikipedia article Other Losses--Woogie10w (talk) 22:15, 13 July 2010 (UTC)


 * For the record here is the thread where I replied to User Stor stark 7 Talk:World War II casualties--Woogie10w (talk) 23:24, 13 July 2010 (UTC)


 * 1.st how is your edit here of value as regards the usability of the link in an article? What is the purpose of your edit here, except as a forum for an attack on me?
 * 2nd. I am fed up with your attacks, I've reported you at ANI.
 * 3rd, unless you are out to provide a biased view you should provide a link to the full discussion that here includes both sides instead of only a link to your text.
 * 4th, For the record, I made a summary of the whole thing.--Stor stark7 Speak 00:27, 14 July 2010 (UTC)

Just to elaborate a bit on my talk page post when I removed this, the fundamental problem with the text is that it's using a single primary source in isolation of both any context (eg, who wrote it? why was it written? why was it sent? was it based on correct information? why was a copy of it preserved and published?, etc) or other sources (most notably, but by no means limited to, any preceding communications, the French Government's response and the US Government's response to this). There might be an interesting journal article or book which could be written from following up this and related documents in a professional manner, but Wikipedia isn't the place for it. It should be noted that collections of government documents such as the volume in question are primary sources - they consist of documents taken from official files and published with no or only minor changes and should be treated in exactly the same way the original documents in the files are treated. These collections are produced by archivists to serve as the foundations of historical research and are rarely accompanied by any commentary (as this is the job of future historians to write based on the documents, and other sources). Nick-D (talk) 11:14, 14 July 2010 (UTC)


 * Looking at this all. Any suggestion what may have happened to the missing 290000 would be original research (ie WP:synth). Not making the claim, but combining sources in such a way that it clearly suggests that something happens: For example the following paragraph would be a synthesis: "740,000 POWs were handed over to the French (ref1) The French were reported to mistreat their POWs (ref2) Finally the remaining 450,000 POWs were released (ref3)" as it implies that the mistreatment has a lot to do with the difference between 740,000 and 450,000. Arnoutf (talk) 11:33, 14 July 2010 (UTC)


 * We do in fact have a reliable secondary source that reports 940,000 German POW were in French custody and 34,000 died in captivity. The source is Dr. Rüdiger Overmans, an associate of the German Armed Forces Military History Research Office until 2004, Overmans provided an official reassessment of German military war dead based on a statistical analysis of German military personnel records. The Overmans research project was supported and funded by the German government


 * Rűdiger Overmans. Deutsche militärische Verluste im Zweiten Weltkrieg. Oldenbourg 2000. ISBN 3-486-56531-1 Page 286--Woogie10w (talk) 11:47, 14 July 2010 (UTC)
 * Which makes it clear that that source is to be used. As it is (a) secondary and (b) provides actual numbers, so no speculation and synthesis needed. Arnoutf (talk) 11:52, 14 July 2010 (UTC)
 * Agreed. The use of the primary source document would only be acceptable if appropriately attributed, and frankly secondary sources would be superior for providing context, analysis, and representing the scholarly consensus (or scholarly disagreements).  Since a secondary source exists, it should be used and ideally such a controversial point should be addressed with multiple secondary sources.  It's always a risk to use a primary source, and it's generally a bad idea if there are secondary sources available.  If there is truly a scholarly debate over what really happened, then document it using scholarly sources - not primary sources.  WLU (t) (c) Wikipedia's rules: simple/complex 12:34, 14 July 2010 (UTC)

WP:OI: image manipulation for "legitimate" reasons
I have a question regarding the stipulation in this policy that manipulated images shouldn't be uploaded to Wikipedia. I understand why the policy makes a stand against this, as I don't want to see, for example, manipulated images of people to show them in a better light or other similar deceptions as much as anyone else who comes here for facts. However, for reasons of privacy, I have in the past used photo editing software to blur vehicle license plate numbers and in some cases peoples faces before uploading such images to the internet. When taking original images to illustrate a subject on Wikipedia it is not always possible to take it in such a way as to avoid potential breaches of privacy which can only be adequately addressed after the fact by manipulating the images. So I guess my question is, can images be uploaded to Wikipedia, and be compliant with WP:OI, if they have been manipulated for reasons of privacy, particularly if the manipulation does not affect the illustration of the subject in question? – Matthew25187 (talk) 00:59, 16 July 2010 (UTC)
 * You are allowed to blur or pixellate for privacy reasons, an example is given of pixellation at commons:Commons:Photographs of identifiable people showing 'a fat girl', or as the original says ' a McDonald's patron'. Other than that you are allowed to do basic improvements like cropping or brightening a picture but should also upload the original and link to it, see How to improve image quality. A summary should be given of any such manipulation like 'cropped and removed blue tinge'. The main injunction in WP:OI is "It is not acceptable for an editor to use photo manipulation to distort the facts or position illustrated by an image." Editors are quite strict in its interpretation on the give an inch and they take a mile principle. Of course one can manipulate an image for the purpose for instance of illustrating manipulation but it should be clearly labeled as such. Dmcq (talk) 05:43, 16 July 2010 (UTC)

Primary Sources: A question regarding policy
Here is a quote from No original research:


 * Primary sources that have been reliably published may be used in Wikipedia, but only with care, because it is easy to misuse them. Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used to make descriptive statements that can be verified by any educated person without specialist knowledge. . . Do not base articles entirely on primary sources.

And one from Verifiability:

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely, with page numbers where appropriate, and must clearly support the material as presented in the article.

Aren't these precepts superseded by common practice, particularly in scientific articles? BruceSwanson (talk) 17:19, 13 July 2010 (UTC)
 * I am not completely sure what you mean here. You show a note which gives a scientific article, which is a reliable source and which gives all the evidence per WP:V you quote.
 * The primary source is usually not a problem in these cases. The more problematic primary sources are for example published speeches of politicians; which should be indeed treated with care. So what is your specific question here? Arnoutf (talk) 17:27, 13 July 2010 (UTC)
 * This obviously refers to the ongoing disupte at Zidovudine (AZT) and Talk:Zidovudine. Policies should document the consensus of the community.  In this case, they do.  "Material" refers to the text of the article - that AZT was synthesized in 1964.  You challenged the Horwitz 1964 citation - since the material was already cited to three sources, all reliable, the challenge was somewhat moot.  There were two peer reviewed journals, with full citations - author, title, year, journal, volume, issue, pages, and now weblinks and PMID numbers - plus an entire film about the development of AZT as a treatment.  The primary source (Horwitz) requires no interpretation since it verifies that he developed a compound, Zidovudine, in 1964.  This is supplemented by a second peer reviewed journal article to substantiate this point - and a secondary source at that - as well as the film.  Does Horwitz, 1964 verify that Zidovudine was synthesized in 1964?  Yes.  Does this require interpretation?  No.  The text itself meets all requirements of WP:PROVEIT - the text was challenged (for not good reason, since it already had a source) and was appropriately attributed to two reliable, published sources in an inline citation with full citation templates (plus a film).  AZT is a clear example of the community's practice, 100% in line with the policy.  The issue, which has been pointed out to you repeatedly, is that you don't believe the citation is adequate (probably because you, as an AIDS denialist, believe AIDS is caused by AZT rather than HIV) and are attempting to skew the article to a more critical stance - that would be POV-pushing of a fringe theory.  You don't appear to like the article because of its stance, and have wasted considerable time on multiple misunderstandings of wikipedia's policies and guidelines, despite numerous editors pointing out your interpretation is incorrect.  More won't hurt I suppose, but since you appear more interested in promoting ideology rather than neutrally reporting on the main stream position that HIV causes AIDS and AZT treats it, I doubt this will ever be sufficient and the goalposts will continue to move.  WLU (t) (c) Wikipedia's rules: simple/complex 17:36, 13 July 2010 (UTC)
 * Which aligns with my comment above that it all looks fine and that I do not understand why this point is brought here (and I did not even know the article) Arnoutf (talk) 17:42, 13 July 2010 (UTC)

Arnoutf, I might be better able to explain if you will identify a scientific article you are familiar with. I'll take a look at it and see if it has any examples of what I'm talking about. BruceSwanson (talk) 18:00, 13 July 2010 (UTC)
 * What about Arnoutf's comment regarding this article, which you find problematic but no-one else does? You asked for a review - it was given.  What do you think of the review and reply?  I think any moderately wikipedia editor looking at this specific article, and any similar article, would agree the AZT article does not present any issues regarding the use of Horwitz, 1964.  WLU (t) (c) Wikipedia's rules: simple/complex 18:14, 13 July 2010 (UTC)
 * (edit conflict) I am familiar with a lot of scientific articles, I even co-authored a dozen or so. What you mean however is something else entirely, you mean a Wikipedia article on a science related topic.
 * Primary sources are fine to identify facts; and most scientific articles in the natural sciences will give a lot of such facts. However less straightforward interpretation of primary sources is synthesis and hence original research. This is especially cumbersome for social sciences, arts and humanities were there are hardly universally agreed upon facts.
 * In the case of the article above a scientific article is quoted to provide dates in a timeline. These would be considered facts even in the social sciences and therefore these could be safely used from a primary source per the relevant OR section
 * A primary source may only be used to make descriptive statements that can be verified by any educated person without specialist knowledge. For example, an article about a novel may cite passages to describe the plot, but any interpretation needs a secondary source. Do not make analytic, synthetic, interpretive, explanatory, or evaluative claims about material found in a primary source.
 * So I really do not see any problem or even disagreement between the sections you quote, and hence no question. What is your question? Arnoutf (talk) 18:19, 13 July 2010 (UTC)

Take a look again at the AZT article by going to this reference. It's attached to the following sentence: Jerome Horwitz of the Barbara Ann Karmanos Cancer Institute and Wayne State University School of Medicine first synthesized AZT in 1964 under a US National Institutes of Health (NIH) grant. Read the two other footnotes as well. Do any of them provide any actual confirmation of the information in the sentence? What I'm arguing is that the editor who posted them, to be in compliance with The source should be cited clearly and precisely, with page numbers where appropriate, and must clearly support the material as presented in the article should have given page numbers and also a contextual quote, to demonstrate that he had in fact accessed those primary sources. That wasn't done, and there is certainly no reason to believe that editor read the source.

Now take a look at footnote 6 and it's accompanying sentence. Read the citation. Is it making "descriptive statements that can be verified by any educated person without specialist knowledge"? How about the rest of the primary sources -- are they making such statements too? Now it may be that deep in the paper the sentence will be corroborated. But the editor should have provided chapter and page information for the reader to follow. That wasn't done. I don't think it was done in any of the primary sources. Shouldn't editors who want to defend the use of such sources do the work of finding the relevant information and putting it in the citation?

Finally, look how many primary sources there are compared to secondary ones. Doesn't it show that the article is in violation of the precept not to base articles on primary sources? BruceSwanson (talk) 18:50, 13 July 2010 (UTC)
 * Ok this is not about sourcing but about lack of assumed good faith WP:AGF. Doubting even these almost trivial facts is very unhelpful and I think the other editors on the article have done more than their fair share in providing proof. If you really want the exact phrasing, just get the papers quoted and figure it out yourself. My advice: Just let it rest as it seems you are trying Wikilawyering to push a non neutral pov WP:NPOV. Arnoutf (talk) 19:13, 13 July 2010 (UTC)
 * This misunderstanding has also been explained to BS. He believes that if he can't immediately verify a statement, then it does not apply.  He does not understand the difference between a web-available abstract and a full text, and that while the web-available version is a convenience, it is not a necessity.  BS objects to having to assume good faith that the editor in question (now three editors) have indeed accurately summarized this incredibly minor and uncontroversial point, yet has not made the trek to a university library, or paid the article access fee, to verify the contents himself.  The statements about page numbers is bizarre, considering Broder and Horwitz both have page numbers - and for articles, which are considered sufficiently short as to not need a single page to verify - this isn't an issue.  If it were a book, it might be a greater problem.  But it's not and the point is both bizarre and absurd.  The footnotes don't need to confirm the information directly within the footnote, they need to verify it by presenting a source. The actual confirmation is in the source, and BS has never made the effort to actually find the sources, review them, and state the sources themselves are inaccurately summarized.
 * Which primary sources aren't reporting basic information? Footnote 6 takes you to Ostertag et al 1974, which has a full-text version, which indeed verifies that Ostertag worked at the Max Planck Institute (author contact information), found that azidothymidine interfered with the Friend virus in mice.  That's quite basic information, in fact that's essentially the whole point of the paper.
 * By the way, journal articles don't have chapters and quotes aren't necessary - they tend to add undue weight to individual citations and clutter up the reflist. BS has failed to point out a single specific error with any of his objections.  The article is not based solely on primary sources, and primary sources are used appropriately.  This has already been explained, but doubtless another spurios point will be raised immediately after.  WLU (t) (c) Wikipedia's rules: simple/complex 19:15, 13 July 2010 (UTC)
 * I fully agree with WLU. Ironic though that proofreading; an article BruceSwanson made constructive edits to recently only list a single source, a (by Wikipedia standards unreliable) company website. You might expect that someone critisising others upholds the same standards for all articles he is editting on. (unless of course there is a point of view to be advanced) Arnoutf (talk) 19:22, 13 July 2010 (UTC)

What point of view do you think I'm advancing? And how is the company website unreliable? Seems as good as this one. BruceSwanson (talk) 20:45, 13 July 2010 (UTC)
 * No, the first one is from a commercial company, the second one from a university. So the second one is more reliable than the first one. Of course faculty pages on university websites are not the best of sources either. If you do not see the difference, I think you have just no clue what a reliable source is. Arnoutf (talk) 20:50, 13 July 2010 (UTC)

I don't see the difference. Neither does Wikipedia. From Verifiability: Reliable Sources:  ''All self-published sources, whether experts or not, are considered reliable as sources on themselves, especially in articles about themselves, subject to certain criteria, though no article should be based primarily on such sources. . . ''

Also from the same section:

''Self-published or questionable sources may be used as sources of information about themselves, especially in articles about themselves, without the requirement that they be published experts in the field, so long as:


 * 1) the material is not unduly self-serving;
 * 2) it does not involve claims about third parties;
 * 3) it does not involve claims about events not directly related to the subject;
 * 4) there is no reasonable doubt as to its authenticity;
 * 5) the article is not based primarily on such sources.

The footnote in question merely illustrates that proprietary typesetting-systems exist and companies use them. But you're right about one thing: the proofreading article is substandard. However, it is labeled as such, twice, so you know right where you stand, and reader beware. Would that AZT -- both the article and the drug -- were so labeled as well. But come to think of it, the drug is so labeled. That just leaves the article. BruceSwanson (talk) 22:00, 13 July 2010 (UTC) BruceSwanson (talk) 22:00, 13 July 2010 (UTC)
 * Bruce, that university page isn't used on AZT. Plus, what the hell are you fighting for?  Do you truly believe AZT wasn't synthesized in 1964?  Do you have any reason to believe this?  What sources used on AZT are self-published?  I see none - mostly journal articles, with some news stories and a couple "misc" that are none the less not self-published.  You are citing irrelevant sections of irrelevant policies showing no understanding of why your points lack merit, despite numerous editors repeatedly pointing out where your understanding is outright wrong.  If you don't see the difference between the sources used and inappropriate sources, the problem is you.  AZT is heavily referenced to reliable sources, and can't in any way be compared to proofreading and doesn't need any tags except perhaps expansion.  We get it, you think AZT is a deadly drug.  And it is.  Were it not for the fact that people with HIV/AIDS would die without it, no-one would use it.  Wikipedia is not here for you to wax eloquent about your personal beliefs, to soapbox about nonsense you can't even criticize because you don't understand.  There are message boards where AIDS denialists can circle-jerk to their heart's content about how AZT causes AIDS.  This is not such a venue.  WLU (t) (c) Wikipedia's rules: simple/complex 01:07, 14 July 2010 (UTC)
 * Indeed, the AZT article is one of the best sourced articles I have seen; I see no indications whatsoever that something is wrong there. Of course the sources and evidence may not align with your personal beliefs, but those personal beliefs have no place on Wikipedia. Let it rest. (PS my reference to the proofreading article had to do with the fact that (from your user page) you seem both knowledgable and ethusiastic about the process of proofreading. The Wikipedia project would benefit a lot from you investing your energy in improving that article rather than figthing against an article like AZT) Arnoutf (talk) 13:23, 14 July 2010 (UTC)


 * I just recently signed on to Wikipedia but am a long-time reader of its scientific articles. I agree with Arnoutf that AZT is an outstandingly sourced article. And WLU's observation earlier that context-quotes would clutter up the reflist is a good practical objection to requiring such quotes in a primary-source citation. Actually I'd even go further and prohibit such quotes because citations without them would probably be discounted by uninformed readers. But give the devil his due: Bruce is right that there are scientific articles based almost completely on primary sources. I would support a change in policy to formalize what is in fact standard practice. The mandate to assume good faith should apply more to primary sources, but while I'm all in favor of good faith, you want to be realistic. It it weren't up to readers to confirm source accuracy -- if editors always had to do it and prove it -- then that might restrict primary-source use and encourage secondary sources in their place, since such sources essentially provide their own quotes and interpretations, however wrong. I don't see how that would improve matters. Instead, reader input should be considered integral to Wikipedia's primary-source practices. And in closing this newbie rant, I agree that basing a citation on an abstract is not unsound, given that, as others have pointed out, the abstract summarizes what is in the article. It doesn't distort it. If it tried to, it wouldn't be published through peer-review. In any case, it is validated by common consent and usage. Operative67 (talk) 18:19, 15 July 2010 (UTC)

This is how primary sources should be handled. BruceSwanson (talk) 01:44, 17 July 2010 (UTC)

Question 2 regarding primary sources

 * 1. When you write an article about a book, is it OK to use the book itself as a reference, e.g. what is written in the preface of the book by publishers and others about the history of the publishing of that edition and previous editions of the book?
 * 2. When you write an article about the book, is it OK to use the book itself as a reference for what it says, e.g. that according to the book so and so many were killed in such and such place?
 * I'm thinking of this book by the way.--Stor stark7 Speak 20:15, 13 July 2010 (UTC)


 * if you read the whole on use of primary sources Our policy: Primary sources that have been reliably published may be used in Wikipedia, but only with care, because it is easy to misuse them. Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used to make descriptive statements that can be verified by any educated person without specialist knowledge. For example, an article about a novel may cite passages to describe the plot, but any interpretation needs a secondary source. Do not make analytic, synthetic, interpretive, explanatory, or evaluative claims about material found in a primary source. Do not base articles entirely on primary sources. Do not add unsourced material from your personal experience, because that would make Wikipedia a primary source of that material.
 * Re 1: It is very unlikely other editors will objects as these facts can easily be verified otherwise.
 * Re 2: That is probably ok as well, if you make sure to cite the exact passages. Speculation about what it means however should be avoided (e.g. In the book X reports that Napolean lost 1,000,000 soldiers in his march to Moscow (page 12-14) which sheds some doubts on other historians (ref, ref) who list the Grande Armee of having been never larer than 500,000 /Alternative/ The author must be mistaken as other historians established the maximum size of the Grande Armee at 500,000 (either version of the last part should be avoided as this is analysis of sources)) Arnoutf (talk) 21:02, 13 July 2010 (UTC)
 * For question 1, I would think that sort of information can be uncontroversially sourced to the book's preface itself, and would only bother attributing it to the book if someone challenges it. If someone really challenges it, then a secondary source may be required to contextualize any controversy.  For the second question, if you're citing a simple "plot summary" (or in this case, primary ideas), you're probably OK to simply state it outright in an appropriately titled section - I wouldn't even reference it since it is clearly describing the contents of the book itself.  But if you are talking about reception, inaccuracies, contradictions, etc. then you would probably need a secondary source about this source and you're probably getting towards territory of a separate article - rather than talking about Taken By Force you'd be editing an article about Rape in World War II or something similar.  An extensive rebuttal would probably be coatrack territory.  WLU (t) (c) Wikipedia's rules: simple/complex 01:13, 14 July 2010 (UTC)

Without commenting on the above questions, it's worth noting that this appears to relate to an ongoing disagreement over the content of the Taken by Force (book) article. The other editor involved wasn't notified of this discussion that I can see - I've just notified them. Nick-D (talk) 11:50, 14 July 2010 (UTC)
 * Sigh, this apparently simple question is clearly provided out of context to advance a point of view (thanks Nick D) for clarifying. For example the phrase "The manuscript to the book was submitted in the U.S. in 2001, but the US publisher chose to suppress the book after 9/11 and it first appeared in 2003 in French translation. " reeks of pov pushing, unless it is a verbatim use of the text in the preface (in which case quotes are needed).
 * To Stork stark7. You can use such materials but only if reported in a neutral matter of fact voice without any analysis, suggestions or implied motives. The way you phrase the text are not as such problematic with regard to the use of the primary source but much more problematic with regard of explicit or implicit synthesis WP:SYNTH made on the basis of these primary sources. Arnoutf (talk) 11:59, 14 July 2010 (UTC)
 * Seconded. Information can be used without attribution or other "kid gloves" approaches only if uncontroversial.  Words like "suppress" are inappropriately emotionally loaded and not uncontroversial.  Certainly, if the publisher deliberately declined to publish because of 9/11, that's worth discussing with appropriate context, but just saying "supressed" has overtones of censorship and conspiracy when it's probably much closer to PR and marketing decisions.  WLU (t) (c) Wikipedia's rules: simple/complex 12:26, 14 July 2010 (UTC)


 * A publisher can't suppress a book unless they've got the author to sign an unusual contract allowing them to refrain from publishing without thereby giving the author the right to terminate the contract & seek publication elesewhere. According to the account above, the author did indeed publish elsewhere, so how could there have been suppression? Peter jackson (talk) 17:12, 14 July 2010 (UTC)
 * They can delay publication and I guess this is what "legally" happened here. In any case this is besides the point, it is about how the source was interpreted/analysed/rephrased which is the problem here, the content debate should be on the relevant article page, not here. Arnoutf (talk) 17:38, 14 July 2010 (UTC)


 * This is an age-old question. Can ANY book on history be treated as reliable source of info in a WP article providing a summary of its own content? – In the hearts of many Wikipedia editors a book description often becomes a collection of facts, because our own guidelines promote that sort of attitude by discouraging expressions of doubt such as "supposed" or "purported" and "semantic cop-outs" represented by the term "allegedly". Take as an example WP article on Neighbors by historian Jan T. Gross. Even now, after an exceptionally long and frantic edit war going back to 2008, it still contains practically inconceivable statements in light of our own NPOV policy, such as that: "depositions produced during a trial conducted in Stalinist Poland, extracted by abusive secret police interrogators [i.e. prolonged beatings and physical torture], are credible in this case." The statement was made by the author... and no-one else. No mention of any serious controversy here, which is a fact. His statement therefore serves as a source of credible info for any unsuspecting reader. -- Fixatif (talk) 19:43, 14 July 2010 (UTC)


 * Well, contrary to what editors such as Nick-D, whose POV should be very evident if you check his edits in the Allied War crimes article, by the way, would have you believe, I did not "invent" the word. It came from an article that was referred to in the preface of the book. The article was written by Richard Drayton, senior lecturer in history at Cambridge University. In it he writes "Five years ago, Robert Lilly, a distinguished American sociologist, prepared a book based on military archives. Taken by Force is a study of the rapes committed by American soldiers in Europe between 1942 and 1945. He submitted his manuscript in 2001. But after September 11, its US publisher suppressed it, and it first appeared in 2003 in a French translation." By the way, you may be interested to know that the editor who Nick-D seems to side with here deleted both the mention of Drayton, and also deleted the link to the article in question once he had deleted the reference to it from the main text. (just as with the other external link by the way, that was also mentioned in the main text before he deleted the reference to it.) But guess this is how Wikipedia works.--Stor stark7 Speak 20:39, 14 July 2010 (UTC)
 * This does not make the case stronger. The 2007 preface is quoting a journal article making it rather circular with a primary source quoting a secondary source, discussing the primary source. If you insist to maintain such a paragraph you should use the original Drayton article. As it stands now the current paragraph reads extremely subjectively. The section needs a complete overhaul, and even then it may put undue WP:UNDUE attention to this issue. In any case, this is all content dispute, as the complexity of the situation goes way beyond the issues that are linked to citing primary sources, and with that we are way beyon the topic of this thread. This should be solved on the article talk page, not here (you may consider getting comments therer through WP:RfC) Arnoutf (talk) 21:20, 14 July 2010 (UTC)


 * I've got no idea why I'm being named in the above post by Stor Stark - I haven't contributed to either this article or the discussion on it beyond the above post observing that this was about an actual disagreement and notifying (the other involved editor in the dispute) about this discussion. Nick-D (talk) 08:49, 15 July 2010 (UTC)
 * This is straying from the original question. IMHO a book can be used as a reliable source for its contents but only as a brief summary; critical reactions and the "truth" of the book requires secondary sources.  The issue of Taken By Force should be dealt with on the talk page - but briefly I would argue against using the Guardian article for simple factual statements because a) it is an editorial and thus more opinion than fact, and b) focussed more on Western atrocities in war than on the book iteslf (though there is obvious overlap).  News stories are never good choices for external links (too short and often insufficient detail) but in this case make an acceptable source for attributed statements (i.e. "Richard Drayton said...") when a full context can be given.  The POV dispute being ported here should be ended.  Basically the article should discuss the book, mentioning the issues discussed in the book only briefly.  However, the book is a perfectly acceptable reliable source on other articles.  The problem here appears to be that the subject of the book is being coatracked into the page about the book.  It's a fine line, but one that should be drawn; the material itself should be used elsewhere. WLU (t) (c) Wikipedia's rules: simple/complex 14:27, 15 July 2010 (UTC)
 * If anyone is interested, I've rewritten the article; I like to think this is a model of an NPOV book page but then again I also like to think I've got a six-pack and am handsomer than the average gentleman. SS7, in my opinion this is the appropriate way to handle a book as a primary source, including controversial topics related to the book.  WLU (t) (c) Wikipedia's rules: simple/complex 15:29, 15 July 2010 (UTC)
 * Although outside of this topic, very strong rewrite. It brings up all the sensitive points but in a factual tone of voice without pointing fingers or implying blame. Thanks! Arnoutf (talk) 15:35, 15 July 2010 (UTC)
 * Thank Tim Vickers, he's my inspiration. Given your comment, I believe it's even more important for SS7 to have a look at that page - we can deal with controversial topics on wikipedia, but there are ways of doing so that are acceptable to the community, within the policies and guidelines, make wikipedia look more professional, and deal with the controversial topic itself in a meaningful way.  It does take experience and very often input from experienced editors as well as examples may be the best way to see how.  It's not easy though, and I don't blame SS7 for both trying to integrate this with wikipedia and being surprised when he hit a bump on the road.  WLU (t) (c) Wikipedia's rules: simple/complex 15:56, 15 July 2010 (UTC)

@WLU, It looks much nicer now, thanks. There are some aspects to be improved though, I left you a message on your talk page.--Stor stark7 Speak 23:17, 15 July 2010 (UTC)

@Nick-D. I went here to get a second opinion on the forced labor issue, and fully provided your rationale. It certainly looks like you feel your original rationale was week and needed political back-up, which was why you "expanded". Were you afraid that left by itself your originally provided rationale would not stand up to scrutiny?

Also above, you endorse woogie10w? You find nothing wrong with his posting above? Where he turned a simple question asking for confirmation of your opinion into a drama and seriously breaking policy with allegations about my intentions. You find nothing objectionable in that and even go in to endorse it? What standards do you have?--Stor stark7 Speak 23:17, 15 July 2010 (UTC)

@Nick-D Why did you feel the need to comment here on the issue of sources when writing about a book? You never edited the book, so obviously you went there after I provided the name of the book. Why did you go there? Was it in hope of finding something you could turn against me? What exactly was the ongoing conflict that you claim? I hadn't edited the article since 2 edits the 13th, and before than not since March. Yes, I was concerned about the rationale given in the edits succeeding mine, which was why I wanted to know about the policy situation. I provided no leading questions, nothing to bias anyone. All I did was ask a question, and even provide the name of the book I was thinking of. Why did you have to invent a conflict and post here, unless it is that you like creating wiki-drama out of serious questions when it can be used against me? Really poor style IMO.--Stor stark7 Speak 23:17, 15 July 2010 (UTC)
 * Can we please stop the content discussions as they refer to a completely different article. I think this issue has been discussed in all the depth needed and possible here and would suggest to close. Thanks Arnoutf (talk) 07:25, 16 July 2010 (UTC)

Simplification for simplifications sake
Are you saying that, if no one had previously documented that grass was green, or that the sky was blue, one may not post such things, based upon their own observations?

Another example might be a website, which has a token character, let's use my website for example. If someone were to make an article on my website and talk about the three characters that are on said site, dispite never being published elsewhere, that would be considdered original research? Or would the website it's self (seeing as it is the source) be evidence enough? 58.106.136.123 (talk) 05:57, 16 July 2010 (UTC)


 * See WP:Common knowledge about things like the sky is blue (except if there are clouds or it is dark or you are on Mars). Supposing your website were notable. That would mean some reliable source like a newspaper thought it was worth doing an article about the site, and it wasn't just paid advertisement, they actually did think it was notable. Then the article could have a reference to the website, possibly some reference to some corporate information there, plus a summary of the information from the reliable sources and links to any pages that were specially mentioned. What they could not do is the work of a newspaper reporter or whoever looking at your site and writing an article about it. Looking at your site and writing what one sees is original research. Dmcq (talk) 06:22, 16 July 2010 (UTC)


 * Ok thanks for the common knowledge explination. Going back to the website scenario for a moment though. Say a news paper or a magazine etc, did do an official article on the website, and mentions that there are 5 main contributers in the websites group, and says that there are three who live in Australia, one who lives in UK and one who lives in Africa, but only names the 3 members from Australia, and then someone went to the website it's self and found our members page and named the other 2. That would be considdered as original research and not allowed? Even though it was taken from the original source?58.106.136.123 (talk) 06:35, 16 July 2010 (UTC)
 * Yes. Basically if a reliable source doesn't notice something or doesn 't think it worth mentioning then Wikipedia doesn't either. Dmcq (talk) 06:55, 16 July 2010 (UTC)
 * Well thats nonsence... How can you justify something like that, and what benefit would it be to you to have innacurate information? One person may not be mentioned for various reasons aside from the paper finding it irrelevant. In the scenario mentioned above, it would cause the article to be innacurate, seeing as the guy from UK is one of the founders. Second of all what if the article is innacurate or changes have been made since? How is the original source not evidence enough? If the article talks about the website being mostly a single colour, but later the site colour changes, that would mean that Wiki is wrong, and not only that, but it can't be corrected until someone else writes an article about it? Thats craziness.58.106.136.123 (talk) 07:03, 16 July 2010 (UTC)
 * Well one can use WP:Common sense but anything more than a minor amount of that is original research. If you think you can set up a better encyclopaedia otherwise you're free to try and you can always argue to change policy. I don't think you'd get far on this though it is pretty fundamental to Wikipedia. Dmcq (talk) 07:12, 16 July 2010 (UTC)

Actually the referencing guidelines WP:V builds heavily on common sense: but in practice not everything need actually be attributed. This policy requires that anything challenged or likely to be challenged, including all quotations, be attributed to a reliable source I usually interpret this as: If it is a clear and simple fact I am not bothered too much about the quality of the source at first (As it is unlikely to be challenged - so here your names of website contributors could apply). BUT (and these bits are essential), I only do this if I think the facts are fairly trivial (unlikely to be challenged) AND nevertheless as soon as anybody, for whatever reason, even without explanation, doubts whether the source is sufficient to support the fact I (which makes it proven to be challenged) ; I DO NOT protest, but go out to look for a better source or agree with removal of the challenged text. For me personally this works very well, as it limits the effort findig sources that nobody asks for, while it guarantees quality as soon as doubts are raised. This demands the willingness to give up on some texts from time to time though, and this seems to be the problem for many Wikipedians. In general such a pragmatic approach takes away a lot of the discussion on principles that are engaged in outside the context of the edit, and does not make a hypothetical case that useful to explore. Arnoutf (talk) 07:39, 16 July 2010 (UTC)
 * I undertsand the issues that pragmatism can evoke, especially when discussing hypothetical scenarios, but I find annalogies help me to understand things better than a list of rules or guidelines. I just cannot comprehend how original material is NOT good enough evidence, it seems counter productive, and I'm seeking an answer. Yes I did set a trap with my example, because I admit we have a rather strange arrangement, which only seems to be getting stranger, Haha. However I'm trying my best not to go into details about my website and only use it as an example, as it's details are easy to draw from, for me at least, to beable to ask questions in these instances. Now the colour question was a little silly and I thought that one up a little off the cuff, but I think the spirit of the question is understandable, and has been answered, despite bringing up many more questions. Like how can fictional character articles continue to be upkept without refering to original material, only secondary sources. Though to an extent the Common Sense article has sort of answered that, but yet still begs the question of how far that can go? If someone were to write a book and publish it independantly, and features characters from other sources, and the independent material is accepted as canon by the riginal creator, and an article is written about it but doesn't go into the progression fo the character's personality and the finalle of the plot. Is the original material (that being the independently published book) valid then to update the characters biography on here (assuming before hand that the character already has a biography)? This has the possabillity of looking too deeply into the entire ssituation, and I am aware that there is a slippery slope ahead of this conversation. Which means I will plainly come full circle return to the confusion. How can there be any better evidence, than the innitial source material? —Preceding unsigned comment added by 58.106.136.123 (talk) 07:51, 16 July 2010 (UTC)


 * Original sources are used sometimes. However an original source is not normally good at assessing the story about itself and saying the notable bits. A reliable secondary source can assess a website and say the important things. For a film or novel plot wikipedia takes the view that they are written to speak for themselves. They are original sources but if they are notable the story can be read from them though one should be careful to retain a neutral point of view and not go too far, so guidelines about them say things like try not to exceed 700 words so editors just keep to the obvious plot. For a web site the only bits like that one would normally use are as I said above the company details - its 'about us' page, also official statements on notable controversies they have been involved in, however all rather sparingly and attributing anything dodgy in that as their opinion, WP:RS gives the guidelines about sources. By the way a self published book would not count as a reliable source normally. And I don't think there is much problem with accepting that the logos and colors on the front page are meant to be noticed and are as shown. Dmcq (talk) 09:06, 16 July 2010 (UTC)


 * Thanks for the help, I'm glad to have this cleared up :) —Preceding unsigned comment added by 58.106.136.123 (talk) 09:18, 16 July 2010 (UTC)

Calculating things using current laws
Example 1: Let's say someone runs the 100 m sprint in a new world record time. And *I* calculate this person's average speed. Is that considered original research if no sources regarding the average speed are mentioned?

Example 2: Light speed. This is about a law (maybe exhibited by a photon) versus observers. If I use that law to calculate other things, things that nobody else has calculated before, does THAT mean I am using original research?

Given time, I could come up with many more examples...

I think the Wikipedia rules about "original research" should be clearer. 173.168.177.217 (talk) 23:54, 16 July 2010 (UTC)


 * The first is an allowed 'routine calculation', see the section in the policy about this. The second wouldn't be allowed as it isn't something straightforward. Exactly what is unclear to you? I thought the section on routine calculation was pretty clear. Dmcq (talk) 00:07, 17 July 2010 (UTC)


 * I'm not sure what the 2nd example actually means, as the English isn't too clear. In case it's relevant, I'll point out here that the speed of light (in vacuo) is now in effect a definition. Peter jackson (talk) 10:30, 17 July 2010 (UTC)

An important discussion on the RSN
The ongoing discussion demonstrates that many Wikipedians do not understand what "primary" source mean in exact sciences. IMO, it would be useful to add some explanations either to policy or to guidelines on that account.--Paul Siebert (talk) 04:06, 20 July 2010 (UTC)
 * I think the WP:PSTS section is fairly clear. But I agree that it can always be improved. Crum375 (talk) 04:25, 20 July 2010 (UTC)
 * It is not. There is an big inconsistency in the WP policy. It states that "a scientific paper is a primary source about the experiments performed by the authors." However, obviously that historical paper written based on some archival research is a secondary source. Nevertheless, the archival data for the historian is the same as, e.g. a raw X-ray photo of the star for the astronomer. Why the historian's article about Rome, written based on historical documents (primary sources) is considered a secondary source, whereas the astronomer's article written based on the Hubble photos is a primary source? Does it mean that some scholars (historians) produce secondary sources whereas their colleagues working in the area of exact sciences produce just primary sources? In connection to that, if we assume that is correct, then we have to conclude that the, e.g. Uncertainty principle article should be deleted, because it is a pure original research: the article is based on the Heisenberg's fundamental article and other articles about scholars' own work, which is explicitly prohibited by the policy. Obviously, that is nonsense, so some modification of the policy is needed.--Paul Siebert (talk) 04:36, 20 July 2010 (UTC)
 * This page has problems. Was it changed recently?  I don't remember it being so contradictory. - Peregrine Fisher (talk) 04:42, 20 July 2010 (UTC)
 * Since the division onto primary and secondary sources is aimed to avoid original research by the Wikipedians, it should be clearly stated that primary sources are all raw data (historical documents, archival data, memoirs, raw scientific observations, tables curves, space photos, figures from the articles and articles' supplementary materials, etc), whereas all data analysed by a professional (during his own original research), as well as all conclusions, theories, hypotheses proposed based on these raw data by this scholar are secondary sources.--Paul Siebert (talk) 04:49, 20 July 2010 (UTC)
 * No, we shouldn't say that because it's wrong. ElKevbo (talk) 05:09, 21 July 2010 (UTC)
 * No, no, no. Historical work based on archival documents are probably primary sources.  There is no contradiction in the policy.
 * I'm disturbed by the continual dismissal of historical and other non-physical science work as being secondary, as if historians and others who work with documents are incapable of conducting original research. ElKevbo (talk) 05:25, 20 July 2010 (UTC)
 * But the historians themselves call the documents, archival data and memoirs they use "primary sources". Can a primary source be created based on other primary sources?--Paul Siebert (talk) 05:28, 20 July 2010 (UTC)
 * Yes, new primary sources absolutely can be created from other primary sources! :) Historians do that all of the time when they analyze primary sources and introduce their own interpretation. And that's the crux of the matter: there is no simple way of classifying documents are primary or secondary without knowing the context of how they're being used and how they were generated because there is no simple, universal movement from primary -> secondary -> tertiary. And the same document can have parts that are primary and some that are secondary. ElKevbo (talk) 05:35, 20 July 2010 (UTC)
 * Re " are incapable of conducting original research. The NOR policy prohibits original research made by the Wikipedians. By contrast any scientific article (or review) can be published only if it contains something new, i.e. is original. While the Wikipedians cannot do OR by themselves they can and should rely on the original works published by professionals.--Paul Siebert (talk) 05:32, 20 July 2010 (UTC)
 * I don't think anyone is arguing that we should allow Wikipedia editors to conduct original research and I apologize if it seems as if I am making that argument. My objection is that insisting that historians and others can only produce secondary sources denigrates them to mere copyists incapable of original research.
 * Incidentally, plenty of "scientific" articles are published that don't contain much or anything new. It's not terribly common for entire articles to be of this type but it happens and they have their place.  But nearly every article - at least in the social sciences - will have some sections that are purely summative of past research.  This is reflected in the typical social science doctoral dissertation in which the entire second (and most of the first) chapter is devoted to reviewing the literature, despite the intent of the dissertation to be an original contribution to human knowledge. ElKevbo (talk) 05:40, 20 July 2010 (UTC)
 * You seem to mix "original research" and "primary source". Primary source are sometimes not a research at all. The raw data used by a physicist, or archival data used by a historian are definitely primary sources, but not necessarily a result of some research, however, the articles written based on that are not primary primary sources, although they definitely are the original research.--Paul Siebert (talk) 06:00, 20 July 2010 (UTC)
 * Speaking as a student of the technical sciences, papers that merely summarize are published in other fields as well. They're godsends when you're plunging into a new area of study, and only slightly less helpful for professionals trying to stay abreast of their field. --erachima talk 05:46, 20 July 2010 (UTC)
 * And these would be examples of primary sources in the natural sciences. A lot of research/brief communications are the same, they report a finding of interest but do not provide in depth interpretation. In the social sciences the speech of a politician (transcript) is clearly a primary source; a political scientist publishing an interpretation of that speech writes a paper that should be acceptable in Wikipedia. A weblog also is a primary source, but the publication about the discourse in that page by a communication scientist should be allwable.
 * Nevertheless I think there is something in this discussion about the place of scientific papers as reliable sources. I notice that a few years ago many editors wanted to use true primary sources (such as a transcribed speeches and weblogs) while the use of scientific articles was rare. This has changed a lot (for the better) by now, and today we are discussing which types of scientific papers can be used. In my view the current guidelines were not developed to deal with scientific papers (as the editors drafting these guidelines probably could only hope that Wikipedia would reach such a high level of sourcing soon). Perhaps it is time to discuss how to deal with "peer reviewed scientific papers" as a reliable source in some more detail.
 * For what it is worth my view would be that (1) All peer reviewed scientific papers are reliable sources (2) That position papers, perspectives, editorials, etc. where the author sketches his/her vision should be regarded as primary sources (and therefore interpreted with utmost care) (3) That paper reporting plain empirical data (such as research reports, but also original historical documents) should be treated as primary source (5) That meta-analyses and literature reviews should be treated as secondary/tertiary source. (6) That empirical papers, that contain new data but also provide analysis can be used but that some care should be taken as the theoretical interpretation of such data may well be a primary idea (especially in the social sciences). But that is of course only my 10cents. Arnoutf (talk) 07:22, 20 July 2010 (UTC)

Oudent. I'm afraid that I don't really understand the point of this discussion. The current policy regarding this issues seems to me to be both clear and appropriate. A Wikipedia Editor should not publish his or her own analysis or interpretation of what primary sources say, see WP:PRIMARY, in particular the extract quoted below. Are there any proposals to modify that section of the policy, either to make it clearer or to change its substance?--Gautier lebon (talk) 08:40, 20 July 2010 (UTC)
 * Reading that statement it seems to make it clear that intepreted claims, alasysis, etc. from primary sources are themselves secondary sources because of the last part "original analysis of the primary-source material" where it contents that the first analysis is that of a primary source, usually by someone closely involved, and thus while it is a secondary source, it is too close to the primary source to have an unbiased opinion. 陣 内 Jinnai 23:55, 20 July 2010 (UTC)

"Educated person"
Yesterday I couldn't spell educated and now I are it. Suggest reasonable person, though I imagine this has been done over and over. It's POV though, I thik I know a thing or two even though I left school at sixteen. Si Trew (talk) 14:00, 20 July 2010 (UTC)


 * That "thik" well I laughed. Si Trew (talk) 14:01, 20 July 2010 (UTC)

Synthesis
I think the policy on synthesis is too severe. I wrote an article called "Timeline of the future" and it was deleted because people said it violated this policy. (See .) The article tried to put into one single article whatever scientific predictions I came across for what would happen in the future. (A few other people made small contributions, but not as many as I would have liked.) Why should such a thing be prohibited? People argued that we can't do that unless we can find an external source giving a timeline that contains all these things. I think that is going to far. It's almost as though we're saying that Wikipedia editors cannot use their own brains. I do not consider what I did to be original research at all. (There was also a complaint that I didn't give original references for the individual predictions, but that's another problem. It seems to me that people should have helped rather than trying to delete the article!) I frankly don't see what motivates people to enforce a policy like "no synthesis" to such an extent. If it's not a good policy (in the way they took it), then why bother enforcing it? Eric Kvaalen (talk) 12:38, 19 July 2010 (UTC)


 * Since the article's already been deleted, I think you're asking a lot by raising the question now, as we can't see the content. But simply the title is problematic. Naturally it will be verifiable that material concerned with certain future events has been published, and it would be possible to cite those in a way that does not violate policy. But an encyclopedia article giving the "timeline of the future"? PL290 (talk) 13:05, 19 July 2010 (UTC)
 * I think one of the big issue with this type of article is that it can includes just about everything - since there's no secondary source making the synthesis, it's basically up to the editors to include whatever they want to include. This leaves the article opens to massive POV and undue weight issues, especially with things like global warning, economic predictions (which states will collapse, which ones will be the next superpowers, etc.), environmental issues and so on. The article will become a massive battleground since there won't be any single authoritative source to refer to and to help decide what should be included or excluded, and what weight should be given to each prediction. So personally, I would completely support the deletion. Laurent (talk) 14:50, 19 July 2010 (UTC)
 * Agreed. Wikipedia is not a do-it-yourself publishing house, where anonymous editors can create new articles about newly invented topics. What we do here is to summarize what reliable sources have already published about existing topics. So in this particular case, the topic would be "timeline of the future", and you'd have to find reliable secondary sources which have a) come up with this specific term, and b) provided a list of items which directly relate to it. If we ourselves invent the term, that's already original research since we are not allowed to invent new terminologies, and if we ourselves decide what to include or exclude from the content we'd be again violating original research, unless each selected item was supported by a source which specifically included it in "timeline of the future". Again, we don't invent new stuff, or even new combinations of existing stuff &mdash; we only summarize what others have published about a given topic. Crum375 (talk) 16:55, 19 July 2010 (UTC)
 * This is a page for discussing what Wikipedia policy should be, not for simply saying "we are not allowed" to do this or that, or that "we'd be violating" this or that policy. Just to respond to WikiLaurent's fears, the article did mention global warming and sea level rise, and I suspect that those played a role in getting the article deleted, even though it didn't say anything that isn't also said in other Wikipedia articles. There was nothing about economic predictions or predictions about what states will become superpowers. It was mostly about things in the far future, having to do with the earth or the universe. There were no battles, and in fact the article didn't attract much attention from other editors, except those who were bent on getting it deleted. The question I raise is whether (without looking at present Wikipedia policy) such an article should be deleted. I think it was a useful article, and should exist, and I don't like the fact that people argued for its deletion on the basis of this "synthesis" policy, without thinking about whether it's a good policy when applied to deleting an article like the one in question. As though their goal in life is to enforce present Wikipedia policy, without thinking about whether that's a worthy goal. Eric Kvaalen (talk) 09:15, 21 July 2010 (UTC)
 * Yes such an article could be valuable for Wiipedia; but we must avoid several pitfalls of which synthesis is only one. We need to avoid proviing a 'crystal ball' in not making our own prediction. We need to provide reliable sources since only if we can attribute it to reliable mainstream sources can we write this (i.e. no fringe or other undue theories). Additionally we need not only source the incidents but also the entire timeline; or allow for a timeline with contradictions. If we start putting a single timeline together from reliably sourced predictions we could (hypothethically) find a prediction that by 2020 the world will be destroyed by meteor strike. We could also find that in 2025 90% of the world population has internet access. These two predictions are not compatible. If Wikipedia makes a choice, that is original research and should be avoided at all cost. So yes it may be valuable but it will be extremely difficult to put together such an article without going way outside what Wikipedia wants to be. Arnoutf (talk) 09:51, 21 July 2010 (UTC)

Clarifying synthesis: one more try
I didn't get much reaction to my last posting on this topic, see, so I thought that I'd give it one more try. The current policy on synthesis, WP:SYNTHESIS, gives two examples. The first example is a clear case of a syllogism, that is a logical deduction. The second example is presented as if it also is an example of a logical deduction, but I do not think that this is correct. In my opinion, the second example is an instance of including material whose relevance is not established. That is, the offending statement in the second example (a reference to the Harvard Manual) should be excluded because there is no citation suggesting that it is relevant for the issue being discussed, but not because the Editor is making a logical inference, which he or she was not doing. That is, I propose to clarify that Synthesis contains two sub-categories, both prohibited: (1) Editors should not draw inferences or logical conclusions and (2) Editors should not include material whose relevance is not established by secondary sources. Is there any support for this proposal to clarify the section?--Gautier lebon (talk) 08:56, 20 July 2010 (UTC)
 * Actually you did get quite a bit response in that and other related threads, only not what you wanted to hear. Bac to the second example of this page (as you do agree with the first one)

(line 1) Smith claimed that Jones committed plagiarism by copying references from another author's book. Jones responded that it is acceptable scholarly practice to use other people's books to find new references (ref). (line 2) If Jones did not consult the original sources, this would be contrary to the practice recommended in the Harvard Writing with Sources manual, which requires citation of the source actually consulted (ref). (Line 3) The Harvard manual does not call violating this rule "plagiarism". Instead, plagiarism is defined as using a source's information, ideas, words, or structure without citing them.
 * The Harvard manual would be the reference, and as it has a chapter on plagiarism it might be relevant to the topic.
 * Line 1 gives a factual representation of the dispute. Line 2 already goes over the line as it links for the first time the acts of Jones to guidelines in Harvard. This is synthesis. Line 3 is factual, but adding it contributes to the implicit synthesis just as in example 1.
 * Your proposal "(2) Editors should not include material whose relevance is not established by secondary sources." basically eliminates the use of primary and tertiary sources altogether (i.e. you can only use sources that are discussed in secondary sources making the use of any other type of source irrelevant). Arnoutf (talk) 09:52, 20 July 2010 (UTC)


 * Actually, if we are still looking for good examples of WP:SYN, this here, courtesy of the Climate Change arbitration case, is a beauty: . -- JN 466  11:07, 20 July 2010 (UTC)
 * I guess you mean the whole section:

(line 1) Watson has claimed on many occasions that the proof of the greenhouse effect can be observed by looking at Mars, Venus and Earth, claiming in a taped debate sponsored by The Guardian (at 76 minutes) "We only need to look at 3 planets: Mars, Venus and Earth and you can explain why there is such a difference, a frigid Mars planet, no greenhouse gases, Venus is absolutely boiling lots of greenhouse gases and earth is by luck somewhere in the middle"[8] (Line 2) This view is in conflict with our basic understanding of Mars. (Line 3) NASA reports in a 'Mars Fact Sheet', stating that the atmosphere of Mars is approximately 95% carbon dioxide,[9] (Line 4) which is confirmed by standard textbooks on the subject.[10]
 * That is indeed a beautiful example.
 * Line 1 is a primary source reflection of an opinion by Watson (which is clearly a synthesis, but since it is presented as a quote that is not very relevant)
 * Line 2 is clear synthesis as no source is given that state the conflict
 * Line 3 is a fact sheet which is not necessarily relevant to any of the above, but gives implicit support for the synthesis
 * Line 4 is just plain silly, as if text books have any quality that gives them any relevance to confirm or disconfirm official NASA communications (in case of disagreements between NASA and text books I would believe NASA, the wording here however implies the opposite). Arnoutf (talk) 11:36, 20 July 2010 (UTC)


 * Yes the whole climate change business has hordes of edit warriors who are RIGHT and their views must be expressed, I have a few of those articles on my watchlist to try and stop stuff like this. This is exactly the sort of stuff WP:OR, and WP:SYNTH in particular, is supposed to stop, and those editors would exploit anything aboout syllogisms to the hilt like they currently try and exploit routine calculation and common knowledge. Dmcq (talk) 11:53, 20 July 2010 (UTC)

The proposed statements are too broad:
 * 1) "Editors should not draw inferences or logical conclusions." This would exclude elementary logical conclusions where no position is being advanced, such as, if vehicles are forbidden from being driven more than 100 km/h on a particular stretch of road, then Honda Accords are forbidden from being driven faster than 100 km/h on the particular stretch of road.
 * 2) "Editors should not include material whose relevance is not established by secondary sources." Exclusionists could interpret this to mean that the secondary source must give the name of the exact publication from which material is to be included, rather than just establishing a connection between the topic of the article and the topic of the source from which material is to be included. Jc3s5h (talk) 12:56, 20 July 2010 (UTC)
 * Your first example of the Honda Accord is indeed synthesis and can therefore be removed. However, this is a fairly trivial example and in practice almost no-one will object. Why should it be covered as original research nevertheless; well the phrase in context might imply that there are rules specifically tailored to Honda (which is the synthesis here). Arnoutf (talk) 13:01, 20 July 2010 (UTC)


 * I doubt I can agree with Arnoutf because our philosophies are too different. I believe that rules should only prohibit that which should always be prohibited; otherwise it should be phrased as guidance that only applies in certain situations. Arnoutf objects to the Honda Civic example because "the phrase in context might imply that there are rules specifically tailored to Honda". If we prohibit every writing practice that might create a problem, we can make the guideline very short: Wikipedia is prohibited! It created an unacceptable chilling effect to prohibit practices that are often acceptable, and then let them in by ignoring breeches of the rule.


 * The title of the relevant section is Synthesis of published material that advances a position. Unless one is advancing some position, straightforward conclusions are allowed. Jc3s5h (talk) 13:38, 20 July 2010 (UTC)
 * I do not think our philosophies are that different. If your deduction is made just to make the neutral fact that Honda cars are not allowed to violate speed limits there is no problem.
 * However if it were phrased in a context like: "The XXX government has posed protectionist measures against Japanese cars to protect the national car industry. Honda Accords are forbidden from being driven faster than 100 km/h on the particular stretches of road." This would be a serious problem.
 * I believe in the common sense of Wikipedians to leave fair use of simple deduction (although I admit having experience with several editors seem overly fond of Wikilawyering on rules for rule sake alone), but we need some simple rule to stop POV pushers who string together a lot of simple deductions to build their own case (and I encounter these editor even more frequently, and this latter category tends to be less open to discussion than the former). So although I agree the rule might be overly strict for some edits, creating exceptions would, in my view, yield too much to POV pushing editors. But I agree this is a bit of a grey area. Arnoutf (talk) 13:51, 20 July 2010 (UTC)
 * I agree with Jc3s5h that this synthesis policy is being taken too far, as though editors should not be allowed to think. See my discussion above at . Eric Kvaalen (talk) 09:15, 21 July 2010 (UTC)

Outdent. I got many comments to my first posting on this topic, few to my penultimate, which is why I opened the topic again. Arnoulf says:--Gautier lebon (talk) 11:04, 21 July 2010 (UTC)

Arnoulf is correct in that I should have proposed "whose relevance is not established by secondary or tertiary sources". Arnoulf appears to state that it is OK to use primary sources which are not discussed in secondary or tertiary sources. But I've seen many interpretations of the current synthesis policy (both in specific cases and on this discussion page) that, in effect, prohibit the use of primary sources that are not cited in secondary or tertiary sources. So I believe that my formulation (when corrected by adding "tertiary") clearly articulates the current policy. If people don't agree with that policy, then the policy needs to be changed, not merely clarified.--Gautier lebon (talk) 11:04, 21 July 2010 (UTC)

The other comments above seem to me to indicate that indeed the policy is not clear, because people interpret it in different ways. At this stage, I'm not proposing to change the policy, I am proposing to rewrite it so that everybody understands clearly what it implies.--Gautier lebon (talk) 11:04, 21 July 2010 (UTC)
 * We can, and should not forbid use of primary sources altogether. For example if I am editing an article on a person, and I can access the civil registry to report date of birth and date of death, I am pretty sure nobody would object. However, civil registry documents are primary sources. Additionally, we can no longer use the score sheets of sports organisations (primary sources) to report outcomes of races. The current policies allow use of primary sources for usch simple facts without any interpretation (but not much else) and I think that is a fair postion on the use of such source. See
 * The problem in my view becomes apparent when editors use secondary sources as primary sources (e.g. stating something about the report rather than the topic) or when it is not clear whether a source is secondary or primary (as is the case with scientific papers). However, changing our policy on use of primary sources will not help, as the problem seems to be that people say something is a secondary source, while they use it as a primary source. If anything we should make the distinction between primary (use of) sources and secondary sources more clear. Arnoutf (talk) 11:18, 21 July 2010 (UTC)
 * Primary / secondary is not an inherent property of a source, it is in the context of it's use. For example, a written analysis is a secondary source regarding it's subject, and a primary source regarding itself.
 * I think that WP assigns too much weight to primary vs. secondary when discussing the suitability of sources. North8000 (talk) 12:56, 21 July 2010 (UTC)
 * I agree, perhaps we should discuss usage of sources rather than types of sources. E.g. Sources for: Undisputed facts (all RS will do), Source for mainstream opinion (some kind of overview source is needed), etcArnoutf (talk) 13:20, 21 July 2010 (UTC)
 * Yes, I agree with Arnoutf. I withdraw my proposal (2) above, and propose instead something along the following lines: "(2) Editors should not include material which is not relevant; relevance can be established by secondary or tertiary sources.  An undisputed fact (which may be attributed to any source, including a primary source) is not necessarily relevant: care should be taken not to introduce facts that imply an inference or conclusion that is not supported by secondary or tertiary sources".--Gautier lebon (talk) 11:27, 22 July 2010 (UTC)


 * In my opinion, "Editors should not include material which is not relevant" would be a backward step from the existing wording, which seems to me to make exactly the right point, clearly and concisely: "Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources." Also, I disagree with your opening suggestion that The first example is a clear case of a syllogism, that is a logical deduction. The first example is,
 * There is no deduction presented, nor is any conclusion stated. Two statements are simply strung together, linked by the word "but", to imply a conclusion not explicitly stated by either source. Sorry, perhaps I am missing something, but to me the existing section does its job well enough, and I don't think it would be improved by anything proposed here.PL290 (talk) 12:34, 22 July 2010 (UTC)
 * Nope you are completely right. We stopped discussing WP:SYNTH and wandered off to discuss WP:PRIMARY another policy. I would not want to change the synthesis policy in this regard. Arnoutf (talk) 13:10, 22 July 2010 (UTC)
 * Nope you are completely right. We stopped discussing WP:SYNTH and wandered off to discuss WP:PRIMARY another policy. I would not want to change the synthesis policy in this regard. Arnoutf (talk) 13:10, 22 July 2010 (UTC)

Oudent. OK, I yield to the majority.--Gautier lebon (talk) 15:45, 22 July 2010 (UTC)

"are supported by"
A primary source may only be used on Wikipedia to make straightforward, descriptive statements that any educated person, with access to the source but without specialist knowledge, will be able to verify are supported by the source I propose changing "are supported by the source" to "corresponds to the source". The reason for this change is that primary source content should be integrated into Wikipedia text (i.e. correspondence) -- not that primary source content is to be used for arguing a disputed point in Wikipedia (i.e. support). patsw (talk) 00:28, 25 July 2010 (UTC)


 * What problems have arisen from the current wording?  RJC  TalkContribs 01:08, 25 July 2010 (UTC)
 * I have just indicated that the current wording lacks clarity. Which of the two reflects the spirit of NOR the best, and with greater precision? patsw (talk) 02:37, 25 July 2010 (UTC)
 * "Supported by" seems to me to be clearer, and closer to the meaning of WP:V and WP:NOR. The "correspondence" wording sounds vague and confusing to me. Crum375 (talk) 02:43, 25 July 2010 (UTC)
 * I agree with Crum. We are allowed to use primary sources to support statements after all (even a disputed point) ... we just have to be careful not to do so in a way that constitutes OR.  That is not always easy, but its certainly possible and allowed. Blueboar (talk) 03:15, 25 July 2010 (UTC)
 * Leave as is, in my opinion. It will get better results, regardless of what it can mean if you think about it too long. - Peregrine Fisher (talk) 03:53, 25 July 2010 (UTC)

What constrains the use of primary source content under WP:NOR?
 * The primary source content is equivalent or similar in character, quantity, quality, origin, structure, function, etc. to the Wikipedia text?
 * The primary source content provides materials, effort, information or endorsement for a specific activity, ideal, or entity in the Wikipedia text? patsw (talk) 04:05, 25 July 2010 (UTC)


 * In general, if it ain't broke, don't fix it. Language lacks the rigidity and specificity that would enable one to communicate perfectly with another person, except where two people have been trained in the use of an artificial language (such as two lawyers in contract law).  So, making it perfectly clear to you will introduce ambiguity to another person (some editors have already testified that this would be the effect on them).  Relying on dictionary definitions cannot get around this problem:  even if you were to prove that their understanding was "wrong," it would still be their understanding.  If people aren't currently confused in actual disputes, further attempts at "clarity" can only either have no effect or be counterproductive.  RJC  TalkContribs 06:40, 25 July 2010 (UTC)


 * I'm all for keeping the 'are supported by the source'. If the reader 'verified' it based on the source it might still be original research. What we want is that the article is reasonably faithful in saying what the source says. Dmcq (talk) 09:10, 25 July 2010 (UTC)
 * The sense of "reasonably faithful" is what corresponds to means. Supports is a conceptually different relationship.patsw (talk) 20:03, 25 July 2010 (UTC)


 * patsw asked: What constrains the use of primary source content under NOR?... And the answer is nothing, as long as you are not using it to support OR.  The policy states: Primary sources that have been reliably published may be used in Wikipedia. (bolding mine)...  This is intentionally worded as a caution, not a limitation.  OR is OR, no matter the source type.  However, because it is very easy to (intentionally or unintentionally) form OR when relying on primary sources, we think it worth a special caution. Blueboar (talk) 14:23, 25 July 2010 (UTC)


 * Patsw states that primary sources should not be used to argue disputed points, but this is not the case. For example, if a review of a book (a secondary source) purported to quote a novel (a primary source) but the quote was incorrect, a citation to the appropriate page in the novel could be provided on the talk page to show the quote was wrong, and the entire review, or that aspect of the review could be omitted from the article. As stated by Blueboar, primary sources must be used with care. For example, the article should not contain any speculation about the motiviation of the reviewer in misquoting the novel unless the speculation was contained in a reliable source. Indeed, any mention of the misquote would probably be undue weight. Jc3s5h (talk) 15:11, 25 July 2010 (UTC)
 * Those examples have utterly nothing to do with making a distinction between supports and corresponds to. patsw (talk) 20:03, 25 July 2010 (UTC)

The arguments for "supports" seem to strip any meaning from WP:NOR with respect to primary sources. Can a Wikipedia editor enter a new, personally authored (i.e. original) argument for a disputed point into an article, which was not actually in the primary source, but merely has found "support" for it in the primary source? Support has a vague meaning: A fact plucked from a primary source might be used to support two or more different disputed points opposed to each other. patsw (talk) 20:03, 25 July 2010 (UTC)


 * Please try and explain your point simply. An example might help. Dmcq (talk) 20:26, 25 July 2010 (UTC)


 * 1) The primary sources are the text of the American Recovery and Reinvestment Act of 2009 and the press release of the Bureau of Labor Statistics
 * 2) The Wikipedia editor's addition is "Obama's $862 billion economic stimulus package, which passed last year, has saved or created 3 million jobs"
 * 3) Those primary sources support the claim. patsw (talk) 23:25, 25 July 2010 (UTC)
 * A primary source must be used carefully, and the Wikipedia material which relies on it may only make descriptive statements about the source. So in this case, if an editor wants to add "Obama's $862 billion economic stimulus package...has saved or created 3 million jobs", he must show where in the primary source it says that. It cannot be something that is only indirectly derived or inferred from the source, as that would not be a "descriptive" statement. In the sources you provide above, I see nothing like the claim about the 3 million jobs, or how they are attributed to a stimulus package. This is where a secondary source is needed, since analysis and interpretation are required. Again, a primary source may only be used descriptively, in other words, it may not be interpreted or analyzed by Wikipedians. Crum375 (talk) 23:36, 25 July 2010 (UTC)
 * I don't think that anyone could in good faith think that they could support the statement that Obama's stimulus package saved or created 3 million jobs with those reports because of the wording in NOR, especially given what is said about syntheses. Changing the wording to deal with tendentious editors is bound to fail, given that they don't really care what the policy says.  RJC  TalkContribs 00:53, 26 July 2010 (UTC)

I don't think there's a need to pursue this further, in my opinion. Everyone seems to think that support from primary sources is to be interpreted in a narrow way. The broad meaning of support would trigger WP:SYN as my example above shows. I disagree with RJC's pessimism regarding tendentious editors. patsw (talk) 01:14, 26 July 2010 (UTC)
 * Your logic is faulty, as you seem to assume that WP:SYN is an option, and if some "support" can be shown in the source, it overrides everything else. But that's plain wrong. We may not violate SYN, for either primary or secondary sources, and primary sources must be used descriptively only, with no interpretation or analysis. That the source must support the material in question is in addition to these requirements. Crum375 (talk) 01:32, 26 July 2010 (UTC)
 * Your understanding of what I just wrote is faulty. I understand how WP:SYN is applied.  We agree. patsw (talk) 02:16, 26 July 2010 (UTC)
 * If you understand that SYN is not optional, and that primary sources must be used descriptively only, then your example does not show the policy needs any changes. Crum375 (talk) 02:34, 26 July 2010 (UTC)

Request for comment un Crum375's pre-proposal
As you may recall, in June Crum275 drafted a pre-proposal which was essentially to merge wp:VER and WP:NOR and essentially make no changes during the merge process. I think that it's a good idea due to the large amount of overlap between the two in both content and in the principles / policies covered. And he asked for input at this stage, before it became even a proposal.

The proposal and its comment/talk page is at: http://en.wikipedia.org/w/index.php?title=User_talk:Crum375/att&oldid=377560470

It also has a rationale page at http://en.wikipedia.org/wiki/User:Crum375/att/rationale

I noticed that it's getting a little quiet over there and thought it might be good to suggest request folks weigh in over there, including whether or not they think that it's a good idea. This is "pre-proposal", so if it went another step that would be to becoming a proposal. North8000 (talk) 20:32, 7 August 2010 (UTC)

Original research and standards
Question: is a Wikipedia article about some subject X limited to what the standards say about it, or are some explanations, logical consequences etc. that provide to the readers (layperson as well as professional) clearer answers than the standards also acceptable WP material?

Primary reason for this question is that some editors consider consistent clarifications of the standards, even when based on material published by reliable sources (peer reviewed), as OR as long as they have not become standard or very widespread. (By their very nature, explanations and clarifications are not likely to become standard, nor systematically copied or otherwise distributed).

Secondary reason is that what the standards say about topic X is by far not the only relevant thing that can be said about X. Should a good encyclopedia article not contain some hermeneutics as well? Boute (talk) 07:27, 5 August 2010 (UTC)


 * Many standards do come with an additional explanation document and if they are complicated other people write things explaining them as well so those could be used as sources. I wouldn't consider it original research to give some straightforward clarification of something which is obvious and free of possible misunderstanding to someone who knows about the area. However one would have to be very careful about not reading more into something than is there. I know of one standard which was read differently by different people without them even realizing and that led to trouble later, the later version tightened it up but that left a number of systems now incompatible. Dmcq (talk) 08:29, 5 August 2010 (UTC)

Nice to meet again (codomain, last June). The standards situation I am talking about is much simpler than the one you mentioned, not likely to cause catastrophes. It concerns the decibel page. The problem is that the definition given there unnecessarily involves power ratios (which is just one of many applications, hence not proper in a definition) and never really tells what dB itself is or means. So it is unsatisfactory for both the interested layperson and the professional. Yet, it can be directly derived from the definition that x dB = 10^{x/10}, and the long-sanding engineering practice in specification sheets writing, for instance, Gain = 30 dB for Gain = 1000, leads to the same conclusion. This pure and simple clarification has also been accepted in a peer-reviewed journal (IEEE Antennas and Propagation Magazine, Dec. 2009). Yet all insertions about this simple equality in the article are reverted by other editors on the grounds that it is a "nonstandard interpretation". Therefore my question was whether Wikipedia articles must adhere only to what is explicit in some standard, or try to be clear for the reader. Boute (talk) 10:07, 5 August 2010 (UTC)


 * While one can make basic, mathematical extrapolations from the source for clarity in the text (e.g., if the source gives the volume in hogsheads but the rest of the article already uses liters), the fact that people are objecting to the change as a "nonstandard interpretation" suggests that it is not as obvious to them as the policy requires. You say, however, that this clarification has been accepted in a peer-reviewed journal.  This means you should be able to cite it with a source, no?  Are they objecting that this "clarification" is not accepted by the broader scientific community?  That is, is this an WP:OR or WP:RS and WP:UNDUE issue?  RJC  TalkContribs 14:46, 5 August 2010 (UTC)


 * Material from reliable sources can be considered for inclusion in an article, but views that have only a small number of adherents need not be mentioned due to WP:UNDUE, even if the view has been mentioned in a small number of peer-reviewed articles. Jc3s5h (talk) 16:27, 5 August 2010 (UTC)
 * (ec) Views of a "tiny/fringe minority" need not be mentioned per WP:NPOV, but a "small minority" is still acceptable, so long that its views have been published by a reliable source in relation to the article topic. One way of objectively determining whether a minority view should be included is to rely on reputable mainstream reliable sources which review the entire field from the outside (as we should be doing), such as major newspapers, like The New York Times, or major encyclopedias like Encyclopedia Britannica. If these mainstream publications mention the minority views when describing the article topic, then we may also. Crum375 (talk) 16:39, 5 August 2010 (UTC)


 * Yeah, I just read Talk:Decibel. This isn't an OR issue but an UNDUE one.  Boute is promoting a redefinition of decibel that hasn't caught on and is trying to put it forward on Wikipedia against the concerns of six editors.  Forum shopping is generally discouraged, especially when the question is posed in a misleading manner.   RJC  TalkContribs 16:33, 5 August 2010 (UTC)

The idea that I am promoting a redefinition of dB or a different view (or anything else to which the notion of "catching on" is applicable) is only an impression that editors opposing the extra explanation are creating to justify removal. The only thing I am trying to do is giving a clearer explanation of the existing definitions (and the existing view), one that also better reflects actual practice (from which one infers that things can be as simple as 30 dB = 1000). The operational definition of the decibel has never caused problems since the dB was introduced nearly a century ago, so a redefinition would be ridiculous. On the other hand, the fact that no two authors explain dB in the same way and the tinkering with the definition in the article (see history) shows that a better conceptual explanation is needed. Boute (talk) 18:48, 5 August 2010 (UTC)

Afterthought: some of the editors explicitly agreed that the explanation in the article is at various points obfuscating and evasive, but maintain that this is intentional because they only want to reflect the actual situation in the field. My concern is that, if better explanations are known, they should be given to the readers. Boute (talk) 19:00, 5 August 2010 (UTC)


 * This is a content dispute and should be settled on the Decibel talk page. It is established that the views expressed in Boute's peer reviewed article do not meet the Wikipedia definition of original research (although they might not merit mention in the article) so this is no longer a matter for this talk page. Jc3s5h (talk) 19:08, 5 August 2010 (UTC)

I fully agree, the content discussion has gone sufficiently far here. Remains my policy question at the start: is it policy that, on a topic about which there happens to be a standard, all presentations in an article are restricted to what the standard says explicitly, and that further explanation/clarification via logical consequences is not suitable WP material? (For topics without standards, no such restriction would apply). Boute (talk) 19:46, 5 August 2010 (UTC)


 * You may not make things clearer than reliable sources make them, even if you rely only upon reliable sources to do so, per WP:SYNTH. State standards as the standards are stated.   RJC  TalkContribs 19:59, 5 August 2010 (UTC)


 * That confirms the impression I got, but it turns Wikipedia from an encyclopedia into a standards mirror site. It seems then that the article should not be called "decibel" but "the decibel standard". Also, if someone decides to "standardize" the sin function, no theorems may appear on Wikipedia unless they are incorporated into the standard.  Incidentally, the definition in the current article does not cite any standard accurately, so it is also an explanation (although better than the standards I downloaded - showing that the need for "deviation" is recognized), just like mine.  The difference in treatment between the current explanation and the (very small) addition to it is not justified. Boute (talk) 05:17, 6 August 2010 (UTC)


 * The IEEE article in question is a primary source; to show notability and relevance, esp. when it relates to a standard, you'd need a secondary source discussing it in relation to that standard. So if you had another article in a reliable publication which mentioned the IEEE article and the new concept when discussing the main topic, then it would be includable. Otherwise, it would be a footnote, at best. Crum375 (talk) 20:13, 5 August 2010 (UTC)


 * If it were about a "new concept", I would agree. However, it is only a logical deduction.  Meanwhile, I reduced the proposed explanation which originally contained many illustrations into a single sentence (the figurative "footnote", as you say), but even this was reverted as "nonstandard", whereas it only clarifies the standard.  Boute (talk) 05:17, 6 August 2010 (UTC)


 * Summarizing: is Wikipedia an encyclopedia or a "standards mirror site"? Also: what are the advantages of a digital online version over a printed one if it has to lag years behind? (of course, a well-understood "reliable sources" principle is essential) Boute (talk) 05:17, 6 August 2010 (UTC)


 * I just glanced through the discussion, and my gut feel is that if you just inserted it as a sentence with a source, you probably wouldn't have had a problem. Looks to me like a straightforward equation, at lease after you distilled it.  Instead your description of it makes it sound like a classic OR violation.  Like "the mainstream world has blown it, and here's a great new idea that will fix it"  North8000 (talk) 20:52, 7 August 2010 (UTC)


 * You are right. To get something published in a scientific journal, you must prove originality; to get the same results into Wikipedia, even after it has been published, you have to prove the opposite to satisfy some editors.   The problem is that in mathematics a correct proof is all that is really needed for verifiability; applying criteria such as secondary sources etc., which are clearly meant for "events" with possible "eyewitnesses", is overdoing it.  Anyway, meanwhile I discovered that the definition of dB by some standard bodies (NIST) admit an infinite number of correct interpretations that differ mainly in the potential for other uses. I also decided to let all this rest for a while and propose later the simplest one-line derivation possible. Boute (talk) 14:32, 8 August 2010 (UTC)
 * If there is a source, there should not be a problem... we allow our sources to conduct original research... we do not allow our editors to conduct original research in Wikipedia. Once something has been published outside Wikipedia, it is no longer original to Wikipedia or a Wikipedia editor.  That said... you do have to be careful not to go beyond the source.   Blueboar (talk) 16:42, 16 August 2010 (UTC)

"For example, a review article that analyzes research papers in a field is a secondary source for the research."
This, along with "a scientific paper is a primary source about the experiments performed by the authors. Historical documents such as diaries are primary sources." creates a gap between primary and secondary sources where most research articles fall. A review article is a separate class of scientific publication which overviews a wide range of artilces written by others, often dealing with some broad subject and devoting only a brief attention to every particular publication. As a rule, such articles are explicitly marked "review" by journals' editorial boards to distinguish them from other articles. However, most other scientific papers also contain a review of the works of the others (usually in the "Introduction" section). In addition, the "Discussion" section also contains a comparative analysis of a scholar's own results with the results of others. In other words, only "Experimental part" and "Results" sections can be considered as a primary source sensu stricto. With regard to history and similar articles, most of them, not only review articles, contain an analysis of vast amount of primary sources. In connection to that, some additional explanations are needed. I propose to replace the text:
 * "For example, a review article that analyzes research papers in a field is a secondary source for the research."

with:
 * "For example, an review article that analyzes research papers in a field is a secondary source for the research. Historical articles that analyse archival documents, diaries, or memoirs are also secondary sources."

--Paul Siebert (talk) 17:07, 11 August 2010 (UTC)

Are "List of" articles exempt?
I think I know the answer... but I need a consensus I can point to. Does WP:NOR (and specifically WP:SYN) apply to list articles? Blueboar (talk) 15:27, 16 August 2010 (UTC)
 * I know you asked a similar (but different enough) question at WP:N, but for purposes of NOR, I think the answer (based on AFD results) is that list content is not exempt from NOR, but the list definition does allow foran amount of synthesis that Wikipedia editors have to assemble topics and articles to best summarize the sources; there's less of a problem if the list definition is sourced to avoid NOR, but we're talking about copyediting distinction of synthesis that's allowed for lists. There's still question if the resulting list is indiscriminate or not (see my response to your question at WP:N), and of course there is means of assembling a definition that violates other policy (such as POV-directed lists). --M ASEM  (t) 15:43, 16 August 2010 (UTC)


 * Actually, the question here and at WP:N are not related (or at least were not intended to be related)... I am in discussions at two separate list articles both of which have policy related issues to resolve.
 * The specific issue that led to the question here relates to List of wars between democracies and whether we should allow use of source A (to cite that nation A is a democracy) and source B (to cite that nation B is a democracy) and reach the conclusion that a war that took place was a "war between democracies". The argument for allowing this is that we are simply adding (one democracy + one democracy = two democracies)... the argument against is that the two sources may be using different definitions of the term "democracy", and thus it is synthetic to add them. Blueboar (talk) 16:12, 16 August 2010 (UTC)
 * I would consider a Wikipedia editor deciding on his own that a war between country A and B was a "war between democracies" as WP:SYN/WP:OR violation, unless there is a source making that direct point for us. This is even if there are separate sources calling A and B "democracies", and a source telling us they are/were at war with each other. I say this because it seems to me that it is (somewhat) controversial (i.e. "advancing a position") to show that there are wars among democracies, and of course the very concept of "democracy" is not well defined (and sometimes even "war" isn't). So as bottom line, get a source making the direct point for you, don't build it up yourself piecemeal. Crum375 (talk) 16:48, 16 August 2010 (UTC)
 * Ok, in this specific case, I would argue this: "war between democracies" is likely going to need a stand alone source to affirm that it is a non-novel term for WP due to POV issues, and that the specific case you describe - where a known war between A and B but that does not call it out as a war between democracies, and only by the addition of two separate sources asserting A and B as democracies - is unallowable synthesis. Here, the phrase "war between democracies" is potentially a loaded POV term (why does the nature of the government structure come into it? Is this a war over ideals of the democratic way? (hypothetical questions)), so the term needs to be well-defined; any war that is included should specifically be called as such; without that term of phrase, you're inferred that a war that happens to be between two democracies is a "war between democracies".  --M ASEM  (t) 16:49, 16 August 2010 (UTC)
 * If it sourced that they are both democracies (at the time of the war) and that they had a war, IMHO it is simple summarization to include that war on this list. I think that such is the least of this article's problems compared to the problems mentioned by others above.  I can't see are reason for such a title/subject except to make a point.  How bout "A list of families bankrupted by taxation" Is that an article or a soapbox for a viewpoint?  North8000 (talk) 19:10, 16 August 2010 (UTC)
 * I agree that if any two (or more) democratically-organized governments go to war, then it's a war between democracies. Editors are not required to turn off their brains, and they are not required to assume that "war between democracies" is a term of the art rather than a plain English description of facts that are simple enough that a schoolchild should be able to verify them.
 * Having argued in favor of editors being able to use their best judgment to determine whether a given entry is "verifiably a member of the listed group", I'd like to see such a list WP:LEAD with a non-synthesis, well-sourced explanation of why people care about this subject. The risk factors for war [e.g., democratically elected governments vs other forms of government] certainly have been written about, and the page should provide that context.  WhatamIdoing (talk) 19:25, 16 August 2010 (UTC)
 * See, I think here its the list definition that's a problem - it can be read as a correlation without causation considering how it is presented. If we're looking at a list that counterpoints Democratic peace theory (as War between democracies redirects there), that article states that "democracies...rarely go to war with one another".  This is not to say that a list of wars between democracies to provide counterpoint for that (which there is a list for) is necessary wrong as long as secondary sources are asserting the conflict as such, making it a valid point to put under "rarely", but simply to say country A and B are fighting and thus must be a counterpoint to Democratic peace theory is a statement that *could* be loaded with POV.  Now, on the other hand, if it is more a curiosity that the list of war of democracies is just a list of such conflicts regardless if they support the Democratic peace theory, then I would say that the combination of sources that assert the war exists, and that A and B are democracies, seems reason but possibly bordering on indiscriminate and would have to be carefully monitored.
 * This is not really a problem in the "families backrupted by taxation" because I cannot see how this requires a synthesis of sources - either an article on a bankrupted family is going to say whether they were bankrupted by taxation or not. --M ASEM (t) 19:33, 16 August 2010 (UTC)
 * The problem with lists is that they don't allow for the full discussion and balancing of viewpoints that an actual article can in order to keep the entries NPOV. Someone above says that if two countries are democracies, and they go to war, then, by definition, it's a war between democracies.  But none of those are really clear terms.  For a blurry example, just look to the most recent Iraq War.  First, who exactly is going to war, there?  Is it the U.S., U.K., etc. vs. Iraq?  Or is it an international coalition against Iraq (in which case, it doesn't seem to apply, as the international coalition itself is not a "democracy).  For that matter, what about Iraq?  Technically, they had elections, albeit sham elections.  Which source gets to make that call?  And what if sources disagree?  In the future, what would happen if a country like Thailand, which vacillates between being a democracy, a monarchy, a military dictatorship, and anarchy were to go to war?  Now, if a source explicitly stated that the Iraq war was a war between democracies, then it should be on the list; otherwise, it starts to look like WP:SYN to me.  Qwyrxian (talk) 22:05, 16 August 2010 (UTC)
 * To me list articles seem to be a way to define a topic which isn't really a topic, and could itself be OR, and in any way that you want so as to pre-determine what the content will be.  I don't see any reason to loosen the rules for list articles compared to regular articles. North8000 (talk) 02:52, 17 August 2010 (UTC)

I agree with Qwyrxian's points. But there are other problems. Most editors would agree that "list of unemployment totals by country", is deceptive to the point of being useless, unless a standard definition of is used because different countries measure unemployment in different ways (often to help the government in power win elections). So no one thinks it wrong to insist that something like the OECD is used and not national definitions for a composite list. However editors who fully understand the implications of that, seem not to understand that with definition of things like terrorism, genocide etc, have exactly the same problems.

A list of terrorist incidents by year for the United Kingdom using the UK's definition of terrorism probably provides some information that is justifiable (is is not a syn, and providing the list is explicit in the lead by explaining the UK government's definition and data is being used and that UK government has a biased definition (eg Bloody Sunday is not included because the UK government's definition does not include its own violence whether or not others think that a terrorist incident)), it that it allows the reader to judge year by year how well the British Government is doing in one of its prime responsibilities (as with any other UK government set of statistics such as GDP etc).

But as soon as editors start to collate incidents from around the world they run into multiple issues with a fruit salad of definitions and such a list is a synthesis because the data collated is not comparing like with like, but is being presented as if it were.

It seems to me that even if one person says that such and such is a war between democracies, if two such wars are collated from different sources, unless the sources explicitly state the definition they are using, for democracy, and for war, and those definitions are presented in the text of the article, then it will be a synful list.

We all know that in practice the way these lists are put together normally involves a pattern match in Google and if it has a pattern match is found in a reliable source, then it gets included with the argument "citation given" even if there is not explanation of the author's definition (back to the unemployment example of why this is a bad idea).

I argued long and hard about this very issue with the term "[aerial] terror bombing" -- which when raids called terror bombing were listed -- was a POV subset of all possible bombing sorties. It depend on whether a Wikipedia editor had found a pattern match of someone somewhere using the term to describe a sortie as a terror bombing raid (not such a problem if all the sources had been using the same definition, but a problem because the sources usually used it in a biased way present the author's political agenda).

The thing was, that no one editor was pushing an explicit POV, each added what they thought to be a good example, but the example was not an examples that came with the author's definition of what the author thought terror bombing to be, and so over time the list became a list of "events described as terror bombing" not a "list of events that helped readers to understand what terror bombing is". So to present the list that was generated it would have to have been necessary to called it "events described as terror bombing" rather than "terror bombing" (as we were force to do with the list List of events named massacres)

It is high time that an essay on this issue was written, but I am not at all sure how it could be presented succinctly as do and don'ts in a guideline. -- PBS (talk) 05:51, 17 August 2010 (UTC)


 * That's why stand-alone lists are still supposed to have a properly written WP:LEAD.
 * First, you define your list. Then you populate it. If you've done your job, no reader will get to the first entry without knowing exactly what criteria you used to choose items for the list -- and no editor will wonder later whether this or that entry belongs to the set. WhatamIdoing (talk) 05:55, 17 August 2010 (UTC)
 * But that doesn't address the problem that giving the list a definition will not fix the problem of slippage of terms between sources. For example, I've recently been watching List of largest empires.  These empires span a vast amount of time and space (all of recorded human history and the whole world), and, of course, relies on multiple sources.  The problem is that I guarantee that these sources do not agree on what the definition of an empire is, how size should be measured, and how the high point of size should be determined.  That is, Research A may have studied European empires, while Research B studied Asian empires, and now the article puts the two side by side and ranks empires in terms of size.  But it's quite likely that since the two researchers use different research paradigms (maybe one is an anthropologist while the other is an economist), such that if Researcher A expanded her work to look at Asia, she'd end up with a radically different set of results from B, thus meaning the two lists fundamentally can't be combined without it being synthesis.  Qwyrxian (talk) 06:19, 17 August 2010 (UTC)
 * But back to what Philip says above, I think that, even better than an essay would be to revise WP:LIST. In particular, the section WP:Source List needs to include more information about WP:OR. I also wouldn't hurt to re-emphasize that lists are equally subject to WP:V and WP:WEIGHT--another problem I run into is lists populated with unreferenced items (I see this alot in lists of people, like supposedly "notable" alumni or "notable" residents), and would be great if the policy explicitly spelled out that that wasn't acceptable either.  So, do you think we should take this over to that talk page?  Or keep it here, since this must be a more highly watched page?Qwyrxian (talk) 06:19, 17 August 2010 (UTC)
 * I think this helps identify the problem. Sometimes our lists are using definitions that are strictly factual with no implied POV or bias.  Let's say we had a list "List of countries with more than $1 Billion in exports" (ignore the fact if it's a good list or not). For one case in this list, we might not find one source that affirms the total, but two equally reliable sources from the same year that comfirm, individually, that two different exports from a country are $750M that same year. Clearly, WP would allow save synthesis for A+B to assert by both sources that that country belongs on that list.  I know that's a bad example, but it's a base one.
 * The counter to this is any list topic that can be read with implied bias or POV - this doesn't mean they are wrong, just we have a different set of rules here. "List of largest empires" is such a case because the "largest" term is POV - this doesn't make it a bad list, just a lot more precautions taken.  Here, synthesis has to be avoided to make an entry fit except in very rare, limited cases, because otherwise it is possible to twist sources enough to make an entry fit that wouldn't otherwise matter. We also have to be more open to alternative definitions unless there's good reason not to.  As Qwyrxian uses, if one source considers only Asian empires, and another only European ones, and the list is generic for any "large empire", then we should include both even if their definitions are far different, making sure to qualify that these are "in Asia" or "in Europe" as given.  A similar example that I'm more familiar with is List of commercial failures in video gaming, where there is no good established definition of "commercial failure" in this field, but instead we allow any reliable source that asserts something as a "commercial failure" to be included - we avoid our own bias and let the secondary sources speak for it themselves.  But regardless, we have to avoid synthesis - if we have to pull multiple sources with some synthesis to establish inclusion for these type of lists, we're probably doing something wrong. For example in the largest empire case - if one source claims "Fakeystan is an empire", and another source stated "Fakeystan is larger than the Roman Empire", that does not make Fakeystan a large empire.
 * So back to List of wars between democracies, it would suggest initially it is a factual list, but the fact of the matter is that the term "war between democracies" is specifically a part of the larger topic of the Democratic peace theory, with the list of wars being the counter to that point. It is a loaded POV term, not as strong as others, but it still can be read with a bias.  I do believe that if it was merely a list without any other context of wars between democratic countries as a strictly factual list, synthesis (3 sources:  A and B are at war; A is a democracy; B is a democracy --> therefore...) would be ok as in the first example I give.  But because of the POV nature of what the list supports, we have to list only those wars that are clearly established by one or more reliable sources without synthesis that they are wars between democracies.  This example is very counter intuitive to most other lists so it is a trickier beast to consider. --M ASEM  (t) 13:34, 17 August 2010 (UTC)

Fresh eyes would be appreciated
Yet another science-related sourcing guideline is being proposed at Identifying reliable sources (science-related articles). Fresh eyes would be appreciated to make sure it does not turn into a sourcing fork of this policy, or an SPOV fork of NPOV. Already my efforts to ensure newspapers are not ruled out as reliable sources have been reverted. SlimVirgin talk| contribs 17:25, 24 August 2010 (UTC)


 * Wonderful, just wonderful... so we now have a precedent for every topic area to carve out its own sourcing guideline. This is not a good idea. Blueboar (talk) 17:37, 24 August 2010 (UTC)


 * This is what has been happening in every area of the project, people insisting that only "their" specialists are the reliable sources. It's leading to routine NPOV violations, because specialists will tend to have somewhat narrow concerns, and will not question their topic or their approach as a whole. Unfortunately each group of editors doing this seems not to care about the effect it has on other areas of the project, which is the essence of good policy writing&mdash;being willing to see how the policies affect areas you care about, and areas you don't care about. And because they tend to edit only in those narrow areas, they themselves will never be confronted with trying to add a reliable source to another topic area, only to have it rejected because not specialist enough.


 * We're losing the concept of the all-round Wikipedian, who writes for a general readership and offers lots of different perspectives, including ones they themselves strongly disagree with. SlimVirgin  talk| contribs 17:51, 24 August 2010 (UTC)
 * we once had someone like that? What happend to him? (I'm only half joking) Blueboar (talk) 03:14, 25 August 2010 (UTC)
 * It was a her. :) SlimVirgin  talk| contribs 03:27, 25 August 2010 (UTC)
 * I think the idea that every area needs its own sourcing guideline is very misguided. At the very least, any such guideline must clarify upfront in no uncertain terms (in big bold letters) that in any contradiction between the guideline and the governing WP:V policy, the policy takes precedence and the guideline must be changed to reflect it. If it were up to me, I would demote all such guidelines to essay status, as attempts to explain the policies to newbies, and not as serious tools for regular editors to rely on. In essence, these guidelines amount to policy POV forks, which is detrimental, if not destructive, for Wikipedia as a whole. Crum375 (talk) 17:56, 24 August 2010 (UTC)
 * I wholeheartedly agree with the above sentiments. The sourcing guidelines in WP:RS are already perfectly adequate for every situation (including medicine, for that matter). We don't need to fork off separate guidelines for each different topic area. WP:RS already says that peer-reviewed academic journals are more reliable than news, websites, SPS, etc -- so why do we need an article that really says nothing more "for this topic, it is best to use peer-reviewed academic sources in this subject area"? -- Jrtayloriv (talk) 18:03, 24 August 2010 (UTC)


 * Because they want to stress that non-specialist sources should not be used. That excludes a great deal of criticism, which often comes from people outside the topic area, or people inside it who have become "intellectual whistleblowers," or critics of some other kind. I remember someone arguing that the Los Angeles Times was not a reliable source on Prem Rawat, because it contained criticism that Prem Rawat supporters felt was unfair and uninformed, so they wanted to insist on academic sources only. The science editors would scoff at that reasoning, but in their own fields deploy it all the time. And so we witness the death-by-specialism of neutrality. SlimVirgin  talk| contribs 19:25, 24 August 2010 (UTC)
 * I agree. I left a note there.  Not that I don't think that sourcing standards in wp:ver/wp:nor don't need work, but a unified standard is where it needs to be.
 * Speaking of unified, if few or nobody is against crum's wp:ver/wp:nor merger-with-no-changes idea, should we start a section discussing moving that to a proposal phase?  North8000 (talk) 19:43, 24 August 2010 (UTC)

WP:SPOV by any other name would smell as sweet. Any attempts to edit this get reverted, so expecting actual substantive changes to keep it in line with WP:NPOV are instantly doomed. Collect (talk) 23:47, 24 August 2010 (UTC)


 * If they won't allow people to edit it, I don't think they can expect it to become a guideline. SlimVirgin  talk| contribs 01:36, 25 August 2010 (UTC)


 * Slim, do you want to tell us why your name doesn't appear in the talk page's history? If your edits are largely getting reverted, then you need to find the proposal's talk page and convince the multiple editors who are opposing (parts of) your changes that your improvements really are improvements.  It's not Bold—Edit War—Whinge elsewhere:  It's Bold—Get Reverted—Come talk to the actual people who are reverting you.  WhatamIdoing (talk) 01:52, 25 August 2010 (UTC)

An article written according to the NPOV policy and an article on the same topic written according to (effectively) SPOV, i.e. NPOV plus WEIGHT being used to filter out unreliable sources. Count Iblis (talk) 02:05, 25 August 2010 (UTC)


 * The Wikiversity article linked contains almost no content, and what it mostly consists of is reasonable skeptics claiming there is no consensus, which is a cherry-picked view of the subject. While I agree that our article on the topic appears to be more about educating than informing, I don't think the Wikiversity article is particularly neutral.  SDY (talk) 02:28, 25 August 2010 (UTC)
 * Note... Wikiversity has different rules than Wikipedia... for example it allows Original Research which we do not. Apples and oranges to compare them. Blueboar (talk) 02:38, 25 August 2010 (UTC)

Agree with the opinions already expressed that it's highly undesirable to have different sourcing rules for different subjects. (Well, I suppose the general principles can be interpreted in specific ways for particular subject areas, but we need all the information in one place: I'm sure there isn't - or shouldn't be - so much information about this as to make WP:Identifying reliable sources too long a page if it were all put there.)--Kotniski (talk) 12:33, 25 August 2010 (UTC)


 * IMHO If we enhance wp:ver/wp:nor sourcing requirements per my previous vague pseudo-proposal, it would cover those special cases better, thus avoiding these issues. Basically it's
 * A. Strength of a citing/source is a combination of suitability under these 4 factors:
 * 1. Existing wp:ver/wp:nor "publishing" related requirements
 * 2. Existing wp:ver/wp:nor "primary/secondary/tertiary" related requirements
 * 3. Objectivity
 * 4. Expertise
 * with respect to the statement/topic that it is citing.


 * B. The more contested / questionable the statement, the stronger the sourcing required. And visa versa


 * This would automatically adapt to these specialized areas, and adapt to the various types of statements within them as Slim discussed (e.g. technical information, viewpoints and dissent etc.) And solve a whole mountain of other issues.


 * Let's do it. (?)
 * Sincerely, North8000 (talk) 13:03, 25 August 2010 (UTC)
 * Keep beating... you never know when that horse may come alive.Blueboar (talk) 13:21, 25 August 2010 (UTC)
 * Not sure about the applicability of the analogy, but I hope so! :-) North8000 (talk) 14:03, 25 August 2010 (UTC)

How many kicks at the can is this? Jayjg (talk) 01:16, 26 August 2010 (UTC)
 * Where's Crum? Short Brigade Harvester Boris (talk) 02:13, 26 August 2010 (UTC)
 * I sent Crum an invite. But maybe is on a beach somewhere without a computer. North8000 (talk) 23:52, 26 August 2010 (UTC)
 * Sorry, I am kind of busy lately with RL, but what is the specific issue that needs addressing here? Crum375 (talk) 00:05, 27 August 2010 (UTC)


 * North, Boris was making a joke (and he missed Crum's earlier contrib as it was buried in the thread). Let me explain the joke in case others missed it. You see, it doesn't matter if Einstein is cloned from a hair sample, or if Elijah returns to post to this thread, the same group of editors will vigorously oppose the evolution and modification of policies to match the maturation of the site and it's future development.  The Goddess herself could emerge from the depths of the Earth to propose the grandest policy change ever envisioned.  It would be rejected. Moses, Buddha, Christ, and Muhammad might try their hand, only to get reverted with the edit summary, "see talk".  Ayn Rand herself might try her hand, only to be told to seek consensus first. And that, my friend, is the joke. Viriditas (talk) 01:19, 27 August 2010 (UTC)
 * Or, perhaps, some of us have very sound, well-reasoned, and well thought out beliefs that elevating "science" sources on all topics vaguely related to science is a type of bias that shouldn't be tolerated. I once thought that science trumps all, that people should respect what scientists on topics above and beyond what any "outsider" or "less educated" person might say.  I no longer hold such an exclusionary perspective.  The scientific point-of-view already enjoys such a powerful bias in our society that working against when and where we can is both important and valuable.  I'm not saying that science isn't better suited to answer certain questions than other paradigms.  But I am saying that this type of proposal will inevitably be too broadly applied (or, at least, people will attempt to apply it too broadly), to silence non-scientific viewpoints.  This is not a case of resisting change just for the sake of resisting change--it's a case of resisting a bad change that's harmful to the encyclopedia.  Qwyrxian (talk) 01:36, 27 August 2010 (UTC)
 * I didn't understand that. When someone talked about kicking the can down the road too long, I thought it was in regards to Crum's proposal. North8000 (talk) 18:19, 28 August 2010 (UTC)
 * Science is not a POV or a viewpoint. It's only a method that allows one to gather accurate knowledge about the world. There are other methods, and I have in fact written about them myself, but an encyclopedia is not intended to rely on divine inspiration or fortune telling as a foundation for knowledge. Viriditas (talk) 02:21, 27 August 2010 (UTC)
 * No, but it can rely on historical methods, textual analysis, rhetorical analysis, news reporting, etc. Science is not a viewpoint, but the belief that science is the best/only source for "accurate information about the world" is a POV. Thus, the attempt to restrict sources to only scientific ones is a POV.  To be honest, I might be inclined to support the proposal if we could very clearly delimit what articles it covers (for the same reason why I think WP:MEDRS is generally a good thing), but I don't foresee that happening.  Instead, I believe it will be used as a tool to dominate many articles that have both scientific and non-scientific aspects.  Qwyrxian (talk) 02:35, 27 August 2010 (UTC)
 * Historical methods, textual analysis, rhetorical analysis, and news reporting often rely on scientific data, studies, and conclusions. They are not mutually exclusive.  Nowhere does it say that "science is the best/only source for accurate information".  Science is a method that allows one to approximate reality from a materialist/reductionist POV.  It does not take that POV.  Science can also work from a system level, or holistic approach.  Materialists and reductionists can also be anti-science, and rely only on religious or political methods for gaining knowledge about the world.  I would be very happy if you could provide a single good example of where a natural science reliable source guideline would dominate an article and detract from a subject.  Every research manual, guide, and handbook recommends using the most accurate sources on a given subject.  Why should Wikipedia differ from this approach? Viriditas (talk) 03:31, 27 August 2010 (UTC)
 * Although it is on the same topic, structurally your question does not match the topic being debated. The question is whether or not to created a they specific guidelines, and whether to use wording that could be used or misused to categorically exclude certain types of sources.  North8000 (talk) 14:42, 29 August 2010 (UTC)
 * My question directly addresses the claim above it, namely, please provide a hypothetical example where it will be "used as a tool to dominate many articles that have both scientific and non-scientific aspects" as Qwyrxian stated. Viriditas (talk) 12:24, 30 August 2010 (UTC)
 * One comment: Phrase and sentence level pieces of policies and guidelines are pervasively used in unintended ways or even contrary to the intent of the policy.guideline which they are pulled out of context from. North8000 (talk) 14:42, 29 August 2010 (UTC)
 * And an example in this case? Viriditas (talk) 12:24, 30 August 2010 (UTC)

WP:ESSAYS
Obviously it goes without saying that WP:ESSAYS are exempt, but shouldn't this be codified? ;-) Hair-splittingly yours,   Thorncrag    02:21, 28 August 2010 (UTC)
 * Technically, it's already codified. From the policy, with my emphasis: "This means that all material added to articles must be attributable to a reliable published source". Other policies, such as in WP:NFCC, have used "articles" to refer to "pages in the main namespace other than Main Page". --Damian Yerrick (talk | stalk) 12:39, 28 August 2010 (UTC)
 * Yes, Our content policies and guidelines apply to articles in main article space. Talk pages, user pages, sandboxes, and anything with "WP:" as a prefix are not articles.  Blueboar (talk) 13:38, 28 August 2010 (UTC)
 * But that's not really true is it. For example, one is still not permitted to post copyvios, attack pages, promotional spam, plagiarism, issue threats, promulgate hoaxes, etc etc etc on user pages or project pages, even though they do not meet one's definition of an "article".  ;-)   Thorncrag    18:44, 28 August 2010 (UTC)
 * These issues are not governed by our content policies and guidelines. They are governed by our legal policies and behavioral policies (which do apply to non-article space). Blueboar (talk) 19:08, 28 August 2010 (UTC)
 * Yes thank you for that; as I said, it goes without saying for established editors, but nuances like these aren't clear to new editors. :-)   Thorncrag    18:50, 29 August 2010 (UTC)
 * Perhaps then the point is that Template:Essay ought to state it, in addition to its existing disclaimers. PL290 (talk) 19:26, 29 August 2010 (UTC)
 * I don't think its necessary, but I would not object to such an edit to Template:Essay. Blueboar (talk) 12:03, 30 August 2010 (UTC)

Abuse of primary sources
Added by me, reverted by Slp1: "; a public document may cite the existence and nature of a court case, but any traditionally private personal details need a secondary source". If (ab)use of primary sources is a topic under NOR, then surely the abuse relating to WP:BLP should be mentioned, especially since BLP refers summarily to this section for its backup authority. Or should it be more generic, like "any details requiring interpretation"? See also WT:BLP. JJB 19:10, 5 September 2010 (UTC)

Talk pages
I was under the impression that OR wasn't allowed on talk pages, but I seem to be wrong. I'm sure I've seen references to it being not allowed though. I don't understand why we would allow OR on talk pages - this quickly becomes a forum style discussion, with editors doing things such as trying to recruit help for their OR or heaping tons of text on talk pages. It can be very confusing and misleading to our readers at times also, as it can be hard for them to distinguish between nonsense and reliable sources. Dougweller (talk) 07:08, 7 September 2010 (UTC)
 * I think that WP:NOTFORUM should already address the problem of discussions going on about the subject. As for confusing our "readers", I must say that I don't understand. The purpose of a talk page is to help editors, not readers.  Technically speaking, someone "reading" the article generally won't even look at the talk page; I'm sure that if they do they'll find all sorts of things confusing, like all of the policies, essays, and arguments we bandy about.  Furthermore, it isn't even possible not to discuss OR on the talk page, as part of what we have to do is to debate whether various contents are or are not OR and what constitutes a reliable source.  Qwyrxian (talk) 10:16, 7 September 2010 (UTC)
 * Of course we have to discuss what is OR, I should have been more careful, I meant editors should not be adding their own OR to talk pages. Dougweller (talk) 16:06, 9 September 2010 (UTC)
 * It really depends on how the OR is discussed. There is nothing wrong with having a civil discussion on an article talk page that ventures into OR ... but pushing one's OR on talk pages can be disruptive.  Blueboar (talk) 16:13, 9 September 2010 (UTC)
 * Yeah, I think I'd rather stick with WP:SOAP and WP:NOTFORUM. My thinking here is that those (alongside, sometimes, WP:CIVIL and WP:NPA) are enough to prevent disruptive presences on the talk page.  But I figure that if someone starts venturing into OR on a talk page (presumably a new editor), that's a great place to steer the conversation back to WP's policies on NOR, verifiability, NPOV, etc., etc.  Even though it's rare, I always hope that the person who jumps on to WP to start spouting their own theories might be "brought into the fold," so to speak.  Or we might be able to pull actually usable, verifiable information out of the rant.  As long as the OR isn't taking over the talk page (in size or frequency), it seems like it could be useful.  Qwyrxian (talk) 21:41, 9 September 2010 (UTC)

"advance a position"
In this policy, what is the meaning of the phrase "advance a position"? Thanks. --Bob K31416 (talk) 02:33, 22 September 2010 (UTC)
 * It generally is interpreted as meaning "asserting a fact or opinion not directly asserted by the source". -- Jrtayloriv (talk) 02:43, 22 September 2010 (UTC)
 * That is an interesting question. The wording in the policy is a bit frustratingly opaque on that. I equate "advancing a position" with "making an assertion." I also think of WP:NPOV because neutrality would be compromised if an unsupported position were "advanced." Bus stop (talk) 04:11, 22 September 2010 (UTC)
 * Let's say I added a sentence to Barn that stated that most barns are typically painted red because red paint is the cheapest paint available, and sourced it with a reliable source. I am advancing the position that most barns are painted red and that it is because red is the cheapest paint. But if that source only strongly supports that most barns are painted red and only hints or suggests that it is because of the cost of red paint, then I am advancing a position not strongly supported by the source.     Thorncrag    04:45, 22 September 2010 (UTC)
 * And to further the analogy, a better example might be to use that same source, but also state in the article that the cost of red paint is so low because so much red paint is produced because barns are always painted red... :-)   Thorncrag    04:47, 22 September 2010 (UTC)
 * Is that distinct from a violation of WP:SYNTHESIS? Bus stop (talk) 04:50, 22 September 2010 (UTC)
 * Yes because synthesis requires making up a conclusion in the article that wasn't explicit in either one of two sources. Synthesis by definition is combining two things to form something else and in this case, taking two facts from two sources to make up a conclusion based on those two facts and insert it into the article.    Thorncrag    05:00, 22 September 2010 (UTC)
 * Sythensis is taking one source that says "red paint is cheap" and another source that says "most barns are painted red" and coming to the conclusion that "most barns are painted red because red paint is cheap". 陣 内 Jinnai 16:29, 22 September 2010 (UTC)
 * In that example, the combined sources don't even support the conclusion (barns may be painted red for an entirely different reason). Consider the following subtler example of synthesis, where the conclusion follows logically from the combined sources although it is not explicit in either one of the two sources: "most barns are painted with cheap paint".  Precis (talk) 02:00, 23 September 2010 (UTC)

Conclusion
Presently in the lead is the sentence, "It also refers to any analysis or synthesis by Wikipedians of published material, where the analysis or synthesis advances a position not advanced by the sources."

Please consider adding "or conclusion" so that the sentence would be, "It also refers to any analysis or synthesis by Wikipedians of published material, where the analysis or synthesis advances a position or conclusion not advanced by the sources."

The word "position" tends to mean  a side in an issue. Often an editor's OR does not involve taking a side in an issue but is rather an unsourced conclusion that the editor makes. Thus, adding "conclusion" would seem worthwhile. --Bob K31416 (talk) 15:54, 23 September 2010 (UTC)
 * I don't object to the addition, but is it really needed? After all... can you state a conclusion about something without taking a position on it?  Aren't they the same thing? Blueboar (talk) 16:25, 23 September 2010 (UTC)
 * Did the definitions below help? --Bob K31416 (talk) 20:25, 23 September 2010 (UTC)
 * I am not enthusiastic about this idea. You can't even rephrase a sentence without drawing a new conclusion that is not explicitly present in the sources, and I have already seen occasional attempts to paint everything other than a straight copyright violation as original research. Like all policies this one must strike the right balance.
 * Is there an actual problem? In my experience Wikipedia is quite good at defending itself against original research, whereas intelligent editors who can work with a number of sources and turn them into a long, coherent text are a very rare commodity. Maybe if you give a specific example I can understand where you are coming from. Hans Adler 16:48, 23 September 2010 (UTC)
 * I was somewhat surprised by your comments since they suggest that WP:NOR presently does not prohibit unpublished conclusions. Please note this excerpt from  the WP:NOR section Using sources,   "Drawing conclusions not evident in the reference is original research regardless of the type of source", and this excerpt from the WP:NOR section WP:SYNTH, "Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources." --Bob K31416 (talk) 20:25, 23 September 2010 (UTC)
 * Wikipedia is full of fundamentalists who want to interpret all policies in the strongest way possible. If you interpret the prohibition of original research in the strongest way possible, then the only way you can write an article is by literal copying. E.g. if you replace a pronoun in the original text by a noun which describes it more precisely, then technically you go beyond the source by assuming that "it" in the second sentence has a specific referent in the first sentence, and by making an assumption about which one it is. I have too often seen claims that we can't replace a word used in a source by an obvious synonym that solves an editorial problem. These claims are not taken particularly seriously, but they come up often enough that people no longer laugh at them. Hans Adler 17:47, 25 September 2010 (UTC)


 * I oppose the change. "Position" implies that several plausible positions exist, and the editor is advancing one of them. "Conclusion" applies equally when there are several plausible positions, and when there is only one, inescapable, trivial conclusion. Drawing trivial conclusions is allowed (e.g. every time a converted amount for a quantity is provided, when the conversion is not present in one of the cited sources, one is synthesizing the cited source with a table of conversion factors to reach a new, trivial, conclusion). —Preceding unsigned comment added by Jc3s5h  (talk • contribs)  17:01, 23 September 2010 (UTC)
 * Did the definitions below help? --Bob K31416 (talk) 20:25, 23 September 2010 (UTC)

Perhaps it would help if we reviewed the definitions of position and conclusion.

Here's some definitions of "position" that can be found from googling: position definition.

And here's some definitions of "conclusion" that can be found by googling: conclusion definition.

--Bob K31416 (talk) 18:18, 23 September 2010 (UTC)


 * NOR is actually quite simple... The ultimate nutshell is: "Don't go beyond the source"... whether we are talking about advancing a position not found in the sources, or drawing a conclusion not reached in the sources, the error is that it goes beyond the source. Applying this concept... a good article should neutrally lay out what the sources say the positions are, and neutrally present those conclusion that the sources have reached. Blueboar (talk) 20:32, 23 September 2010 (UTC)
 * As you can see for yourself, the present lead doesn't mention conclusions but the policy makes significant statements regarding conclusions, as I mentioned previously. The lead is thus deficient in that regard and hence "or conclusion" should be added as indicated in the opening message of this Talk section. Thanks. --Bob K31416 (talk) 23:55, 23 September 2010 (UTC)

Well, it looks like this page has fallen silent since I posted my responses a day ago. Regarding the addition of "or conclusion" as suggested at the top of this section, perhaps editors could give a simple support or oppose, for my own information, so that can I see what the present feeling is here regarding the addition. Thanks. --Bob K31416 (talk) 21:49, 24 September 2010 (UTC)
 * Neither... I don't object to the addition, but I also don't think it is really needed. Blueboar (talk) 21:53, 24 September 2010 (UTC)


 * I don't see the need for a change. The phrase is clear enough as it is. SlimVirgin  talk| contribs 21:56, 24 September 2010 (UTC)
 * I think an editor in the previous section "advance a position" summarized the situation well with the comments,
 * "The wording in the policy is a bit frustratingly opaque on that. I equate 'advancing a position' with 'making an assertion.' I also think of WP:NPOV because neutrality would be compromised if an unsupported position were 'advanced.' "
 * Although a reader could make a connection between "position" and "conclusion" it isn't comfortable reading because the word "position" doesn't quite have that definition. I think we have to be sensitive to the general reader of this policy who may not be familiar with the terminology and use of words in Wikipedia, such as "position", that aren't clearly defined in Wikipedia and have somewhat different meanings than what is common outside of Wikipedia. (Please see the definitions of position and conclusion above.)


 * I suppose one might argue that "conclusion" is a type of "position" but from the definitions above, "position" implies a side in an issue. A conclusion may not involve an issue. It may simply be something that an editor has reasoned to be correct but may or may not be, especially since it hasn't been published in a reliable source. That's why I suggested adding "or conclusion" so that the sentence would be,
 * "It also refers to any analysis or synthesis by Wikipedians of published material, where the analysis or synthesis advances a position or conclusion not advanced by the sources."
 * Not trying to change policy, just trying in a small way to have a less "opaque" policy page, i.e. as the editor put it in referring to the meaning of "advance a position",
 * "The wording in the policy is a bit frustratingly opaque on that."
 * Regards, --Bob K31416 (talk) 16:16, 25 September 2010 (UTC)


 * I think you are trying to define things that should not be defined because there is no definition that fits all situations. And anyway, if there is a specific problem with the policy it's really time to give an example now. Hans Adler 18:03, 25 September 2010 (UTC)
 * Hans, There seems to be some miscommunication between us.
 * Re "And anyway, if there is a specific problem with the policy it's really time to give an example now." - AFAICT, I'm not raising an issue regarding a problem with policy.
 * So let me ask you, would  adding the phrase "or conclusion"  be a policy change in your opinion? --Bob K31416 (talk) 22:29, 28 September 2010 (UTC)


 * There's no point in continuing to ask the same question, Bob. Position encompasses conclusion, just as flower encompasses rose. Please give a real example that shows "advancing a position" is causing confusion, or leave it be. SlimVirgin  talk| contribs 00:40, 29 September 2010 (UTC)
 * SlimVirgin, Since it's not a policy change but only an improvement in clarity, I don't think that's an appropriate request for something that is only an improvement in the writing of WP:NOR, not a change in policy.   And please be careful not to suppress useful changes with requests for real examples and consider whether you can give real examples to justify all your attempts at improving the writing of WP:NOR that aren't changes in policy.


 * When I read "advance a position" there is a question in my mind as to what it means because the dictionary definitions seem to be different than what you think the definition is.  Please review my previous comments  here and here   regarding the meanings of "position" and "conclusion".      --Bob K31416 (talk) 10:58, 29 September 2010 (UTC)


 * Support Clarity is required on a core policy. And this is certainly not clear as it stands. --Michael C. Price talk 01:06, 29 September 2010 (UTC)
 * Thanks for your support. I tried to add it but it was reverted, so unless there is more support I won't pursue it. Thanks again. Regards, --Bob K31416 (talk) 14:19, 30 September 2010 (UTC)

Synth - "true"
Here's the the beginning of WP:V,
 * "The threshold for inclusion in Wikipedia is verifiability, not truth&mdash;whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true."

The discussion of the first example in the SYNTH section of WP:NOR says,
 * "Although no conclusion is drawn and both parts of the sentence are true, it implies that the UN has failed to maintain world peace."

This seems to miss the point that WP:SYNTH is about combining material from reliable sources. --Bob K31416 (talk) 05:56, 1 October 2010 (UTC)


 * I tweaked it to refer to sources rather than truth. SlimVirgin  talk| contribs 06:08, 1 October 2010 (UTC)

Explicitism and Implicitism in debates on SYNTH/OR
Over at WP:NOR/N there was a discussion of SYNTH/OR in a topical article; where the sub-topics were reliably sourced as existing in themselves, but not reliably sourced as part of a literature of the topical article. Another editor's post got me thinking about two irreconcilable editorial views on this topic, unrelated to content, as irreconcilable as deletionism and inclusionism.

As an example "Disagreements between Dogs and Cats" could be reliably sourced from Garfield (passim.). One topic which Garfield doesn't cover, are the issues that dogs tend to hump your legs, cats don't. We can reliably source both of these points, but should they be included in the topical article?

Explicitism says that sub-topics may only be included where RS whose subject is the topic associate the sub-topic directly to the topic. No dogs humping legs in "Disagreements…" no cats not humping legs in "Disagreements…" Implicitism says that sub-topics which are reliably sourced in themselves may be included in a topical article regardless of RS specifically associating them to the topic. In this version "Disagreement…" contains dogs humping, and cats not.

Explicitist editors view an implicitist article as SYNTH/OR, and would seek to have the sub-topics not associated in RS with the topic removed. Implicitist editors view an explicitist article as an incomplete coverage of the topic. As an explicitist, I view Judaism and violence as needing to be blanked and rewritten from Firestone (2004), the only article appearing to explicitly discuss the article's topic in a scholarly manner. As an explicitist, I view the first four and eighth section of Mass killings under communist regimes as reliably sourced for the link to the subject (explicit scholarly and popular comparisons of alleged mass killings in states self-describing as communist); I view sections five through seven as SYNTH/OR, as while each point is reliably sourced, they are not reliably sourced as part of comparisons of mass killing.

I view these editorial positions as currently irreconcilable within the policy matrix of wikipedia. Can someone advise me on any policy directing an editorial position in this matter?

I'd like to acknowledge editor Nuujinn and the editors I've interacted with on Mass killings under communist regimes for inspiring me to think this thought. Fifelfoo (talk) 04:27, 28 September 2010 (UTC)
 * As you define it, I've been preaching explicitism for 1.5 years at longevity traditions and related articles. A published self-proclaimed expert wants certain bullet points, which documentedly arise from himself, to be article subheads, even though no source whatsoever (except WP:SELFPUB) uses either his categories or his terms for the categories. Further, even the alternate title ("longevity myths") appears in only one throwaway source, which is why I'm arguing (in mediation cabal) for the "traditions" title, which is explicit in several on-point sources (cf. WP:RNPOV). Without looking at your articles, I'd argue for just a little leeway from legalist explicitism, in that an otherwise reliable source need not be excluded simply because not "scholarly", nor because it fails to take a comparative view, and I'd also doubt Firestone 2004 is the only scholarly source; but your general principle is right on. A number of sources under those two heads would not be excluded under OR, because they probably do relate to the titles as given ("and" is typically a wide-open compromise title word), but they might be excluded under WP:UNDUE instead because the topics lend themselves to biased writing (such bias can be attributed as "A notable fictional cat says" etc.). But if the topic explicitly uses the words "Judaism" and "violence" in close context, that's my explicitism. In the other example, if it were titled "Differences" instead of "Disagreements", there is no OR in citing different cat and dog behaviors as a difference; but for it to rise to the level of disagreement (or, say, controversy) would require sourcing saying somehow that people had disagreed or controverted. Nonexplicit sources can be spun out into titles that they explicitly relate to.
 * I am all for language making explicitism explicit, but I will not write it. JJB 15:50, 28 September 2010 (UTC)
 * This is quite confusing for me... I am not at all sure what the issue is. Would someone simplify the issue? Blueboar (talk) 02:27, 29 September 2010 (UTC)
 * Synthesis can be a difficult problem. This arises primarily in the area of comparative articles, such as "A and B".  Ought sub-topics in the article "A and B" be produced from a bunch of sources talking only about A, and a bunch of sources talking only about B; or must sub-topics be produced exclusively from sources talking about "A and B"? Fifelfoo (talk) 02:33, 29 September 2010 (UTC)
 * Well, first of all, WP:Article titles strongly discourages articles with "and" in the title (such as Judaism and violence. This is because such titles are so often a red flag that the article has WP:NPOV and WP:NOR issues.  Thus, as a first step... I would suggest a discussion on how to re-title Judaism and violence ... I suspect that such a discussion will help highlight where any POV and OR exists. Blueboar (talk) 02:58, 29 September 2010 (UTC)


 * The problem being referred to arises when an article is initiated on a topic that is not satisfactorily sourced. Sources should be present describing and even defining a topic. Editors should not be taking it upon themselves to create subjects for articles if those subjects cannot be demonstrated to preexist. Sources should be telling us what that topic is. Sources may indicate how the term representing the topic is generally used. Sources should probably be available indicating common components of that topic. A well-sourced topic can be expected to be spoken about in an introductory way. That introduction serves as an outline to guide editors. It serves as a reference. It indicates what may be included and what may generally be excluded from the area that is the topic of the article. All sources need not be in complete agreement on all points, but a general "ballpark" picture should emerge from the general commentary that sources provide on a topic. Bus stop (talk) 03:34, 29 September 2010 (UTC)
 * Hear, hear! And if a topic is agreeably titled "A and B" on unequally yoked topics, why, main sources should absolutely be about the relationship between A and B! JJB 04:46, 29 September 2010 (UTC)
 * Another example is Mass killings under communist regimes; which I am not trying to discuss here, but merely use as an example of my concern. Incidents like the Ukrainian famine, or the GuLag, or Pol Pot's policies are firmly documented.  There are even a very small handful of reliable sources which say that Communist regimes engaged in mass killing for reasons specifically and only bound up with their communist nature (there are many sources which say, for example, "Totalitarian reasons for mass killing, for example Germany, Soviet Union".)  But the article doesn't need a section on GuLag (a single incident, non-comparative) when the article's topic is the comparison of mass killings.  We have a whole article for GuLag for its significance in and of itself.  Fifelfoo (talk) 04:04, 29 September 2010 (UTC)
 * See, but with this one you're getting into content where theory becomes much fuzzier. "Comparison" is not in the article title. A source about gulag is a source about the titled topic. Whether it gets its own section would depend on its place in sourcing, and on first glance, I'd favor continuing to enfold gulag cites into the chrono Soviet history, AND/OR move them to gulag. If someone insists that "gulag" needs its own section, I'd agree when sufficient sourcing materialized. Explicitism is valid but doesn't apply to exclude that. JJB 04:46, 29 September 2010 (UTC)

Content is the key issue, I think, relating to OR. Fifelfoo asked 'Ought sub-topics in the article "A and B" be produced from a bunch of sources talking only about A, and a bunch of sources talking only about B; or must sub-topics be produced exclusively from sources talking about "A and B"?' I think the answer to that would be generally yes, but it's not so simple. Using a current fav of mine, United States presidents with facial hair, the A and B are US Presidents and facial hair. Let us say for the sake of discussion that the newly elected president Fraknickle has a bread, and someone cites a source that discussed this beard and the public's reaction to is, but does not treat the issue of US Presidents and Facial in general at all. In that circumstance, are we allowed to make the link between between Fraknickle and US Presidents (based on common knowledge, similar to paris is in france) and thus treat Franknickle as a A1, or is Fraknickle an X? I would argue the former is premissable, although I know some editors would likely disagree. --Nuujinn (talk) 14:02, 29 September 2010 (UTC)
 * I like the Fraknickle example (I'm assuming there's a literature of Presidential Beards of course :). This seems like a trivial implication.  Consider another example  Anarchism and Marxism is a good subject topical article, they've been in a 150 year shit fight, and are widely compared.  Using Jones on "Anarchism and Marxism on the State" to write a section on "the State" would be fine.  Using Lief on "Anarchism and War" and using Susanson on "Marxism and War" to write a section "on War" wouldn't be, in my explicitist opinion. Fifelfoo (talk) 14:28, 29 September 2010 (UTC)
 * Exactly - this needs to be clearly stated and is, in of course my h'opinion, a basic part of our NOR policy. Dougweller (talk) 14:55, 29 September 2010 (UTC)
 * I concur generally, since in the last example Lief would be an A, and Susanson would be a B, and to combine them directly to produce a C(onclusion) would be SYNTH. But if Jones makes the connection, would it not be permissible to include Lief and Susanson in that section, so long as the latter two are not combined to reach a conclusion, and esp. if the points they make illuminate some aspect touched upon in Jones? --Nuujinn (talk) 16:12, 29 September 2010 (UTC)


 * Nope. You are conceptualizing this in much too complicated a way. Any premise that serves as a foundation for an article has to have a reasonable preexistence outside of Wikipedia. Certain points need to be made about that topic. It is not going to be the same in different cases. But sources have to substantially establish that the topic is recognized outside of Wikipedia. This might be accomplished by some sorts of statements about the topic. Those statements might say that the topic was first articulated by some individual. Or that the topic in question emerged from some debates in some time and place. (These are just examples.) It is by means of talking about the subject that sources acknowledge its existence as not just a few words in passing but rather a concept that assumes a standard that can be referred to time and again, with the reasonable expectation that its meaning remains somewhat constant. Adequate sourcing is essential to article creation. Article creation in the absence of sources sufficiently supporting and even defining the topic is really not what this project is about. We should always be asking ourselves: are the sources available going to be genuinely helpful to me in writing this article? Bus stop (talk) 16:29, 29 September 2010 (UTC)
 * And there is the fundamental disagreement. --Nuujinn (talk) 18:17, 29 September 2010 (UTC)


 * There is a disagreement. I can only state my position, and I'm trying to do so succinctly, despite my sometimes long-windedness. My position is that articles should be generally sourced relating to their overall scope. Bus stop (talk) 18:23, 29 September 2010 (UTC)

I don't see the disagreement as being much. If Jones sufficiently demonstrates (one source is not always sufficient) that "A & M on the State" is a sufficient academic topic, and Lief writes on "A on War", he can be mentioned under the State subhead on his portions that relate to the State, IF it doesn't imbalance the article. In my example (which is also a real one and thus shows my bias), I hold that the topic "longevity myths" and several subheads like "village elder myths" should be demonstrated to be real topics in sufficient sources (which I don't believe has materialized) and that one should be shown a subtopic of the other. If I find that more sources use the title "longevity traditions" and more of them name as a subcategory "politically motivated claims", then I can talk about some things that use the insufficiently established subtopic "village elder myths" if sources classify them as politically motivated claims, but I would not talk about others if there is no demonstration that "village elder myths" and "longevity myths" are sufficiently deletion-proof and in topic-subtopic relation. But I think I hear sufficient agreement that explicitism per se includes enough wiggle room to be promoted to policy. JJB 19:43, 29 September 2010 (UTC)

Wait, you think it's inappropriate to talk about the Red Terror or the Great Leap Forward in an article titled Mass killings under Communist regimes? john k (talk) 01:52, 30 September 2010 (UTC)
 * The Red Terror (1917-1923) was a society specific incident. The scholarly literature discussing it assigns Russian Bolshevik, rather than -Communist par Communist- causes.  The Great Leap Forward was a society specific incident.  The scholarly literature discussing it assigns Chinese Communist Party, rather than -Communist par Communist- causes.  The literature of Communist causes of mass killings discusses a number of examples, but is a political science literature dedicated to determining underlying causes, and discusses things like this, "Communist causes for mass killing lie in class factors, for example, rural dispossession and class change was actualised by mass killings, for example the GLF, Ukrainian Famine, Pol Pot."  Such an article shouldn't have a raw section discussing the GLF:  the GLF should be discussed in the section "Mass killing to cause class dispossession and class change" to the extent the GLF is used as an example in the literature of theories of communist causes of mass killing.  The current article contains a short theory section and then a sequence of incidents presented in a manner not present in the literature on "Communist mass killings", but rather as a compilation of the literature of History of China, History of the Soviet Union, etc.  Fifelfoo (talk) 02:11, 30 September 2010 (UTC)
 * I do not see how the title of the article implies that it can only discuss killings that are a result of "Communist par Communist" causes. In fact, the title of the article says nothing about causes, whatever.  Articles like this are always worth keeping a close eye on, because they are great vehicles for partisan propaganda.  But I don't think the solution is mass removal of relevant material on dubious OR grounds.  Maybe it's just because I'm a historian rather than a political scientist, so when I see an article titled that, my expectation is that it would include brief summaries of the major incidents of mass killings under communist regimes.  But I think that's kind of common sense.  And I absolutely do not think it is OR.  Presumably the political science literature on "killings under communist regimes" does talk about things like the Red Terror and the Great Leap Forward as examples of mass killings by communist regimes.  So it's not OR to have neutral summaries of those incidents in this article - that is synthesis, but not synthesis to advance a novel argument, so long as we are careful, at least. john k (talk) 14:34, 30 September 2010 (UTC)

This discussion makes me uneasy because it strikes me as a symptom of fundamentalism that has lost sight of the original reason for our prohibition of "original research". Original research in the non-technical sense (and I can't use it in a technical sense because this debate is precisely about what that should be) is absolutely necessary for writing a good encyclopedia. We have thousands of articles full of original research in this loose sense, which could only be "fixed" by deletion or extreme pruning that makes them incomplete. If I give concrete examples here I probably risk attacks by fundamentalists on the articles in question, so I will give you an example from the past: The first link is to an article that appeared in DYK, and which was widely discussed especially in its AfD, to which several editors of relatively high profile contributed. The issue was not OR or SYN. In this respect the article was perfectly fine. The issue was a suspicion that the subject of the article was a hoax. It turned out that it wasn't so much a hoax as a kind of urban myth in the aerial photography community. The photos are authentic, but to the best of my knowledge (and I have become one of the few experts on the subject) the term "Bavarian Pigeon Corps" came up through a huge telephone game that probably went like this: Aerial photography by pigeon was invented by Julius Neubronner from Kronberg im Taunus. (Correct) -> It was invented by a Bavarian. (False – Kronberg (Lower Bavaria) is not related.) Aerial photography by pigeon was tried out by the [Prussian] German military during World War I. (Correct) -> The Bavarian army tried it out (nope), so it must have had a pigeon corps (nope), i.e. the "Bavarian Pigeon Corps" (nonsense). One "reliable" source even claims that this happened in "the Bavarian War" – a war unknown to historians.
 * Bavarian Pigeon Corps, 12 June 2009.
 * Pigeons in Aerial Photography, 16 July 2009.
 * Pigeon photographer, today.

The second link is what I first turned the article into: A huge piece of OR and SYN, but essentially correct and informative. As I did not have any post-1920 sources about the topic that gave an overview, the article would have been extremely vulnerable to anti-SYN fundamentalists at the time.

The third link is to the article as it is today. I now have an original copy of the inventor's autobiography and most importantly I have an in-depth article by the one scholar of history of photography who has researched the topic. (Based on this I had to correct a few minor details, but found that the earlier version was fundamentally correct.) She didn't find every detail though, or maybe suppressed some, so our article is more complete. The most extreme example of original research currently in the article is the first two paragraphs of the section on World War II. I had to piece this together from various sources, because so far nobody has written about the fate of pigeon photographers in World War II, presumably because they didn't really play a role. Yet the idea was still in people's minds, and that's a crucial piece of this article.

That we can have articles such as this one is one of the strengths of Wikipedia. Contrary to what you could think when you only look at the noticeboards, Wikipedia is not primarily a place for battles between political viewpoints, various nationalists, or between crackpots and reasonable people. An article such as my example would be very hard to write in a contentious topic area. That does not mean that it should not be allowed to write such articles in those areas where it is possible.

For the moment this article is safe, since nobody disagrees with anything it says. But a lot of Wikipedians are very young Americans, and therefore prone to fundamentalism. If we ever get a silly ideological split similar to that between "inclusionists" and "deletionists", then not just this article will be in danger but a large part of our accurate content in mathematics and other completely non-contentious areas.

To get back to the point: Prohibitions of OR/SYN are tools for getting rid of crackpots and deciding disputes. They are not part of Wikipedias goals, and should not be. Every discussion of OR and SYN needs to keep this in mind. Hans Adler 15:59, 30 September 2010 (UTC)


 * It seems that the part of WP:NOR that applies here is,
 * "To demonstrate that you are not adding original research, you must be able to cite reliable published sources that are both directly related to the topic of the article, and that directly support the material as presented."


 * In the case of the article Judaism and violence, a source should contain discussion of both Judaism and violence, otherwise the source is not directly related to the topic of the article. Looking at the list of references for the article, there appears to be enough sources available on the topic of the article, i.e. there appears to be enough sources that each discuss both Judaism and violence. Here are some examples from that list that probably includes many others that are on the topic of the article,
 * "Violence, Scripture, and Textual Practice in Early Judaism and Christianity"
 * "Abraham's Curse: The Roots of Violence in Judaism, Christianity, and Islam"
 * "The destructive power of religion: violence in Judaism, Christianity, and Islam"
 * "Beyond violence: religious sources of social transformation in Judaism, Christianity, and Islam"
 * "The just war and jihad: violence in Judaism, Christianity, and Islam"
 * If a source doesn't contain a discussion of both Judaism and violence, then it could be used if there was a good reason according to WP:IAR,
 * "If a rule prevents you from improving or maintaining Wikipedia, ignore it."
 * --Bob K31416 (talk) 20:50, 30 September 2010 (UTC)


 * Jumping in here expressing a quite different perspective (pragmatism) without having any answers: many articles on WP are deficient in many more ways than can be covered by a list of policies. The problem is reaching a consensus about a plausible and acceptable goal state for the articles, which should cover both explicitism and implicitism as per above. Much of the question in the aforementioned discussion is: could a certain article reach a viable state that is plausible, logical, coherent and reliable? Can we do it or are we too ignorant? If the topic concocts a topic or notion that doesn't exist, that may not necessarily immediately obvious to all editors. The problem is then convincing other editors of the lack of viability.


 * If the topic is viable, such as for example being a real debate topic in politics, then the explicitism vs. implicitism perspectives gives different answers of how to deal with an unbalance in "sound reason" vs totally "undue synthesis". There should be some balance between the conflicting 'coherence' (of discource) vs. 'citability' in the article while it is redeveloped to a better state containing both kinds of 'coherence' and 'citability'. Such a balance might be achieved by moving items to and from the talk page, accompanied by evolved discussions to reach some basic degree of consensus on the talk pages. Otherwise I'm inclined to believe the development/discussion processes on Wikipedia are pretty inefficient so that errors and conconcts are fairly longlived on WP. Rursus dixit. ( m bork3 !) 17:06, 3 October 2010 (UTC)
 * I think that's very well put, and akin to how I view the balance between inclusionists and deletionists--if all parties regardless of orientation are willing to work together to achieve balance and consensus, we'll have a decent article. Also, in a different venue, Sjakkalle expressed something that rings very true to me regarding SYNTH: "Don't get too entangled with WP:SYNTH. That policy is meant to prohibit statements containing original ideas and original deductions about the subject, it was never meant to be a straightjacket to prevent editors from organizing the material in the manner they think will convey the information in the best possible fashion. WP:NOR prohibits original research, but it does not prohibit research or original writing." We engage in research and make subjective judgments all of the time--we evaluate sources, judge relative weight, choose our wording, and the like. Guidelines and policies should inform these decisions, but we must also choose which apply and how to apply them, since a rule treats a case, but we much decide which case we are faced with (this comes out of Wittgenstein and the pragmatism of Pierce). For some topics, explicitism alone may be perfectly adequate--for example, in the hard sciences. But in other area, implicitism as Fifelfoo has defined it, seem more adequate than strict explicitism, since the latter would prevent us from writing a decent article on the topic. I think Judaism and violence falls in that category, but please note I am referring to the content of the article--good arguments can be made as to whether or not we need that particular article, or whether or not it has the best possible title, but it seems to me that the content is largely well sourced. But it is the tension between views that brings us to have decent articles, so long as we are willing to reach consensus (and that means to some degree or another we are willing to compromise). Sorry to be long winded, but I've been thinking about this a lot the last few weeks, and I'll get of the soapbox and shut up now. --Nuujinn (talk) 17:44, 3 October 2010 (UTC)

To answer a question above, I must diverge from Fifelfoo when asked should GLF be listed under "Mass killings under communist regimes". Of course it should, because it is one, both in sources and in common definitions of the words (albeit sometimes stringing together definitions is SYN, because not all "conscientious" "objectors" are "conscientious objectors"). If the article were titled "Communist causes of mass killings", Fifelfoo's explicitism would be absolutely correct unless a source said the GLF was caused by communist rather than Chicom rationales (but communist v. Chicom is such a hair-splitting argument that it should be cheerfully yielded by either side early on in its formulation). WP has room for such a parallel article without forking. So yes, I still think explicit language in policy on this point is called for. JJB 18:01, 3 October 2010 (UTC)
 * Yeees... but in practice we're left with a List of Nasty Communist State Violence, varying from Kampuchea through to a small pack of bastards in the Balkans who stretched it to top 10K dead. And when we went through listification, it failed list notability critera.
 * The article on Mass Killings hinges on a real academic literature that links Communist states with a tendency towards mass violence, varying from Lemkin (everything's Genocide!, anti-communists are my funding agency!) through to Valentino's article (the typology varies from his monograph, which lacks a category). At the margins there are hysteric fringe views such as the "Introduction" to the Black Book where communists kill people because... they're godless communists, (from what I've read, the Black Book's chapters are single society case studies with some quite respectable authors).  Through to edge classifications such as Democide which assign deaths to a tyrannical cause, and communism as a subset.
 * The meat of the academic and notable popular literature on Mass Killings in Communist states is all about cross cultural attempts to formulate reasons; the RS (and the few HQRS I've located) is all about causation and comparison. The link between the GLF and Red Terror and Ethiopia is through Valentino, or Lemkin, or Democide theory.  The SYNTH is placing the exemplars in the literature side by side and eliminating the actual narrative in the scholarly sources: the narrative is theories of causation, "Tyranny" explains GLF, 5YP, Yr0 due to central state apparatus, "Dispossessive Mass Killings" explains 5YP, Yr0, and explains the absence of mass killing in the DDR due to..., "Godlessness" claims in a clear FRINGE that M&E were godless bastards and only with Christ can we avoid mass killing, Courtois claims that only God can orient political thought, but deals primarily with the Soviet Union and Stalin.
 * At the moment we have "Blue" containing a list of things which are blue, placed side by side, as if this substantiates the topic; and a short failed discussion of authors who wrote about Blue without raising their theories of Blue. What we need is Blue: Theory of Blue in the context of Light; additive Blue (examples); subtractive Blue (examples); and, production of Blue in process, dye or emitted light.
 * I'm not jumping into the swamp of that article again, or its odious sources, or the horror of its examples. But the key process when writing about an academically covered topic in the humanities and social sciences is to identify the narrative and structure of the already existing scholarly original research in the sources.  Producing a structure by conjuction is SYNTH original research, much like my mousepad, a folder, and my bicycle are all blue. Fifelfoo (talk) 04:18, 5 October 2010 (UTC)

"analysis and synthesis"
I think that the following sentence in the lead could be improved by not having the phrase "analysis or synthesis" repeated.


 * "It also refers to any analysis or synthesis by Wikipedians of published material, where the analysis or synthesis advances a position not advanced by the sources."

I tried but was unsuccessful and it was reverted. Perhaps someone might find an acceptable way of doing it. --Bob K31416 (talk) 06:51, 1 October 2010 (UTC)


 * Trimmed. PL290 (talk) 09:24, 1 October 2010 (UTC)
 * I quite liked Bob's initial trim; it was more plain english. Fifelfoo (talk) 10:13, 1 October 2010 (UTC)
 * Me too. Johnbod (talk) 17:53, 3 October 2010 (UTC)


 * The thing is, it goes wider than the Wikipedia editor. Such analysis or synthesis could in theory be by any party other than the cited WP:RSs, not just the Wikipedia editor. For instance, a dubious, other published source, or an individual who never edits WP, and is perhaps a friend or relative of the editor, etc., etc. That's why I trimmed it further. PL290 (talk) 18:22, 3 October 2010 (UTC)


 * Dubious sources are not covered under OR, they are handled by wp:RS. friends and relatives are irrelevant: it is the person editing the article who is guilty of synthesis, even if they ogt the idea from someone else.  I don't think this objection is quite on target.  -- Ludwigs 2  18:29, 3 October 2010 (UTC)


 * Not objecting, just explaining. Another editor reverted Bob's initial trim; I tried another, which has stood since, and which I think is tighter for the reason just given. But I'm fine with Bob's too, if that turns out to be accepted by others and it's felt necessary to include those extra words. PL290 (talk) 18:50, 3 October 2010 (UTC)

Serious NPOV/SYNTH problem with List of films considered the worst and best
I've started a talk page discussion here. If anyone here is well-versed in the NPOV policy, please participate if you can, because I think these articles' need to be hammered out, possibly with deletion. Nightscream (talk) 02:52, 6 October 2010 (UTC)

Documentaries (again)
I went to the policy looking for information on "documentaries" and didn't find any. Then I looked at the talk archives, and wow - almost every volume of the archive has a dispute over whether referencing a documentary is usage of a primary source or a secondary source, and if a primary source, does quoting or otherwise referring to content from a documentary constitute original research.

Can there be a policy for this? Or does it need to be handled on a case-by-case basis each time? Documentaries are not simple compilations of first person accounts but through the process of compilation, editing, narration, and production constitute an new analytic work. patsw (talk) 16:36, 8 October 2010 (UTC)


 * What distinguishes a documentary from a biography? PL290 (talk) 18:55, 8 October 2010 (UTC)
 * Typically, a biography is about a single person. Biographies can range from reference (or standard) to a psychological or personal interpretation.  Documentaries can be biographical or cover an place, a period or a theme and also vary from being reference to being interpretation. patsw (talk) 20:47, 8 October 2010 (UTC)


 * Would it therefore be reasonable to conclude that it's not the fact of being a documentary per se, but rather, the nature, reliability and neutrality of the authors/producers/publishers, that determines its suitability as a source? PL290 (talk) 21:14, 8 October 2010 (UTC)
 * We evaluate many things when determining reliability... we look at the reputation of the author and of the publisher, we see what sort of reception the work itself received; we see if other sources use it as a reference. Etc. Video sources (such as documentaries) are no different... as is true of print sources, some video sources are high quality works of journalism, history or science, etc.... others are absolute crap.  We can not say "All documentaries are allowed" or "All documentaries are not allowed"... because it depends on the documentary.
 * That said... we do need to be careful when citing video sources, as it is very easy to end up discussing our own interpretation or analysis of what we see in a video source ... and that is OR. In this respect, video sources are similar to primary print sources... they are fine if used carefully, but they are easily abused. Blueboar (talk) 13:34, 9 October 2010 (UTC)

I'm not coming here with a live dispute to resolve. It just seems indicated to me that the discussion of documentaries appears in something like 15 out of 53 volumes of archives of WT:OR, and after all those pixels, there's no discussion of them actually in the policy. One would think some policy would be crafted as a consequence of those disputes. patsw (talk) 22:24, 9 October 2010 (UTC)
 * I just don't see the need... when someone misuses a video source such as a documentary to support OR, the flaw is in the act of OR, not in the fact that the source is a documentary. Documentaries can be used correctly or incorrectly, no different than any other source. Blueboar (talk) 01:00, 10 October 2010 (UTC)

Defining documents as sources
I have copied this from Wikipedia talk:Verifiability because JTSchreiber states that it belongs here rather than in Verifiability.

When making technical claims, formal defining documents are frequently the most reliable sources. In particular, secondary sources for computer instruction sets and operating systems frequently have significant errors. I believe that it would be desirable for Wikipedia to encourage the use of primary documents as sources of technical information in at least the following cases:


 * Computer Architectures
 * Wording of laws
 * Proprietary languages
 * Proprietary operating systems
 * Standards

I would not consider it appropriate to cite the above to establish notability. Shmuel (Seymour J.) Metz Username:Chatul (talk) 16:39, 7 October 2010 (UTC)


 * I would usually give more weight to a primary defining source, as Chatul explains, if a secondary source says something like "Standard XXX states ..." and then misquotes standard xxx. However, sometimes the same words have several meanings, and good secondary sources are useful for pointing out contradictions or ambiguities among several standards and specifications.


 * In the case of laws, different acts passed at different times by different levels of government often conflict, and they may be completely or partially nullified by court decisions. Thus secondary sources are better for determining how all the potential conflicts have been resolved. Jc3s5h (talk) 16:48, 7 October 2010 (UTC)


 * To clarify, I was suggesting the use of, e.g., http://thomas.loc.gov, as a source for the text of a bill at a particular point in time, or as a source for the history of the bill; I agree with you that it is not a good source for judicial and regulatory issues related to the bill. Shmuel (Seymour J.) Metz Username:Chatul (talk) 17:08, 7 October 2010 (UTC)


 * I will support this for electronics-related datasheets at least; I refer to these often, they are uncontroversial, and they have never (in my experience) been wrong. Awickert (talk) 16:53, 7 October 2010 (UTC)


 * Chatul, this discussion belongs on Wikipedia talk:No original research, as I wrote to you on other talk pages. "No original research", not "Verifiability", is the Wikipedia policy that sets the standard on the use of primary -vs- secondary sources.  -- JTSchreiber (talk) 05:17, 8 October 2010 (UTC) This comment was originally posted on Wikipedia talk:Verifiability.  Now that the discussion has been moved, the comment is no longer needed.  -- JTSchreiber (talk) 04:41, 9 October 2010 (UTC)


 * Agreed, for some of the technical domains you mention, when it comes to citing an elemental fact, why go anywhere else except the established single source (when it exists) (e.g. w3.org for HTML standards, oasis.org for certain XML standards) Secondary sources inherently carry at least a small risk of inaccuracy - their commentary might be based on a previous version, they might have gotten it 'wrong', they may have purposefully selected only portions of the source material - all to meet some need other than being the single repository of that information.   A site that could be identified as a true 'mirror', perhaps, but again, why risk it when trying to establish a reference that can survive with wikipedia for years/decades/etc.?  The primary source would probably not be a good meta-reference - market share, acceptance, criticisms, etc. of the domain at hand.  (e.g. facebook wouldn't be a good self source for it's popularity, but is the only true reliable source for specifications related FBML it's own markup language...)  Should this admonition be added at WP:Primary? Cander0000 (talk) 19:23, 8 October 2010 (UTC)

The title of this section is misleading; it seems to be based on a misunderstanding or at least ignorance of our standard terminology (also known as wikispeak). Technically you are arguing that certain types of primary sources should have precedence over secondary sources, rather than vice versa. Most of us have a natural (and normally also healthy) tendency in that direction, especially if we are well qualified to read the primary sources. The problem is that some primary sources are hard to evaluate and some editors are particularly poor at evaluating such sources or at understanding their own limitations. E.g. some technical documents have been superseded by others that may be harder to find. An internet standard (RFC) may have been replaced by a new one. An old social insurance law may be technically still in force in a country for a limited number of citizens, while for the majority a new law holds. A national or supranational (e.g. in the case of the EU) court may have decided that a law is not applicable, or that it must be interpreted in a certain counterintuitive way to make it constitutional. What looks like an innocuous word of normal English may turn out to be a technical term with a precise definition -- sometimes with surprising consequences. Duplicated signature after my talk was split. Hans Adler 13:34, 8 October 2010 (UTC)


 * No. The title of this section expresses exactly what I meant it to express; the issue that I am raising is how to classify defining documents used as sources in a specific context. I could have used a longer title, e.g., Defining documents as sources for exact wording and technical details, but it's not clear that a longer title would have been clearer. Nor am I arguing that primary sources should have primacy over secondary sources, rather, I am arguing that for specific types of citations the primary sources should have primacy. AFAIK there in nothing in wikispeak corresponding to the notion of defining document. Shmuel (Seymour J.) Metz Username:Chatul (talk) 15:44, 8 October 2010 (UTC)


 * I see. Apparently you meant "(Defining documents) as sources", not "Defining (documents as sources)". But that's not clear at all (try a Google search for "defining document" and you will see what I mean), and I didn't even understand it after reading what you really wanted. Must have been a bit dense. It might help to use a more common expression instead of "defining document". Hans Adler 07:43, 9 October 2010 (UTC)


 * I'm not sure what other terms would be appropriate for, for example, the relationship of the document IEEE Std 802.11-2007 to the IEEE 802.11 standard for wireless LANs. Given a choice between what that document says about WLANs, and what a secondary source says about WLANs, I would, in most cases, probably trust what that document says, not what the secondary source says.  Similarly, I would be more inclined to trust IBM's z/Architecture Principles of Operation, than a secondary source such as a book about z/Architecture, as an authoritative source about what instructions are supported in z/Architecture and how those instructions behave.  Both of the documents I cited are explicitly intended to be authoritative references for the standards/specifications in question, and are issued by organizations in a position to be considered the authorities on what the standards/specifications describe, so I, at least, would view them as different from, for example, a person's autobiography, where the person could credit themselves with the invention of special and general relativity, TCP/IP, and the artificial heart - the latter is a case where I could see a primary source not being trusted.


 * So, yes, this is one case where I think "we have [enough] editors with more than enough technical competence to evaluate primary sources" and where the primary source should be allowed. Guy Harris (talk) 05:47, 11 October 2010 (UTC)

Preferring secondary sources solves some of these problems because generally less interpretation is needed to reformulate things in an encyclopedic way. If a secondary source gets the content of a primary source obviously wrong, we can use our editorial discretion: ignore it and use better secondary sources instead. In some areas we have a large number of editors with more than enough technical competence to evaluate primary sources. In these areas we typically have a constructive editing atmosphere, often with occasional unanimous decisions to ignore policy and just do the right thing, which can mean ignoring the letter of a content policy and just writing the best article possible. It seems to me that we don't need explicit exceptions for special types of primary sources. Hans Adler 13:34, 8 October 2010 (UTC)


 * Secondary sources have the same issues of timeliness as primary sources. In many cases the secondary sources are also less accessible than primary sources, e.g., out of print books discussing standards or computer architectures.


 * A secondary issue that I should have raised is that when an article cites a defining document as a source, there is a case for requiring a link to the search facility of the issuing organization for checking whether more recent versions exist. Of course, in some cases an older version may be more appropriate than a newer one, e.g., when discussing the history of a product line. Shmuel (Seymour J.) Metz Username:Chatul (talk) 15:44, 8 October 2010 (UTC)


 * It all boils down to a special case of the old and tedious primary sources vs. secondary sources debate. I have always felt uneasy about the distinction because in my opinion these categories are simply not helpful in the Wikipedia context. We could just as well have a rule that favours hard-cover books over paperbacks or state-owned media over private media. Hans Adler 07:43, 9 October 2010 (UTC)
 * I think it is important to remind editors that this policy explicitly states that primary sources are allowed. We simply need to be very careful when using them ... we must avoid inserting our own analysis or interpretation of the primary source, and not draw our own conclusions from them.  Blueboar (talk) 13:43, 9 October 2010 (UTC)


 * We must also "avoid inserting our own analysis or interpretation" of secondary sources, and "not draw our own conclusions from them", right? Guy Harris (talk) 05:49, 11 October 2010 (UTC)


 * Usually, I think that we expect people to WP:IAR here. Unquestionably authoritative information, whether it is a "primary source" or even "self-published", is acceptable.  (After all, are you going to tell the IEEE that they don't know what their own standards are?  Or a person that he's not gay, because the statement on his official website hasn't been confirmed by a third-party secondary source?)  WhatamIdoing (talk) 21:19, 10 October 2010 (UTC)


 * The first of those two examples is pretty clear; a secondary source isn't going to be more authoritative about IEEE 802.11 than, well, the primary source IEEE 802.11-2007. One might be able to argue, in the second case, that if 99 44/100% of a person's sex life has been with members of the opposite sex, and that continues to be the case, a claim on their part that they're "gay" should either be taken with a grain of salt or, at minimum, would require a clear definition of what "gay" means to them.  (As per my earlier example, I could put up a Web site in which I claim to be the originator of special and general relativity, TCP/IP, and the artificial heart, but if somebody created a Web page for me claiming that I was the originator of all of those, citing my Web site as a reference, it would be quite appropriate to add "citation needed" to those claims.) Guy Harris (talk) 05:55, 11 October 2010 (UTC)

Requiring outside sources to comply with NOR
This keeps coming up: A completely unrelated source (e.g., a website) publishes material -- say, pictures and a description of a place or event. An editor cites the source. Some other editor removes the source, saying that the website failed to comply with this policy by failing to cite the source's sources, or by writing as an eyewitness.

Can we get a section here, or at least a short paragraph, that says "This policy does not require a published source to cite its sources. Unlike Wikipedia editors, outside sources are allowed to publish information that is not based on someone else's publications."? WhatamIdoing (talk) 21:39, 21 October 2010 (UTC)


 * Sadly, I think this is needed... I have run across the same issue. Blueboar (talk) 01:20, 22 October 2010 (UTC)


 * This isn't a NOR issue, but a RS issue. If you have people challenging websites which are of value, and which you believe are reliable, but have been removed due to lack of citations, take it to WP:RS/N.  Some sources ought to cite to achieve reliability (histories, for example), some sources don't need to.  But sources we use ought to conduct original research.  Sources which do not conduct original research are either PRIMARY sources (data, evidence) or TERTIARY sources (what wikipedia is, and dog shouldn't eat dog). Fifelfoo (talk) 01:24, 22 October 2010 (UTC)


 * What WhatamIdoing is talking about is a common misunderstanding of what we mean when we say "no original research". it stems from the fact that, way back when, they chose a somewhat confusing name for this policy.  Now we are stuck with that choice.
 * If you go back to some of the early versions of this policy, we used to have a statement that explained this better... we talked about not making Wikipedia "the first place of publication" for any information, idea or thought. That one line made what we are talking about clear.  That line got lost somewhere.  I think it would help to bring it back. Blueboar (talk) 01:41, 22 October 2010 (UTC)
 * I agree it’s an RS issue. I don’t quite agree with the primary sources bit, but that’s beside the point. Blueboar, are you thinking of the “If you are able to discover something new…” sentence?--SaskatchewanSenator (talk) 01:52, 22 October 2010 (UTC)
 * No... I remember something that said Wikipedia should not be the "first place of publication" or very similar words. I will see if I can find it. The point being that if it was published somewhere else first, then it did not "originate" on Wikipedia, and thus was not WP:OR. Blueboar (talk) 02:02, 22 October 2010 (UTC)
 * Let us know if you find it. It might be worth bringing back.--SaskatchewanSenator (talk) 02:59, 22 October 2010 (UTC)
 * This is not a reliable source issue. This is an "editors misunderstanding the title" issue.  Editors keep saying things like, "That non-Wikipedia writer physically went to the site and looked at the object, and published a magazine article about it.  Going to the site and writing about what you see is "original research", and Wikipedia has a 'No Original Research' policy."  WhatamIdoing (talk) 02:12, 22 October 2010 (UTC)
 * Yes, and those editors are quite clearly wrong. Where do you want to get your trout reserves for the copious slapping required from? Fifelfoo (talk) 02:22, 22 October 2010 (UTC)
 * No matter how well-deserved it is, trout delivery takes time and energy. I'd been hoping we could figure out a way to put this explicitly into the policy, so that I could just point them at a sentence.  Typing 'WP:YOUREWRONG' seems so much easier than dealing with the trouts.  ;-)  WhatamIdoing (talk) 04:44, 22 October 2010 (UTC)

peer-reviewed journals by other than university presses
I think peer-reviewed journals published by other than university presses should be considered reliable (if otherwise reliable). A substantial number are published by non-university presses, yet, typically, they're managed, edited, and peer-reviewed by professors. However, the policy presently states, "In general, the most reliable sources are peer-reviewed journals and books published by university presses; . . . ." I propose editing to "In general, the most reliable sources are peer-reviewed journals; books published by university presses; . . . ." It's possible that a Tony's Pizzeria puts out a folio with two chefs across town as peers but such journals are probably rare enough that "[i]n general" as the opening clause sufficiently protects WP. Nick Levinson (talk) 03:05, 29 October 2010 (UTC)
 * I think you're reading the parenthetical clauses wrong, but your proposed clarification immediately clears up any possibility of misreading. BTW, Tony's Pizzeria Of Political Economy regularly puts out journals.  Some are academically peer reviewed, some aren't.  They're generally honest about labelling them as such.  I would accept, for example, peer reviewed articles from the socialist or the ultra-right wing capitalist presses as genuinely peer reviewed if they state as such.
 * I support this editorial change, and am implementing it. Fifelfoo (talk) 03:21, 29 October 2010 (UTC)

Transitive synthesis
Is there any sort of transitive rule of synthesis? For instance, if source #1 says that A is B, and source #2 says that B is C, is it synthesis to say that A is C? For a more concrete example, if source #1 says that "all mammals are warm blooded", and source #2 says that "cats are a type of mammal", would it be considered synthesis to say that "cats are warm blooded"? A bit confused as to whether these types of transitive scenarios were meant to be excluded as synthesis. ← George talk 18:13, 29 October 2010 (UTC)


 * Yes... such statements are synthetic OR. Now, the reality is that we often overlook and ignore such OR when dealing with a very obvious conclusion in a non-controversial article, while we won't do so if the conclusion is not that obvious or if the topic is controversial (and it is important to remember that what may be obvious to you, may not be obvious to someone else). If a conclusion is worth mentioning in an article, some source somewhere will state it... explicitly... if not, it probably isn't worth saying. Blueboar (talk) 03:57, 30 October 2010 (UTC)
 * second Blueboar generally. In humanities and social sciences one should avoid at all costs the synthesis of A is B in author one and B is C in author two thus A is C. This is because humanities and social sciences definitions of objects of analysis are necessarily imprecise and one of the primary original research responsibilities of specialists is to ensure that like objects are being compared when making such an evaluation. Fifelfoo_m (talk) 04:57, 30 October 2010 (UTC)


 * Thanks for the feedback guys. The specific scenario that is currently under discussion is this: Between the West Bank, Golan Heights, and East Jerusalem, there are a couple hundred Israeli settlements. There's been a lot of edit warring in articles about individual settlements about how to describe their legality. We've been attempting to build consensus for a single, succinct sentence describing the legal status of such settlements. Citing this source, we came up with a couple proposals including "Israeli settlements in (the Golan Heights/the West Bank/East Jerusalem) are considered illegal by the international community, but the Israeli government disputes this" and "The international community considers Israeli settlements in (the Golan Heights/the West Bank/East Jerusalem) illegal under international law, but the Israeli government disputes this." This phrasing is quite similar to the BBC's own style guidelines, which state in part: "When writing a story about settlements we can aim, where relevant, to include context to the effect that 'all settlements in the West Bank, including East Jerusalem, are considered illegal under international law, though Israel disputes this'."
 * Now then, back to my original question, would it be synthesis to include our proposed sentences in articles about specific settlements, when the source discusses the legal status of all Israeli settlements (in the West Bank, Golan Heights, and East Jerusalem) broadly, not listing them by name? We were coming up with proposals thinking that, because reliable sources label all Israeli settlements in these three areas illegal, we could include the line in the articles of individual settlements in those areas. Will we need to find individual sources for each of those couple hundred settlements, explicitly mentioning the settlement by name? ← George talk 07:24, 30 October 2010 (UTC)
 * Well, I think this is WP:NPOV rather than an WP:OR issue, since there is no synthesis involved in applying the statement "every X is Y" to every X. The problem comes in attributing blanket statements to the international community.  You can say that UN Resolution XYZ says something about the settlements, but not that the UN speaks for the international community.  And it would be a synthesis to build a statement about the consensus of the international community out of a finding by a UN agency, a statement by the Arab League, etc.  You could say that the New York Times says that the international community says something, but then we run into a problem of WP:UNDUE as every statement for or against the settlements would have to be included in every article.  It may be best to note in settlement articles that, as with every settlement, the legality of settlement X is disputed, or just that the settlement is part of the larger settlement dispute, with a link to a more comprehensive, centralized treatment of the issue.  This would also prevent each article from being a WP:POVFORK or WP:COATRACK.   RJC  TalkContribs 14:40, 30 October 2010 (UTC)
 * Hmm, I've actually been concerned about WP:UNDUE, but from the other side of the coin. The problem we have is that almost every group we have sources for considers the settlements to be illegal - the United Nations, the International Court of Justice, the International Red Cross - as well as almost every country in the United Nations (by resolution vote, or per individual sources). The only country that has consistently argued that the settlements are legal under international law is Israel. So my concern is that by excluding the (super)majority view that the settlements are illegal, or simply summarizing it as a dispute (which gives equal weight to both sides of that argument), we would be violating WP:UNDUE. This probably isn't the best venue, but I'm interested in the thoughts of other uninvolved editors on the subject. Maybe stop by my talk page if you have a chance? ← George talk 18:07, 30 October 2010 (UTC)


 * I don't think it would be OR to include this sentence—"The international community considers Israeli settlements in (the Golan Heights/the West Bank/East Jerusalem) illegal under international law, but the Israeli government disputes this"—in articles about settlements not singled out by name by the sources. It is not a contentious issue in the sense of widely disputed. One way round it, though, would simply be to link to Israeli settlement, rather than add the illegal status to each article, or perhaps have an infobox on each article that explains the legality issue. SlimVirgin  talk| contribs 18:21, 30 October 2010 (UTC)


 * I didn't mean that the relative weight of each side on the settlement debate created the problem with WP:UNDUE, but rather the space devoted to the issue in each article on every settlement. Otherwise, we run into articles like, "Kiryat X is a settlement located in the Golan Heights with a population of 103.  The legality of Kiryat X is condemned by ..." followed by a long discussion, ending in "Israel disagrees."  An article on a settlement should discuss the settlement, pointing to the broader dispute without rehearsing it in all of its details.  RJC  TalkContribs 18:44, 30 October 2010 (UTC)


 * Yes, good point. It could all be left to the infobox. There's Infobox settlement, or someone could create Infobox Israeli settlement.  SlimVirgin  talk| contribs 18:51, 30 October 2010 (UTC)


 * Great feedback folks. Really we're just trying to find some boilerplate solution that stops edit warring at what may become literally hundreds of contentious articles on Israeli settlements. I think adding it to the infobox instead may be a really good option, so I'll try to put together a proposal for that. If editors are interested, the ongoing discussion can be found here. Cheers. ← George talk 19:16, 30 October 2010 (UTC)

George has misrepresented this issue here, he said: "if source #1 says that A is B, and source #2 says that B is C, is it synthesis to say that A is C?" which is not what we want to do. --Supreme Deliciousness (talk) 21:39, 30 October 2010 (UTC)
 * Just a general note ... it incredibly easy to create an improper synthesis without meaning to do so.  I have no idea if this relates to the current discussion, but I say it in case it might.  It helps when "accusations of OR" start flying.  Those challenging the material need to remember that OR may not be the intent... and those adding the material need to realize that others are not mind readers and so won't know the intent unless you explain it to them.Blueboar (talk) 22:14, 30 October 2010 (UTC)
 * Hoping not to defocus this discussion here too much, on the side-issue of support for assertions of governmental legitimacy leading to assertions of OR & POV, I'll suggest that anyone interested look at the Timeline of Philippine Sovereignty article. Comments on talk page there, please. Wtmitchell (talk) (earlier Boracay Bill) 23:38, 31 October 2010 (UTC)

OR or weight or what?
AT WP:VPP I contended that putting in an ethnic group without having anyone saying it was interesting was original research on the basis that the first person to show an interest in a subject is doing original research. This in effect means that a primary source like a government form could not be used because they do not particularly single a person out. For the same reason one should not say they have blue eyes just because they have blue eyes in a photo, the photo does not comment specially on the blue eyes even if it is a primary source. And we should not use court orders that have not been commented on, searching them out is original research. However this might all also come under WP:WEIGHT, I feel though it is an area that seems to be coming up more frequently and we need a good direction in the policies about it. Is original research the right place and if so which bit? or what policy deals with this? or do we say any old government record is allowable as a reliable source for making up bits of articles even if no-one thought them interesting before? Dmcq (talk) 11:05, 7 November 2010 (UTC)
 * You seem to be arguing that we can't use primary sources on Wikipedia, which is specifically not the case when primary sources can be interpreted by any reader without background knowledge. For example, in most cases, an ordinary photo of a person with blue eyes would be sufficient to prove that the person has blue eyes.  Now, I wouldn't actually include that in a Wikipedia article, because it's not sufficiently important to include, but it does meet the criteria for WP:V and WP:NOR.  Qwyrxian (talk) 11:26, 7 November 2010 (UTC)
 * Of course it is verifiable. A lot of original research is verifiable if one follows the original researcher. The other argument you gave is weight, but perhaps the fact discovered by the original research would have weight once displayed like some politician took drugs. But should WIkipedia do that before anyone else? Both I think are covered by original research. Dmcq (talk) 11:54, 7 November 2010 (UTC)


 * "A lot of original research is verifiable if one follows the original researcher"? I don't understand what that is supposed to mean. The kinds of original research that is not acceptable are
 * Experiments, analysis, etc., by a Wikipedia editor because Wikipedia editors are not reliable sources
 * Experiments, analysis, etc., accessible to a Wikipedia editor but which cannot be verified because the source is not accessible to the public (that is, published)
 * Experiments, analysis, etc., by sources that are not much more reliable that a Wikipedia editor (more precisely, sources which fail the criteria at WP:RS
 * Deciding if a statement which appears in a reliable source should be added to an article is a matter of neutral point of view, weight, and good writing. It is not original research to decide that a statement in a reliable source is worth adding to an article (unless it is added in a way that makes an implication not stated in the source). Jc3s5h (talk) 16:50, 7 November 2010 (UTC)


 * By "A lot of original research is verifiable if one follows the original researcher" I mean if somebody looks at a photo and writes down that a person's eyes are blue then yes it is verifiable that the eyes are blue. The photo is a primary source. I'm saying the person who wrote down they had blue eyes has done original research. As to your example you seem to be saying it is not original research to dig around in a persons past investigating and then come up with some record that they smoked hash and publish that on Wikipedia. You say that the only reason to not stick that in is neutral point of view, weight and good writing. I'm saying it is original research if nobody has noted it as interesting. A primary source can indicate something is of interest but a secondary source is much better for that. Some record amongst a hundred thousand others with no particular significance attached to it except by a researcher is not noted as interesting. Dmcq (talk) 17:08, 7 November 2010 (UTC)


 * Deciding what is interesting is the essence of writing. Claiming that editors can't decide which information, derived from reliable sources, is interesting enough to add to an article is saying editors can't write at all and Wikipedia should be destroyed. This claim is a threat to the existence of Wikipedia and must be resisted by every allowable means. Jc3s5h (talk) 18:10, 7 November 2010 (UTC)


 * Exactly what kinds of things are there where an editor decides it is interesting when nobody else says so that are so important to the existence of Wikipedia? You sound to me like one of those Dr Who fans I mention below. Dmcq (talk) 19:49, 7 November 2010 (UTC)


 * A nit-pick re the idea that a photograph verifies that someone has blue eyes. Not quite... the person in the photo could be wearing colored contact lenses.  Alternatively, the photo could be doctored.  Because of this, it is not verifiable that the person has blue eyes (only that this photo shows them with blue eyes). Blueboar (talk) 18:20, 7 November 2010 (UTC)


 * True, but there's lots of stuff like that about where some fan notices that Dr Who wore something in an episode or suchlike and insists it must go into Wikipedia and if you watch the episode carefully you'll see it so they have a reliable source.My point about that is that not only is it not interesting because no reliable source has noticed it but also that it is original research because the episode was not made with that as even a tiny minor piece of its plot. Dmcq (talk) 19:49, 7 November 2010 (UTC)


 * Terminology is getting mixed up here. Original research is not about who did the research first in Wikipedia. The statement above that "the first person to show an interest in a subject is doing original research" is incorrect. Every article originated at some time or other and every fact in an article originated, and that's by simple definition of chronology. That isn't objected to under Wikipedia policy or there wouldn't be a Wikipedia because the first article would never have been posted regardless of content. Rather, if a source reports that a person was born in northern Europe and from that alone an editor concluded (without a photo) that the person is blue-eyed, even if the person really is blue-eyed, the editor's conclusion is original research, because the source doesn't say so and so the statement originated with the Wikipedia editor.


 * Suppose you wrote an article about a basketball player, and notable ones tend to be tall. You read in a source this sentence: "The most boring fact about this player is height: seven feet exactly." To the source's author and maybe to all the readers of the source, the fact is boring. Maybe everyone on the team is that height. But most people are not that tall and it probably helped that person's career. In Wikipedia, "[t]he question of content coverage within a given page is governed by the principle of due weight and other content policies." So to Wikipedia's readers it might be relevant. It likely can be reported in Wikipedia. Whether a fact is reportable in WP requires that it not be trivial, but in this case it probably wouldn't be trivial.


 * Court orders and other government records can be used under the policy for using primary sources. For example, if you find a criminal conviction on a major charge for a person not reported in Wikipedia as guilty of anything, the conviction you dug out is generally reportable even if no one wrote a secondary source about it, although being a primary source it must be handled with more care, e.g., quoting rather than interpreting (the care requirement is one reason why it's often better to seek out a secondary source if one exists on the point). On the other hand, a police officer's notebook showing that the officer gave a warning against jaywalking to someone when no traffic was coming and the lack of a secondary source about it means the warning is probably too trivial to report in Wikipedia, even if the person was dead from unrelated causes so that the BLP policy restrictions didn't apply.


 * The critique of a photo as evidence of someone having blue eyes is correct regarding contact lenses but is indistinguishable from critiques of written sources as to retouching: The highest-quality written sources are susceptible to error resulting from the textual equivalent of photo-retouching. For a news investigation about a person who liked to sue the press, a reporter was told by an editor not to report that the person has one head and two hands unless the reporter counted them personally. The lead came back, "The one-headed, two-handed [person] ...." The editor took out the words but said thanks for counting.


 * Nick Levinson (talk) 21:33, 7 November 2010 (UTC)
 * Some caveats. Court orders, transcripts and government records cannot be used in BLP articles . We do not want people digging about looking for criminal convictions; if no secondary source has noted it as a significant event in a person's life, WP does not either.
 * Nick is right that text sources are susceptible to retouching error too, but they remain much less prone than photos to original research than permitting editors to draw their own conclusions about eye colour, for example. Or skin colour.  I can't remember where it was but there was recently a long discussion about determining the race of article subject based a photograph. The consensus was that we shouldn't.   --Slp1 (talk) 22:33, 7 November 2010 (UTC)


 * I don't understand "Every article originated at some time or other and every fact in an article originated, and that's by simple definition of chronology". This was in relation to my assertion that the first person showing an interest in something is doing original research. Perhaps they considered original research as a Wikipedia only term? I was saying that we should only be putting things into Wikipedia that have been noted in reliable sources and the reliable sources will have done the original research. Original research that is not allowed is that done by Wikipedia editors. Of course Wikipedia editors get interested in things and write about them, but they should only write about things reliable sources have noticed before them. Dmcq (talk) 23:00, 7 November 2010 (UTC)


 * I think we now agree on the essential issues.


 * The correction of my comment is accepted for now. I recall that a conviction could be used but not a charge, but I don't recall where it said that, so I can't compare the statements, and so I accept the ban. And certainly court transcripts are much more problematic than are final judgments.


 * Thank you. Nick Levinson (talk) 23:29, 7 November 2010 (UTC)