Wikipedia talk:Verifiability/Archive 24

Fee-based source citations
In the section on citing sources with examples, there should be something said about whether a citation, which is only available as a fee-based online subscription is appropriate. For example, there are now online "publications" which have no free physical counterpart. The New York Times you can read freely at a library, so providing a link to the fee-service of it would be appropriate to me as you can also get it freely in physical form in some way (provided the citation specifies the date, page, etc). However there are websites who publish entirely online, for which you must pay to view, and have no free counterpart. We should address this area. Wjhonson (talk) 04:40, 26 December 2007 (UTC)
 * What specifically do you feel needs to be addressed? I see two possible issues and I'll give you my perspective on each, but I think this discussion belongs elsewhere:
 * Their actual use: I see no problem with this use. Just because something isn't publicly available without a fee does not mean it isn't verifiable. As a matter of fact, you may need to go to a far away land and read an ancient manuscript to verify something on Wikipedia, but it is still verifiable. Furthermore, citations do not need web links, so links to pay sites are simply links that you need to pay to access (your choice). But, if we are talking about reliability, I suggest we head over to the WP:RS talk page and have this discussion there...
 * Format of the citation: There is no set standard, per se, for citation use, so specifying that should be talked about on the WP:CITE talk page. If you want some sort of warning that the link is to a pay cite, then you may need to visit the template talk page.
 * ...well, my thoughts on the subject anyway. — BQZip01 —  talk 07:01, 26 December 2007 (UTC)
 * Wjhonson, could you provide a few examples? I'm not aware of a fee-based source that would be considered a reliable source that is not available in libraries free of charge. Vassyana (talk) 21:07, 26 December 2007 (UTC)
 * BQZip01, an unpublished ancient manuscript (what you describe) would generally not be considered a reasonably verifiable source. Extremely rare sources or otherwise exceedingly difficult to verify sources are rejected in practice, as far as I have seen. Very often, people state that the source is not reasonably verifiable. However, people also (correctly IMO) assert that such a rare and/or obscure source is counter to NPOV, since our purpose is to summarize the general body of published material, not to gather together obscure and extreme minority references. Vassyana (talk) 21:07, 26 December 2007 (UTC)
 * Everything has a cost. Going to a library (near or far depending upon the material) costs you time/money and the library probably paid for the material (whether paid for by charity or you).  Libraries do get some donations which they or volunteers have to process into their collection.  What matters is whether the information is verifiable, not whether a specific person can get it without payment.  -- SEWilco (talk) 21:18, 26 December 2007 (UTC)
 * My point exactly SEWilco. Thank you. As for obscure texts, it all depends on what you are referring to. The actual print of the Magna Carta would be an interesting thing to add if drafts were found. Getting ahold of these to check would be exceedingly difficult, but not impossible. If you want to say, the draft of the Magna Carta said XYZ and have pictures as proof on some pay website, that's fine by me. Saying that it is what was originally intended is not acceptable. In short, it all depends on how it is used. — BQZip01 —  talk 07:21, 27 December 2007 (UTC)
 * Yes, how the material is used does matter. I stumbled upon an image of Bradford's history of the Plymouth colony (Image:Of Plimoth Plantation First 1900.jpg) which is older than the state's image, but I used it merely as a better illustration for the article.  The image shows the document has a different title than our Of Plymouth Plantation article, but I'm not using the image as a source.  I will be writing a paragraph about the various names given to the document, but the source will be the text of the book (actually two books) rather than the image.  Such an image could be supporting material, but photographic evidence is not required, we only require that the source of information be provided.  -- SEWilco (talk) 17:25, 2 January 2008 (UTC)


 * In general, I see 'verifiable' as meaning that some reasonable sub-set of Wikipedia editors/readers, who are not members of some restricted group, can access and read the source. Fee-based on-line sources are like books and periodicals, someone has to pay for them. If the fee-based services, books or periodicals are available in at least some major city/university libraries, then they are presumably accessible to enough otherwise un-related Wikipedians to be acceptable as sources (whether the sources are reliable is another consideration, independent of the question of accessibility). Payment of a reasonable fee for access to an on-line resource is comparable to buying a book to use as a source on Wikipedia (which I have done several times). There will always be marginal examples to debate, but I think the general rule should be: if the on-line resource is available free to visitors in a number of major city and university libraries, and the fee for access to individual articles is comparable to the price of books and/or hard-copy periodicals, then such sources are acceptable if they otherwise meet the criteria for reliable sources. -- Donald Albury 13:17, 27 December 2007 (UTC)

The specific reference is to a website called http://www.stirnet.com This site is being used as a reliable source, that is, it is being cited inline to support certain statements. The site purports to base it's own individual reconstructions upon underlying sources which themselves are generally available. However the problem is, that unless you pay to access their reconstruction pages, you cannot know what statements are there made, nor can you know what sources were used to make those statements. And so using this site, to support any statements here, is not verifiable in the sense you all above are supporting. That is, you *cannot* go to a library anywhere in the world and view what the site is saying or not saying and therefore you cannot verify the link, or the statement, as coming from this source. It's not merely a question of whether the web site adequately represents the underlying primary source, the question is, can we allow support from a site, which we cannot independently verify in any way, except by paying them? I say no. If an editor, who HAS paid, is willing to cite the underlying freely-available source and perhaps *in conjunction* that they got it from this pay site, I'd support that, but it's a bit long-winded isn't it? Wjhonson (talk) 09:02, 30 December 2007 (UTC)

I think we need to distinguish two issues here: accessibility and reliability. All secondary and tertiary sources are presenting material from other sources that has been selected, edited, rewritten and/or analyzed. We don't require that all of the prior sources used in a cited source be accessible to the Wikipedia reader. The material that stirnet is using for its reconstructions is probably from primary sources, which we couldn't cite in WP. The issue with stirnet is whether it is a reliable source. I think that is debatable, but having access to the sources used in their reconstructions would not necessarily make the site 'reliable'. -- Donald Albury 11:14, 30 December 2007 (UTC)


 * I'm pretty sure this has been raised somewhere before, probably WP:AN or WP:ANI, unless areference is to an entirely free website, there is always some sort of fee, be it the time to go to the library, buy a book or whatever. There is no requirement that all references be immediately checkable by all users, or we would be restricted to websites that do not charge.  I would suspect that a number of members of Wikipedia would maintain a subscription (either on their own account or via a library of whatever sort or their place of work) to the particular site you are questioning - try asking at Wikipedia talk:WikiProject Baronetcies and there is somewhere (the name escapes me for the moment) a project or page whose sole purpose is to allow you to request verification of subscription sources.  If after exhausting all avenues of enquiry no-one can be found to verify it, may be there would be an argument for deleting the reference, but the mere fact that it is a subscription site is not sufficent.  Pay sites should not be listed in an "External links" section, but links being used as references are specifically exempted from WP:EL.  David Underdown (talk) 18:15, 30 December 2007 (UTC)


 * WikiProject Resource Exchange is the project I was thinking about above. David Underdown (talk) 18:23, 30 December 2007 (UTC)

Firstly stirnet does not cite primary material. It's citation are all to secondary works, such as compiled visitations, Burke's Commoners, or The Complete Peerage, etc. Secondly, the distinction David is trying to drawn is not-wikistandard. We are all in agreement that a person could go to a library with a work, or a public source with a work and consult it. The problem with stirnet is that there is *no* alternative to paying them. None. Zero. Zilch. It doesn't exist anywhere. The only way to get at their information is to pay them. I believe that violates the free nature of our work. Free does not mean, always and everywhere free. It means sometimes and somewhere free. Stirnet is nowhere and notime free. Wjhonson (talk) 20:35, 30 December 2007 (UTC)


 * "The problem with stirnet is that there is *no* alternative to paying them. None. Zero. Zilch.  It doesn't exist anywhere.  The only way to get at their information is to pay them.  I believe that violates the free nature of our work.  Free does not mean, always and everywhere free.  It means sometimes and somewhere free." I'd be interested where you get that opinion from Wikipedia policies/guidelines. Near as I can tell, it is your opinion exclusively. Your opinions notwithstanding, Wikipedia is still free even if the sources are not. This is standard for many academic/professional publications.  — BQZip01 —  talk 02:50, 31 December 2007 (UTC)

You are studiously avoiding the point. Verifiability requires the ability to verify the source material. Wikipedia is not in the business of creating new commercial opportunities for some website, we are not here to drive dollars to some website. A website whose data can only be viewed by paying them, i.e. which has no free counterpart is not part of our mission. Wjhonson (talk) 04:29, 2 January 2008 (UTC)


 * As several people have noted, accessing information often costs money. Until very recently, the New York Times online archives were available online only to paying customers. There were plenty of Wikipedia articles with NYT citations, where the link was to an article behind the paid firewall. Theoretically, you could have verified those articles for free in a library, but I suspect that very few Wikipedians could have reached such a library without incurring travel costs. No one ever suggested that the fee-based nature of the source had anything to do with the reliability of the New York Times.


 * So, in my view, if a source is reliable, the fact that you need to pay for access to it is irrelevant. Marc Shepherd (talk) 13:24, 2 January 2008 (UTC)


 * So far as I can see, not one person agrees with your basic stance, that a fee only service is automatically a non-starter for use in referencing. Unless or until you can get a consensus for this stance, please stop removing references from Gore Baronets, if you have removed other refrenes from other articles, I think you should replace them there as well.  David Underdown (talk) 14:08, 2 January 2008 (UTC)


 * Let me play devils advocate and throw in something like scientific journals. Personal subscriptions to these are easily close to $1000/year, and while libraries can get reduced rates, the cost is prohibitive enough that outside of university libraries (which are not publicly accessible), most community-based libraries aren't going to spring for that information.  Articles can be obtained on the web but for cost ($12-25 or so for an item, depending).  Do we prevent such sources being used since they are not "freely available"?  Of course, counter to this is that I can verify freely that such the article does exist and likely can also check the abstract to make sure the material is relevant to the WP article in question; other on-line services do not have this feature (you may have to be a paying member even to search).  --MASEM  14:32, 2 January 2008 (UTC)


 * Well, in the particular case of Stirnet, if you click on the links, you can see the article for a couple of seconds, and it is then replaced by something saying you need to log in (or at least that was my experience earlier), so you can indeed at least check the relevance. David Underdown (talk) 15:46, 2 January 2008 (UTC)


 * Legal citations often end up pointing at commercial legal databases which are available only to subscribers. Should a Wikipedia article not include a citation which can be used to find information, even if to a fee-based service such as Westlaw (using West American Digest System citation) or formerly-for-fee New York Times archives?  The NYT citations have become more useful after the fee was removed, and the Wall Street Journal might soon be doing the same...should WSJ archive links be forbidden because there presently is a fee?  I think stirnet is OK.  The above description is that it provides descriptions without citations, with citations available upon paying a fee; I view this as being similar to providing free abstracts.  But what is important is that verifiable sources are available even if for a fee.  Actually, in the case of some information such as who is a child of whom, that is a simple fact for which no interpretation is needed and a primary source document should be acceptable as well.  -- SEWilco (talk) 17:07, 2 January 2008 (UTC)


 * If the content happens to be on a fee-required web site should not be relevant. I searched for the words "stratigraphic abnormality" and got links to SpringerLink, Blackwell Synergy, and Elsevier.  I might be able to touch one document within 2 hours and two of them within 24 hours at costs of $7 and $400; I don't know if I can find the third document except through the online service.  Such commercial publication sites could easily create online fee-based services with some investment (risk) of their reputation (whether they would be profitable is irrelevant to this discussion).  The fee should not be an issue for citation, all we require is that the information be verifiable.  -- SEWilco (talk) 17:38, 2 January 2008 (UTC)


 * Which I think agrees with the point I make above: as long as the existence of an article and a summary of what it is about can be verified for "free" (searching for it is not something that is charged by the providing service), though the article itself may be behind a monetary barrier, it is still a usable source; one of course must assume good faith that the editor that included that source has access to the text of the source and thus can summarize as needed for the purposes of the article. This would then disclude services that prevent you from even searching its archives unless you pay for a subscription or the like.  --MASEM  17:44, 2 January 2008 (UTC)

There is *no* free alternative to the site mentioned. NONE. None at all. No free alternative. Can we all understand this point please? All of the examples above are moot because each one of them *does* have a free alternative. In this case however this is NO free alternative. Can someone address this point exactly? Thanks. Wjhonson (talk) 21:18, 2 January 2008 (UTC)
 * But even for sites where there may be a "free" alternative, there is a cost in verifying it, that's why most of us don't seem to see this as an insurmountable barrier. Personally, I have previously paid for a reprint of a journal article which I intend to use as a reference (the article writing has got a bit stalled, see User:David Underdown/Last Night of the Proms).  In the particular case taht prompted all this, for £5 you get 7 days access to the sight, which would be more than sufficient (I'd have thought) to verify all places the site may be used in Wikipedia.  Even if an equivalent free source were available in e.g. my local library, it wold cost me £1.80 to get there and back, and or a reasonable amount of time.  Free in the Wikipedia ense means that the information is free to be re-used once we have it here, not that there is no cost in placing that infomration here.  David Underdown (talk) 21:48, 2 January 2008 (UTC)


 * No one is disagreeing that there is a cost to exist. You have to eat, so there's a basic cost just for you to be here.  That's moot. We do not, have not, and will most likely not, be in a position to support a fee-based site like stirnet.  It's just not gonna happen.  If or when stirnet decides to post it's source list independently of it's compiled pedigrees, in such a fashion that the specific cited sources *per pedigree* can be consulted without fee, then I'd change my mind.  Any editor who already *has* access to stirnet can stop being lazy and post the detailed citations *to* the underlying sources using stirnet as an overlying source if they choose which would be fine as well, including the lazy-link to stirnet.  I'd have no problem with that either.  However, the problem lies in simply posting a link to a stirnet pedigree, without any other indication of how to retrieve the information for-free.  For free does not mean *absolutely free* it means relatively free.  That is, you have to drive, you have to eat, you have to breathe and wear clothes and take a bath, but you can get it from the library.  Library equals free.  Every thinking person agrees with me ;) Wjhonson (talk) 23:17, 2 January 2008 (UTC)
 * Wikipedia creates free content, there has never been an obligation that we only cite free content. We prefer citations to free content, as they are generally more useful, but if all available sources are fee-based, then that is what we are going to have to cite.  There is nothing wrong with referencing fee-based sites when neccesary.  Dragons flight (talk) 23:27, 2 January 2008 (UTC)
 * There seems to be a fairly strong consensus that, from a policy perspective, there is no categorical objection to sources that are hidden behind a fee-based firewall. You are raising some additional questions, essentially, "How do we know it's reliable if we can't see the source for ourselves?" Those are probably questions that cannot be resolved on this talk page to your satisfaction. Marc Shepherd (talk) 23:31, 2 January 2008 (UTC)
 * Have you tried asking the person who originally added the citations if they could give more detail (there's no guarantee that they still have the article watched)? Equally, if they could not themselves immediately verify the original sources, then they should not list that as their source because they cannot gurantee that the information is actually there.  Indeed, if you look closely at the site FAQs, there is a fee free option - if you make "a meaningful contribution to the site" you can be granted a free subscription period.  David Underdown (talk) 08:43, 3 January 2008 (UTC)

It's a little much to ask our editors to contribute work to a fee-based subscription service just for the priviledge of verifying it's sources. Don't you think?Wjhonson (talk) 07:30, 6 January 2008 (UTC)

Excessive doubt cast on self-publication
From the policy page:


 * Anyone can create a website or pay to have a book published, then claim to be an expert in a certain field. For that reason, self-published books, newsletters, personal websites, open wikis, and blogs are largely not acceptable as sources.

This seems to be understood as claiming that books and sites that are not self-published are considered more reliable than those which are. But the standard used by commercial publishers, in deciding what works to publish, is not one of truth or verifiability.

The distinction between a self-published book and a conventionally published book is that in the latter case, the publisher has invested money in the publication, under the belief that it will turn a profit on the sales of the book. The publisher's choice does represent a decision independent from the author's about the quality of the work -- but the quality thus judged is profitability, not honesty or accuracy.

From Wikipedia's perspective, the profit of publishers is a remarkably arbitrary criterion on which to judge whether a work is worth using as a reference. In the days when print publication was the only option, this would not have been such a big deal -- but today, when Web-based publication is a large fraction of all "publication", it is a huge big deal.

Consider the following ways that a work might be published today:


 * 1) The author writes a book or article and it is accepted for conventional publication because the publisher thinks they can make a profit selling it.
 * 2) The author becomes a paid contributor to a large profitable advertising-funded Web site, such as About.com or Salon.com, and the work is published there.
 * 3) The author publishes the work him- or herself on his or her own advertising-funded Web site, such as BadAstronomy.com.
 * 4) The author submits the work to a non-profit edited Web site, such as TalkOrigins.org, which publishes it.
 * 5) The author publishes the work on their own non-profit Web site.
 * 6) The author publishes the work on other electronic media, such as a regularly posted FAQ document on a Usenet newsgroup.
 * 7) The author releases the work as open-source content and it is republished by many people, both for-profit and not, as with the documentation that accompanies an open-source project such as OpenBSD or the Linux kernel.

In which of these cases do we really think that the method of publication has all that much to do with the reliability or verifiability of the work? The only case, it seems to me, where publication definitely has something to do with verifiability is a totally different one: peer-reviewed publications, where the decision to publish is based not on profit but on review of the work by independent experts in the field.

Other than that specific case, I do not think that we are on solid ground discriminating against self-published works. It's possible that we want to get at the idea of factual review of a work; but this is entirely separate from the business model (or lack thereof) under which the work is published. --FOo (talk) 03:12, 30 December 2007 (UTC)


 * the point is that a publisher has a reputation--how much of one depends upon the publisher,and therefore has some degree of editorial control over the material. In the case of a major academic or society publish,they have a great deal of control--for books and journals, almost always serious formal peer reviewing. For a commercial publisher, they have at least fact checking--the reliability and the quality depends on he publisher.For fiction even, they have a certain established reputation for quality or at least public interest. self-publishing normally has none of these-the entire responsibility is upon the author. But you are right that the details matter: not all peer-reviewed journals are of equal quality, and some websites maintain a high reputation. But it remains a good general rule. In particular, for published books in conventional formats, self-published does tend to mean unreliable.  DGG (talk) 04:49, 30 December 2007 (UTC)


 * You seem to be saying a few different things at once. First off, I completely agree with you regarding peer-reviewed academic publication. In peer-reviewed journals, "publication" is not just about getting your words in front of a market; it's about having experts in the field vet those words for both truthfulness and importance. There are huge, obvious reasons that we should prefer peer-reviewed sources over ones that are not so.


 * I don't see a very large problem over self-published printed works, simply because there really aren't that many of them; not very much work is being excluded by excluding them. Among works on the Internet, it is a much bigger issue, since there is so very much work out there. And there are (as I tried to outline above) so many styles of online publication.


 * The policy, as written, gives the impression that self-publication trumps all other measures of a source's reliability, rendering it categorically unusable regardless of all other qualities. This is just plain nuts when dealing with online publications, where there are just so many high-quality ones that also happen to be (for some definition of the word) self-published. Allowing (the absence of) a publisher's business model to trump the actual quality of the work or the author's expertise just doesn't make a hell of a lot of sense.


 * Why am I bothering to raise this issue? In discussion around two somewhat controversial subjects on Wikipedia (namely, evolution and Scientology) and to a lesser extent in some other fields dealing with science and pseudoscience, I've seen claims of "self-publication" used to attack sources that are otherwise high-quality and reputable but which disagree with someone's POV. One of the examples, I mentioned above: the TalkOrigins.org site, a truly remarkable collection of articles about evolution -- but neither an academic publication nor a commercial one with a publisher's business model behind it. But articles like this one and that one are well-written, well-sourced, and just plain high-quality articles that we should not be afraid to reference.


 * But with the current "self-publication" policy, all a person who disagrees with them needs to do is to argue that it is "self-published" and that is supposed to suddenly trump any other concern? I don't buy it. That isn't how we actually do operate in dealing with online sources, and it isn't how we should operate, either. --FOo (talk) 06:22, 30 December 2007 (UTC)


 * The problem is that if the person who disagrees cannot delete on this basis, you now have a dispute on the factuality of the source, and assuming that the SPS is the only source available, which it likely is, how can that dispute be resolved? (I hope it's okay to insert here, my apologies, if not)Slofstra (talk) 19:51, 11 February 2008 (UTC)


 * It's a useful filter. Most publishers in fact apply more considerations to accepting a book for publication that just potential profit. Self-financed publication, book or web site, is a sign that no established publisher sees enough merit in the work. Publishing-house financed publication does not guarantee quality, but there is a better chance that such books are reliable than for self-publication. -- Donald Albury 11:14, 30 December 2007 (UTC)


 * For what it's worth, I entirely agree with FOo. The statement that self-published sources are "largely unacceptable as sources" is amazingly narrow-minded and myopic. It is especially remarkable to find it here, a new-media encyclopedia that is collectively self-published. Talk about the pot calling the kettle black!


 * You need to look at the totality of the circumstances. Sometimes, material is self-published because no traditional publisher would find it profitable. Today, there are avenues for publication that simply didn't exist a generation ago. The fact that the author paid the freight himself does not categorically make the work unreliable. Instead, one must look to other indicia of unreliability, such as the absence of footnotes or citations, remarkable claims that fly in the face of known consensus, lack of citations of that work in other sources, etc.


 * One must also consider what other sources are available. A topic like The Holocaust is copiously documented in traditional peer-reviewed publications. The topic is also a well known lightning rod for crackpots and revisionist historians. It's doubtful that a self-published source is our best source for anything worth saying on that subject.


 * But a self-published biography of a second-tier nineteenth century opera singer might indeed be the best current source of information about that singer. Marc Shepherd (talk) 21:19, 30 December 2007 (UTC)
 * I disagree with this. If the biography has not been peer reviewed or passed muster by a publishing house it may not be credible. OTOH, if the biography has properly cited it's primary sources, then I should think those primary sources (letters, journals, diaries) could be used to cite the info brought into wikipedia, as long as the subject passed WP:NOT.  I'm tentatively suggesting a strategy here which would allow an SPS to be used and still follow wiki policies. I may not be correct as the question has been of concern to me, and the policies contain conflicting stmts on primary sources. Slofstra (talk) 19:58, 11 February 2008 (UTC)


 * I do not disagree if used this way. I am thinking more of several types of material I often deal with: Self published fiction, which is asserted for notability--and except for the very rare case of a documented best seller, we should ordinarily reject entirely--and yes, there have been exception historically. Second, the use of such sources in alternative medicine and similar topics, where we should also reject them entirely, unless they are cited in accepted reliable sources. Third:the publications of an institution about itself--which again is not evidence for notability, though it is for uncontroversial details.
 * alternative scholarship in developing fields where such sources are the best or the only ones is another matter, and we should accept it--provided it is possible to show that the author is actually recognized as an authority. 08:45, 31 December 2007 (UTC)


 * Regarding self-published fiction, you're talking about notability, whereas the issue here is whether a work can be used as a verifiable reference. On the matter of alternative medicine, I have to agree with you that there are going to be plenty of bad sources that are self-published ... but also that there are plenty of bad sources that are commercially published, too. The representation of tiny fringe views is dealt with by our undue weight rules.


 * Again, my concern is that if an editor identifies a source which is otherwise good -- that is, it is informative, relevant, represents expertise in the subject matter -- we should not disqualify it as a source merely because it's self-published. --FOo (talk) 21:13, 31 December 2007 (UTC)

Very interesting discussion! I think the time has come to take blogs seriously, tho identifying which ones will never be easy. One I've cited and will continue to cite is ScienceBlogs. These are serious people, mostly scientists, who can establish a blog there only on invitation from the community (presently 65 bloggers). Of course, most entries there cite the articles and studies they are discussing, but often the educated interpretation (and criticism) found in such entries is much more informative than the cited article(s) itself. Another type of online publishing is, for example, PLoS ONE, "An interactive open-access journal for the communication of all peer-reviewed scientific and medical research". These types of publishing are fast becoming very important and ought not be ignored. --Hordaland (talk) 15:34, 31 December 2007 (UTC)


 * Even for ScienceBlogs I would encourage some caution. Some of their blogs come from experts who do a very good job of communicating the established science.  Others, such as the "denialism blog", get heavily used to advance the authors' pet cause, and suffer many of the problems of third-party verfication that plague other self-published sources.  (In the denialism case, the authors apply a concept of "denialism" that is defined by the authors' own non-peer review manuscripts, and at times it has much broader scope than the way other scientists might use the same terminology.)  Dragons flight (talk) 17:44, 2 January 2008 (UTC)


 * Agreed. Blogs are sources for what notable bloggers say; some kinds of blogs can also offer original factual reporting, which can be useful in articles about natural disasters or war zones or whatever. It isn't clear that they are any more biased than many "mainstream" news sources. But there is as much variability in blogs as there is in (say) TV shows; some are entertainment, some are news, some are opinion, and some are a mix among these.


 * But the larger issue is that it's just ridiculous to exclude a whole class of sources based on their business model rather than on anything having to do with their practices, the expertise of their writers, their own use of reliable sources, or any other measures of their credibility. --FOo (talk) 05:44, 3 January 2008 (UTC)

Excessive facts tags?
I inserted numerous tags on the Red Line (MBTA) article, and another editor reverted them, with a comment in the main page's history "That seems a bit excessive. Perhaps you were looking for the article's "talk" page? Reverting.)"; the other editor does not appear to have made any effort to improve my edit, simply removing it completely without adding any note to the talk page.  What is the best way of dealing with this?  JNW2 (talk) 04:00, 1 January 2008 (UTC)


 * There seems to be a general sense that drive-by tagging of an article is not terribly useful, and it comes across as a little bit rude. If you have specific concerns about an article, it is usually more productive to express them on the talk page rather than to spew "numerous" tags across the article. --FOo (talk) 10:58, 1 January 2008 (UTC)


 * Or, alternately, use a single Unreferenced or refimprove at the top of the article, which goes the same point across with less... Invasiveness. Grand  master  ka  11:04, 1 January 2008 (UTC)


 * I'd like to stand up for the opposing position that adding fact tags for uncited material is improving Wikipedia and is usually appropriate, and that removing them rather than adding cites is quite rude, inappropriate, and is failing to improve Wikipedia. -- Writtenonsand (talk) 11:51, 1 January 2008 (UTC)


 * Adding citations for uncited material is improving Wikipedia; adding appropriate tags to encourage this is one of the ways of doing so; adding tags for multiple sentence of parts of sentences in a single paragraph over the same issue is usually more irritating than helpful. The particular tags related to documenting the probably noncontroversial dates of opening of various branches and introduction of various services. This could have been better handled with one or two tags, explained on the talk page. It's not as if the article has major problems with these issues. DGG (talk) 12:32, 1 January 2008 (UTC)


 * I agree that drive-by tagging is not productive, and does not reflect well on the taggers. A strategically placed fact tag or two should suffice if there are facts that really need citations. An overall tag in the references section (or on top if the problem is really serious) is better than a scattering of unsightly fact tags. Anything more than that is overkill. Remember that Wikipedia is here to serve the readers' education needs, not the editors' ego needs. If you really want to improve Wikipedia, WP:SOFIXIT yourself. If you can't, discuss it on the talk page. But by all means be objectively constructive, rather than subjectively critical. Dhaluza (talk) 01:39, 5 January 2008 (UTC)

Sentence that is a bit much "in your face"
The reliable sources section includes the following sentence:

"Because policies take precedence over guidelines, in the case of an inconsistency between this page and that one, this page has priority, and WP:RS should be updated accordingly."

I think this sentence should be removed. Even if you accept that policies take precedence over guidelines (and the linked page is more nuanced than that in my opinion - it does not use the word "precedence,") we do not put this sort of "creepy" instuction on policy pages: there is no equivalent sentence on WP:V referring to WP:N, for example. Does anyone take strong exception to removing this sentence? UnitedStatesian (talk) 14:25, 2 January 2008 (UTC)
 * Given that WP:RS is essentially a sub-page of WP:V (explaining what WP:V means when it says we have to use reliable sources)... there should not be any inconsistency between this page and WP:RS. However, it is an unfortunate fact that sometimes such inconsistency does happen.  When it does, we should encourage editors to raise the issue on both pages, so that the inconsistency can be discussed and corrected.  I would expect that 99 times out of 100, this will result in updating RS to match V... but there will be occational times when V will have to be updated to match a strong consusus at RS.
 * Perhaps we should say something more along the lines of "This policy is further explained in the guideline: WP:RS. Editors should be careful to make sure that there is no inconsistency between the two pages. Should you notice an inconsistency, please raise the issue on the talk pages at both locations." ... a similar statement should be placed on WP:RS directing people here. Blueboar (talk) 17:49, 2 January 2008 (UTC)


 * Thanks for the feedback, BB. Your language is superior to a delete.  I'll wait a day or so for other input, and then will make the change. UnitedStatesian (talk) 23:09, 2 January 2008 (UTC)


 * Yes, I agree that this is a better approach. Consensus is reached through discussion, not precedence. I have noticed a lot of this sort of authoritarian approach to policies WP:CREEPing up lately, contrary to the original spirit of the wiki. We should be on the lookout for this sort of thing, and question it when we find it. Dhaluza (talk) 11:59, 4 January 2008 (UTC)

moved the following discussion from below

 * US, re your recent edit, V is the policy, and RS has to be consistent with it, not the other way round. SlimVirgin  (talk) (contribs) 01:27, 4 January 2008 (UTC)
 * SV, I raised it above, at Wikipedia talk:Verifiability, and I thought BB's language was a big improvement. I think we should direct people to raise the issue on talk pages, since I think we all agree that it makes sense to do so - one person acting unilaterally on their perception of "inconsistency" may not make be the best outcome (edit wars, etc.). Can you suggest alternate language rather than a revert? UnitedStatesian (talk) 01:35, 4 January 2008 (UTC)
 * To explain... I have seen at least two rounds of RS being edited so it no longer is consistent with WP:V. In both cases it took a lot of time and effort to bring it back in line. I would prefer to scrap RS completely, and deal with the reliability issue in V... but as long as we are going to have a seperate guideline, I would like the active editors at V to keep tabs on it.  My suggested language that people to raise the issue of any inconsistency at both locations is so more WP:V editors will become awair that there is an inconsistency and fix it.  SV is correct in saying that WP:V is the policy, and 99 times out of 100 (and probably more often than that) when there is an inconsistency it is RS that needs to be ammended to match WP:V.  However, I also leave open the remote possibility that someone in the future may come up with an issue at RS that was not thought of and addressed in V.  They would have to make an extraordinarily good argument in favor of changing V to match RS... but it could potentially happen. Blueboar (talk) 02:09, 4 January 2008 (UTC)
 * Well, I think we did once, and that was with "extraordinary claims require extraordinary sources." SlimVirgin  (talk) (contribs) 01:47, 5 January 2008 (UTC)

Images
I have a general question about this policy. Does it apply to images? I mean, do images have to be published by a reliable source per WP:V?

If WP:V doesn't apply to images, is there any policy that does? If not, does this mean that anyone can produce an image and make a claim regarding it? What if there are conflicting claims?Bless sins (talk) 04:57, 3 January 2008 (UTC)


 * Please see WP:NOR for the primary exception clause regarding use of images. Additionally, please assume good faith regarding such uploads, in general. If someone makes an extraordinary claim with an image, i.e. This photo claims Bush knew about the attacks on 9/11!!!, then it needs to be backed up accordingly. Basically, just accept an image unless it can be proven to be false/misleading. Please realize that captions for said pictures are another matter altogether. Is there a specific image you have in mind? — BQZip01 —  talk 05:20, 3 January 2008 (UTC)

Reliable sources
My view is that this section needs to be seriously reviewed, and for a number of reasons. University presses have, for some decades been in dire financial straits and have as a result been publishing all kinds of works, some highly questionable, and others a disgrace. Marxists, even communists, have their 'works' published by some university presses. As for this quaint phrase (which I take to be American) "peer-reviewed", does that mean that if you can locate half a dozen fellow-travellers of any reasonable note who will speak highly of your essay/book/whatever that it is somehow not only good but equivalent to the Holy Bible as a source? I'm sorry but that won't do. Reviews are two-a-penny.

In addition, many great scholars prior to the Second World War were forced to self-publish for the simple reason that the subjects they were writing about were so specialised that commercial publishers would not entertain them. As WP:reliable sources currently stands all these tens of thousands of expert publications appear excluded. That cannot be right.

Since about 1950 many biographies and even history books by so-called celebrities or notorious historians are so obviously published for their sensational value rather than academic content. It seems ludicrous to narrow the orbit of publishers of some books, whilst millions of other trashy books fall into that orbit because they have been published by mainstream publishers.

For me at least a reliable source is ideally a book and one that can be found in the National Libraries or which can be called up by your local library from the national repositories. That then covers all the categories I have discussed, above, good and bad. It is absolutely crucial that we don't fall into the censorship trap whereby Wikipedia starts defining, via their jargon, authorised and non-authorised authors. That is heading down the Nazis/Soviet road. David Lauder (talk) 14:16, 3 January 2008 (UTC)


 * David, I agree with many of your points, but it's hard to see how to change things. There are a number of editors who set a great deal of store by the fact that a book was published by a university press and was "peer-reviewed," even though as you say, a lot of nonsense is published that way. However, what we do at Wikipedia is report what's being published by sources that other people trust, and people do still trust university presses.


 * Perhaps you could write a footnote about the pre-WWII scholarship that had to be self-published? SlimVirgin  (talk) (contribs) 18:54, 3 January 2008 (UTC)


 * I fear you misunderstand what I have written. At the moment the manner in which WP:Reliable sources is phrased it is actually very narrow. This can be used, like so many other WP guidelines, adversely. It needs to be more broadly-based. Regards, David Lauder (talk) 22:03, 3 January 2008 (UTC)


 * RS is not really an active page, David. The policy page is WP:V, though it makes virtually the same points. I'm not sure what you mean by "narrow." SlimVirgin  (talk) (contribs) 22:16, 3 January 2008 (UTC)


 * SV, I dispute your assertion that WP:RS is not an active page (perhaps you were thinking of WP:ATT?) WP:RS is a guideline, not a policy, but it is very much active (and in fact WP:V refers to it repeatedly - I don't know how that policy can work without the guideline).  What is next, your asserting that Notability is not an active page? UnitedStatesian (talk) 23:10, 3 January 2008 (UTC)


 * It's not active because it doesn't say anything. SlimVirgin  (talk) (contribs) 01:27, 4 January 2008 (UTC)

We've gotten a bit distracted from David Lauder's original point. And I do agree with David that the current policy and guideline (whichever version of it) overstates the reliability of "peer-reviewed" material, and unreasonably denigrates other kinds of sources that are often just as reliable.

I've felt for a long time that WP:V and WP:RS should more broadly discuss the various factors that tend to indicate reliability (or the lack thereof), rather than implying rather simplistically (as they now do) that "peer reviewed = good" and "all others = not so good". Marc Shepherd (talk) 01:07, 4 January 2008 (UTC)


 * Marc, the reason for that is that a group of editors arrived at this page a few months ago and tried to force the policy to say that only peer-reviewed material was acceptable in certain articles (science, history, medicine). They caused a tremendous fuss about it, and although we managed to avoid them stating that only peer-reviewed sources were acceptable, they nevertheless managed to stress the value of peer-reviewed material. I agree that it's not a good idea, but I'd be hesitant to destabilize the page again, and it's hard to come up with a description of "reliability" that everyone would accept. SlimVirgin  (talk) (contribs) 01:30, 4 January 2008 (UTC)

I can only reiterate what I said at the opening of this discussion. "Peer-reviewed" in my opinion is a meaningless expression. Reliability in science and medicine is generally self-evident (but just look at the number of scientists and doctors who scorn each other!). History is far more contentious (see David Irving). At the moment we have a scenario whereby valid and highly respected historians of many centuries are being dumped by modernists and nationalist-leaning editors who naturally support each other. Therefore their "peer-reviewed" books become "reviewed by fellow travellers and so it is good/better than anyone else". I have no objection to the alternate opinions of the various historians being displayed. If in those explanations it is clear that one is wrong, so be it. But the road we are progressing down is supporting bias and WP:reliable sources is being cited. A broader publication base is required, as I said above. Whether or not this has been discussed before or not, it again needs (constant?) reviewing. I thought continuing concensus was a Wikipedia foundation stone? Regards, David Lauder (talk) 10:39, 4 January 2008 (UTC)


 * All processes have weaknesses, but the process of sending articles anonymously to anonymous reviewers, which is what happens during a standard peer-review, ensures a degree of evaluation which lay-contributors to Wikipedia will be hard-pressed to match. Obviously journals with "Marxism" or "psycho-analysis" in their titles are likely to have reviewers who are likely to accept particular paradigm but such approaches still are regarded as acceotable within sociology or similar subjects. Certainly I would regard peer-reviewed journals as more respactable and more reliable than the introductions to translations of books published in the early 20th century, as you were favouring over at Houston Stewart Chamberlain at its talk page unless you are using those as primary sources.--Peter cohen (talk) 10:53, 4 January 2008 (UTC)
 * As someone who has had very considerable involvement in publishing I feel bound to say I have never ever heard of sending proofs anonymously to anonymous reviewers. All the publications I have worked with have sent proofs knowingly to selected reviewers. You cited my concerns re Chamberlain: yes, when an introduction to a book is written by someone who has clearly read the work concerned and is someone who is reliable and notable I see that as equating to what is loosely called here "peer-review" and I am sorry you don't. The whole problem with authors over the last forty or so years is that many of them and, more to the point, many of their disciples, carry an arrogance and superiority complex over all those previous historians who worked just as hard as they purport to have done which is mind-boggling. Last but not least, where Modern History is concerned, we have counless authors writing from the perspective of the year in which they write and with that mentality. Thats why we need wider sources than this "peer-review" jingoistic business which clearly is deliberately narrowing the field. What then of newspapers and TV as sources. Are they "peer-reviewed"? Hardly. They are largely sensational stories written by journalists. Some sort of graphic inconsistancy here. All I am calling for here is balance. Regards, David Lauder (talk) 17:22, 4 January 2008 (UTC)


 * I have become an editor here, of late mostly because I have been the victim of what I will call a peer review cabal by very famous scientists (I am a scientist as well, less known). This is a group of university professors trying to outright steal my invented technology and publishing broadly lies that it was predated by a scientist, long dead who only generally discussed an actual implementation of the technology but never devised it nor never really enabled it within his writings in the least. By many of them publishing together in unison of the same lies, all of them direct competitors they have successfully convinced the scientific community that credit should be given to the long dead scientist and that I should be shunned as they go about infringing it (using all this for cover). They have as well began to treat this scientist like god of the technology and have used all his terminology in the extensive writings about my technology, deliberately supplanting my terminology with the dead scientist's as they describe their various infringing projects of my technology at various prestigious universities about. There are a large number of universities who would lose out if the truth were known as they now have extensive projects upon the technology and have organized this meticulously to the letter.


 * The reason I take this up here is I noticed that discussion on the reliability of "peer review" was had. One should also consider corruption of such and ferret out and bust any peer review that is a conspiratorial act of scientific misconduct, it would seem. As these guys are very good at this I would submit to you seasoned editors herein an invitation to bust this huge rat's nest of deceit and have the truth be known, as difficult as that might be to extrapolate. By the way, this is very important technology and none of these crooks have managed to properly develop it and after failing themselves have made public statements as to the inviability of the technology even though it has long been proven otherwise by my very many prototypes. An example of these public attacks and mechinations can be seen in the primary principle's book here, devised for the sole purpose of deceit and containing nothing but generalities:
 * Also note that they have surpassed the legal limit in the number of words used for fair use of a copyrighted material and are thus infringing (in addition to commiting libel). I feel that work should be banned from Wikipedia and seek consensus of that here). Fraberj (talk) 13:00, 5 January 2008 (UTC)

Very interesting. I wonder who first introduced this "peer-reviewed" term to Wikipedia? In real terms in means a review by fellow collegues. If you were under peer pressure at school it would be pressure from those around you. Therefore this terminology is clearly flawed as it dramatically narrows source material to that accepted by cabals of others. It is much harder to go back before the Second World War or 1900 and trace other writers of the same Ilk, and then to see if they had any comment on others alive at the same time. Such an exercise is time consuming, to say the least. So that leaves us with contempory sources and their reviews. How very cosy. Regards, David Lauder (talk) 13:45, 5 January 2008 (UTC)


 * It's not quite as bad as that; see the article on it at Peer review. Also, not all sources in Wikipedia have to be peer-reviewed. SlimVirgin  (talk) (contribs) 15:11, 5 January 2008 (UTC)


 * David has expressed the view that works of history written in 2007 or 1997 or 1987 should be given much the same weight, or perhaps less, than those written in 1907 or 1897 or 1887. Were the ideas advanced in works of a century past still the view of modern historians, those views would be recapitulated in modern surveys. The question therefore is, should views which were advanced by historians long ago, but which are no longer supported, be included. If current practice on Wikipedia is any guide, probably not.
 * David's comment here is perhaps relevant: "I would believe both Joseph Bain and Anderson before I'd believe someone called Fiona Watson (whoever she is)". If that makes sense to others, it makes none to me. What Dr Watson did to offend David, apart perhaps from disagreeing with his family's stories about themselves, is not immediately obvious. Why her name should disqualify her views is equally unclear. Modern studies disagree with many of David's beliefs about the past, but that's neither here nor there. We're here to regurgitate what academics write. If what they write now conflicts with older views then we might mention that views have changed, but no more than that is necessary or desirable. Angus McLellan (Talk) 20:33, 5 January 2008 (UTC)
 * Angus misses the point entirely. Firstly, at the time I wrote that comment I had no idea who Fiona Watson is, but I still disagree with her on the subject I was speaking about. Angus previously sneered at me for saying Malcolm Canmore had spent 14 years at the English Court as though it were crap or my invention and he or those writers he prefers knew better, but David Donald Murison, MA., BA., writing in The Scottish Tradition, edited by Professor G. W. S. Barrow, cites it and has no problem with it (p.71). Angus also sneers at "family stories", in this case many of them confirmed by people such as Sir Thomas Dick Lauder universally regarded as a substantive scholar and author. But Angus fails to understand that in medicine, science, and history, the scholars often disagree with each other and sometimes violently. With science and medicine it is different as things are constantly moving on. This is not necessarily always the case with history. But like the other subjects there will always be those who wish to reinvestigate and reinterpret history and, indeed, to denounce older scholars. That is their (and their disciples) right in a free society, but it is not their right to have it written in stone, the publications of all those scholars before them burnt and the New Bible written by them. Such rigidity and using Wikipedia guidelines to enforce it should not be permitted. As I have constantly said, by all means cite the latest opinions, especially where new evidence has been uncovered which leads to certainties, but don't dismiss the hundreds if not thousands of historians and great scholars who have gone before us. History is not science and medicine, and there is far more POV. Thats why balance is required. Regards, David Lauder (talk) 21:25, 5 January 2008 (UTC)


 * Who's sneering? I have no idea what professional historians feel about their predecessors, but I'd think it would be on the lines of Newton's comment about standing on the shoulders of giants. The fact that you can hardly scratch the surface of Insular medieval history without relying on books written long before WWII should be sufficient proof of that. The rest of your comments stand in no need of commentary from me. Angus McLellan (Talk) 22:42, 5 January 2008 (UTC)
 * It is important to stick to the subject of this page instead of making highly personal attacks. Too many people on Wikipedia fall foul of WP:OWN citing narrow guidelines. I am attempting here to broaden sources without avoiding scholarship. David Lauder (talk) 10:46, 6 January 2008 (UTC)

Need to know whether a source should be used
Hello. There's a been an ongoing dispute at Talk: Super Smash Bros. Melee about the relevance of a gaming forum to SSBM in relation to gaming tournaments. Sources have came to light today from MLG&mdash;a gaming tournament website. However, the writer of the sources&mdash;who's trying to assert Smashboards's notability&mdash;is a actually an administrator at Smashboards. This obviously makes him biased, but does it invalidate the source? Thanks. Ashnard Talk  Contribs  19:41, 4 January 2008 (UTC)


 * I'd say it does. If it's that important, then there should be another source for the information; or, at least, editors vouching for it who have no personal connection. Marc Shepherd (talk) 20:29, 4 January 2008 (UTC)
 * There are other authors who have written about the competitive Smash community and mentioned Smashboards, including other authors from MLG, Nintendo Power, and other eSports websites. All of the authors who have ever written anything about competitive Smash (including Nintendo Power authors) have accounts on Smashboards because Smashboards is where the majority of competitive players inhabit. To not have an account there and write about the community would be poor journalism. Ashnards assertion of bias should have no effect on the validity of a source, especially when the articles were written for, paid for, edited, and published by MLG. MLG is a major source on competitive gaming as it is the largest gaming league in the USA, Ashnard has agreed that MLG is an authority, yet he selectively discredits my articles for my affiliation with Smashboards, an affiliation that if I lacked I would have no right to write about competitive Smash. People don't call into question the validity of a historian who is a republican writing about the importance/history/term of president Clinton. There is suspicion of bias based solely on my membership at a website, but there is no proof of bias.  As an author I feel I must defend my work. I agree that me pushing for Smashboards inclusion in the tournament section of the SSBM wiki would be biased, however I'm not the person pushing, I'm merely supplying the people who were pushing for it with sources so they can back up their assertion, yet when I presented my work and others as sources Ashnard discredited them based solely on bias with no proof. Alphazealot (talk) 16:17, 6 January 2008 (UTC)


 * I see that you're pulling the "I need to be a member to report accurately" card. To use the analogies that you love so much, it's like saying that a journalist must be a member of a particular football club to truly report on that club. As for the Clinton analogy, no open Republican should be used as source if that source is trying to certify the relevance of Clinton accurately. Ashnard  Talk  Contribs  18:46, 6 January 2008 (UTC)
 * I would be a member regardless because I am an avid fan of SSBM and Smashboards is the largest message board dedicated solely to the game. A republican writing about Clinton would be no more or less biased than a democrat writing about Clinton, which is why its assumed authors who have an expertise in a certain field write about that field with as little bias as possible. Are you saying that historians have no political parties and have never voted, which is why they write without bias? Absurd. A journalist should go to whatever lengths are necessary to accurately report on something. In the case of Smash, you either go through Smashboards or through MLG, eventually at the end of the day you'll be associated and have dealt with people from both (as evidenced in numerous articles I have linked to you).Alphazealot (talk) 21:16, 6 January 2008 (UTC)


 * No. There is a difference between voting for a political party and being an active member in its functions. By the way, why defend your position as a journalist as if I'm attacking it? I'm just questioning your position as an author of a source to verify a particular assertion. But you just don't seem to understand. Ashnard  Talk  Contribs  21:51, 6 January 2008 (UTC)

Private communication
Is it fair to say that is not an acceptable Wikipedia reference under any circumstances? Wanderer57 (talk) 23:25, 4 January 2008 (UTC)


 * Yes, that's right. See WP:NOR too. SlimVirgin  (talk) (contribs) 01:45, 5 January 2008 (UTC)

Simplifying the convoluted wording
I've taken a stab at copy-editing the policy for understandability - mainly the lead and where it links to for further explanation. If I've changed the meaning or context please let me know. The Transhumanist (talk) 21:20, 6 February 2008 (UTC)


 * Transhumanist, you're editing months old wording that has been quoted thousands of times. Citations for "quotations and for any material that is challenged or is likely to be challenged" is as central as wording gets. That doesn't make it perfect. That doesn't mean it can't be discussed. But you can't just shoot first on this policy. Non-trivial changes are reverted first at core policy, and questions asked later.


 * I'm honestly sorry we just went to three reverts, but I was surprised. Marskell (talk) 21:27, 6 February 2008 (UTC)

Basically, I changed the lead and top section to this (though I used level 2 headings instead of level 3):

The threshold for inclusion in Wikipedia is verifiability, not truth: readers should be able to check each fact on Wikipedia against a reliable source. Editors should cite a reliable source for material they add to the encyclopedia - unsourced claims may be removed.

Verifiability is one of Wikipedia's core content policies. The others are No original research and Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in Wikipedia articles. They should not be interpreted in isolation from one another, and editors should try to familiarize themselves with all three.

Removal of unsourced claims
If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it.

Any edit lacking a reliable source may be removed, but editors may object if you remove material without giving them a chance to provide references. If you want to request a source for an unsourced statement, consider moving it to the talk page. Alternatively, you may tag a sentence by adding the fact template, a section with unreferencedsection, or the article with refimprove or unreferenced. Use the edit summary to give an explanation of your edit. You may also leave a note on the talk page or an invisible HTML comment on the article page.

Do not leave unsourced information in articles for too long, or at all in the case of information about living persons. As Jimmy Wales has put it:

Burden of evidence

 * For how to write citations, see Citing sources

Discussion of above proposed changes
Please let me know if anything is wrong with it. The Transhumanist 21:35, 6 February 2008 (UTC)


 * Please revert yourself, discussing first. In the absence of that, this is disruption. Marskell (talk) 21:39, 6 February 2008 (UTC)
 * How'd you come to that conclusion? Good faith edits to improve a page, including a policy page, are not disruption.  I'm here and willing to talk to you in a civil discourse, and all you are doing is making demands.  Why not discuss the changes themselves, so I can improve them to accommodate your specific objections?  The Transhumanist  21:45, 6 February 2008 (UTC)


 * My intent is not to change the meaning, only the readability of the page. So if I have changed the meaning or the context, please point this out so we can fix it. The Transhumanist  21:47, 6 February 2008 (UTC)


 * Oh, gee. My only demand is that you not change a central policy page without discussion. You had engaged in no discussion prior to altering the policy. Altering policy without discussion is disruption. Not hard.


 * But it doesn't matter much, because you'll be reverted ultimately, over a few days or a week. Do as you will, for tonight. Marskell (talk) 21:54, 6 February 2008 (UTC)


 * There's no policy against improving the readability of a page directly, even if it is a policy. I've posted the changes above, and look forward to feedback.  The Transhumanist  22:06, 6 February 2008 (UTC)

I've underlined the text I changed in the passage above. The section "Removal of unsourced claims" was as the end of "Burden of evidience", without the heading. The Transhumanist

Can we avoid the petty "you didn't discuss the changes before editing so therefore they're inherently invalid" line? Not touching third base is insufficient grounds for objection. Don't revert content because somebody might disagree with it. Revert if you do disagree with it, and explain why. Otherwise you're just wasting people's time with empty process. Phil Sandifer (talk) 00:37, 7 February 2008 (UTC)


 * Thank you. By the way, did you mean to write "before editing" instead of "before reverting"? The Transhumanist  03:32, 7 February 2008 (UTC)
 * Actually, I agree that Transhumanist should have discussed it first. That is the accepted practice on policy pages. In any case, it is water under the bridge.  Now that we are discussing it... does anyone see a problem with the edits?  I don't see any major ones. Blueboar (talk) 03:45, 7 February 2008 (UTC)
 * It doesn't appear to be all that accepted. But I'm glad you don't disapprove of the draft above.  I was thinking that the link to WP:RS is awkward in the lead, since that page quickly refers the reader back to the reliable sources section on this page (where reliable sources are actually defined).  WP:RS covers specific types of reliable sources, while this page establishes the concept, so maybe a VER section link is called for rather than jumping off the page. The Transhumanist  04:54, 7 February 2008 (UTC)
 * I did mean before editing (and changed my comment above), but discussing changes before editing a policy page is always wise too. Phil Sandifer (talk) 05:42, 7 February 2008 (UTC)

(OD) Can we address your changes, one-by-one? It would be most helpful, if you do not use . It's ugly and likely to frighten small children and elderly persons. Instead use. Quote the old wording, quote the new wording, explain your reasoning. Try to discuss small increments so we can see clearly what you're stating, instead of us having to cut-and-paste large paragraphs and do text-compares to see what you're doing. Thanks.Wjhonson (talk) 05:49, 7 February 2008 (UTC)


 * First, I'm sorry if I seemed aggressive, but if someone makes significant changes to stable policy wording without using talk, they should be reverted, end of story. That's long-standing practice, particularly on the "big three."


 * Re the specifics, "each fact" would be a very poor choice of wording. The erroneous assumption that we need to cite "each fact" is one of the things that makes people hate our sourcing policies. More generally, I'm a fan of stability, because when I quote something on article talk, I like to know it hasn't changed. The paragraph in question, and the headline "Burden of evidence," date to '06. What's wrong with them? I don't like linking to RS because RS sucks, but that's a problem we need to solve over there. Marskell (talk) 08:21, 7 February 2008 (UTC)


 * Do you mean that some facts don't need to be sourced? The policy doesn't convey that.  What facts don't need to have sources cited after them?


 * "may be removed" is very confusing, once you try it: the policy doesn't mention that if you remove an unsourced section (or sections) from an article that a bot will come along and revert you automatically and imply that you vandalized the article! Instructions on how to remove unsourced material needs to be provided.  I still don't know how to get around the bot thing.


 * The Transhumanist 12:35, 7 February 2008 (UTC)


 * The only facts that this policy states must be sourced are those that are challenged, likely to be challenged, or quotes. Other material must be verifiable, but there is no requirement that it be sourced whatsoever. This is matches the common practice across the wiki that has developed over a period of years. Policy describes practice; it would simply be false, not policy, to claim that all facts must be sourced. &mdash; Carl (CBM · talk) 13:12, 7 February 2008 (UTC)
 * What about proactive policy enforcers who follow Jimbo's mandate (which he "can't emphasize enough") to aggressively remove all unsourced information? The policy also clearly states "Do not leave unsourced information in articles for too long..." I've started a new thread on Jimbo's quote from the policy, below. The Transhumanist  15:46, 7 February 2008 (UTC)
 * There is no consensus to aggressively, broadly remove unsourced information. In fact, it's quite the opposite: unsourced information is just fine provided it's verifiable. It would be inappropriate to begin any large-scale campaign to remove unsourced information across large numbers of articles. Of course it's fine to challenge particular statements; that's part of the normal editing process. &mdash; Carl (CBM · talk) 17:54, 7 February 2008 (UTC)


 * On the bot issue... Most bots only act if you remove large sections of text in one single edit (they register large deletions as vandalism)... If you stick to removing specific statements, one at at time, they usually won't regard the edits as something they must correct. Leaving good edit summaries can help. In the event that a bot still counters your edits... most of them have "report false positives" links to where you can leave the bot's manager a message, telling him or her that the bot acted incorrectly.  The manager will revert the bot. Blueboar (talk) 13:41, 7 February 2008 (UTC)

As said, I object to the fundamental change of the WP:V policy that we would no longer require, as a policy-level requirement, that content has already been published before it is used in Wikipedia. TH's proposal removes that requirement from the WP:V policy page.

TH's characterisation "My intent is not to change the meaning, only the readability of the page" is therefore incorrect: --Francis Schonken (talk) 17:41, 7 February 2008 (UTC)
 * "has already been published" has no readability problem;
 * removing this clause from the page alters the meaning of the policy.
 * No, just the result wasn't as intended. Which is why I asked for feedback.  Thank you for it.  The Transhumanist  18:56, 7 February 2008 (UTC)


 * "Has already been published" is included or implied in the definition of "reliable source". No worries either way. The Transhumanist  20:08, 7 February 2008 (UTC)

I've withdrawn this proposal, and will (re)submit proposals for changes one-by-one as requested by Wjhonson above. The Transhumanist 20:08, 7 February 2008 (UTC)

Proposal: change the name of this policy to "Verification is required"
The current title "Verifiability" is ambiguous. Being verifiable and presenting verifying references in the article itself are distinctly different things (the latter is a subset of the former). In common English, an article is verifiable if you can go out and find sources for it (whether or not references are present in the article, and regardless of whether or not you actually go out and look). Simply being verifiable isn't enough, so this word (the very title of the policy) is inadequate and doesn't capture the essence of the matter.

Verification, on the other hand is the act of verifying, or the state of being authenticated, and is much clearer.

The Transhumanist 12:35, 7 February 2008 (UTC)


 * I'm afraid that you misunderstand the policy. The claim that "verification is required" disagrees not only with the wording of the policy, but with our established practice across the wiki. Your definition of verifiable is exactly what we require - that, in principle, the material in our article is included in reliable third-party sources, whether or not it is explicitly sourced in the article. That is the standard that editors are expected to follow. &mdash; Carl (CBM · talk) 13:08, 7 February 2008 (UTC)


 * I think the ambiguity is somewhat intentional. The ambiguity is a nice compromise between deletionists and inclusionists. The idea is that if something is verifiable (ie it is likely that it could be verified, but hasn't been verified yet) it does not completely violate this policy... it simply needs to be verified.  It covers the grey zone between deleting unsourced statements immediately, and leaving them tagged with a  tags forever. Blueboar (talk) 13:27, 7 February 2008 (UTC)


 * A lot of frustration follows from that, however. New editors see articles without citations, and think it is OK to add anything they want to an article. Then they get upset when their contributions are challenged or reverted. I also think there are editors who feel that something is verifiable simply because they remember reading or hearing it somewhere, or just simply think it is true. I know that this policy is not going to move far from its present position anytime soon, but I would push for making it clearer that pro-actively citing sources would make editing Wikipedia easier for everyone. -- Donald Albury 13:33, 7 February 2008 (UTC)


 * (Re Blueboar) There is no reason why unsourced material needs to be verified just for the sake of having a inline reference after it. It is perfectly acceptable for articles to contain unsourced information indefinitely, provided it is not challenged, likely to be challenged or a quotation. Such unsourced information doesn't violate this policy at all, which is a much stronger thing than 'not completely violating it'.
 * (Re Donald Albury) In general, we want new editors to add things to articles - provided those things are verifiable, npov, relevant, etc. It does take time and patience to explain our policies to new editors. &mdash; Carl (CBM · talk) 13:38, 7 February 2008 (UTC)


 * (ec) The wording from WP:ATT: "Although everything in Wikipedia must be attributable, in practice not all material is attributed." I should be able to verify everything I add, but I need not always do so. (We should actually bring that wording over here.) Marskell (talk) 13:47, 7 February 2008 (UTC)


 * The "in practice" part of that has always struck me as odd. "Although, in practice, everything in Wikipedia must be attributable, in practice not all material is attributed" and "Although everything in Wikipedia must be attributable, not all material needs to be attributed" are both better in reducing the chance that the "in practice" is misread. I didn't mind that line in ATT, but I would rather see it a little clearer if it is going to be here. &mdash; Carl (CBM · talk) —Preceding comment was added at 14:02, 7 February 2008 (UTC)


 * Carl, almost anything that is unsourced is likely to be challenged under this policy, because not only does the policy strongly imply throughout itself that unsourced material should be removed, it explicity states "Do not leave unsourced information in articles for too long." Also, the very nature of policies is that they are intended to be enforced, which provides an implicit imperative for editors to challenge unsourced articles or unsourced content. The subtleties in wording are simply overridden by these implications, but they still cause a great deal of confusion in discussions and waste a lot of editors' time in arguments at AfD, etc. So basically, any information that doesn't include citations of reliable sources is in violation of this policy. Therefore, the convoluted wording should be replaced so it is easy to understand, or the policy should be changed to confer your interpretation clearly. At this time, the policy does neither. The Transhumanist  18:46, 7 February 2008 (UTC)


 * What does Jimbo mean by "sourced"?:

"It should be removed, aggressively, unless it can be sourced. This is true of all information..."


 * That's explicitly clear. Why would Jimbo state that a cite tag wasn't good enough if he wasn't referring to providing citations? I think Jimbo's statement is pretty hard to interpret any other way than "if it doesn't have citations, then it should be removed aggressively".


 * The Transhumanist 15:09, 7 February 2008 (UTC)
 * There is a difference between the "some sort of random speculative 'I heard it somewhere' pseudo information" that Jimbo is talking about, and information that is likely to be verifiable, but isn't yet verified. Obviously we should remove nonsense and challenge statements that are questionable... but, as Carl states, not everything needs to be given an inline citation - and some statements are so basic that they do not need citation at all (these are frequently referred to as "Water is wet" or "The sky is blue" statements... statements of uncontroverted and obvious fact).  There is a ballance here between "get rid of it if it is not cited" and "let people say what ever they want". Blueboar (talk) 16:06, 7 February 2008 (UTC)
 * But pertaining to the removal of unsourced material, he said "This is true of all information".  The Transhumanist  16:31, 7 February 2008 (UTC)


 * You are pointing out is that Jimbo's statement is different than both established practice and its description here in WP:V; you're right. The subject that Jimbo used for that email was "Zero information is preferred to misleading or false information" - a sign of the type of claims he is referring to. It's informative to read the rest of the email thread in the archives, as it shows a diversity of opinions. &mdash; Carl (CBM · talk) 16:52, 7 February 2008 (UTC)
 * Transhumanist seems to be a hard liner where it comes to verifiability... that is OK. Carl seems to be a soft liner (if that is a term)...  That is OK too.  Me... I fall somewhere in the middle.  That is also OK.  Indeed, the entire point of the current language is that it accomodates all points of view on verification.  If any of us sees something that we think needs a citation, we can challenge it.  If a citation is not provided after a reasonable time we may delete the unsourced statement.  The question then becomes: what is a reasonable amount of time?... and that depends on the nature of the statement and what kind of article it appears in... In some cases (such as an unsubstatiated negative comment a BLP) we should delete immediately.  In others we can wait several months.  We use our good judgement.  If there is a disagreement over whether the challenge is realistic or not, we discuss and reach a compromise.  All that fits with the current language... which is why I prefer it. Blueboar (talk) 17:32, 7 February 2008 (UTC)
 * Yes, that's very well said. &mdash; Carl (CBM · talk) 17:48, 7 February 2008 (UTC)


 * Actually, I'm firmly in the inclusionist camp. I'm just trying to understand all sides so I can try to solve some problems.  The policy causes much confusion and needless arguments across Wikipedia.  And then there's the rude awakening when newcomers reach their disillusionment. A clearer presentation in the policy (and elsewhere) of the point described by Blueboar above may help to prevent some of these problems.  The Transhumanist  19:41, 7 February 2008 (UTC)
 * No, not really. Like the popular proverb says, a pig is still a pig even if you put makeup on it. Changing the name of the policy from "verifiability" to "verification is required" doesn't really address anything, as the arguments come from people who a) don't know about the policy (hence don't care about the name) and b) people who know about the policy, but try to skirt around it. Tito xd (?!? - cool stuff) 19:47, 7 February 2008 (UTC)
 * A clearer description may help some cope better with the two types of people you just mentioned. But I'm withdrawing my proposal, and will make new ones as they come to me.

Images, quotes
Guys, is there any reason we can't some pictures and big quotes to liven this page up a bit? David tried to add an image and it was removed within seconds, and now my big Tacitus quote has gone to NPOV, even though it's clearly relevant to V. I think we need a few eye-catching things on this page to make it easier to look at and to convey the message. SlimVirgin (talk) (contribs) 07:23, 2 February 2008 (UTC)


 * Personally I kinda like the Tacitus quote. Wjhonson (talk) 09:10, 2 February 2008 (UTC)
 * It's already one of the liveliest pages in policy spage. Eleven shortcuts, a Jimbo quote, a nutshell, three policy box templates, footnotes, and an audio link.  It's setting off my ADHD just to look at it.  As a suggestion maybe we could set this quote off with a text box instead of a section header so it doesn't appear in the TOC, and read, as a part of the official policy rule.  Also, an internal or external link to a commentator who interprets this quote for some of us not-so-smart mortals might help.  Wikidemo (talk) 13:49, 3 February 2008 (UTC)


 * And why do I want to go play DnD right about now?Wjhonson (talk) 05:37, 7 February 2008 (UTC)
 * You look at dice-shaped objects and you get the urge? Now that's being addicted.  The Transhumanist  14:52, 9 February 2008 (UTC)

E-mail exchanges as a source
On several occasions I have had e-mail exchanges with primary researchers or historically important figures (in terms of the article). Three examples are illustrative.

After writing an article on the seminal Star Trek (text game) (not here), I was contacted by the original author of the program and told that the history that came down through the microcomputer era was in fact totally wrong. Since correcting my article, the real history has slowly made its way around the 'net, including the wiki article here. More recently, while writing the HiPER article I contacted the primary project lead and asked him a few questions about some technical specifics. After a short exchange he proofed the article and said it was great. Yet it failed GA because of this. Finally, while writing the Nova laser article, I noticed that there were a number of broken pages and dead links on the LNLL web site, so I contacted them about this. In the exchange I also asked about an odd historical discrepancy that I had noticed: in public sources from the era it is always shown using what is known as "direct drive", yet technical references released later show it using "indirect drive". In an e-mail reply, I was told that the public sources were deliberately misled in order to keep the indirect drive method somewhat private. This interesting historical fact is not published anywhere, nor would it be, so the only way to find it out is to talk with someone involved. And as a result of doing so, the Nova article here is, by far, the most technically and historically accurate article on the topic that is generally available to the public.

Yet according to the wiki guidelines, all of this should be removed. In fact, as it is clearly stated here in V, it is better that the article be incorrect and misleading. Why? Because e-mails cannot be V by all and sundry. E-mails, phone calls and face-to-face meetings are the only way to verify many facts in the real world. These are the basis of the process that makes up the vast majority of historical research in the "real world". But here on the wiki, they are disallowed.

So that's why I'm here. We need clear policy on how to deal with these sorts of sources. Really really really need it. Simply put, there's no policy at all. There are some ad-hoc policies, but none of them are useful -- no, sending the e-mails into the black-hole database is not good enough, because I can't send my phone calls into it (yes, I have tracked down people and phoned them to verify facts, see Bede BD-10). Worse, when push comes to shove, the only V comes from "on high" with no explanation.

So again, we need to have some sort of published policy on this topic.

Maury (talk) 15:13, 2 February 2008 (UTC)


 * I think the policy is quite clear; until the research has been published by a reliable, third-party source, we can't use it. -- <b style="color:navy;">Donald Albury</b> 17:12, 2 February 2008 (UTC)
 * Ummm, yeah, that's what I said. The policy needs to be expanded to address this. Maury (talk) 18:03, 2 February 2008 (UTC)


 * Maury... are you asking that we expand the policy to allow this... that isn't going to happen. The policy is clear, you can not use these emails.  Using the emails constitutes more than just a violation of WP:V, it is also a violation of WP:NOR. The only way you could get this into wikipedia is if you write up your story, with transcripts of the emails, and get it published somewhere.  Then people can cite to that published story. Blueboar (talk) 01:26, 3 February 2008 (UTC)


 * Agree with blueboar here. Another point above was the phone calling to verify.  That is fine and commendable, but we can't use that as a source.  We can make phone calls to verify what the article states, but we can't turn around and cite our own phone call as the verification for what it states.  I.E. we can't use that process, within the article.Wjhonson (talk) 01:46, 3 February 2008 (UTC)


 * Maury, there are registered users and a lot of unregistered. Unfortunately, many, many, many of them are vandals, POV pushers, liars, or just not competent enough to understand and evaluate sources and properly report what they said. If we allowed unpublished sources then I'm sure a huge amount of nonsense and crap with alleged unpublished sources would be added. I'm not saying you would do it but we don't have a system to judge reliability and competence of editors, or to track who added something without non-trivial work. Currently we can delete unpublished claims and I strongly think we should continue being able to do that. Many reliable sources want to publish notable things so if something has not been published then it's likely to be either non-notable or unreliable. There are exceptions but living with them is much better than opening the floodgates to unpublished junk. By the way, Template:Cite email was deleted at Templates for deletion/Log/2007 April 19 for being unverifiable. You say "there's no policy at all", but there is a clear policy. You just don't like it but it's unlikely to change. PrimeHunter (talk) 12:47, 3 February 2008 (UTC)
 * Also, it's good to keep in mind&mdash;if you do find evidence that a published source is in error, you could always approach that source and ask them to correct. Many will, and if they do, their correction becomes a perfectly reliable and usable source. Seraphimblade Talk to me 13:03, 3 February 2008 (UTC)
 * One limitation is that Wikipedia is many things to many people, but it cannot be all things to all people. One thing we are not terribly good at is to take on the role of a research institution, publisher, etc.  We don't have the editorial oversight or management to meet those standards.  If you're reviewing original sources and corresponding with the people involved, I think the best place to get that into the record would be a blog (which, under the right circumstances, could be a reliable source to the extent you are an expert in the field or it publishes material such as email written by third parties).  You could also ask the source if they would in fact self-publish their comments or submit them somewhere.  If you're more ambitious and want to be more solid, you could contribute freelance articles with these results to newsletters, papers, and so on.  Some reliable place where information of that sort gets published.  I know that's tedious, but perhaps it's tedious because the work of creating knowledge is tedious: for every minute of actually writing new stuff there's an hour of verifying, editing, reviewing, and convincing someone else it is notable and meets publication standards.  The Nova Laser article is a perfect example - that may be the most accurate depiction anywhere, but if someone wants to research this for a book, news article, etc., and sees that Wikipedia contradicts published sources or contains information not verifiable anywhere, they can't really use it.  Perhaps the Lawrence Livermore National Laboratory or the scientific or engineering groups involved have a history newsletter that would be interested in an article on this tidbit.  Anyway, I think it's perfectly fine to use (without citing) primary material to inform your decision on how to edit an article and what to emphasize.  If you hear from the source that an article is inaccurate, it's probably better to do something about it than to leave an untruth in the encyclopedia.  I would look for a contrary source and try to resolve the contradictions, or look more closely at the sources and change an "X is true" statement to a "Y claimed X" one, and failing everything else, leave a note about all this on the talk page to at least get it on the record and create a foundation from which you or others could pick up the matter later. — Preceding unsigned comment added by Wikidemo (talk • contribs)
 * Argh! Please don't suggest that people 'self-publish' their material so it can be put into Wikipedia. I think I know what you meant (a reputable entity putting up a web page, a government agency publishing a study, etc.), but 'self-publish' opens up a floodgate to potentially bad material. -- <b style="color:navy;">Donald Albury</b> 13:46, 3 February 2008 (UTC)
 * If you self-publish it would be a conflict of interest to put that in Wikipedia. But self-publication is a good alternative to Wikipedia in dealing with original research.  The question of what to do about unpublished emails is interesting, though.  The content reliability is the same wherever they are published (though perhaps a little more concern about accuracy of reproduction and authenticity).Wikidemo (talk) 14:45, 3 February 2008 (UTC)

It seems Wikidemo is the only one here that is addressing what I was actually talking about: the existence of published untruths vs. unpublished truths.

Let me re-state an example. The history of the Star Trek game, widely spread around, is a fable that Mike dreamed up to sound like a "real company". Centerline Engineering was a name he made up for fun, and put into the REM statements of the program. Those statements is its recorded history. Had I not been contacted by Mike, this fable would be up here on the wiki, passed off as truth, with lots of references to ensure that this "truth" was unchangeable.

Surely the collection of brains here can come up with something better than "no". Can't anyone think of something here? Please don't quote regs at me, I avoid beaurocrats in the real world, I'm going to do the same here.

Maury (talk) 21:55, 9 February 2008 (UTC)


 * There is a reason why "verifiability, not truth" is used as a catchphrase in this policy. The truth is often murky, and one man's "truth" is often another man's "lie". We cite reliable sources so that when someone says, "I don't believe this", we can say, "we're not making this up, we got it from this reliable source that you can check for yourself." -- <b style="color:navy;">Donald Albury</b> 23:59, 9 February 2008 (UTC)


 * Anyone can claim they were "contacted by Mike". Our consensus is that we want to be able to verify the statements that are added to Wikipedia.  We cannot verify that you were contacted by Mike.Wjhonson (talk) 01:27, 10 February 2008 (UTC)


 * Maury, if this person is a well-known researcher/expert in his field, he can be used as a self-published source under this policy, so long as he's willing to publish the material on his website, for example, and so long as it's not too contentious or about living persons. SlimVirgin  (talk) (contribs) 01:30, 10 February 2008 (UTC)


 * This is the relevant section. Many editors feel this should be weakened further for pop culture, which is what this topic is, although others strongly disagree with any weakening. SlimVirgin  (talk) (contribs) 01:33, 10 February 2008 (UTC)

Proposal: split section by adding new heading, then switch order of resultant sections
I propose to divide the section "Burden of evidence" in two, by adding the heading "Removing unsourced claims" right before the sentence "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it." Then move the new section just above "Burden of evidence".

It'll improve order of presentation, and make the article easier to navigate via the TOC. I look forward to your responses.

The Transhumanist  20:21, 7 February 2008 (UTC)


 * Splitting that section changes the meaning. The section first says which claims must have sources - those that are challenged, likely to be challenged, and quotations. Then it says what to do about those particular sorts of claims if they are unsourced. By splitting it, you make it seem as if all unsourced claims should be removed, which has never been policy. It is true that editors may remove unsourced claims at their discretion, but the policy should not advise them to do things that will only get them in trouble, such as going around removing lots of information only because its unsourced?


 * I'm not sure what's motivating you to worry so much about this. Our policy on verifiability is very well established, and I think it is very unlikely you will find consensus to make it more strict. &mdash; Carl (CBM · talk) 13:26, 8 February 2008 (UTC)


 * My motivation is as a copy-editor. The semantics of this policy are in its interpretations, and aren't included on the policy page itself where they belong!  The way it is written, this policy is nearly incomprehensible except to the reader who is already familiar with Wikipedia politics.  This policy doesn't explain things very well, leaving the reader to learn what it means by trial and error.  For example, you stated "but the policy should not advise them to do things that will only get them in trouble, such as going around removing lots of information only because it's unsourced."  But the policy strongly implies that unsourced material can and should be removed.  If there is a rule that you shouldn't go around removing information only because it's unsourced, and if you can get in trouble for breaking it, then we should have that rule written down somewhere.  By the way, what policy would editors be violating if they did go around removing (lots of) unsourced information?  The Transhumanist  00:29, 9 February 2008 (UTC)


 * The language in WP:V is a compromise between various positions. It would prove difficult to move it one direction or another. If you started removing lots of material, someone would ask you to get consensus for doing so first. &mdash; Carl (CBM · talk) 00:48, 9 February 2008 (UTC)


 * But you didn't answer my question. What policy would you be violating if you removed (lots of) unsourced information?  And  I have another question for you: what kind of "trouble" were you referring to above?  The Transhumanist  08:22, 9 February 2008 (UTC)


 * Agree with Carl. Further, regarding "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it" (which in TH's option would be the first sentence under the "Removal"/"Removing" header):
 * That sentence isn't exclusively about "removal": it also indicates to look for a different subject if you're thinking about starting a new article for which no third-party sources can be found. Which has nothing to do with "removal", the proposed section header would give a very narrow slant on a sentence that has a much broader meaning: inviting people to look for article topics that have sufficient third-party sources available.
 * That sentence isn't exclusively about "removal" as it also only applies to "articles", so to article namespace minus redirect pages and disambiguation pages that are agreed-upon exceptions to "non-articles" in article namespace.

For instance also stand-alone non-disambiguation-page navigational lists that don't comply to this verifiability requirement would not necessarily be "removed", they can live in perfect harmony in portal namespace, as TH knows very well. Portal pages, more or less as a rule, comply to this sort of description (as they're not "articles"). WP:V does not give a mandate to "remove" such pages from Portal namespace for that reason (only: they can't be called articles any more). So, also for this, putting that sentence under a "removal" type of section header would be inappropriate.
 * --Francis Schonken (talk) 15:12, 8 February 2008 (UTC)


 * Peace ("living in perfect harmony") lasts only as long as it lasts. That WP:VER doesn't apply to portal space is your interpretation.  I've not encountered that interpretation anywhere else.  The main difference of portal space is that it can include self-references that should be avoided in the main namespace.  Factual material in a portal is still subject to WP:VER, whether it is in the form of article excerpts or lists.  The policy remains to be tested in a major way in that regard (as far as I know), and no exceptions to WP:VER for material in portals have been officialized.  To assume material is safe in portal space may be providing nothing but a false sense of security. A conflict involving such material's verifiability could arise at any time, the applicability of WP:VER would naturally be debated at that time, and consensus could go either way.  The Transhumanist  09:45, 9 February 2008 (UTC)
 * TH, please don't cut through my talk page comments. If you want to comment differently on two parts of a single paragraph, you can do as follows, under someone else's comment:
 * Re. " "
 * Re. " "
 * --Francis Schonken (talk) 15:05, 9 February 2008 (UTC)
 * I agree with your assessment of the scope of "that sentence". The Transhumanist  09:45, 9 February 2008 (UTC)

rationale for omitting facts
Above, at, I argued Reliable Sources should be independent of vested interests. No one replied to that directly; Wjhonson, thank you for your addition, I agree with that. &#151; Xiutwel ♫☺♥♪ (talk) 07:45, 9 February 2008 (UTC)

Narrative based fact selection
In several articles, I have tried to add facts, which I deemed important to include to balance the article, making sure it is WP:NPOV. There is however a mechanism at work which distorts this. Let me continue on Wikipedia_Talk:NPOV &#151; Xiutwel ♫☺♥♪ (talk) 07:45, 9 February 2008 (UTC)

AMA and NIH
Hi,

Are the American Medical Association and the National Institutes of Health websites good reliable sources for medical related articles? If they say something, can we say they are WP:V sources? 74.45.251.26 (talk) 23:31, 8 February 2008 (UTC)


 * Yes, they are good sources, though they might need other sources to balance them if there are neutrality concerns, but in essence, they are definitely reliable sources. SlimVirgin  (talk) (contribs) 03:45, 9 February 2008 (UTC)

These are excellent sources and can be used in particular to support statements on scientific consensus. Tim Vickers (talk) 18:51, 9 February 2008 (UTC)


 * Excellent and fair-minded consensus statements for topics of practical medicine can in my opinion best be found at Medline Plus. As they are PD, we might sometimes do well simply to copy them with attribution. DGG (talk) 02:58, 11 February 2008 (UTC)

Do sources need to be primarily about the subject of an article?
For an article to count as notable, the impression I've gotten is that it must have a source that is primarily about the subject matter. Thus, an article with sources that are only indirectly about the subject or mention it in passing are not enough to establish notability, even if the sources are reliable. Is this true, and if so, is there a Wikipedia policy or guideline verifying this? --Explodicle (talk) 20:25, 11 February 2008 (UTC)


 * No, the source does not need to be primarily about the subject matter. It simply needs to have significant coverage of the topic. WP:Notability (which you link to) clarifies this... it says:
 * "A topic is presumed to be notable if it has received significant coverage in reliable sources that are independent of the subject."
 * And it goes on to say:
 * "Significant coverage" means that sources address the subject directly in detail, and no original research is needed to extract the content. Significant coverage is more than trivial but may be less than exclusive.
 * You may also want to check out WP:Fringe to see if that applies. Blueboar (talk) 20:47, 11 February 2008 (UTC)
 * Thanks! --Explodicle (talk) 01:34, 12 February 2008 (UTC)

Unreferenced maps a major problem
Increasingly I stumble upon unreferenced maps, even in well-referenced, FA class articles. Such maps certainly violate our policies, from V to OR and often, to NPOV. I don't advocate removal of them, but we need to raise awarness among map makers that they too need to cite their sources. How to do it? I think a good first step would be a creation of templates, inline for captions of unreferenced maps, and general for the image pages of unreferenced maps, and of course we would need a corresponding category. See also my post at Template_talk:Fact. Comments much appreciated. PS. One of many examples of a completely unreferenced map used on Wikipedia. --Piotr Konieczny aka Prokonsul Piotrus 23:17, 11 February 2008 (UTC)
 * We don't require sources for images in the same way we do for text. If you think the image is incorrect, you should start by asking the author what data it is based on. &mdash; Carl (CBM · talk) 23:33, 11 February 2008 (UTC)


 * Images, including maps, should accurately represent things that are stated in the article... those statements are, presumably, sourced. Blueboar (talk) 02:58, 12 February 2008 (UTC)

Third-party sources
I removed "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it." because I think it basically turns WP:N and WP:RS into policy. I know there is a discussion to merge WP:RS into WP:V, but it hasn't happened yet.

The lead of WP:V doesn't mention third-party sources. It does link to WP:RS, but that still is not a policy. I think that third-party sources may be reliable, but a source can be reliable even if it's not third-party. "Verifiable" means that readers should be able to check that material added to Wikipedia has already been published by a reliable source. I don't think that the reliable source has to be a third-party source. A topic may have many reliable sources but no third-party sources &mdash; I don't think that means Wikipedia should not have an article on it. The lengths to which an editor searched for third-party sources is also a factor. --Pixelface (talk) 23:16, 31 January 2008 (UTC)


 * I'm afraid you'll be reverted, because you have not shown that you have attempted to first gain consensus here. This page is very significant and changes here can not be so presumptive.  Please explain exactly what you think a third-party source is, perhaps if we hear you state it in your own words, with an example, we can clarify your particular situation. Wjhonson (talk) 23:30, 31 January 2008 (UTC)


 * I understand, although my change was already reverted. I would consider a third-party source as a source that's independent of a subject. Many of the articles in Category:Integers cite no third-party sources. The article 53 (number) cites no third-party sources, but I think Wikipedia should still have an article on it. The article Jean Valjean cites no third-party sources, but I think Wikipedia should still have an article on it. For that character, the author Victor Hugo is a reliable source, but Victor Hugo is not a third-party source.


 * WP:RS equates "reliable source" with "third-party reliable source", but it is still just a guideline. --Pixelface (talk) 23:46, 31 January 2008 (UTC)

On the Category:Integers, most of the statements within those articles are trivial inferences, or simple deductions which can be verified by anyone with skill in that area. We are allowed to make those sort of statements without sources. On Jean Valjean, although the article does not cite any sources, that does not mean that there are none. In particular, Cliff's Notes exist for this work, which is a third-party source and could be consulted. Assuming good faith we can conclude that whoever wrote this synopsis, did so using some source, or is purely descriptive of the this is a cat variety. Any editor could challenge a statement there by simply adding a template. Pixelface, these articles appear to be hypothetical situations, what particular article however, brought you here?Wjhonson (talk) 23:53, 31 January 2008 (UTC)
 * Furthermore, most articles on integers would be in principle sourceable from John Conway's Book of Numbers, Paulo Ribenboim, and similar sources.


 * Nevertheless I would support a slight weakening of the statement. It is at least in principle possible that a subject could have reliable sources which are entirely first party and still be both verifiable and of encyclopedic interest, although I'd have trouble naming one; we should not make absolute statements policy unless we mean them absolutely. The chief function of the sentence is to resound in AfD debates; it will be almost equally resonant if we include probably, or even except in extraordinary cases. Septentrionalis PMAnderson 03:56, 1 February 2008 (UTC)


 * Could you give us an example of a subject for which there are no reliable third-party sources, but which nevertheless should be in Wikipedia? I am opposed to weakening this provision on a purely hypothetical basis. -- <b style="color:navy;">Donald Albury</b> 10:56, 1 February 2008 (UTC)
 * I didn't have one in mind; but I foresee the blanket statement being used more strongly than it was intended. I don't much care, personally, whether we have an article on every television episode ever made; but it would be a bad thing if we had an AfD war on the subject. Why should the defenders have to hunt up a TV Guide listing for each episode to prove notability when the episodes are published sources for themselves? Septentrionalis PMAnderson 14:04, 1 February 2008 (UTC)
 * If the only source covering an episode is TV Guide, then I would question its notability. Are you saying that the mere existence of an episode of a TV show establishes its own notability? As a secondary note, why wouldn't TV Guide be considered a "third-party" source. -- <b style="color:navy;">Donald Albury</b> 15:38, 1 February 2008 (UTC)
 * No, I'm saying that the episode is a reliable source for its own plot and characters; and that a single sentence on this page should not set policy on whether we should have such articles without much further discussion and many more editors. We do in fact have such articles, so it does not describe consensus practice. Septentrionalis PMAnderson 15:46, 1 February 2008 (UTC)

I agree with Pixelface on this - enshrining notability at the policy level is a terrible idea. Our notability policies are badly broken, robotized policies that alienate new contributors as effectively as, if not more effectively than, they improve the encyclopedia. It's a badly broken guideline, and should not be used to turn this into a badly broken policy. Phil Sandifer (talk) 15:42, 1 February 2008 (UTC)
 * I do, however, support saying something. One of the marks of a vanity article is that it has no third-party sources, and we have too many of those. I therefore propose, since the categorical language plainly has no consensus:
 * If no reliable, third-party sources can be found for an article topic, it is worth asking whether Wikipedia should have an article on it at all.
 * Does anyone deny that the question should be asked? Septentrionalis PMAnderson 16:07, 1 February 2008 (UTC)
 * Of course it should be asked, but not within a policy, unless an answer can also be supplied. Questions "to ponder" / "food for thought" are for talk pages and essays; not policies. Policies are to give people clear answers. If an answer hasn't been arrived at yet, we don't simply stick the question in instead. <small style="font:bold 10px Arial;display:inline;border:#009 1px dashed;padding:1px 6px 2px 7px;white-space:nowrap"><font color="#000">Equazcion •✗/C • 16:13, 1 Feb 2008 (UTC)
 * This is one of the many situations in which it depends (on many factors). There is no universal solution; but that does not mean we have no policy on the subject. We do; and it is and ought to be: to require a reason why we should include something with no third-party source. Septentrionalis PMAnderson 16:19, 1 February 2008 (UTC)
 * Then perhaps that's an addition to consider: "Third-party sources are required except in extenuating circumstances. If non are provided, explicit reasoning must be supplied." That might be a good solution, but it would still be a significant policy change, so we should wait to see what others think before adding it. <small style="font:bold 10px Arial;display:inline;border:#009 1px dashed;padding:1px 6px 2px 7px;white-space:nowrap"><font color="#000">Equazcion •✗/C • 16:24, 1 Feb 2008 (UTC)


 * What about a purely descriptive rather than prescriptive statement? "Articles on topics about which no third party sources can be found are often deleted." Phil Sandifer (talk) 16:39, 1 February 2008 (UTC)
 * I like this proposed wording from PS. UnitedStatesian (talk) 16:41, 1 February 2008 (UTC)
 * What would that mean in practice though? It wouldn't be something people could cite at AfD as delete rationale. We'd be saying those articles "are often" deleted, but how would they actually get deleted? <small style="font:bold 10px Arial;display:inline;border:#009 1px dashed;padding:1px 6px 2px 7px;white-space:nowrap"><font color="#000">Equazcion •✗/C • 16:45, 1 Feb 2008 (UTC)
 * We are not discussing (at the moment) deleting all of our notability guidelines, which are routinely cited in these matters. This would not affect AfD except inasmuch as it would avoid elevating notability from guideline level to policy level. In other words, AfD would function as it currently does - based off of a guideline of notability. Phil Sandifer (talk) 19:10, 1 February 2008 (UTC)
 * Removing the requirement for third-party sources is a very bad idea... it opens the door for all sorts of abuse. Sure, there are potential situations where that rule should be ignored... but we have policy that allows for those rare situations... it is called WP:IAR.  There are occasional exceptions to ALL of our policies and guidelines,  that does not mean that we should scrap the policy or guideline. Blueboar (talk) 16:46, 1 February 2008 (UTC)
 * Agreed. Exceptions are already handled universally. Omitting the explicit requirement for third-party sources would mean people would have all the more reason to create, and a better chance at keeping, articles on things that just aren't notable. I don't think anyone disagrees that topics that haven't been picked up by third-party sources shouldn't be kept on Wikipedia -- except, again, in extenuating circumstances. <small style="font:bold 10px Arial;display:inline;border:#009 1px dashed;padding:1px 6px 2px 7px;white-space:nowrap"><font color="#000">Equazcion •✗/C • 17:00, 1 Feb 2008 (UTC)
 * At the moment, however, we say nothing. How about If no third-party reliable sources exist on a topic, it requires some extraordinary reason to have an article on it? We can quote that at Afd, with "I see none here". Septentrionalis PMAnderson 17:55, 1 February 2008 (UTC)
 * FYI... I have reverted back to the original stronger language (simply saying that we should not have an article at all). I really don't think we need to state an "exception" in the policy for hypothetical and very rare situations.  Let the original language stand, and deal with any IAR exceptions if and when they arise... on a case by case basis. Blueboar (talk) 18:03, 1 February 2008 (UTC)
 * You act as though removing this line from the policy page would undue all of the notability guideline pages. This is not a discussion about whether the third-party source guideline exists - it's a discussion of whether it should be elevated from the looser guideline status to the hard and fast policy status. Given the problems it has at the guideline level, the answer to this should be a resounding no. Phil Sandifer (talk) 19:10, 1 February 2008 (UTC)
 * The language requiring third party sources has been in this policy for over a year and a half, so this isn't a case of it being "elevated". I am very reluctant to remove such long standing language without a lot more input from and discussion by the broader community. Personally, I like the current language.  I think it is good policy, and should not be changed... we should require that a topic has been discussed by at lease one reliable third-party source before we have an article on it.  Nor to I see a reason to bother stating that there might be exceptions to the rule ... it is assumed that there might be exceptions to every policy statement ... that is what IAR is for.  Blueboar (talk) 21:00, 1 February 2008 (UTC)
 * I agree. I don't think we should weaken this policy, which is what this proposal would do. -- <b style="color:navy;">Donald Albury</b> 21:51, 1 February 2008 (UTC)
 * I do not think that switching to a descriptive line about our deletion process over enshrining a guideline at the policy level is a weakening of the policy. Phil Sandifer (talk) 23:21, 1 February 2008 (UTC)

I suggest "If no reliable, third-party sources can be found 'or are likely to exist for an article topic, Wikipedia should not have an article on it."Geni 00:15, 2 February 2008 (UTC)


 * That does largely patch the problem by which this escalates a problematic guideline to the policy level. Phil Sandifer (talk) 00:16, 2 February 2008 (UTC)


 * Phil, what makes you think this came from RS? SlimVirgin  (talk) (contribs) 00:32, 2 February 2008 (UTC)
 * Nothing - it comes from WP:N. WP:RS was, at least the last time I looked, actually not a festering abomination anymore. Phil Sandifer (talk) 01:17, 2 February 2008 (UTC)
 * Oh, I see, thanks. I wonder if it does come from WP:N, though. I remember some version of the third-party wording being in V for a long time -- articles should rely on third-party sources, or such like. SlimVirgin  (talk) (contribs) 01:21, 2 February 2008 (UTC)
 * So suggest a wording, Phil. I'll support if I can. Septentrionalis PMAnderson 01:36, 2 February 2008 (UTC)


 * I suggested a descriptive wording to the effect of "articles with no independent third-party sources are often deleted," which notes how things are instead of being prescriptive. I like Geni's language above as well. Phil Sandifer (talk) 03:34, 2 February 2008 (UTC)


 * The prasing in question (requiring third-party sources or not having an article), was added on April 27, 2006 by Robert A West ... seethis edit. Prior to that edit the section had similar language but required "reputable" sources, without specifying "third-pary".  However, other areas of the policy did talk about relying on third-party sources in general... and had for quite some time prior to the change.
 * The change was heavily discussed on the talk page at the time of insertion ... see the discussion thread entitled: Do articles require third party sources? found in Archive 8 here. Whether the language was originally inspired by WP:N or not... it has been an uncontested part of this policy for over a year and a half.  We can certainly revisit the issue now, but let's not pretend that this is new language or a recent "elevation" of a guideline statement to policy status. Blueboar (talk) 02:10, 2 February 2008 (UTC)
 * I realize the language has been in WP:V for a long time. How it is currently being applied is a concern of mine. Articles like 88 or the 88 (number) are pretty much invalidated by such language. --Pixelface (talk) 00:57, 3 February 2008 (UTC)
 * Pixelface I already addressed that above. The statements in 88 are trivial and verifiable by anyone with skill in that area.  We are allowed to make such statements.  We do not need a source to state the the sun is hot and you can't walk on it.Wjhonson (talk) 01:03, 3 February 2008 (UTC)
 * The statement "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it." still applies to the 88 article. Can you find third-party sources for every year? Every number from 1 to 100? --Pixelface (talk) 01:07, 3 February 2008 (UTC)

(OD) If an entire article is simply a collection of trivial inferences and deductions which anyone skilled in that art could verify, should Wikipedia have the article? Your question isn't V at all, it's N. If you're concerned about 88, submit an AfD. Otherwise it's a red herring argument imho.Wjhonson (talk) 01:10, 3 February 2008 (UTC)
 * I'm saying the part about third-party sources should be removed from WP:V. Information has to be verifiable, information does not have to be verifiable in third-party sources. Why are third-party sources (which are used to indicate notability} part of a policy? I'm not concerned about the year articles. I'm saying they are the exception that proves the statement about third-party sources wrong. It's not a red herring. WP:V does not benefit from such generalized statements. I favor the statement Phil Sandifer came up with, "Articles on topics about which no third party sources can be found are often deleted." over "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it." The second statement is simply not true. --Pixelface (talk) 11:20, 11 February 2008 (UTC)

The edit Blueboar linked to has a different purpose than the present sentence, which I find more supportable; but then I brought Robert West into Wikipedia.

That was intended to end the dialogue that keeps happening in verifiability discussions:
 * A:"Find some third-party sources on this, please"
 * B: "But this subject is so special there aren't any third-party sources. You have to understand our special needs."

by replying "There are no such special fields". Out of the original context, it's a deletion criterion, which really does not belong in WP:Verifiability. Septentrionalis PMAnderson 04:09, 3 February 2008 (UTC)
 * Alright...well, then, there are no such special fields. So let's say so. And if it makes cleanup of unsourced or primary-source-only articles easier, I'm doubly all for it. Realistically, by writing unsourced or primary-source-only articles, especially when no secondary sources exist (as opposed to do exist but aren't cited yet), we're violating 3 core policies: NPOV, NOR, and V. How so? If we write about something that reliable independent sources have chosen not to write about, we're giving that subject more weight than the sources do, and that's their decision, not ours. Seraphimblade Talk to me 10:46, 3 February 2008 (UTC)

I definitely agree with removing "third party" from the verifiability policy, a policy which should be about verifiability, not about attention received (that is something for notability). Having something about reliability is OK (if the source used is unreliable, the info is not truly verified). Independent sources can be better at NPOV issues but many times, the non-independent source is more reliable than the independent one. For example, I have written a few articles about railways, and facts and figures are often inaccurate in an independent newspaper, while they are good in the rail company's website. Another example: If there is an article about a town somewhere, and the source used to support everything in the article is the town government's website, then I would call that article verified. Sjakkalle (Check!)  12:09, 12 February 2008 (UTC)


 * It may be useful to distinguish information by whether or not it is controversial. I think we need to insist on third-party sources for anything controversial. That seems to be the way it works in practice; we tolerate all sorts of marginal sources for stuff that no one feels passionate about, it's the stuff people argue over that we want to provide the best (and that includes being third-party) sources for. -- <b style="color:navy;">Donald Albury</b> 13:23, 12 February 2008 (UTC)

Use of word "academic" in WP:V
I qualified the phrase "academic" community in this section, and because this has gone back and forth a couple of times, discussion appears to be in order. The reason for qualifying is not to allow every quack-medicine or flat-earth claim; it's simply to try to ensure that "non-academic" is not casually equated with "extraordinary". The community expressly rejected WP:SPOV, and accordingly there is a consensus against the idea that Wikipedia expressly favors academic points of view vis-a-vis non-academic points of view. It seems to me that WP:V needs to be consistent with consensus on this matter. The key issue, it seems to me, is to prevent misrepresentation: the problem here is claims which represent themselves as scientific/academic theories or facts which verifiably aren't. We should make clear that this subsection doesn't threaten Star Trek, Ablution, Pizza, Libertarianism, and any of the hundreds of thousands of articles on subjects which are simply not scientific, medical, historical, BLP or similar in nature. (I'd be happy if some of the popular culture articles had any 3rd party sources, requiring academic sources seems excessive and without consensus); similarly, in mixed articles like Bible, the key issue is that religious and academic viewpoints need to be identified as such, the rejection of WP:SPOV means there is no favoritism for one viewpoint over another. Labeling non-academic subjects and viewpoints as "extraordinary" in a general, across-the-board fashion strikes me as risking an end run around the community's rejection of WP:SPOV and hence against consensus. Best, --Shirahadasha (talk) 01:50, 4 February 2008 (UTC)


 * It should be noted that in the United States, there are many issues in which current social policies are different from what most academics believe. For example, the composition of the current United States Supreme Court is much more conservative than most law school faculties, and the Supreme Court on several occassions rejected viewpoints that virtually the whole academic community supported. It also strikes me that in many matters of social policy, politics, morality, etc. where scientific viewpoints are relevant, other viewpoints are also relevant and significant and can easily be reliably sourced. Arguing the merits isn't the point; the fact is, rightly or wrongly, WP:SPOV isn't policy. Best, --Shirahadasha (talk) 01:55, 4 February 2008 (UTC)


 * This is a way of looking at it that I had not considered. I see your point. Thanks Brimba (talk) 02:08, 4 February 2008 (UTC)


 * I completely agree with the reasoning of Shirahadasha's removal of the word "academic"... however, can we come up with a better term than "community". People understand what "academic community" means... but "community" on its own is more vague.  You could say that three or four people who push a particular extraordinary claim form a "community" ... which would defeat the whole intent of the bullet point. Blueboar (talk) 03:37, 4 February 2008 (UTC)


 * We could just say "specialists in the relevant field," which leaves open whether they're academic or other specialists. So that part would begin: "claims that are contradicted by the prevailing view among specialists in the relevant field ..." SlimVirgin  (talk) (contribs) 03:46, 4 February 2008 (UTC)


 * "Experts" might be better, actually. "Specialists" has an academic ring to it. SlimVirgin  (talk) (contribs) 03:49, 4 February 2008 (UTC)


 * My concern is as follows. Imagine there is a group of 10 people who decide the sky is green. They publish their own Green Sky journal, let's say, they consider themselves as experts in their field, and they each act as peer for the others, so their publications are "peer reviewed". Aside from the fact that WP:UNDUE would suppress their views as extreme minority, I would like to see their claim also fall under "exceptional". But one could argue that there is no "red flag" here, since within their own community of green sky "specialists" or "experts", the prevailing view is that the sky is green. This is why I think we need to emphasize the mainstream, so we don't end up relying on tiny self-reinforcing groups (except when writing about those groups). We need to examine claims relative to the mainstream of the entire global community, not just the narrow "specialists" or "experts". Crum375 (talk) 04:08, 4 February 2008 (UTC)

(OD) Here is what it currently states on the policy page:"'*claims that are contradicted by the prevailing view within the relevant community'"

Possibilities:
 * 1) Claims that are contradicted by other reliable sources
 * 2) Claims that are contradicted by experts within the relevant community

I'm having a bit of a problem with how we determine "...prevailing view..." Maybe that's just a nit we can't pick.Wjhonson (talk) 04:21, 4 February 2008 (UTC)


 * I would be OK with just saying "contradicted by mainstream experts." When we get into "relevant communities", it becomes harder to define what that community is, and hence what the prevailing view is. Crum375 (talk) 04:29, 4 February 2008 (UTC)


 * If its defined too tightly, then you get into cases such as “he is nearly the only historian to have worked on this subject.” or some other “community” of one or two or six individuals. Anything with expert in it, or similar will allow someone to alter their definition rather than their source. It may be a catch-22 situation, I don’t know. Brimba (talk) 04:43, 4 February 2008 (UTC)


 * We already say "surprising or important claims not covered by mainstream sources". If we were to add "mainstream" to the contradiction area, that would effectively neuter undue weight.  We'd be in-effect saying, that we only relate mainstream sources.  That's basically why SPOV was rejected.  We're really here to report all moderately interesting sources, all moderately important sources, even when they contradict each other.  In fact those situations make articles more interesting imho.Wjhonson (talk) 04:49, 4 February 2008 (UTC)


 * I should have read Crum’s rational instead of just skimming it. His concern is the same as mine, just expressed more clearly, although I am more incline towards the “Claims that are contradicted by other reliable sources” wording. I think his statement of “and they each act as peer for the others, so their publications are "peer reviewed"” is particularly on the spot. Brimba (talk) 04:59, 4 February 2008 (UTC)


 * As I understand it, the concern is a group who are not part of the "usual suspects", and what we're discussing is how to articulate what we mean by the "usual suspects" without being either too narrow ot too vague. One possibility might be a phrase like "accepted authorities". Another possibility might be language specifically covering claims of expertise in novel fields, which I understand the example represents (this might be too specific, since understand the example doesn't necessarily convey all the possible issues.) Best, --Shirahadasha (talk) 08:10, 4 February 2008 (UTC)


 * There's not much point in defining this in too detailed a way, because all the section is saying is that, where an exceptional claim is made, editors should make sure the source is extra reliable. But we should be doing that anyway. SlimVirgin  (talk) (contribs) 08:50, 4 February 2008 (UTC)


 * Especially as, no matter how we word this, we will have to work out how it applies in each case that arises. It will come down to a discussion of whether any individual claim is 'exceptional', and what sources are reliable enough to support it, and we can't prevent that by loading down the policy with details. -- <b style="color:navy;">Donald Albury</b> 12:13, 4 February 2008 (UTC)


 * I think I'm kinda OK with the way it reads right now to tell you the truth. I do understand Blueboar's concern that community by itself is vague.  Can we say "accepted authorites within the relevant community" ?  At least we're not saying academic ;) Wjhonson (talk) 10:55, 4 February 2008 (UTC)
 * You know, I've read the original suggestion again, and the only example that is really relevant is that in articles on Bible-related subjects, religious and academic viewpoints both need to be given space. This is a bad example: religious studies departments exist, and are more than capable of presenting coherent summaries of mainstream religious views. Indeed, most conceivable examples - with the sole exception of pop culture articles - will have the same objection. Can someone come up with an example of something academics will not study? Relata refero (talk) 13:45, 4 February 2008 (UTC)


 * To continue the religion example, the Catholic Church maintains credentials for individuals qualified to teach the Catholic perspective on Theology and compiles secondary and tertiary sources on the subject that it regards as reliable. An academic without Church credentials can offer an academic viewpoint on the subject, but is not considered qualified or peer-reviewed to offer a genuinely Catholic one. If authorities with Church credentials generally agree on something, it is their consensus which represents the dominant view on what the Church teaches and an academic who proposes something inconsistent has to have his or her opinion attributed as such (i.e. as representing the academic's view, rather than as being an authority on the Church's view). One may personally disagree with the Catholic Church's credentialling process and believe that non-credentialled individuals are "more than capable", but it's not Wikipedia's place to do so. Academics have different criteria than the Church for what makes a theory worthy of publication (Academia tends to favor novelty, for example, in a way a religion may not). Catholic authorities have the same say for whether a claim is legitimate or fringe Catholicism as scientists do for whether a claim is legitimate of fringe science. WP:NPOV means no preference. Best, --Shirahadasha (talk) 23:28, 12 February 2008 (UTC)

Tel Aviv
Hi. Tel Aviv almost reached FA status yesterday but failed because the sourcing isnt considered to be good enough.(Featured article candidates/Tel Aviv/archive2) I wonder if there is someone here who would be able to help solve this/tell me where I can find someone who is able to do so. Thanks in advance. Flymeoutofhere (talk) 08:59, 12 February 2008 (UTC)
 * The statements complained of include this: Israel has the highest number of museums per capita of any country, three of the largest of which are in Tel Aviv. sourced only from travel guides and Israeli tourist literature. This was badly put by the reviewers: On the one hand, these are the only sources likely to make such claims; on the other hand, we are not a tourist guide; we should not be including it either. See WP:PEACOCK; name the museums of Tel Aviv by all means, but don't make this sort of debateable claim at all. Septentrionalis PMAnderson 20:47, 12 February 2008 (UTC)
 * Tiny little Andorra has at least 12 museums (see this site) with a population of just under 72,000... that gives us a per-capita ratio of one museum for every 6,000 people. Israel has a population of just over 7,000,000... which would mean that Israel would have to have something like one museum for every 600,000 people for the claim to be accurate. Does it? OK, I know that my quick calculation is OR, but I'm not trying to add it to any article. I just wanted to give an example of why the claim should be taken with a grain of salt. Blueboar (talk) 15:21, 13 February 2008 (UTC)
 * Your arithmetic has a typo. You mean Israel would need more than 1 per 6,000, or 1168 museums. DGG (talk) 21:41, 13 February 2008 (UTC)
 * sorry about that... I stand corrected. Blueboar (talk) 00:21, 14 February 2008 (UTC)

court documents?
I am currently having an issue with a passage in an article that cites a court judgement. I have reason to doubt that what is said is accurate. However I have no way of getting to check that the document says what the article says it does except at an unacceptable cost. Obviously a court document is reliable per se, but it is effectively unreliable if you can't get to see it for yourself. What is the best way to deal with this? ireneshusband (talk) 15:26, 13 February 2008 (UTC)


 * "Obviously a court document is reliable per se", so we treat it as reliable. A document is not less reliable if it is not easily [intentionally vague meaning] available to one person.  -- SEWilco (talk) 15:40, 13 February 2008 (UTC)


 * Indeed, the docuemnt is still verifiable... the fact that you find the cost unacceptable is irrelevant to that fact. It is no different than if someone cited a very expensive book.  If I wish to double check what is said in the book, I may have to go out and purchase the book.  I might not like that, and may find the cost unacceptable... but that is my choice and does not change the fact that I could purchase it if I chose to do so. Blueboar (talk) 15:55, 13 February 2008 (UTC)


 * A court document is considered a primary source, which is generally unacceptable for use in Wikipedia unless accompanied by a secondary source interpreting it. Crum375 (talk) 15:58, 13 February 2008 (UTC)
 * This is especially true in BLP cases, as your case seems to be. Crum375 (talk) 16:02, 13 February 2008 (UTC)


 * Um... primary documents are acceptable... they just need to be used with caution. Certainly any analysis, interpretation or conclusionary statements should not be based upon the primary document (as that is a violation of WP:NOR), and such statements do need to be accompanied by a secondary source... but if all you are doing is repeating what is stated in the primary document then the document is a reliable source.  To take a judicial ruling as an example... We can quote (and cite) the ruling, but that is as far as we can go.  We may not interpret it or say what it means unless we have an independant source to back that interpretation. Blueboar (talk) 17:25, 13 February 2008 (UTC)


 * If it is available only at great cost, or with exceptional difficulty, and someone relies on it for something which is in good faith open to question, it is appropriate to ask for a quotation of the key text. It is incorrect to insert a reference without actually having seen the item, so the person inserting it must have had it available.  DGG (talk) 21:38, 13 February 2008 (UTC)

And depositions are not reliable sources; that's why we have perjury laws and crossexaminations. Is the judgment quoting one? Has it been appealed? Is it being appealed? Septentrionalis PMAnderson 05:28, 14 February 2008 (UTC)


 * A primary source such as a judgment is acceptable as a quote, without any interpretation or analysis. Once we start adding any interpretation beyond the original words, we are into WP:OR territory. This is why we generally need secondary sources for legal cases. Crum375 (talk) 05:54, 14 February 2008 (UTC)


 * There seem to be two issues here. One is the expense. Expense basically isn't a problem, as long as sources are considered reliable it doesn't matter if they are expensive or hard to access. User:Blueboar and User:DGG have suggested some steps to take. The second issue involves the use of court documents. I agree court documents can be reliable, but for limited purposes. In general, only documents by the judge (judments, opinions, etc.) should be used, documents that come from the parties (motions, depositions, etc.) should not. Finally, not all lawsuits are notable and nuisance lawsuits are a fact of life, so particularly where WP:BLP issues are involved, a lawsuit (particularly one that has not reached a judgment) should generally be mentioned in the first place only if it has been reported on by newspapers, law journals, an appelate court, or other reliable third-party sources indicating that the lawsuit is notable. If the lawsuit is notable, one can then go to court records for additional detail. --Shirahadasha (talk) 06:09, 14 February 2008 (UTC)

Do we accept sources that are _accessible_ but not _publicly_ accessible?
Initially I've asked a similar question at the policy village pump. Now it is split in two with one part being this (about the verifiability aspect of that particular question) and the other being about the reliability aspect (going on here on the Reliable Sources Noticeboard). After the original discussion an editor suggested to ask here for expert answers on the verifiability aspect. Here goes. In the EVE Online article an editor recently posted a link to a forum topic in the EVE Aurora forums. Aurora is an organization that helps the developers and game masters of EVE Online organize in-game events. The Aurora forums are accessible. However they are not publicly accessible. Members of Aurora have access to them. The average reader would have to become a member himself first if he wanted to access and verify the source. Since it's not open for the public it has not been published so to speak. However, what do you think about its verifiability? A fellow editor has asked: "How much effort should a random person have to expend to verify a source [...]?" Is there such a cut-off point? Or can we safely consider this particular source to be not published since it's not available to the public?. If there is such a cut-off point it will be useful to know the effort necessary in this case: A reader who wanted to access and verify the source had to apply for a position at Aurora. Which in turn would require him to play te game. Does that make the source verifiable? Is it verifiable enough? Do we have to care? It may very well be considered unpublished and therefore be not acceptable. What do you think? -- Aexus (talk) 07:38, 22 January 2008 (UTC)
 * No, citing to such a forum is not acceptable. Sources must be published. Blueboar (talk) 13:28, 22 January 2008 (UTC)
 * But isn't the same true about many professional publications that are cited on Wikipedia? You either need to subscribe to them or to a paid library service. My local newspaper, for example, is publicly accessible to me because I can walk into it's office and ask to look at it's back library, but because it's not online its completely inaccessible to the majority of Wikipedia editors and readers . I live on the East coast, you live on the West coast, that newspaper might as well be on a protected forumfor all you're concerned. I'd say, yes for basic and non-contentious information and no for anything redflag. Especially no for anything on future developments in the industry. - perfectblue (talk) 18:44, 22 January 2008 (UTC)


 * If anybody can register to get access to a source then I would consider it verifiable (though sources not requiring registration are better). But if you have to apply for registration and can easily be rejected then I don't consider it verifiable. The latter appears to be the case here. PrimeHunter (talk) 19:41, 22 January 2008 (UTC)


 * Newspapers are available in public libraries; there is no requirement that content is available online whatsoever. But using any online forum as if it was a reliable publication is probably unwise regardless of whether registration is required. &mdash; Carl (CBM · talk) 19:51, 22 January 2008 (UTC)


 * I didn't mean that a source should be web accessible. The point that I was trying to make is that a great many of the sources cited on Wikipedia are technically publicly assessable because they are stored in public places such as libraries, but if that public place happens to be a 5 hour bus ride away from the next town then they are not actuallyassessable to the majority of people. For example, how many pages site foreign books or newspapers? The same is true for out of print books and magazines. You might be able to access them, but can anybody else.


 * Take this as an example. Can you actually go out and find me a copy of the Berkendale Gazette from March 5th 2005 which details how resident of Berkendale (Australia) were menaced by a pack of dogs that injured two children? Can you even demonstrate whether of not there is such a paper as the Berkendale Gazette? - perfectblue (talk) 21:21, 22 January 2008 (UTC)


 * We've had this very discussion a number of times over the years. The interpretation of accessibility is that, the item must have the ability to be accessed by any member of the public without undue effort.  In general, imho, this would mean that you do not need to be approved by the subject, author or publisher in any fashion in order to verify the item.  This might be a good way of expressly phrasing the distinction between a library and other repositorys that require owner-approval of your request. Wjhonson (talk) 22:13, 22 January 2008 (UTC)
 * How would define "ability to be accessed by any member of the public"? Who qualifies as a "member of the public"? Do I qualify? (Cautionary note: I live on Boracay island in the Philippines) -- Boracay Bill (talk) 23:29, 22 January 2008 (UTC)


 * I think this is the perfect statement, and should be moved into the policy. DGG (talk) 22:41, 22 January 2008 (UTC)


 * For the particular case of proving whether a newspaper exists at all. There are resources whose only function is to list all newspapers in the world. In addition if a Google search turns up no references to the newspaper (whatsoever) that's a big giant red flag right there. Wjhonson (talk) 22:15, 22 January 2008 (UTC)
 * for current material only, & in countries where the newspapers are likely to be online.DGG (talk) 22:41, 22 January 2008 (UTC)


 * I think that you want to think about "theoretical accessibility" rather than "actual accessibility". I'm not trying to be snide, but most material in the world is inaccessible to most people for one reason or the other: foreign language, not physically available, not available online, available only online and therefore not available to people who are not online, written in technical jargon that is uninterpretable without specialized knowledge, available only in written english not audio and so not available to the blind or illiterate, and so on. The concept is closer to the legal concept of "publicly available". --Lquilter (talk) 22:19, 22 January 2008 (UTC)

I think this is an easy case. It's a web forum, it's not published, so it's not an RS, whether or not registration is reqired (as said above). Period. CBM has already stated the general case, "there is no requirement that content is available online whatsoever." Long out-of-print books or non-digitized newspapers are verifiable, and are allowed. WP:V does not mean WP:ICANFINDITONGOOGLE. Like DGG, I think Wjhonson drew the perfect distinction between public libraries and private repositorys requiring owner consent to access. Cool Hand Luke 23:09, 22 January 2008 (UTC)

I can't think of any reasonable definition of "published" under which such a forum posting would fall, so on what grounds could this ever be considered a reliable, verifiable source? --Tango (talk) 01:29, 23 January 2008 (UTC)


 * Well published only means something like "made available to the public to view or hear". I don't think rejecting on published is the way to go here.  Also rejecting on accessible wouldn't work as anyone can sit at a coffee shop and Google.  Or even ask a friend to do it and mail the results to them.  Not a horrible amount of effort. Scotch my remarks I forget we were discussing a *private* forum space.Wjhonson (talk) 20:51, 23 January 2008 (UTC)

For what it's worth, my take on this would be that it doesn't matter how easily available any particular source is to any particular individual as long as someone here can verify it. WP has the advantage over traditional encyclopedias in this respect. We also assume good faith by default, so it's only where that falls down that the actual source needs to be checked. I remember seeing a page somewhere hereabouts where people with access to certain "rare" sources have volunteered to check them for the public good. I can't find that page right now and I don't recall if the intention was that the editor who has checked the source then signs the citation to "authenticate" it in some way, but if not that sounds like a good idea. Mind you, that would then tie verifiability to the checking editor's standing here - is that what we would want? Anyway there you are, just a few thoughts. —S MALL JIM   16:14, 25 January 2008 (UTC)


 * I think all sources should be publicly available to people willing to go where they are located, or to register for them or buy them without having to be approved by somebody. If a source is only internally available to employees of a certain company then it should be disallowed no matter how many employees there are. Are you saying we should allow classified documents if a few editors happen to personally have access and be able to add it to Wikipedia without violating the law? PrimeHunter (talk) 17:12, 25 January 2008 (UTC)


 * Yes, the sources should be available to some reasonable number of WP editors/users who are not members of some restricted group. We can tolerate reasonable access fees. We should not tolerate restrictions on access based on employment, citizenship or other such factors. -- <b style="color:navy;">Donald Albury</b> 17:36, 25 January 2008 (UTC)


 * I agree with you both: I wasn't thinking of internal documents that are not intended for public access - I should have made it clear I agree that the core criterion should be "publicly available". But the fact that the document in question may be an old, limited circulation, journal only held by a few university libraries (which do not generally allow public access) should not stop the content of that journal from being considered verifiable, even though to 99.999% of WP's readership it is effectively unavailable. Anyway, I've found the page I was thinking of: it's WikiProject Resource Exchange/Shared Resources. Not as impressive a list as I had remembered and with a slightly different intent, but there's some hard-to-get-at stuff there. —S MALL  JIM   18:38, 25 January 2008 (UTC)
 * Yes I would have an issue quoting a source that was not "publicly" available. If a university, has a collection of material, of which they only allow a hand-picked 12 people to look at, that is not publicly available.  Most large university libraries *do* allow public access, and even do so to their rare collections.  You'd need to post a specific case for us to discuss.Wjhonson (talk) 21:37, 25 January 2008 (UTC)
 * Rather than addressing specific examples, I was agreeing with Lquilter's and perfectblue's interpretations of verifiability being linked to the concept of "theoretical accessibility" as opposed to being "reasonably accessible". What's reasonably accessible to someone living in a university city is very different to what's reasonably accessible to someone living on Pitcairn Island, so I don't think it's helpful to think of verifiability in that way. Then extending the concept of theoretical accessibility into the real world leads us to the idea of one person checking a source for someone else, which in turn leads to the concepts of AGF and trust. —S MALL  JIM   11:29, 26 January 2008 (UTC)
 * The problem with "theoretical accessibility" is that it doesn't help us with practice. Maybe it will help to distinguish 'who' from 'where' when talking about accessibility. As for 'who', I would say that an item is accessible if anyone who shows up at a library or other location where the item is held can freely review it. As for 'where', I think an item must be available to anyone who shows up at more than one location in more than one city and, hopefully, in more than one country. For one thing, the last part would exclude unpublished archives, municipal records, etc. I might also add that if a book is so rare that only one or two copies exist, it is likely that if the book really is relevant to an article, there will have been commentaries and analyses of it published, and we can cite those derivitive works. -- <b style="color:navy;">Donald Albury</b> 11:56, 26 January 2008 (UTC)
 * I have no problem with citing sources that are not widely accessible. What if you cite an internet page but 5 months later that page is deleted and can never be found again? Is that still a viable reference...I believe yes. That is why you have to put the date that you saw the information. It is the ame reason why you have to write the edition number of a book, because one day that book will be reprinted and it is possible that information will be lost. It is still a reliable source even though it is no longer verifiable. There is an article Aquinas College, Perth that makes a lot of use of sources that are only available via the password protected school intranet...so unless you are a student you can't verify it. Does that make it unverifiable...I don't think so. 60.229.28.247 (talk) 01:10, 27 January 2008 (UTC)
 * If an external link no longer works, we have two choices; link to an archive of the page (such as on the Wayback machine) if one exists, or remove the link from Wikipedia. If a web page is gone and a copy does not exist in an accessible archive, it is no longer usable as a source. The whole point of verifiability is that editors and other readers can go to the cited source to verify what it says for themselves. If the web page is no longer available, verifiability has been broken. So, yes, if nobody can see the source any more, it is unverifiable. -- <b style="color:navy;">Donald Albury</b> 14:22, 27 January 2008 (UTC)


 * Please note there is actually a third option: If the link to a newspaper or magazine article is no longer available because the article has, for example, gone into the pay archive, we simply remove the URL but retain the article reference. A Los Angeles Times article citation about a movie producer remains a Los Angeles Times article citation about a movie producer. The article still exists, and this is, in fact, the kind of citation done in print publications such as bopoks and scholarly journals. --Tenebrae (talk) 17:39, 27 January 2008 (UTC)


 * Good clarification. -- <b style="color:navy;">Donald Albury</b> 21:32, 27 January 2008 (UTC)
 * Yes I agree with Donald Albury here, if a citation is to web-only information, and that page is later discovered to be a dead-link, then any editor can both remove the link, and the statement it references. Or mark it with a  tag to alert other page-watchers that this statement's original reference is now unavailable. Wjhonson (talk) 02:33, 28 January 2008 (UTC)
 * I don't agree. Just because you can't access a web page doesn't mean it won't come back. Web servers aren't particularly reliable. And if it's a long-standing reference people may well have checked it, we shouldn't delete verified information without due process.- (User) WolfKeeper (Talk) 04:14, 28 January 2008 (UTC)
 * This policy does not say that information has to have been verified, it says that information has to be verifiable. Once the link is gone and there is no archived version and no paper published version, it is gone, it is unverifiable. Dead-links are very unlikely to come back. And if a statement in an article was supported only by that one web site, it is questionable whether it should have been there in the first place. -- <b style="color:navy;">Donald Albury</b> 12:51, 28 January 2008 (UTC)
 * Yes, but the fact that it *has* gone away is OR. How many times do you check? Check it once and delete? No. There has to be a black and white guideline on how to deal with this issue.- (User) WolfKeeper (Talk) 17:50, 31 January 2008 (UTC)
 * I don't think we can have a firm and fast rule on this. It depends on the situation, and the statement being sourced to the now dead weblink.  For one thing it depends on how controvercial the statement is... if it is uncontrovercial, then there is no harm in leaving the statement in the article for a while.  Remove the dead link (leaving a nice edit summary to explain what happened) and simply request a new citation.  Either someone will provide one or, eventually, the statement will be taken out.  No rush.  But, if the statement is controvercial, then more direct action needs to be taken.  It is true that we can not assume that the information is unverifiable... so the first thing to do is seriously check to see if there is another reliable website that repeats the controvercial information (and also check that the original website has not simply changed its address).  If, after a serious search, the information can not be located... then we should remove the link and the statement (again with good explanations and probably a note on the talk page as well).  It is up to those who wish to include the controvercial information to find a new source. Blueboar (talk) 18:52, 31 January 2008 (UTC)
 * No, don't remove the dead link. Just because it is not accessible to you does not mean it is not a usable citation.  There are several ways information can be retrieved by using only a URL.  Or the URL might only temporarily be nonfunctional.  If you want to ask for another source, just put the request next to the old citation.  -- SEWilco (talk) 19:32, 31 January 2008 (UTC)

Thank you everybody for your opinions so far! What I've learned is that the forums in question do not meet Wikipedia's verifiability standards, namely public availability. This source cannot be considered publicly available as it's not up to the reader's discretion to access and verify it. Readers have to be approved by Aurora, the organization that operates the forums in question. This conclusion I take with me and back into the EVE Online article. Thanks so far! -- Aexus (talk) 20:21, 31 January 2008 (UTC)

For clarification
So a question for clarification... What about sources that are only available online *and* require a subscription access to view the entire article? I'm thinking of a magazine that used to be in print many years ago, but ten years ago (or so) moved to a web-only publication format. For a subscription fee of $20/year, you get access to the online magazine (new issues/content) as well as its archives (which includes all of the original paper content as well). If I want to quote an article from this magazine, it's certainly verifiable (at least by anyone who has a subscription, or wants to pay for it). --Craw-daddy | T | 17:12, 3 February 2008 (UTC)

-- Aexus (talk) 00:03, 4 February 2008 (UTC)
 * According to the definition that arose in our discussion this magazine meets the verifiability standards. The reader of the Wiki article can decide whether he wants to access and verify the web-only magazine or not. He doesn't have to approved. It's up to his discretion. If he's willing to pay the fee than that's all that matters. It's like citing a book that's only available at a library in another city. You'd have to spend money on a bus ticket to access and verify the source for yourself but that's your decision. Nobody stops you from doing it. Both this fictitious book and the web-only magazine have to be considered publicly accessible. From that standpoint they meet Wikipedia's verifiability standards.


 * I disagree that we reached this consensus. Imho, if there is no relatively-free version, where you need no approval, and need not pay the originator in order to verify the content, then to me, it is not accessible, and thus vails V. Wjhonson (talk) 00:52, 4 February 2008 (UTC)
 * I've always understood it to be a fundamental, long-settled issue that we can source factual claims to pay sites, offline sites, etc. The mode of access simply doesn't figure in to verifiability but everything else being the same it's best to use an accessible site, courtesy link, etc. Wikidemo (talk) 01:30, 4 February 2008 (UTC)
 * Imho, it's not fundamental and not long settled that we can source claims to *solely pay sites*. We can source to pay sites, if there is a relatively-free alternative which does not require payment to the originator of the data.  We can source to a pay site like the New York Times online, because you can also go to the library and read it for free.  We can source to a pay site like some obscure magazine online if there is a possibility that you can get a copy of that item without the need to pay the author or publisher anything.  That does not mean it's free to you, if you have to send a letter or make a phone call to some remote librarian or colleague.  It means there's no direct interaction required between you and the author or publisher.  That's the effective language we have discussed.Wjhonson (talk) 01:35, 4 February 2008 (UTC)

-- Aexus (talk) 12:34, 4 February 2008 (UTC)
 * Ah alright, my mistake. I answered Craw-daddy's question with the discussion about private sources in mind. I thought that a pay site is verifiable as it's my own decision if I want to pay the fee.


 * This is all getting very complicated. It seems there are few if any sources that are both reliable and inaccessible, because the nature of scientific and academic research, news reporting, etc., is that they all have their various own versions of verifiability, and facts are not trusted unless they are subject to acceptance by a community.  I can imagine a few special cases, say, religious research on a secretive sect, military secrets, police investigations, sealed court records...but for the most part the two questions boil down to the same thing and we can usually reject the source as unreliable.  Regarding the "Aurora" site that sparked this discussion I would simply say that the material is either unreliable or original research and leave it at that.  But yes, I see how if a site is accessible only to a small exclusive group it does not help for purposes of verification.  As an analog, if I added a comment to an article that the tribal elders of X practice a form of magic in which Y, and I cited it to a tablet Z kept by the chief of the tribe (and not studied, described elsewhere, or otherwise available), I think you can say that we would dismiss the claim as unverifiable.Wikidemo (talk) 13:31, 4 February 2008 (UTC)

1. If I can ask a librarian to check a "pay-only" citation for me, and that librarian has access to the "pay-only" website as part of his job, does that make the source verifiable? 2. What does "verifiability" really mean? Does it mean actually seeing a copy of the source myself, or will having someone I trust tell me that the source agrees with the passage in the WP article be enough? —S MALL JIM   16:16, 4 February 2008 (UTC)
 * Two questions about your last post, Wjhonson, if I may:

I think the bottom line has to be that the concept of verifiability cannot depend in any way upon the person who requires the verification. Would anyone disagree with that? —S MALL JIM   14:14, 4 February 2008 (UTC)

On reflection I'd better clarify that because it is possible for the concept of verifiability to be tied to a hypothetical man on the Clapham omnibus. What I mean is that the verifiability of any particular source cannot depend in any way upon the individual who requires that source to be verified. —S MALL JIM   14:45, 4 February 2008 (UTC)


 * I think so, too. It shouldn't matter who wanted to access and verify a source. If I want to access the source I should be allowed to do so. As soon as it's not up to me anymore the source becomes unverifiable for me. Thus it fails one of Wikipedia's standards which in turn means that we consider it invalid. -- Aexus (talk) 17:49, 7 February 2008 (UTC)

One thing that occurs to me is that although you yourself may not have access to a source, if you know that a lot of other people do, then that is some reason to be confident of its reliablity. For instance you may not live near a university library or have online access to a particular academic journal, but you can be reasonably confident that enough editors do have access that one of them will be able to verify any reference to it and ammend wikipedia accordingly.

On the other hand, to give the example that has brought me passing by, if your source is a court judgement somewhere, you would require the help of an editor in the city where the court is located who has the time and the inclination to visit the courthouse him or herself. This is clearly not a huge demographic, making the document hard to verify. ireneshusband (talk) 09:41, 14 February 2008 (UTC)

A different question
I can understand saying that a pay to view sight is OK... I am less happy about the idea that you have to be approved in some way. What is the approval process? Any one can pay, but will anyone be approved? In any case... There is another issue that may come into play here. Do members of this forum post anonymously, or do they post using their own names? If the former, I don't think we can call it a reliable source, since there is no verification as to who says what or the qualification of any given poster to say what they say. If the latter, what is the policy at the site... how do they verify that someone is who they claim to be? Blueboar (talk) 01:15, 4 February 2008 (UTC)

-- Aexus (talk) 12:28, 4 February 2008 (UTC)
 * The discussion about reliability of these forums goes on here on the Reliable Sources Noticeboard. In short: members of Aurora post with their in-game name. Thus they are not anonymous. Also, we can consider forum posts by some members reliable. For example posts by the leader and his substitutes. They are notable. Even if it's a forum we could then consider the source itself reliable as the statement in question was made by a notable person. The reliability issues a resolved. In my opinion the forums don't meet Wikipedia's verifiability standards. There's no payment involved in this case. At least not directly towards Aurora. If I wanted to access the source I had to apply for a position at Aurora. And to do that I'd have to start playing EVE Online. Which in turn costs a monthly fee of around 15,- USD. A little less if you buy access for several months in advance. As for other criteria I can't give a definite answer. New players probably aren't accepted. One had to play for, say, six months first. That way Aurora can ensure that members know at least some aspects of the game.
 * Reliability is a factor of a source's verifiability, so a discussion of its reliability here is appropriate. In any case... for multiple reasons, I don't think the site is usable as a citation on wikipedia. Blueboar (talk) 15:18, 4 February 2008 (UTC)

Images
I haven't run into this yet but a couple recent posts got me thinking... what is the Verifiability process for images in an age of photoshop that can make almost any image appear "real". For content, we push the V policy, but then we promote and even require "free" images created by users. We have none of the reliable source requirements for image content - no fact checking. An image could easily be doctored for bias (I can think of several examples in recent media where such has happened). Anyway.. this has likely been discussed but it popped into my head so I thought I would bring it up. Morphh  (talk) 3:56, 14 February 2008 (UTC)
 * We don't really have a verification process for images. Part of the problem is that, due to copywrite and similar issues, many verifiable published images can not be used on Wikipedia... which means that we often have to rely on images that are self-created by our editors.  We have to assume good faith and assume that an image portrays what the creator says it portrays.  That said, images can be challenged just like any other material in wikipedia.  If you have reason to believe that an image has been doctored, challenge it. Or better yet, find a better image to illustrate the article and substitute it.  Blueboar (talk) 19:59, 14 February 2008 (UTC)

Third party publishers
Shouldn't it be defined in the article what it means to be published by a third party? What DOES it mean? Snake item (talk) 16:40, 14 February 2008 (UTC)


 * Snake, please don't change the policy; it needs to be stable.


 * The meaning of third party depends on context. Usually, it refers to a source other than the subject of the article. SlimVirgin  (talk) (contribs) 16:47, 14 February 2008 (UTC)
 * In this context: "Articles should rely on reliable sources, published by a third-party that has a reputation for fact-checking and accuracy." What does a third-party publisher mean? Snake item (talk) 16:53, 14 February 2008 (UTC)


 * I think you changed it. It says: "Articles should rely on reliable, third-party published sources ..." It means material published by someone other than the subject of the article. SlimVirgin  (talk) (contribs) 17:19, 14 February 2008 (UTC)


 * Also, there's no need to keep adding FAQs. The sentence says "self-published material," so clearly it would include self-published FAQs. SlimVirgin  (talk) (contribs) 17:19, 14 February 2008 (UTC)

Another thing. That sentence "Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy" is saying that the SOURCES should have a reputation for fact-checking and accuracy. Shouldn't it be saying that the PUBLISHERS have a reputation for fact-checking and accuracy? It's bad writing isn't it, or does the writer actually mean that the SOURCES should have a reputation for fact-checking and accuracy? Snake item (talk) 17:00, 14 February 2008 (UTC)


 * See above. It's not bad writing. It's written that way deliberately because a source can be a writer or a publisher. SlimVirgin  (talk) (contribs) 17:20, 14 February 2008 (UTC)
 * Wow. This is one bad article. There is so much awkward writing and ambiguity in this thing. I leave it to you guys who are guarding it to deal with. So long. Snake item (talk) 17:37, 14 February 2008 (UTC)
 * Thank you for your kind words and encouragement. SlimVirgin  (talk) (contribs) 18:50, 14 February 2008 (UTC)
 * No problem. I could work on this article alone, and it would be impeccable in just a couple days. This article has been worked on for here for years has it not? And this is all you have to show for it? Pretty pathetic. I think Wikipedia is a case of too many cooks spoiling the broth. Snake item (talk) 18:56, 14 February 2008 (UTC)
 * Snake, if you think you can clarify the wording, we encourage you to make the attempt... the accepted method is to propose the changed language here on the talk page first. I can not guarentee that your language will be accepted, but I can guarentee that people will tell you why the language isn't accepted.   And who knows, it is always possible that people will react by saying: "Yes, that clarifies the policy perfectly... why didn't we think of that?".  Blueboar (talk) 19:50, 14 February 2008 (UTC)


 * Ah well, there you have it. I also think that if I could work on it on my own, it would be much better. Perhaps we all think that. But this is the result of many people's input, which is why at times it's not perfectly worded, but also why it's quite insightful. SlimVirgin  (talk) (contribs) 19:54, 14 February 2008 (UTC)

Extremist sources
There is a discussion of the correct application of this policy to extremist sources at Reliable sources/Noticeboard. Tim Vickers (talk) 17:32, 9 February 2008 (UTC)


 * This discussion seems to have come to consensus, does anybody else have any comments? Tim Vickers (talk) 19:32, 11 February 2008 (UTC)

This problem has now been resolved. Thank you to everybody who commented. Tim Vickers (talk) 16:54, 12 February 2008 (UTC)


 * Apparently this issue has not been resolved as Crum375 has stated today that "I see no "consensus" there." and replaced the Animal Liberation Front links in the Animal testing article diff. Could people please examine this question again? Further comments would be much appreciated. Tim Vickers (talk) 18:58, 17 February 2008 (UTC)
 * I believe this issue should be discussed on the talk pages of the relevant articles. We do accept published material from sources about themselves, and the images in question were taken by the sources as eyewitnesses. Also, the "Britches" video is displayed on PETA's PetaTV, which is not "extremist" by Wikipedia's definition, as PETA has nearly two million members, with many celebrity supporters. Crum375 (talk) 20:12, 17 February 2008 (UTC)

common error
i'm not a frequent wikipedia editor, but wanted to point out that nearly every single article on a film or tv actor has a first sentence with an unverified/uncited assertion, something like: "Skeletor is a Norweigan actor best known for his performance in the television series, He-Man and the Masters of the Universe." ...best known to whom? I dunno if there's some kind of wikipedia insider community, but this might be pointed out to folks working mostly in tv/movies...

98.163.232.140 (talk) 09:27, 17 February 2008 (UTC)


 * If there is a dispute about the "best known for" part, the wording gets changed. Usually this is covered in the body of the article in some way, based on commentators discussing the actor's most famous or breakthrough role. When sources say some actor "went up a step" due to some recent role, or explicitly say some role made the actor famous, I think it's fair to use the "best known for" phrase. Gimmetrow 20:44, 17 February 2008 (UTC)

Policy and practice at odds
The purpose of our content policies is to set forth what material should be challenged. This policy isn't clear on that point.

Whether or not a fact is challenged for lack of a source is superfluous to determining whether it should be challenged. Policies set forth what is acceptable or not, but if the acceptability is defined in terms of the enforcement of the policy, then it's circular and unnecessarily redundant and confusing.

Wikipedia's content policies set forth what should be removed (or challenged, as the removal process most often includes challenging as a step). This policy defines what should be removed in terms of what is in the process of being removed (via challenge). Statements like 'Enforce this policy if it is already being enforced' or 'remove material (without sources) only if someone wants it removed (for not having sources)' or "any material that is challenged or is likely to be challenged may be removed", are worthless. The basis of challenge is this verifiability policy (which is the inspiration of challenges like "this material violates "Verifiability"), so the policy shouldn't define what should be challenged in terms of challenge. That's redundant, and is confusing because it loops back on itself.

The policy should clearly define what should be challenged in the first place. It should also make clear if material can be removed without challenging it first. And it should specify which processes of removal are appropriate and how challenge fits into the process. We have available to us the delete key, discussion on the talk page, tagging, and nominating for deletion at AfD. When should each be used? As this policy is where it all starts, it should put things in proper (clear) perspective. The Transhumanist 20:58, 10 February 2008 (UTC)
 * I hear you, but it would be a lot of work to define when a verifiability challenge is appropriate or not. As is we leave up to a consensus and common sense approach.  Sometimes a procedural solution that lets people work things out on the fly is better than an attempt to codify the substance of when an edit can be made or not.  At the extreme, people who add "fact" templates or remove material by way of wikigaming or being contentious are considered behavior problems, but we're open to any good faith questioning of a fact.  I don't think unwarranted removal of obvious facts has become a big problem here.  If the fact is not so obvious then it isn't a whole lot to ask that someone who wants to defend it to cite it.  A bigger issue is when people challenge facts in order to change the POV of an article.   That only happens regularly in a small class of controversial subjects, and where it does the controversy itself is a good sign that people who make edits should be extra careful to cite their sources.Wikidemo (talk) 21:10, 10 February 2008 (UTC)
 * It wouldn't take much work to define when a verifiability challenge is appropriate or not: you did it just now. It's merely a grammatical issue (and not one of codification).  The current policy is poorly written, writing it well wouldn't be too difficult, as you've just shown (your explanation above is clearer than the policy is).  Note that consensus is often reached based on what the participants of discussions believe this policy means.  And if the policy isn't clear, then the formation of consensus based on it may be misguided.  The Transhumanist  22:50, 10 February 2008 (UTC)


 * TH, above you said: "The policy should clearly define what should be challenged in the first place." and now "…consensus is often reached based on what the participants of discussions believe this policy means." The first of those statements is arguable, the second is probably as it should be. Some people hold that the purpose of Wikipedia's policies is to document what happens in practice, not to prescribe closely what should and should not be done (except for NPOV). If this view is true (as I believe it is, broadly) then because there is a wide range of opinions on every aspect of every policy, it is right that the policies reflect this wide range by using unexplicit wording. That, I understand, is why they carefully use the word "should" more often than "must", for instance. So far from being poorly written, the policies are as they are for very good reasons. —S MALL  JIM   00:30, 11 February 2008 (UTC)
 * I've read through the above debate and would like to add a couple of points regarding 'Verification is required':
 * In submitting (or stated more precisely, blundering through) a recent AFD I found the policies as I read them were at odds with the arguments used by those editors who monitor AFDs, and the stated policies are also at odds with the direction of the debate above. I don't think this was a case of me misreading the policies toward my own POV.  This inconsistency should be corrected in one direction or the other, pretty please.  And the inconsistencies I experienced were EXACTLY those documented by Transhumanist in the discussion above, so his points have really resonated with this editor who has learned many things the hard way.Slofstra (talk) 19:18, 11 February 2008 (UTC)
 * When the scope of wiki was largely similar to existing encyclopedia's, basically within the scope of existing secondary sources, it was probably okay to let in-line citations ride and things would work themselves out, the sources were readily available. These days most of wiki is new tertiary and secondary material, much of it with few or no secondary or even primary sources.  So IMO much more rigid enforcement of WP:V is required, to maintain wiki's credibility in the public view.Slofstra (talk) 19:18, 11 February 2008 (UTC)
 * Is it fair to let editors add material letting citations slide, then expect other editors to do the heavy lifting later? And is it feasible to operate in this general manner, because in the case of new-scope material I've worked on the citations never come.Slofstra (talk) 19:18, 11 February 2008 (UTC)
 * It's much easier to delete erroneous material based on a lack of citations than to debate factuality. If more emphasis is placed on the latter as a modus operandus you may lose credible editors who don't enjoy edit wars.Slofstra (talk) 19:18, 11 February 2008 (UTC)
 * Can an editor delete material that is uncited, without having to debate whether the material is not factual or not 'verifi-able'? (I stand corrected but I believe that much false information can be considered to be verifiable as the word is defined above.) Slofstra (talk) 19:29, 11 February 2008 (UTC)
 * There was also some confusion in the AFD discussion as to whether providing references means simply listing references at the base of the article or citing each fact individually. Slofstra (talk) 19:29, 11 February 2008 (UTC)
 * Citing at the base is fine as a minimum standard, and is acceptable when everything in the article is taken from one or two sources. In line citation for each fact is preferred for most articles, especially those on controvercial topics.  In other words both are allowed. Blueboar (talk) 02:13, 12 February 2008 (UTC)
 * The problem is that an article starts that way, and then other uncited edits are added, and you cannot tell what came from where. So basically, what started reasonably clean is polluted and you cannot tell which is which.  Perhaps this is not widespread.  I see it though in the area I work in.  Slofstra (talk) 02:43, 12 February 2008 (UTC)
 * So true! Editors will stick a new statement in immediately before an in-line citation. If not caught right away, it then looks like that statement is covered by the citation. We also have editors who 'update' information, such as census data, without changing the citations. The only way I know to handle these cases is to immediately revert to the version that is actually supported by the citation(s). Of course, these cases grade from well-intentioned, but naive, editors through various sorts of vandals to some very deliberate attempts to slant articles. -- <b style="color:navy;">Donald Albury</b> 13:33, 12 February 2008 (UTC)
 * This is going to occur whether we cite at the bottom or in-line... editors who are not familiar with this policy will add information that is not backed by the sources no matter where we list them. The only way to combat this is for those who do know the sources to be vigilant.  But we are getting off topic.  The question is whether we should include a firm definition of what can and can not be challenged.  I think this would be a mistake.  As we are now, we have flexibility.  Anything can be challenged, either because it is unsourced or because it does not correctly reflect what the source says.  However, it also allows for the retention of information that is likely to be verifiable but not yet verfied.  We leave it up to the good judgement of our editors. Blueboar (talk) 16:20, 12 February 2008 (UTC)
 * Where I'm feeling a little conflicted is that according to the policy I can only challenge a statement if it is not 'verifiable'. The interpretation by 'Carl' above is that as long as third party reliable sources exist, a statement is 'verifiable' and passes wiki policy. So, if I'm unsure of an uncited statement I cannot delete it, because third party reliable sources may very well exist. I'm loathe to delete things without a wiki policy to back me up.Slofstra (talk) 22:07, 12 February 2008 (UTC)


 * If the statements have been tagged and no one has bothered to provide a citation, you are free to delete the statements. It is polite to wait a bit, but I would say that if a citation has not been provided within a couple of days of the notice being placed, it is unlikely to be provided, even if you wait a year. Personally, I tend to prioritize; the less likely I feel that a good reliable source exists, the quicker I am to delete. But I think the policy is clear; if a statement has been challenged and no citation to a reliable source is forthcoming, the statement can be removed. -- <b style="color:navy;">Donald Albury</b> 22:27, 12 February 2008 (UTC)


 * On the other hand, the motivation for permitting challenges is to allow editors to remove material which they suspect may be incorrect. A "challenge" shouldn't be made only because there is no reference, for information which you don't suspect to be incorrect. I think this is the main source of confusion about cn tags. Although some people think it would be nice, there is no policy or goal that every sentence will eventually carry an inline citation. The things that need inline citation are the things that are attributed to an individual, controversial, often misunderstood, etc. &mdash; Carl (CBM · talk) 12:48, 13 February 2008 (UTC)


 * So what would be your criteria be for suspect may be incorrect? In subjects that I know something about I see all kinds of nonsense added to Wikipedia. In an article about something with which I'm unfamiliar, how can I trust any unsourced material? If I don't know whether or not the material in an article is correct, then I think I'm justified in suspecting that it may be incorrect. -- <b style="color:navy;">Donald Albury</b> 15:22, 13 February 2008 (UTC)


 * My general practice has been, in the absence of WP:BLP issues, vandalism, criticism or essay-style content especially likely to be personal opinion, a highly contentious subject, or similar issues where sourcing is especially important, to tag unsourced, questionable content and to give the editor a courtesy period of time to supply a source, which can be weeks or months for plausible content in non-contentious articles. I also try to mention the matter on the editor's talk page if it is a recent addition. I believe providing a notice and delaying deletion provides a reasonable balance of enforcing the policies while not unduly frustrating newbies. Nonetheless, at some point, a source has to be supplied or the content deleted. --Shirahadasha (talk) 23:05, 12 February 2008 (UTC)


 * Thanks for the affirmation, but just to belabour the point, and I do feel sheepish adding this. But in light of what is said, it would seem that Transhumanist is correct, "Content must be verified (not just verifiable)."  It seems to me a strange thing to allow by policy some types of content to be both entered and deleted.  That being unverified verifiable content.  Actually once such stmts have been deleted, the writer is perfectly justified in adding them back without sources, it would seem. Slofstra (talk) 00:33, 13 February 2008 (UTC)


 * My position has been that the policy should be clear and straightforward, not convoluted. I've never been for verification over verifiability, except in that I thought that's what the policy (and by extension the Wikipedia community) required.  But after analyzing the policy's semantics and the politics revealed by the discussions above and elsewhere, I've decided to support my own position...


 * The danger of a cut and dried policy requiring verification is that most of Wikipedia is currently unsourced. If source citations were to become mandatory, then most of Wikipedia's content would be subject to immediate removal.  That would be a nightmare.  So verifiability is a good standard - that is, the requirement that there are reliable sources "out there".  But, the policy is both unclear and uncommitted on this point, which causes much confusion and unnecessary conflict in AfDs, etc.


 * It's unclear in that verifiability implies verification and is often interpreted as verification. This is evident in AfDs in which articles are nominated for deletion for being unreferenced rather than for lacking sourceability.


 * It's uncommitted in that the policy oscillates between its support of verifiability and verification, requiring verification (i.e., source citations to be included in the article itself) in cases when either the factuality of material is challenged or the unsourced nature of the material is challenged. This is bad, because it provides everyone with a sword by which they can (legitimately attempt to) cut almost anything from the encyclopedia which they personally simply do not like (on the grounds that it lacks references).  In cases of challenge, I think it should be enough for participants at AfD to check the sources (on the internet) to verify verifiability.  Deletion nominators would then be more likely to check for sources on the internet themselves to be sure their nominations wouldn't be dismissed out of hand.  They would also be chastised for submitting frivolous nominations when they failed to do so.  In practice, consensus is often reached on the basis of there being sources out there, without being cited in the article, disregarding the policy.  At which point it is up to the closing admin to decide whether he wants to enforce the policy over consensus, which admins are allowed (and at least in word are often expected) to do.


 * As it stands, the policy, if read carefully, supports verification over verifiability. That is, you can remove material that lacks source citations because it lacks source citations. But the structure of Wikipedia combined with the policy supports non-verification.  That is, you can add material to Wikipedia without sources or source citations, and material isn't in violation of the policy until it is challenged.  The policy doesn't explicitly require such material to be challenged, and so it grows and grows.  If such material is simply removed from an article, it can be restored because it is safely stored in the edit history.  To be removed over objection requires obtaining consensus, which is an uphill battle (and serves to slow rampant deleters down).


 * The problem with the above system is that it wastes human resources, which could be prevented if the policy simply explained verifiability clearly in the first place and enforced verification only when reliable sources cannot be found on the internet.


 * The political reality is that by not actively promoting the enforcement of verifiability what you wind up with is unverified content for the most part. Some Wikipedians want us to go one step further than verifiability by providing citations, because they believe those are needed in order to acquire public credibility, and the current system does pressure editors into providing citations, even though this will never result in citations catching up with new material. In their view, some citations are better than none. But requiring citations up-front across-the-board for factual statements (excluding navigation resources) would make writing Wikipedia articles much more tedious and much less enjoyable.


 * But public credibility would grow over time if verifiability were aggressively enforced, even in the absence of citations, because verifiability is verifiable.


 * Well, that's my take on the matter. I hope the above explanation helps.


 * The Transhumanist 10:34, 13 February 2008 (UTC)


 * Like just about everything, Wikipedia is full of contradictions. It seems to be surviving, and even growing, however. 'Verifiability' is less desirable than 'verification', simply because a reader cannot be sure that a statement is 'verifiable' unless they know where it came from. Even if a reader/editor has verified a statement from a source, that doesn't help anyone else unless a citation to the source has been left in the article. The credibility of Wikipedia would indeed be improved by stricter verification, but I don't see any drastic moves in that direction succeeding any time soon. In the meantime, the current wording of this article encourages more reliance on quality sourcing without unduly splitting the community. So, how would changes to this policy improve Wikipedia without causing other problems? As for the effort of providing sources up-front, I do that in all the articles I create or add substantially to, and I find doing the research to be a great part of the fun in creating or adding to articles. I realize that many will not agree, but I think working that way is more creative and satisfying than relying on (often fallible) memory. -- <b style="color:navy;">Donald Albury</b> 11:45, 13 February 2008 (UTC)


 * One reason for needing a balance here is that we don't want pedantic users slapping tags on obvious statements (or statements that are obvious to anyone in the field) just for the sake of it. It would tie things up in knots. There needs to be a genuine question. That's why we don't want to prohibit unsourced statements entirely, just unsourced non-obvious statements. As with everything else in this business, "obvious" can't really be defined, the boundary can't be precisely articulated, so the policy has to be taken with a grain of salt and some common sense. Best, --Shirahadasha (talk) 13:38, 13 February 2008 (UTC)


 * I've been working in a small area with controversy disproportionate to its importance. The approach I use is to remove material I believe to be false or opinionated, and do it on the basis of WP:V.  I have no wish to get into opinion related discussions, although I don't mind consensus building discussion around matters of fact. So my tach has been that I want verification upheld. Because of the nature of my experience here, I have been slow to understand how some editors wish to expand 'wiki' with useful information that is obscure or obvious and thus difficult to source. But I don't agree, Transhumanist, that the answer is to relax WP:V although we agree the practice and policy are inconsistent. For me personally I would just give up if WP:V was relaxed.  But it strikes me now that policies and practices should take note of the fluidity of wikipedia. They seem to assume that the text emerges at once in full bloom.  And that is how things work in the academic world; you don't publish until you meet academic criteria. Whereas in practice here, new articles and material have fairly relaxed verification standards. It should be expected that as an article improves over time new edits should be held to a tighter standard.  Maybe more event related guidelines would be useful: conditions required before you add, before you remove text, before you submit an AFD.  I found out the hard way that for an article to just exist does not require much. Or maybe an article's level of integrity should be assessed and tagged - as an article's level increases, the standard is tougher.  Maybe we should stop tagging the bad articles and start tagging the good ones (I mean in terms of a formal published standard, not a nebulous peer award, or maybe the existing awards should be integrated with policy in some way).  I'm getting a bit wordy, but I'm not far enough along to do better than take a few general stabs here. Fully aware that every solution has its drawbacks, and I don't have a broad-based sense of the problem to be more specific.  What about a scheme where an article is assessed at a certain level of integrity based on objective standards at each level. You would apply for an increase in the same way you apply for AFD or anything else.  The idea that at a certain level, certain kinds of edits just aren't allowed - for example, at the highest level once the entire non-obvious content has been cited, no uncited edits are allowed.  More relaxed rules at lower levels.  At lower levels you have to focus purely on factuality, not verifiability.  The biggest boon would be to readers who get an instant idea of the objectivity of the article. This could also reduce frivolous edits, edit wars, page protection requests, and all kinds of policing which has emerged ad hoc. I can already anticipate that there are 100 things wrong with this idea, but perhaps someone with more experience and knowledge can bake some more of this half-baked cake. Slofstra (talk) 18:28, 13 February 2008 (UTC)

It's perfectly possible for an article in which every single fact is properly cited from a reliable source to be very bad:


 * Its selection of facts can be biased. Eg an article about Christianity in South Africa might mention that the Dutch Reformed Church supported apartheid, or that Tutu & Boessak opposed it. To mention one fact without the other would be biased.
 * "Reliable sources" not infrequently contradict each other, or even themselves. Eg the Macmillan Encyclopedia of Buddhism gives contradictory statements on pages 33 & 82f, in articles written by different scholars. Neither makes any mention of the other's views & states its own position as unquestioned fact.

There's no answer to this except endless research. Peter jackson (talk) 12:01, 19 February 2008 (UTC)

Verifiable sources having known but unverifiable errors
Requoting a recent snippet from WP:RS: "I had a case once where a reputable source contained an error. That error was quoted by another editor in making their case. I contacted the author and verified the error. And posted the author's response. The opposing editor insisted there were all sorts of Wikiproceduralthingies I had to do (including forcing the author to write to Wikipedia) otherwise the source would stand until the next edition was printed and the error retracted in print. (After all, otherwise, it was just my word.) Some suggested practices on how to handle errors in reputable sources would probably help prevent such nonsense in the future. —PētersV (talk) 01:54, 20 February 2008 (UTC)"

Is this situation sense or nonsense? I'm not clear on what "Wikiproceduralthingies" were suggested, but it seems to me that having the author of the cited source containing the error write to wikipedia would not satisfy the verifiability, not truth threshold for inclusion criteria. -- Boracay Bill (talk) 05:55, 20 February 2008 (UTC)


 * We're expected and permitted to exercise common sense. The key here is that "verifiability, not truth" is the threshold for appearing on Wikipedia. We can perform original research and use sources that don't explicitly satisfy RS to form arguments that content should not appear on Wikipedia. And if there is a good consensus that such content should not appear, and that enforcing the rules would actually harm Wikipedia, we can invoke editorial discretion. This option is always available for these sorts of situations. Someguy1221 (talk) 06:18, 20 February 2008 (UTC)


 * If it is indeed an error, there seem to be two likely situations. In one, it's a claim discussed by other reliable sources. In that case, either the other sources make the same claim or present a contrary claim. In the other, it's a claim not discussed in other references. In both cases, if the claim is exclusive to that single author/work, we should treat it an extreme minority claim. If other sources repeat the same (allegedly erroneous) claim, then the author's opposing word alone would still be an extreme minority claim. Either way, we should stick to what the general body of reputable references says about a topic. (In essence, the verifiability question is solved or superseded by the requirements of NPOV.) Just an opinion. Vassyana (talk) 06:25, 20 February 2008 (UTC)
 * I think we can mention the correction and explain the relative lack of authority. But then I am not generally confident with the way we interpret "extreme minority"--I think it can be used as a way of expressing POV. DGG (talk) 06:35, 20 February 2008 (UTC) But what is the particular example at issue? DGG (talk) 06:35, 20 February 2008 (UTC)
 * As it happens, I've just come across another instance of this situation. See this exchange of edits. -- Boracay Bill (talk) 06:58, 20 February 2008 (UTC)
 * interesting example: many of the figures in that population table needed some qualification or reconcilement. When we say V not truth, we do not mean to say that V=error, not truth, when a clear error can be shown and documented. when there's an obvious arithmetical error in a source, correct it and report the original in a footnote. 18:02, 20 February 2008 (UTC)

Self-published sources not ok for notability, but can they be used for factual additional info ?
I think this is an important question. It is evident that it would be inappropriate to use a self-published source to assert notability, or to make any substantive claim about the subject, which is of course subjective. However, would/should it not be acceptable, to use self-published sources in order to find facts that are objective and that do not relate, in any way whatsoever, to the subject's notability or importance ?

For example, let's say a writer, or an actor, or a scientist, has an article on wikipedia thanks to third-party verifiable sources, which confirm notability and which tell facts about the subject's work. But such sources don't say, for example, the date of birth or place of birth, or whether the person has children, etc. Would it not be acceptable to take such objective and factual information, totally unrelated to notability, from a self-published source ?

Let's suppose actor or author X has a website, which states among the following:
 * 1. that she/he is a leading expert in the subject
 * 2. that her/his works are best-selling
 * 3. that her/his works have been crucial to influence the position of key politicians
 * 4. that she/he has received positive feedback on his work from prominent people
 * 5. that her/his date of birth is 10 May 1955
 * 6. that she/he lives in Chicago with wife W and two children
 * 7. that she/he is a friend close associate or colleagues of Y and Z and they often meet in a certain bar to discuss their work (where relevant or significant).
 * 8. that her/his favourite drink is tequila and she/he does not smoke

In my view, (1) to (4) are clearly not acceptable - and if they are true, they must have been published by a third-party source, which is what wikipedia is all about. However, I don't see why we cannot use (5) to (8). A good biography should indeed have information about the person such as date of birth and family or place of residence, and why not, some info about the person's personal life. I personally think that it would be ok to refer to a self-published source to confirm date or birth or favourite drink, as such information is: (a) important for biographies; (b) irrelervant to notability or substance about the subject; (c) something the subject has no reason to lie about, and no scope to be biased about.

I hope to get your comments. Cheers, --CCorward (talk) 13:36, 17 February 2008 (UTC)


 * Hi CC, we can indeed use self-published sources in articles about themselves, with some restrictions, which more or less mirror what you wrote above. See this section. SlimVirgin  (talk) (contribs) 13:54, 17 February 2008 (UTC)


 * Verifiability is one thing, relevance of content is another. Wikipedia is neither a social networking site nor an indiscriminate collection of information. The fact that one's favourite drink is tequila clearly does not fit in a Wikipedia article (unless this fact is remarkable for some reasons). This also applies to your item 7. So, favourite drink: a definitive no no! --Edcolins (talk) 18:05, 17 February 2008 (UTC)
 * See also Relevance of content. --Edcolins (talk) 18:06, 17 February 2008 (UTC)


 * Thanks a lot Ed. I was being deliberately a bit over the top about "favourite drink tequila" (yes, that may be relevant sometimes, e.g. Boris Yeltsin liked vodka, and Rasputin liked Madeira (wine), but I was not suggesting we should spell out favourite drink for each author, actor, etc).  However, I think administrator SlimVirgin agrees with my overall idea, as reflected indeed in this section.  So I think that date/place of birth, information about spouse and children or about sexual orientation (provided the self-published source is trustworthy !), tutors during unversity, people who wrote a blurb in someone's book, etc., as long as this is kept to factual information in a subjective manner not open to contention or biased wording, then would be ok.  I stand by to be counter-argued on this.  I think it's an interesting discussion, besides the case that you and I have in mind.  Cheers,--CCorward (talk) 18:38, 17 February 2008 (UTC)

'''Any more inputs on this ? We only have two opinions so far ? Cheers,'''--CCorward (talk) 08:16, 18 February 2008 (UTC)


 * I would basically agree with Edcolins. Fans of pop stars sometimes want to have the biographies list all their friends and favorite movies or books or whatever. Even if that's verifiable, it's relevant to myspace pages, but not relevant here. Other self published content may be relevant, though: that someone is working on something for future release (book, movie), for instance. Be alert that self-published info can be wrong, as sometimes happens with seemingly self-published birthdates conflicting with a third-party reliable source. Sometimes the way to solve those conflicts is to list both bits of info with sources, and anger both sides. Gimmetrow 08:34, 18 February 2008 (UTC)


 * CCorward, you already know my opinion on the inclusion of "information about .. people who wrote a blurb in someone's book.." in a Wikipedia article about the author of the book. I don't think the information adds neutrality to the article because blurbs are by definition promotional. The "importance"/"fame" of the authors of the blurbs is part of the promotion. So, in my opinion, no, we should avoid adding such information in articles as a general rule. There may be exceptions, but I don't see one for the case at stake, i.e. Talk:Maurizio Giuliano (More on blurbs). --Edcolins (talk) 20:02, 18 February 2008 (UTC)

"Leading" is a peacock. "Best-selling" is a peacock unless supported by the list it topped. (A selfpublished source may not be unduly self-serving). Influence is a claim about a third party, so a secondary source must support. (selfpubs may not make claims about third parties). Positive feedback should be noted in context. Does it name a third party? See last comment. Does it peacock the notability of the persons without naming them? See WP:PEA. DOB is fine, related to notability if secondary sources confirm subject is non-fictional :) Having family is fine is secondary sources confirm that the subject is human. :D Fine to say they frequent a bar if alcohol use or other related observation is noted by a third-party. "Collegue" must be a verified relationship if claimed (see above about third party claims). Favorite drink? This shouldn't be notable. What's their source, Facebook??

"Related to the subjects notability" means related to topics noted by accepted third party sources, nothing more. See WP:SELFPUB and feel free to suggest modifications to that policy if you foresee that they will make Wikipedia a more authoritative source of information. ClaudeReigns (talk) 20:21, 18 February 2008 (UTC)

Trying to capture consensus on which self-published information may be used
Thanks all ! Ok... If I may try to capture the above into consensus, this is what I gather: a self-published source may indeed be used, with caution and discretion, and where no third-party sources exist, provided that the following conditions are met:
 * (1) This is not significantly relevant to the overall notability or importance of the article, i.e. the information is not "unduly self-serving", and it adds information about the subject whilst not unduly adding to the claim of notability;
 * (2) The information is all of the following:
 * About factual and objective facts and data, e.g. date or place of birth, family, political opinions; or about a claim (hence, a fact about a claim) duly attributed to the person, e.g. forthcoming works, etc.;
 * About something the self-publisher has no reason or scope to lie about, e.g. date of birth (though, true, an actress may lie about her date of birth...);
 * Relevant to a biography, e.g. a person's date of birth or political opinions is relevant, while a favourite drink (with very few exceptions, e.g. notable alcoholist) would not be relevant;


 * (3) There is no reasonable doubt that the self-published source may be fake, e.g. an author's or actor's personal website is likely to be true, while a blog might lead, in some cases, to a reasonable doubt that it may be fake.

So by means of example, we may extract the following from an actor's or an author's website: We may not instead extract the following:
 * Her/his date and place of birth are...
 * She/he has spouse named X of nationality Y and W children named so and so...
 * She/he is a supporter of the X party, her/his religion is Y, and she/he is against tobacco;
 * She/he is currently working on subject X for her/his next book or movie;
 * She/he is a friend close associate or colleague of non-particularly-notable colleague X, provided the claim is relevant and significant as well as credible- see below.
 * Her/his favourite holiday destination is Japan (where relevant, e.g. about a notable traveller, otherwise it would usually be irrelevant).
 * Her/his children are very good at school and sing very well (self-serving and in most cases irrelevant);
 * She/he is notable for her/his involvement in political party X or for being an anti-tobacco campaigner (self-serving);
 * She/he likes sushi (irrelevant, except e.g. for an author who writes books on food);
 * She/he is a close friend and colleague of prominent person X (self-serving, unless e.g. the claim is attributed with caution and reasonable, e.g. because the two people are proven to work in the same Organization);
 * She/he has travelled to all of the Pacific islands (scope to lie about).

Please do share thoughts and correct me on above. I think it may be a good way to refine the policy, if applicable.--CCorward (talk) 21:16, 18 February 2008 (UTC)


 * This is a nice refinement. Can you draft a guideline from this? It is 95% in line with the way I already practice the policy as I read it. The difference being I wouldn't have allowed the statement about the non-particularly notable colleague before, but I can see it is fairly harmless and preserves information if that colleague would redlink. ClaudeReigns (talk) 06:06, 19 February 2008 (UTC)
 * I don't really like the "friend" part either; it doesn't seem to me to match practice very well, and sounds too much like myspace. Gimmetrow 06:14, 19 February 2008 (UTC)
 * The thing I had objected to was making a claim about a third party, which the current policy seems to forbid. Marriage and children are substantial affiliations which public records could disprove (or modify in the case of divorce). These are, or potential are, lifelong relationships. You may call me a friend or colleague, but I may not concur. A single cooperative effort may join one organization to another immortally on our pages though the partnership could have been most temporal. The nature of people seeking notability is to attempt to create the appearance of associations which may or may not in fact exist.
 * I guess I'm still at 95% after all. Although. Specific praise in dedication to third parties where non-listy may be a passing loophole, e.g.: "For Seán P. F. Harris, getaway driver and foul-weather friend," the dedication to Harry Potter and the Chamber of Secrets. We could derive that 'notable author dedicated notable books to non-notable unrelated persons' without it being much of a stink about verifiability or notability. One tends not to devote several months of one's life to a project and dedicate it vain. We could test this at J. K. Rowling and see if it flies. If not then shut the loophole. Or would that be WP:POINT? :D ClaudeReigns (talk) 11:12, 19 February 2008 (UTC)


 * I now added above that the claim should be relevant or significant, and credible, and I reworded it a bit. So, for example, if it is relevant that Ernest Hemingway and Pablo Neruda were friends and often met in Bodeguita del Medio to discuss their work, then it's also relevant that current poet X and journalist Y meet every Sunday in pub Y, while discussing how to put a journalistic story on the Kosovo genocide into poetry - those are relevant and significant facts, and unless the notability of the two individuals is very different (e.g. not a first-time author vs. a Nobel prize laureate), a self-published source may be credible.  Perhaps a better example: political writer Y was tutored at university by scholars W and Z, and feels that she/he is influenced by the scholars' writings in her/his own work.  This I find is relevant and significant, and if we already know that these people were indeed at the same university at the same time, then the claim is also fully credible and with little scope to be unduly self-serving.--CCorward (talk) 21:53, 21 February 2008 (UTC)

What exactly constitutes self-published information anyway ?
I think we all agree that a person's website, or blog, or Amazon.com list, is a self-published source. However, would the same apply to a person's book or movie ? True, a person can pay to get a book published, but for that matter, then this would also apply for a person to get articles written about herself/himself. So unless there is a reasonable doubt that this happened, I don't think that a book by the person is a self-published source, more than an article about the person, as long as the book is the subejct of third-party sources - and otherwise, it would not be in wikipedia anyway.

It may be good to expand on this.--CCorward (talk) 21:16, 18 February 2008 (UTC)


 * At Grey DeLisle it was established that the only reliable source for D.O.B. was the actor/artist's official MySpace page. IMDb, numerous other sources (which had cited us) were all incorrect and it was only proven through Grey's letter at the social networking site which led to a check at databroker Intelius finally debunking the false DOB and supporting (but without subscription it could not verify) the true DOB. This has kind of a disturbing impact on the rules, really, because it's exceptional that a social networking site will provide the most credible and comprehensive verification of data. This DOB issue, and issues citing a childbirth this year (for which a primary should be most useful) have been problematic because editors rightly read our policy on sources, but fail to notice the article's talk page. Perhaps a boilerplate related to WP:IAR accessible only to a high authority (is there something above admin?) could be used in such situations. In theory I agree with the unreliability and unverifiability of such sources. ClaudeReigns (talk) 05:54, 19 February 2008 (UTC)
 * Hmmm... I think we should apply the rationale of reasonable doubt here, i.e. an actress may lie about her DOB, while it is less likely that a scholar would do so.  And perhaps more important, then the text can be phrased accordingly, e.g. according to one source, X has two children named so and so.  And anyway, about the credibility of self-published sources, well... what about the credibility of media ?  Media misunderstand, misinterpret, and misrepresent on a daily basis, and the fact that something was written in print does not mean it is true...--CCorward (talk) 21:59, 21 February 2008 (UTC)
 * Yes, an actress who would appear on-screen, absolutely has good reason to lie. Grey DeLisle is a voice actress. The verifiability of these things should go back to a source, and imdb isn't one... it's just a collection of contributions like us--except they don't provide sources to verify. When the MySpace page disputed our DOB for her, it didn't just dispute it, it produced what it represented as a letter from Grey. The search at Intelius under DeLisle's birth name verified our error. When the MySpace page announced the birth of a son to Grey and Murry, they didn't just announce it, they showed pictures (great verification) and quoted DeLisle's announcement. So whereas imdb doesn't give sources to verify, this social networking site at least described and displayed primary sources. Normally we must treat such a site with deep skepticism. But I am more skeptical of those who make claims without backing them up, whatever their URL or POV. ClaudeReigns (talk) 23:50, 21 February 2008 (UTC)

I don't think it's at all surprising that an example can be found where an "unreliable" source turns out to be more accurate than the "reliable" ones, but I also don't think there's any reason rethink our policies because of it. No human source is true all the time. Nothing is completely reliable. "Reliable" sources are simply thosed believed to be accurate most of the time. On average, the things people say about themselve on their MySpace page are less reliable than then things printed in regular media. But only on average. --Shirahadasha (talk) 03:14, 22 February 2008 (UTC)
 * Total agreement. I think that's one reason WP:IAR exists. Trouble is, everytime the subject comes up, the whole sourcing history has to be repeated. I suppose I should just link this once it's archived. ClaudeReigns (talk) 12:39, 23 February 2008 (UTC)

Case study: blurbs in someone's book
Edcolins knows what I am referring to, which I think is an interesting case study, left aside that specific case. Essentially, an author has published a book, and the book and author are in wikipedia because third-party reviews of the book abound. Personally, I would think that it is relevant to state information about the book not only in terms of contents, but also in terms of who else may have contributed, e.g. was there a co-author or co-editor, or a foreword by someone ? Not very different from stating who the publisher was, in my view. Yes, a foreword by a prominent person is used to sell the book, true, but so is the writing by the author too, and so is the name of the publisher (e.g. a book by Oxford University Press might be considered automatically notable by virtue of its publisher)... I think that "the book has a foreword by ..." is factual information about the book, relevant for wikipedia as long as this is not a reason that led to notability (i.e. the book must have been there in the first place thanks to third-party sources). And the same in my view applies to blurbs, which are essentially the same as a foreword, except that they are shorter, and usually on the back cover or before the first pages of actual book. Incidentally, in the book I am referring to, you may notice that the blurb authors are also listed as commentators in Amazon.com, signifying - to me - that their contribution to the book is part of the book. I think their names (which are after all some academics and diplomats, hence not much more notable than the author) are relevant to the book and hence to the author. And after all, if there were a foreword by a very prominent individual, e.g. the President of the United States, would this not also be pertinent ? I stand by to be supported or counter-argued, by Edcolins of course, but ideally also by others. ;-) Cheers,--CCorward (talk) 21:16, 18 February 2008 (UTC)