Wikipedia talk:Verifiability/Archive 10

Archive 5

←

Archive 8

Archive 9

Archive 10

Archive 11

Archive 12

→

Archive 15

Question

Are these "personal websites"?

• http://penelope.uchicago.edu/Thayer/E/Roman/home.html (LacusCurtius)
• http://www.tyndale.cam.ac.uk/Egypt/ptolemies/chron/roman/chron_rom_intro.htm (Chris Bennett's Roman chronology)
• http://www.af.lu.se/~fogwall/satie.html (Satie Home page)
• http://www.gfn.com/

Could someone give a definition of "personal website"? Or at least explain what would make the examples above "personal website" or not? --Francis Schonken 07:25, 25 June 2006 (UTC)

I would say quickly that I've always felt .edu pages are not personal websites. The university or organization in question has copyright on all of it, yes? Marskell 11:48, 25 June 2006 (UTC)

Well that, for instance, is something that's not always clear:

• The LacusCurtius website (.edu) contains a lot of material that has entered the public domain (so, not "copyrightable") - this website has some other complex copyright provisions, visible at http://penelope.uchicago.edu/Thayer/E/HELP/Copyright/home.html - with at least some part of it "copyright © William P. Thayer" (which, FYI, is user:Bill Thayer, so in this case you could simply ask him on his talk page);
• Chris Bennett's pages at cam.ac.uk (".ac.uk" being equivalent of .edu in the UK; ".cam" being the University of Cambridge, see http://www.cam.ac.uk/ ) has "Website © Chris Bennett, 2001-2005 -- All rights reserved" on its pages

So, no, I don't think it can be assumed that such subsites of University websites are all under the copyright of their respective Universities.

Further, I'd like to mention this additional example, in which case the copyright seems to be covered by the Colombia University's general copyright notice exclusively: http://www.math.columbia.edu/~woit/wordpress/archives/000077.html - now, this is a blog page, so the fact of being copyrighted by a university does not seem to be the hallmark of reliability for ".edu" pages: LacusCurtius and Chris Bennett's pages (neither copyrighted by the university they depend from) appear to be considered as much more reliable than that other page that falls under the general University's copyright, afaik. --Francis Schonken 12:28, 25 June 2006 (UTC)

University and school sites in general have different areas. Those developed and loaded by the University are not personal; those by different departments or research groups are not personal; most also allow for personal sub-pages or sub-sites written by individual professors, instructors, personnel, and students. Some of the professors and instructors include a cirriculum vitae, published papers, and so on - these are not necessarily "personal". Others are more personal, and student pages are generally speaking always personal. You have to actually look and see what is being presented and who is presenting it. Actual effort, but I am sure anyone with a modicum of intelligence and a few minutes is up to the task. KillerChihuahua^?!? 14:26, 25 June 2006 (UTC)

The question is misguided. "Personal website" in wikispeak means "website without editorial oversight". What we must ask when dealing with scientific subjects isn't what kind of editorial policy any source has, we must ask how the source is regarded by subject experts. Yes, well-regarded sources almost always are peer-reviewed, but we should guard ourselves against the attitude of "it's refereed, it must be true" and "it's not refereed, it can't be true". Dr Zak 14:40, 25 June 2006 (UTC)

• Thayer's site is a personal website. The clue is at the very top of the page and then the very bottom of the page links to [1] which states: "All material on this site ... is copyright William P. Thayer". It is owned and (apparently) its responsibility is Thayer's alone. Wonderful site !
• Chris Benett's site is copyrighted, created and owned by Bennett. It appears in every particular to be a personal website.
• Satie's Home Page appears to be a personal site. Thayer's site is more explicit about copyrights and responsibility for content, but none of these claim to have legal staff which checks for liability, nor is its content opined to be "authoritative" beyond what is viewed on the page and it doesn't say it hired experts who check the site's content for accuracy.
• [2] is a business. It sells its site space to advertisers for profit, it has a Terms of Service, therefore has legal liability for its content. A business has legal responsibilities beyond a personal presentation information, a business owner (even an individual) has obligated himself beyond the responsibilities of a private citizen producing a personal website. * The guideline could use a really clear definition of "personal website" This question comes up again and again in different guises, by different editors. A really clear definition would save a lot of discussion page chatter. Terryeo 16:19, 25 June 2006 (UTC)
  • I agree with Terryeo's analysis -- the first three are personal web sites. -- Dragonfiend 16:36, 25 June 2006 (UTC)

A personal website is very easy to spot. Anyone can make that judgement just by looking at the page. If there is a website that may be borderline or that you are not sure about, discuss on talk page and ask other editors to take a look as well. Note that there is nothing inherently wrong with a personal website, providing it is used as sources of information about themselves in articles about themselves, so long as there is no reasonable doubt about who wrote it as per WP:RS. ≈ jossi ≈ t • @ 23:56, 25 June 2006 (UTC)

The last example is interesting. I don't think a business's website should be considered any better a source than a "personal" website. The reason we strongly favour newspapers and publishing houses is that we can have some certainty that they exercise editorial control over the material they publish, and will make some effort at fact checking. That cannot be considered true of businesses in general, although they are a degree more reliable than individuals on account of being more scared of being sued! They also tend to have their content checked, but you cannot count on this to the degree you can with the NYT. I'd say that the policy should reflect that (although it's purely an application of common sense) by saying that a business's website can only be relied on to be a source for statements about itself, in precisely the same way as a personal website. The plus for a business website is that there can be little doubt that it wrote the stuff about itself. Grace Note 02:52, 26 June 2006 (UTC)

And about the .edu domain. Even if the university claims copyright over the site, that does not imply that it checks the material on it. I think that academics' sites, unless they reproduce material published elsewhere, should be treated in precisely the same way as anyone else's site. Grace Note 02:54, 26 June 2006 (UTC)

I must agree here. This page, for example, is a personal essay on Russian history by an economist. It shares the difficulties of personal websites: it is obviously advocacy, and it disagrees with reliable printed sources on such major details as the years of the famine. Prof. Caplan happened to put this up on his office website, which is .edu; we should treat it exactly as though it had been on his home site. Septentrionalis 17:01, 5 July 2006 (UTC)

Read this page Wikipedia:No original research#Expert editors and when these other sites are viewed, the question should be, is this person an expert in their subject matter? Do these pages themselves quote reliable sources? Wjhonson 01:24, 17 July 2006 (UTC)

Protected

There seems to be an awful lot of reverting here. Can you all try to sort it out at this talk page? It looks bad to have an edit war over a policy page. I've protected the page for the moment. AnnH ♫ 00:48, 26 June 2006 (UTC)

Thanks ML. There's a strong consensus for protection above: #Ten changes a day?. Jayjg ^(talk) 00:55, 26 June 2006 (UTC)

Heck, I don't even think most of us care that whether the wrong version was protected. Stop the madness!  ;-) Robert A.West (Talk) 01:00, 26 June 2006 (UTC)

Same here. Thank you, Ann. SlimVirgin ^(talk) 01:19, 26 June 2006 (UTC)

Thanks Ann. I think the version we have here is relatively stable and we really should talk it out before making changes. Grace Note 02:47, 26 June 2006 (UTC)

I would prefer the 3 policies, this NPOV and NOR be protected all of the time and changes only happen to them after concensus on their discussion page. Of course, that would require about a once a day visit by someone who could implement agreed upon changes. With so many editors speaking so many languages, some organization is better than none, some stability better than none. Terryeo 19:26, 26 June 2006 (UTC)

I second the motion, FWIW. Robert A.West (Talk) 00:23, 27 June 2006 (UTC)

I also agree. -Will Beback 00:34, 27 June 2006 (UTC)

I agree as well. ≈ jossi ≈ t • @ 05:28, 27 June 2006 (UTC)

I'm on board. FeloniousMonk 05:43, 27 June 2006 (UTC)

Count me in. AvB ÷ talk 08:59, 27 June 2006 (UTC)

I also agree with this idea. Yamaguchi先生 09:25, 27 June 2006 (UTC)

I agree with it too. This makes it a more formal (and polite) procees as well. --Siva1979^{Talk to me} 18:44, 7 July 2006 (UTC)

I agree too. The policy pages are themselves cited so often in disputes all over Wikipedia that we cannot afford to have their content messed up in edit wars arising from those disputes! --Coolcaesar 20:18, 10 July 2006 (UTC)

• I also agree. Just the main policy pages. SilkTork 12:30, 18 July 2006 (UTC)

proposed addition

In the section on "sources" perhaps we should add, at the outset, that "When adding a verifiable content it is essential to provide enough information that the source can be found and checked. The more precise the content, the more precise the source must be." or something like that. Slrubenstein | Talk 13:57, 28 June 2006 (UTC)

What would this addition address, Slrubenstein? Is that not implied already in the existing text? ≈ jossi ≈ t • @ 14:32, 28 June 2006 (UTC)

I agree it is implied. I just wonder if we could benefit from something more explicit. it is not a big deal and if no one is enthusiasitc about this we can drop it. Slrubenstein | Talk 15:38, 28 June 2006 (UTC)

Student Newspaper

Is a student newspaper a valid primary source of a racial slur in regards to a living persons page? A student newspaper has said that a particular media personality made a racial slur to a teen. This quote is not attributed at any other location other then blogs that link back to the student newspaper. Is this a valid primary source under these conditions? --zero faults ^{|sockpuppets|} 16:41, 28 June 2006 (UTC)

This source will be only valid on an article about this specific student's newspaper. Most definitively not as a source for a biography of a living person. See: WP:BLP. Also note that personal blogs are not acceptable as secondary sources. ≈ jossi ≈ t • @ 17:10, 28 June 2006 (UTC)

Conflicting Sources

What should we do when multiple verifiable and reliable sources contain conflicting information? Peteresch 17:24, 28 June 2006 (UTC)

We report the conficting information by citing these sources and attributing the conficlting POVs to those that hold them. Read WP:NPOV#The_neutral_point_of_view. ≈ jossi ≈ t • @ 17:26, 28 June 2006 (UTC)

Almost true. If it is a scientific or medical subject we report what the scientific consensus is, citing a recent, authoritative review. If no consensus has emerged yet we report that fact and state who holds what view (and why). We also report significant minority views but avoid giving undue weight to them. See WP:NPOV#Undue_weight. Dr Zak 17:43, 28 June 2006 (UTC)

Exactly. Well said. ≈ jossi ≈ t • @ 21:28, 29 June 2006 (UTC)

First steps toward revising the introduction

Is anyone wed to the "threshhold of is verifiability, not truth" line in the introduction? It serves to obscure the matter by a catchy slogan rather than just directly stating what the policy is. —Centrx→talk • 20:32, 29 June 2006 (UTC)

I like it because it gets straight to the heart of the policy, viz. we report what reliable sources say, not what is true. SlimVirgin ^(talk) 21:16, 29 June 2006 (UTC)

The policy specifies the standard for included information, that the standard is not individuals asserting "I think this is true". Abstractly, Wikipedia is to include some sort of truth because it is about knowledge in an encyclopedia, but the way that knowledge is determined is by reliable sourcing, and everything in the encyclopedia is true in its basis on authoritative sources. Wikipedia is certainly not supposed to be for false information, it's just that without verifiability the criteria for inclusion is what individual editors think is true or false. Choosing not to have the cryptic statement "The threshhold is verifiability, not truth" in the first line of the page, does not mean that the page entails that Wikipedia is trying to find the ultimate, Platonic, in-itself, absolute truth of the universe. —Centrx→talk • 22:55, 29 June 2006 (UTC)

Unlike SlimVirgin, I don't like it because it has the unintentional side effect of implying that truth and verifiability are mutually exclusive, which runs contrary to what most people believe. It was confusing and seemingly contradictory the first time I read it; my immediate thought was "huh? how can you verify something without ascertaining its truth/general correctness?" The meaning was adequately explained elsewhere in the policy, but it would be better to just make it clear from the outset that the kind of verifiability WP cares about is that which doesn't use the commonly-held criterion of truth as the standard against which statements are "verified", but rather uses the criterion of "reliable sources" as defined and continually refined by the WP:RS guidelines.

If the policy were mine to rewrite, it would be reduced to a set of simple, normative statements as bullet points, along with an overarching rationale that justifies the policy and explains the goal that the policy is intended to work toward. Rationales and informative explanations of each individual point would be easily distinguished from the normative statements. Anything making a statement about the relative reliability of specific types of sources in specific circumstances (e.g., self-publishers, and articles about them) would be completely out of scope since it belongs in WP:RS, not here, IMHO. —mjb 00:41, 30 June 2006 (UTC)

The point of using "verifiability, not truth" is that too many editors want to add what they "know" to be "true" to an article, even if they can't provide references. "Truth" can be a very slippery concept. We are much better off insisting on including only what can be found in reliable sources, and not trying to define "truth". -- Donald Albury^(Talk) 02:14, 30 June 2006 (UTC)

I understand that, and I'm not arguing against that. The phrasing is problematic because "verifiability" has connotations of "truth"-ascertaining. Haven't you been following the voluminous prior discussion on this point over the past few weeks?—mjb 07:13, 30 June 2006 (UTC)

I've been following the discussion here off and on for more than a few weeks, and I haven't seen anything said here recently that hasn't been said before. The point is, if we concentrate on making sure that only verifiable information is included in Wikipedia, it will come a lot closer to being an accurate source of information than if we let editors insert what they "know" to be "true" without requiring verifiable sources. So, our position has to be that verifiability trumps truth; truth does not trump verifiability. -- Donald Albury^(Talk) 10:53, 30 June 2006 (UTC)

The way I see it, "threshhold is verifiability, not truth" gives editors trying to learn more about WP's modus operandi two well-defined points on this continuum. This is where they learn which of these points they should use as cut-off when dealing with a scale that as a whole is, indeed, slippery at best. AvB ÷ talk 11:58, 30 June 2006 (UTC)

This is all the more important since many new editors seem to arrive here with lofty notions about an encyclopedia that tells "The Truth." I presume some of them got those ideas using Wikipedia for a while and becoming impressed with its encyclopedic treatment of subjects, while others simply clicked on the edit this page button to correct something they "knew" to be "wrong". AvB ÷ talk 12:06, 30 June 2006 (UTC)

I think the "policy in a nutshell" should, with the some modifications, be the basis for a first paragraph. This is the purpose of an introduction. WP:NPOV just repeats the "policy in a nutshell" in its introduction. The first paragraph of WP:NOR would not make much sense to an average person without reading the "policy in a nutshell", which on that page serves as the introduction. An introduction should be direct and a summary, which is what the "policy in a nutshell" is trying to be. Either the "policy in a nutshell" has to be recognized as the first paragraph of the article, bolded and separated as the over-all description, and the rest of the introduction doesn't need to repeat it, or the policy in a nutshell is just a duplicate of the first paragraph of the introduction. —Centrx→talk • 22:03, 1 July 2006 (UTC)

In my view, the centre of the dispute over this page and over this policy as a whole is that "verifiable" most commonly means "checkable as true". Its meaning contains an idea of truth that we don't want to convey. This would be resolved, I think, simply by renaming the policy WP:Checkability and rewriting it to be clear that what we want is that any information/material/stuff/"facts"/views/whatever in our articles must be checkable against reputable sources without reference to whether it is true. Again, this is a solution that is so stunningly simple and obvious that it has no hope whatsoever of being accepted because it's just so much more fun to battle over the most meaningless bollocks we can come up with. Grace Note 04:58, 5 July 2006 (UTC)

First, verifiability just as much means checkable as "accurate" or "real". Further, it doesn't just mean "checkable anywhere", it means checkable at reliable sources, sources probably true and accurate. Just what is wrong with truth? It still means checkable, and it excludes editors from inserting their individual notions of truth. —Centrx→talk • 08:33, 5 July 2006 (UTC)

That's the problem. "Truth" doesn't prevent the Fooian editors from insisting that their POV is The Truth, and all contrary information is Barian lies. For an example, see Talk:Imbros and Tenedos, where all too much of the discussion has consisted of Greek editors saying: The island's name is Tenedos [implied because it's Greek]; and Turkish editors saying: The island's name is Bozcaada, because that's the official name [implied: because it's Turkish]. Septentrionalis 17:12, 5 July 2006 (UTC)

Regardless of the "verifiability, not truth" phrase, the policy still states that information on Wikipedia requires citations to reliable sources, and that an editor's individual opinion is subject to those sources. The particular island example is probably not so much the purview of WP:Verifiability as much as WP:NPOV or some guideline about naming places: probably, reliable Greek sources are prevalent for the first name and reliable Turkish sources are relevant for the other. —Centrx→talk • 02:07, 6 July 2006 (UTC)

Sorry, I don't understand your response. "Verifiability" clearly has a broader meaning than "checkability". They overlap but the latter is narrower. And I did not say it means "checkable" anywhere. That's a rather silly comment to make because obviously the same quibble would apply to "verifiability" (one cannot "verify" our articles anywhere either). What you are not grasping, Centrx, is that while "verifiability" might mean what you claim in Wikispeak, it does not in the common parlance, and it would be better to write our policies in the common language. Not that we ever will. I'm not fool enough to think that. Grace Note 08:13, 7 July 2006 (UTC)

How about this?

All information included in Wikipedia must be verifiable. The criterion for inclusion is verifiability, not truth. We seek truth but do not lead the way. "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research. Fred Bauder 02:19, 6 July 2006 (UTC)

Erm, no. We do not seek truth, we publish beliefs. The Wikipedia of 2000 years ago was Pliny's Natural History, who reported that the Blemmyes lived in Africa and, well, other things, that he found in his sources. Dr Zak 02:35, 6 July 2006 (UTC)

Perhaps "report" would be better, but perhaps the policy should not talk about "truth" at all, except to discourage editors from thinking they have the truth. However, if the phrase "verifiability, not truth" is kept, it needs some clarification which possibly requires statements about truth, and still, we do in a way seek the truth. If there were reliable accounts of Africa that discount the existence of Blemmyes there, then those would be included as well, and a statement about Pliny's history would be, as is still done: "According to Pliny,...". —Centrx→talk • 03:09, 6 July 2006 (UTC)

I agree with Dr. Zak. "We seek truth" is a huge can of worms, as it will force us to define what "truth" is. Better to stick with what we do best: describing the assertions of truth by others as published by reliable sources. ≈ jossi ≈ t • @ 03:03, 6 July 2006 (UTC)

Yes, we publish knowledge, well-founded belief. Fred Bauder 04:39, 6 July 2006 (UTC)

"Publishing well-founded belief" isn't the same as "seeking truth". I'm not so sure about the "truth" bit either, as Wikipedia is in the business of knowledge. To make that quite clear, how about "We do not seek truth, instead we publish well-founded belief". Dr Zak 04:57, 6 July 2006 (UTC)

I am not comfortable with "We do not seek truth", after all, trying to find out what has happened is the point of consulting a compilation of published knowledge. Fred Bauder 11:11, 6 July 2006 (UTC)

Ideally, Wikipedia conveys knowledge, which is true. Verifiability establishes that such knowledge must be found in reliable sources and not individual heads. —Centrx→talk • 11:17, 6 July 2006 (UTC)

Jeezus. Instead of writing "verifiable in this context means...", why not use a word that does not need its context defined? Fred, I don't think we should mention the t-word at all. We do not seek the truth. That would be a vain quest, because as has been noted, there are often several different truths! We just report what others have claimed the truth is. All that this policy needs to say is that you must be able to check that "stuff" in here has appeared elsewhere, where "stuff" is whatever word we come up with for the substance of what articles contain. Grace Note 08:13, 7 July 2006 (UTC)

Stop using the t-word! :) Why not just say "we synthecize what reliable sources have said". Wjhonson 01:32, 17 July 2006 (UTC)

Try again; how about this?

All information included in Wikipedia must be verifiable. The criterion for inclusion is verifiability. "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source. Material which has not yet been published in a reliable source may not be included in a Wikipedia article.

Wikipedia:Verifiability is one of Wikipedia's three content-guiding policies. The other two are Wikipedia:No original research and Wikipedia:Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. They should not be interpreted in isolation from one another, and editors should therefore try to familiarize themselves with all three. These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles. Fred Bauder 17:31, 6 July 2006 (UTC)

I think you're going to have a problem saying "any reader must be able". What if a reliable source is obscure and only found in fifteen libraries in the world? Someone will claim that it's in-practice impossible to verify. Wjhonson 01:35, 17 July 2006 (UTC)

Removal of truth

Did I miss something? I don't recall seeing a consensus for removing truth from the intro. I won't revert yet, but I bet someone else will shortly. -- Donald Albury^(Talk) 17:46, 6 July 2006 (UTC)

Early on in my Wikipedia experience I edited some at truth. Not a good move, as I was soon butting heads with Larry Sanger. I find the criticisms above regarding the use of truth in the introduction pale shadows of the difficulties which can ensue. I find it a commonplace that some things which are true are simply not a part of "knowledge", published knowledge at least. Many of these insights arise from personal experience. Fred Bauder 17:53, 6 July 2006 (UTC)

"We seek truth but do not lead the way" was a weird thing to have in the opening paragraph. Good edit to remove it. I suggest that "Verifiability, not truth" is an important thing to stress, and I wouldn't want to see it disappear. Jkelly

"Verifiability, not truth" is a good capsule summary of the policy, marred only by our somewhat idiosyncratic and frequently misunderstood definition of verifiability. We certainly do not "seek truth." Jimbo at one point criticized QuakeAid for having too much "investigative reporting" in it. (QuakeAid is or was a dubious enterprise claiming to be a charity; as originally created the article had the appearance of self-promotion; editors proceded to give it a neutral point of view... all sourced to the hilt... but with excessive vigor). We glory in articles on Psychokinesis, Bates Method, and the Adams motor. We report what others have published about these topics, trying to give reasonable balance. "Seeking truth" is counterproductive to what we are trying to do. We seek an encyclopedia that can be trusted both by those who do not believe that the Adams motor is a genuine perpetual-motion device and those who do. Dpbsmith (talk) 18:54, 6 July 2006 (UTC)

I think that ""Verifiability, not truth" is an important thing to stress, and I wouldn't want to see it disappear." perhaps could be dealt with in a section further down the page. Verifiability, not truth, could perhaps be the header. Fred Bauder 19:06, 6 July 2006 (UTC)

This is, in fact, the situation, Wikipedia:Verifiability#Verifiability.2C_not_truth. Fred Bauder 19:08, 6 July 2006 (UTC)

Usually, VNT means that there are true statements that cannot be included; however, there is a persistent minority that seem to feel that VNT means that falsehoods can be included, just because someone said them. Is there a way to emphasize more clearly that, for the most part, verifiability is a stronger constraint than truth? Robert A.West (Talk) 23:07, 6 July 2006 (UTC)

I agree. If everyone insists that this slogan is included, it must explain that the threshhold of verifiability is higher than the threshhold of truth. —Centrx→talk • 23:14, 6 July 2006 (UTC)

We will have information about the Virgin Birth and Mary, Mother of God. Fred Bauder 23:39, 6 July 2006 (UTC)

Huh? You mean like this: Virgin Birth and Mary, mother of Jesus? -- Donald Albury^(Talk) 23:56, 6 July 2006 (UTC)

Actually, it's a good illustration. The Virgin Birth is either true, or it is not. Neither position can be asserted as fact in Wikipedia, because neither is verifiable: therefore, Wikipedia policy prevents a true statement (as well as a false one) from being included. The statement that the Roman Catholic Church has made the Virgin Birth an Article of Faith is both true and verifiable, and may be included. Robert A.West (Talk) 00:51, 7 July 2006 (UTC)

Falsehoods certainly are included "just because someone said them."

That is, they are included in the form "A said B about C," where B may well be false, but "A said B about C" has been published by a reliable source and is thus quite likely true.

See Hollow Earth theory for an example that is probably less contentious than most.

The person who says B has to be important, of course, and it must be demonstrated that B represents a belief system held by a reasonably large number of people. And we tell the reader who said B, identify the published source, and give the reader sufficient apparatus to make a judgement for themselves. ("A said B about C. Also, D said A is a nutcase.") Dpbsmith (talk) 02:01, 7 July 2006 (UTC)

But, the truth-value of the direct statement "A said B about C" is completely unrelated to the truth-value of the indirect statement, B. Since the direct statement, not the indirect, is the actual assertion, no falsehood is communicated unless A did not say B about C, in which case the assertion should fail verifiability. Of course, we may be relying on Q, who quotes A as saying B about C, and Q may be wrong, but if we believe Q, no reasonable policy will guard against its inclusion. Robert A.West (Talk) 02:57, 7 July 2006 (UTC)

Mary, mother of Jesus is actually a very bad title, using, as it does, the heretical doctrine of the Nestorians, a decidedly minority viewpoint. Fred Bauder 03:33, 7 July 2006 (UTC)

This is an NPOV, not a Verifiability issue. The one attribute of Mary that Catholics, Protestants, Nestorians and unbelievers agree on is that she is the mother of Jesus of Nazareth in all forms and interpretations of the narrative. But, again, this is a distraction. Robert A.West (Talk) 16:44, 7 July 2006 (UTC)

Editing while protected

Could admins please quit abusing their privilege and editing this article while it is protected? The rest of us have to try to sort out the content here. To watch you impose your views when we cannot sticks in the craw. Grace Note 08:15, 7 July 2006 (UTC)

What formulation do you prefer and why? Fred Bauder 12:06, 7 July 2006 (UTC)

Can someone unlock this so that Fred's bizarre removal of Wikipedia's best know policy phrase can be reverted. Marskell 13:16, 7 July 2006 (UTC)

OK, that came out snarky. But per Grace, it seems to be gaming the system against non-admins to have this locked down and then have admins altering the intro. Marskell 13:28, 7 July 2006 (UTC)

Changes must be agreed on the discussion page before implementing. —Centrx→talk • 13:32, 7 July 2006 (UTC)

I don't see agreement. All the edits accomplished were introducing a redundant new sentence ("All information included in Wikipedia must be verifiable. The criterion for inclusion is verifiability."--we don't need both) and the removal of "not truth". Marskell 13:41, 7 July 2006 (UTC)

I agree. I don't like admin-bashing, but in this case Fred Bauder, acting without a clear consensus, has used his admin status to impose his preferences on a policy while it is protected. -- Donald Albury^(Talk) 13:54, 7 July 2006 (UTC)

This is a pretty blatant case of admin abuse. I don't think they come much clear-cut than this. Q: Is the page protected? A: yes. Q: Is an admin editing the page (other than removing vandalism)? A: yes. Pity that wikipedia doesn't have any real way of dealing with this sort of abuse. - O^O

Perhaps Fred could revert himself. Marskell 15:14, 7 July 2006 (UTC)

What I will do is post a note on wikien-l and see if we can get some more input. Fred Bauder 18:19, 7 July 2006 (UTC)

Justify this

Changed my mind and reverted, but please explain why:

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source, because Wikipedia does not publish original thought or original research.

Wikipedia:Verifiability is one of Wikipedia's three content-guiding policies. The other two are Wikipedia:No original research and Wikipedia:Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. They should not be interpreted in isolation from one another, and editors should therefore try to familiarize themselves with all three. These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles.

is better than:

All information included in Wikipedia must be verifiable. The criterion for inclusion is verifiability. "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source. Material which has not yet been published in a reliable source may not be included in a Wikipedia article.

Wikipedia:Verifiability is one of Wikipedia's three content-guiding policies. The other two are Wikipedia:No original research and Wikipedia:Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. They should not be interpreted in isolation from one another, and editors should therefore try to familiarize themselves with all three. These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles.

Fred Bauder 18:35, 7 July 2006 (UTC)

(For those who find collation a nuisance, here's the diff. The second paragraph is unchanged.)

• I don't particularly care about threshold, but criterion is not really an improvement. If all of our editors knew what "criterion" means, we would have fewer problems.
• I like not truth; most of the cranks I meet don't care about details like citation or evidence because they are telling the world The Truth and using Wikipedia as their megaphone.
• Material which has not yet been published in a reliable source may not be included in a Wikipedia article may be justifiable as repetition for emphasis; but it displaces the reasons for verifiability. Both might be reasonable. I suppose the explicit prohibition may be valuable. Septentrionalis 19:17, 7 July 2006 (UTC)

As above re: 'threshold' and 'truth'. The mention of truth is important to retain, because the people who need to read this policy tend to be under the mistaken impression that they can publish anything they believe to be true. I just object to the juxtaposition of "verifiability" with "not truth" in an overly terse phrase, for reasons stated earlier. More verbose phrasing would fix the counterintuitive implication that verifiability and truth are mutually exclusive, but moving toward verbosity rubs some people the wrong way and tends to elicit cries of 'redundant', a rejection that seems to have the effect of dismissing the cause for concern altogether. :/ —mjb 20:47, 7 July 2006 (UTC)

I object to saying "The criterion", because Verifiability is not the sole criterion: the statement must avoid original research and be NPOV, at minimum, and must earn its keep with style, relevance, etc. While "The threshold" is not perfect in this regard, it suggests that we let verifiable statements in the door for an interview, but that we are not obliged to let them take up residence unless they meet other criteria. Robert A.West (Talk) 21:03, 7 July 2006 (UTC)

That's a good way of putting it, Robert. :-D SlimVirgin ^(talk) 21:07, 10 July 2006 (UTC)

"These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles."

Huh? Was there a coup d'etat while I was out working? Kim Bruning 22:46, 11 July 2006 (UTC)

New template

I have created a new template, {{failed verification}} that bears the same relation to {{citecheck}} that {{fact}} bears to {{verify}}. This will help distinguish at a glance among three cases:

1. You provided no source.
2. I have a source that disputes yours.
3. Your source does not support your edit.

I can think of a couple of articles where this template could help focus on the actual point of dispute. Robert A.West (Talk) 19:27, 11 July 2006 (UTC)

No binding decisions

I mentioned this above already, but the page appears to be protected so I'll have to ask someone else to fix it.

"These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles."

Could someone look at this? There's some rather misleading text on there at the moment. Things like non-negotiability and no editing this page and other such nonsense that don't really go well with Wikipedia:No binding decisions. Even the foundation issues are theoretically negotiable (though admittedly not so much in practice).

Kim Bruning 20:06, 12 July 2006 (UTC)

• How about:

The principles of these three policies are non-negotiable and cannot be superseded by any other guidelines. Consensus by the editors of an article does not supersede these policies; all articles must follow them.

—Centrx→talk • 20:40, 12 July 2006 (UTC)

They are quite negotiable, and can be superseded by anything. Consensus by the editors of an article can go a long way too.

You might want things to be different, but guidelines are descriptive, not prescriptive, so you can't actually just make things up.

Of these guidelines, NPOV is the hardest to argue with, but at times even the neutral point of view guideline has been discussed. Kim Bruning 20:58, 12 July 2006 (UTC)

Wikipedia is an encyclopedia. This is not a mere guideline; this is the fundamental, essential property of Wikipedia. Encyclopedias in general, and this encyclopedia specifically, are not written from various biased points of view, nor do they premiere research, and they are based on authoritative sources.

This is the theory, this is the practice, this is the enforced policy. The overwhelming majority of Wikipedia editors follow these principles, though they may not necessarily agree on their exact practical implications. If an article is not written according to these policies, it is revised to be in accord with these policies. If a topic is a theory out of someone's head, or if it is impossible to verify, its article is deleted. If a tract cannot be rewritten to be neutral, it is deleted. This is done by editors, this is done by administrators, this is done by the owners of the servers. See also [3] and [4].

If you mean that it is "negotiable" insofar as a few people do try to flout or negotiate it on articles, the end result is still that these policies are followed; on a wiki the article may for a time not follow the policy, but this does not mean that the policy is overridden. If you mean that it is "negotiable" insofar as editors refine the policy pages, this does not mean that the principle itself is not followed. "Discussion" of the policy or principle does not mean that is not the policy and principle.

This page must convey to article editors that they are obliged to follow this policy; it must convey to policy editors that they are obliged to follow the principle. Any revision of these principles would not be the same process by which this policy page is revised, and it would need to actually convince those thousands upon thousands of editors who agree with it, among them those well-respected, long-standing members of the community who administer and take part in administering the site and the encyclopedia. —Centrx→talk • 21:45, 12 July 2006 (UTC)

Wikipedia is a volunteer organisation. How do you propose to enforce anything on volunteers? Kim Bruning 22:43, 12 July 2006 (UTC)

You revert their edits and stop allowing them to contribute. Titoxd^(?!?) 22:46, 12 July 2006 (UTC)

Volunteers are a precious commodity. So that's probably not the best option in the long run. Any other ideas? Kim Bruning 22:49, 12 July 2006 (UTC)

An organization that doesn't stick to its minimum standards won't attract many volunteers. Any ideas from you? Titoxd^(?!?) 22:53, 12 July 2006 (UTC)

I might have one or two. Kim Bruning 23:19, 12 July 2006 (UTC)

Supporting Kim on this one. Also the current "freezing" of the WP:V page in a way that only sysops can edit it seems in contradiction with the outcome of the poll held not so long ago at Wikipedia:Editing policy pages: no, it is not OK to make policy pages only editable by sysops on a long-term base. WP:NPOV can do without it (I don't think it was ever protected for more than a few weeks), I don't see why WP:V couldn't.

See also my last comment at Wikipedia talk:Reliable sources#Bulletin boards, wikis, and posts to Usenet. --Francis Schonken 13:31, 13 July 2006 (UTC)

It was largely, if not entirely, because of you that the page had to be protected. SlimVirgin ^(talk) 14:33, 13 July 2006 (UTC)

I agree with Centrx and Titoxd above. An unverifiable encyclopedia is worthless. WP:NBD does not contradict WP:V, as WP:NBD says "some decisions [by Jimmy Wales, arbcom, and the Foundation] are binding until those who made the decision recall it," Jimmy Wales says WP:NPOV is non-negotiable, and WP:V complements and is required by WP:NPOV. As far as editors who continually introduce unverifiable POV material being a "precious commodity," I disagree. -- Dragonfiend 14:59, 13 July 2006 (UTC)

Yes, I don't see what would be lost by enforcing policy. Anyone who edits and inserts their personal opinion is not volunteering for Wikipedia. Are you proposing that each article be fiefdoms of editors who can do whatever they want? That's what happens if there is no policy. —Centrx→talk • 07:14, 14 July 2006 (UTC)

If you scroll back, you will see that I "proposed" some policy.^[1] REAL policy in fact, of the take-it-or-leave-it-in-practice variety.^[2]

Verifiability is a very important content guideline, certainly. :-) I'm just opposed to inflating it past the level of fundamental policy. :-P Kim Bruning 12:24, 14 July 2006 (UTC)

[1]I've mentioned foundation issues, a written policy and consensus across all mediawiki projects, and therefore also the actual mandate for wikipedia; I've also mentioned no binding descisions, an unignorable meta-rule that applies on practically all public wikis, including wikipedia.

[2] In theory, these policies are still negotiable. Technically, that's what you're trying to do now, I suppose. I'm just not letting you get away with it. ;-)

Once the principles become negotiable, however, it becomes no longer an "encyclopedia". Insofar as the principles may be negotiated out, Wikipedia becomes something essentially other than what it is, always has been, and was intended to be. —Centrx→talk • 20:12, 15 July 2006 (UTC)

I agree that saying "non-negotiable" violates NBD. Any policy is negotiable. I do not agree that the policies require each other, that seems more to be an opinion, perhaps based on logic, perhaps not. The main point being that any policy and certainly the *wording* of the policy can be discussed and modified at *any* time. That is the nature of wiki in general. I'm not seeing a consensus in these comments that we can't discuss the policy. Wjhonson 01:51, 17 July 2006 (UTC)

Well no, saying they have to be read in conjunction is based on an understanding of them. The basic policies can't be overturned. Jimbo has explicitly stated that NPOV is non-negotiable, and I can't ever envisage a situation where we'll allow people to add their own unsourced opinions to articles. SlimVirgin ^(talk) 01:56, 17 July 2006 (UTC)

Well perhaps Kim's reading was overbroad. The text within the policy can be edited, the basic general concept of the policy is non-negotiable. The specifics of what that means in practice can be modified. The history of the pages bears this out pretty well. Wjhonson 02:11, 17 July 2006 (UTC)

Slimvirgin: You'll find that NPOV is essentially non-negotiable, though specifics are still negotiated every day. This is because it's a foundation issue. Of course it's always possible to fork (and people have indeed forked over NPOV). The other guidelines you state are not as solid. Nor do I believe they should be. Making them non-negotiable will at times cause them to collide with the foundation issues. If you want to have additional (semi-)non-negotiable guidelines, you could always fork, if you like.

Note that the problem is not that these guidelines aren't very important, because they are, and that's good.

The problem is that even if you could tie things down so strongly on a wiki, and even if you should be prescriptive, you still can't forsee the future.

Some essential flexibility is nescessary, so that we can act reliably now and in future.

(Theoretical basis: Hard rules can be very brittle: you can have wild differences in outcomes based on small differences in starting condition. We want robust rules, where wild differences in starting condition still lead to the same result: an encyclopedia.)

Kim Bruning 10:20, 17 July 2006 (UTC)

As I said, I can't envisage a situation in which insisting that edits be based on reliable sources could possibly "clash with a Foundation issue." On the contrary, the push for this comes from the Foundation, because the better our sources, the more Wikipedia is protected. SlimVirgin ^(talk) 10:35, 17 July 2006 (UTC)

The foundation is rightly pushing for reliable sources in articles about living people, because libel is an evil thing.

In many other situations, insisting on reliable sources right off the bat may lead to articles being written more slowly or never being written at all. This was the situation I encountered a lot when I was editing on science/technologyarticles. Especially for new pages, people can often safely start off just from memory. In science/technology you'll often find that they will get pretty close to what the references say, because science/technology fields are typically logically coherent and internally consistent.

In these fields it is not even always necessarily original research to (re-)derive a formula, algorithm, genetic code etc... from scratch, especially if you can be pretty sure that documentation is going to provide just about the same answer anyway.

For example: I had great fun with the color articles when converting RGB to CMYK and HSV values. These values are very easily verifiable. If I say Blue in CMYK (0-1 normalised) is 0.0, 1.0, 0.0, 0.5 and you mix your inks in those proportions and find a dark/dirty magenta, you'll be pretty sure that either you're colorblind, or I'm wrong :-P (A correct value is 1.0, 1.0, 0.0, 0.0) .

Kim Bruning 11:44, 17 July 2006 (UTC)

And that would be OR. Sorry, but sources need to be included from the beginning of any article. If someone is consulting sources when they write an article, it is very easy to cite those sources. If someone is writing an article without consulting sources, they may well be getting things wrong. -- Donald Albury^(Talk) 13:07, 17 July 2006 (UTC)

No that would be common knowledge. OR is when you are *creating* new facts that did not previously exist. I can start an article on Frontier Village, Washington a neighborhood name in Lake Stevens, Washington, not simply because I have personal knowledge of the neighborhood, but because such knowledge is common to the inhabitants. I'm not creating brand-new knowledge of how to properly bake a cake that has never existed before. Just like anyone can edit an article on dogs and say "they have four legs". You don't need a cited source for common knowledge. Now if someone challenges me and says "Oh no the first store at Frontier Village was Bob's Dry Cleaners" then a source should be dug out. However the article shouldn't be AfD simply because sources aren't cited. If so, we'd AfD half the encyclopaedia, actually more. These issues only come up in confrontational articles, most of the time. Really there is a whole spectrum of usage on wikipedia. It would be interesting if the rules could be flexible to match the reality of what we do. Not sure how possible it is. Wjhonson 14:08, 17 July 2006 (UTC)

Original research is defined in Wikipedia:No original research as:

a term used on Wikipedia to refer to material placed into articles by Wikipedia editors that has not been previously published by a reliable source. It includes unpublished theories, data, statements, concepts, arguments, and ideas; or any new interpretation, analysis, or synthesis of published data, statements, concepts, or arguments that appears to advance a position or, in the words of Wikipedia's co-founder Jimbo Wales, would amount to a "novel narrative or historical interpretation."

If it has been previously published, then it is easy to cite the source. If you don't cite the source, how does a reader know that it isn't OR? -- Donald Albury^(Talk) 14:34, 17 July 2006 (UTC)

undent. If something is common knowledge, then it has been published obviously. The mere fact that you cannot lay your hands on the citation at the moment for something like "The Sun is Shiny" doesn't mean its OR. You seemed to be implying (see up) that anything unsourced was OR and that's just not true. Wjhonson 14:48, 17 July 2006 (UTC)

If it's been published, then what is the objection to providing the source? Of course, another question is why would you want to say in an article that the Sun is shiny? If something is worth saying in an article, then it is worth providing a source. Trying to carve out an exception for "obvious" facts not needing references just doesn't work for me. What is obvious to one editor/reader may not be obvious to another. Having said that, I'm not out tagging every single little unsourced fact in every article with {{citation needed}}, but I also strive very hard to provide a useful reference for everything I add to an article, and occasionally for what other editors have added, if I think its important enough. -- Donald Albury^(Talk) 19:06, 17 July 2006 (UTC)

You are talking around me. I was only responding to your comment that anything unsourced was OR, which its not. The point originally made was that a new article should not start off insisting on RS "right-off-the-bat" and shouldn't be AfD'd for not having RS. Rather the article should be *constructively* and not *destructively* modified to come into closer approximation of a sourced article. Too many editors use RS as a bat to beat people over the head who are trying to contribute. Using is a much better technique to get articles into better shape. I left a tag up for a week before I reverted, I think that's fair to everyone. Of course outrageous claims wouldn't need a week :) But saying "President Bush hates broccoli" or "Mother Theresa died at the age of 86" doesn't seem outrageous. Maybe not factual, which means you should use to try to elicit a citation. At any rate, just pointing out that OR doesn't necessarily apply to what is obvious, common, familiar to the group that has expertise in that area. Wjhonson 19:17, 17 July 2006 (UTC)

I didn't use the CMYK example by accident. Most people will agree that blue is blue, you see, even without a specific text source. Some things are just common knowlege in a particular field. Many things are very easy to verify by trivial experiment. If an article starts out unsourced, it's no big deal. We can add sources later. It's often unlikely to be very wrong, and we'll have more articles, which is good! Removing information just because it's not referenced is probably not as great an idea as you might think either.

Of course, by the time we try to make the article featured, all statements do need to be referenced.

The one place where we need to be careful to make our statements verifiable is when we talk about living persons. It's important. Living persons might be harmed in their daily lives by what we write. Kim Bruning 23:18, 17 July 2006 (UTC)

Regarding whether WP:V is non-negotiable, Jimbo has just made a related comment on the mailing list in response to someone who suggested: "As WP:POLICY makes well clear, the nature of the wiki is such that nothing is immutable; were most frequent contributors, for example, to determine that we should no longer require verifiability, it's likely that Wikipedia would (d)evolve in a fashion consistent with community consensus ..."

Jimbo responded: "Actually, I consider WP:V to be so central to Wikipedia that if there were ever a significant majority of contributors who wanted to do away with it, we would have an internal war on our hands that would make the userbox wars look simple by comparison." [5] SlimVirgin ^(talk) 10:39, 18 July 2006 (UTC)

That's unsurprising. It's a very important guideline, especially in certain areas.

It's certainly open to negotiation and interpretation though.

The way I interpret non-negotiable suggests that things are "set in stone" or "unchangable". However, especially in areas like hard sciences, there can be interesting and useful interpretations of what is verifiable, what is verifiability and etc. (as I've also put somewhere higher up in this discussion)

Even discounting that, even if this particular workflow is considered complete and perfect for all eternity, people might still have interesting ideas as to how to phrase the wording of the guideline to make it clearer.

Fortunately (or unfortunately), in reality things are never perfect, the world changes a little more every day, and we need to continuously adapt. Kim Bruning 11:32, 18 July 2006 (UTC)

It's not any particular wording that's non-negotiable, but the principle; and therefore we have to be careful that the principle isn't changed or undermined by careless wording. Same with NPOV. It's regarded as non-negotiable, but that doesn't mean the page can't be edited. It means only that any editing has to be done carefully. SlimVirgin ^(talk) 11:39, 18 July 2006 (UTC)

They are not truely non-negotiable of course, in theory they are still open to discussion. It's very important to leave those "few inches of give" in there, in case we discover a superior approach, or discover a fatal flaw. This is the difference between designing things for short term or designing for long term. I expect wikipedia to be around for 100 years, so it does have to be possible to abandon a guideline altogether. I have no idea why we'd want to be able to abandon this guideline right now, but when designing things, I've stepped into the trap of making things unchangable many times. Sooner or later, murphys law[1] predicts that the one thing you made unchangable and solid turns out to be the lynchpin in situation you'd never envisioned when you started out. This is the point where you must either expend an utterly heroic effort , or -failing that- abandon the project entirely.

Kim Bruning 14:21, 18 July 2006 (UTC) [1] Murphys law is actually backed up by statistics, when applied to longer time periods. Given sufficient time, every possible combination of circumstances will occur at least once. This includes precicely those combinations you were praying would never happen. ;-)

Yes, point taken. :-) SlimVirgin ^(talk) 14:49, 18 July 2006 (UTC)

Confusion. I was told here a month ago, either by you or with no objections from you (SV) (I think it might've been Francis who said it), that there is no separation between the wording and the principle; the way the policy is phrased is no more negotiable than the policy itself because they are one and the same. This was a reason given for reverting some of my edits that were intended to adjust the phrasing to more clearly express the policy. In effect, the policy page was already locked. —mjb 19:08, 20 July 2006 (UTC)

Your wording may have changed the policy slightly. I'd need to see a diff. SlimVirgin ^(talk) 19:21, 20 July 2006 (UTC)

OK, I went looking, and now I see that it was AvB, paraphrasing Francis, who made the original declaration about the inseparability of phrasing and principle. As I look over the history now, it seems there is no evidence of a direct connection between my attempted edits of May 30/June 2 and this statement on June 10, but the point remains: we are getting mixed messages here. Is the phrasing/examples/presentation of the policy 'negotiable', or not? —mjb 22:54, 20 July 2006 (UTC)

Negotiable. Those who disagree can start their own darn wiki! ;-) Kim Bruning 23:10, 20 July 2006 (UTC)

Ut oh. <Looks uncomfortable>. What was the reason to protect atm? --Kim Bruning 21:57, 20 July 2006 (UTC)

A series of revisions and reversions between June 19 and June 25, mostly between SlimVirgin and Francis Schoenken, relating to the phrasing of the intro/nutshell and the section on self-published sources, I believe. —mjb 22:54, 20 July 2006 (UTC)

It was Francis Schoenken against pretty much everyone on the page, as I recall. Which was the edit you wanted to ask about, Mjb? SlimVirgin ^(talk) 23:23, 20 July 2006 (UTC)

I'm not asking about my edits; I'm only asking about the inseparability of the WP:V policy and its phrasing, and the non-negotiability of the phrasing. Francis provided a satsifactory explanation below. At some future date I'll re-bring up the issues I was attempting to address with my edits. They partly have to do with topics being addressed in current discussion, so I'm awaiting the outcome of that, for now. —mjb 22:20, 21 July 2006 (UTC)

@MjB: AvB's paraphrasing was a misquote. I was speaking about WP:NPOV ("absolute and non-negotiable" per Jimbo [6][7]), not about WP:V. And AvB had mutilated the quote: the discussion where it was taken from can be found here: Wikipedia talk:Neutral point of view/Archive 021#Infallibility - a meta-question re non-negotiable. The discussion on that talk page continued in Wikipedia talk:Neutral point of view/Archive 021#Non-negotiable, where I referred to User:Jimbo Wales/Statement of principles, point 6: "The mailing list will remain open, well-advertised, and will be regarded as the place for meta-discussions about the nature of Wikipedia. Very limited meta-discussion of the nature of the Wikipedia should be placed on the site itself."

I had that quote of Jimbo's principles followed by this:

---

Returning to the proposed text ("The fundamental principles of the three policies are non-negotiable and their policy pages may only be edited to better reflect practical explanation and application of these principles."):

1. Jimbo does not pre-emptively restrict editing of policy pages to merely changes that "better reflect practical explanation and application of these principles" and nothing else. On the contrary, if some change regarding the "nature of Wikipedia" would be decided via the mailing list (there are several mailing lists now, choose the appropriate one at Wikipedia:Mailing lists), then it would be Doing The Right Thing to bring policy pages in line with such decision.
2. Please move meta-discussions regarding the nature of the non-negotiability of NPOV to the appropriate Mailing list. Above I said some things about NPOV. This was spread to some talk pages of other policies. I consider that quite inappropriate: I wasn't talking about these other policies. The only place where this could be taken outside this talk page (if this is more than a "very limited meta-discussion" regarding the NPOV policy) is the appropriate mailing list. But then again, when quoting what I wrote above no confusion should be spread that I would have been talking about anything else than the NPOV policy. tx. --Francis Schonken 08:28, 15 June 2006 (UTC)

---

--Francis Schonken 07:33, 21 July 2006 (UTC)

Thanks for the clarification. I'm glad it was only a misquote. —mjb 22:20, 21 July 2006 (UTC)

Negotiability

To perhaps clarify some things about "non-negotiable" policies and principles (with care to distinguish between these words):

• No policy can be repealed for a particular article by consensus in that article's discussion, but especially with these policies and even moreso for their principles, there is no wiggle room.
• Insofar as the vast majority of editors approve of these principles and they are furthermore mandated by the Foundation as essential properties of the encyclopedia, these principles cannot be altered or repealed by even by the relatively broad consensus that could be reached on the policy discussion page. It would require much broader and more long-lasting discussion, and even then would be subject to decision by the Foundation.

To respond to one point with regard to Verifiability specifically:

• This principle and policy requires only that it must be possible to check a fact. It is one step detached from Reliable Sources, permitting even in principle the free wiki contributions that do not specifically cite the source. This policy does not and should not discourage reasonable contributions by passers-by that do not specifically cite sources, even beyond the de facto situation we have with other policies like NPOV, where a POV section may be added and then later transformed to be NPOV.

—Centrx→talk • 20:45, 17 July 2006 (UTC)

Actually, people can do and actually do do whatever is necessary to improve the encyclopdia. They do need to show that what they do is an improvement. If you're not sure what makes an improvement, we provide guidelines to give you a rough idea on what we're thinking about.

We should patrol our guidelines more often. ^^;; Kim Bruning 23:23, 17 July 2006 (UTC)

I have to say that I find the "non-negotiability" aspect preposterous. The policy as currently written (which I find in a shocking state compared to the versions of just six months or so ago) isn't adhered to in practice and indeed if it was we'd hardly be able to write a single sentence without some piddling citation. It's not non-negotiable at all, and indeed (as older versions used to say) must not be interpreted in isolation. I suggest that it might not be a bad idea to roll back to an earlier version of this policy before it got messed up. --Tony Sidaway 23:22, 20 July 2006 (UTC)

Which aspects of the policy are not adhered to in practise, Tony? I find this largely is adhered to (by the editors who adhere to anything) on the pages I edit. SlimVirgin ^(talk) 13:56, 21 July 2006 (UTC)

Facts in our articles are not universally sourced--far from it. For common examples, see (I stress that these articles were chosen randomly) Centipede, Minotaur, The Dream of Gerontius (poem) and cream cheese. I think this is a good thing. Overly fastidious sourcing would reduce the quality of writing and make editing articles, if not impossible, very unpleasant. --Tony Sidaway 19:31, 21 July 2006 (UTC)

It is true that our articles are not universally sourced. I think this is a bad thing. I've always believed this was understood to be a necessary evil to get the encyclopedia started, not an acceptable state for articles to remain in.

As for making editing articles impossible, I don't see it. It certainly requires more work for me to cite a source than to toss in stuff I'm pretty sure I remember reading someplace. But is work unpleasant? If not, what is "unpleasant" about citing sources? Dpbsmith (talk) 19:53, 21 July 2006 (UTC)

It's a pipe dream, and a poisonous one. Adding references in the manner dictated by a strict reading of this policy produces an unreadable mess that is bound to omit important context, synthesis and balancing simply because such material is a matter of authorship and as such is not sourceable. No encyclopedia is written this way--certainly not Wikipedia, which at present is the most widely cited and consulted encyclopedia in the world. --Tony Sidaway 20:08, 21 July 2006 (UTC)

The readability issue is a red herring. If this is a problem, it can be handled by a technical fix. At present, I don't think there are enough articles densely sourced enough for it to be a problem.

Recently, many nonfiction books—Lauren Hillenbrandt's Seabiscuit: An American Legend is a good example—have started to use an elegant and simple system. I don't know what the name for this system is. I call it "invisible footnotes." In the case of Seabiscuit, this is a 446-page book with 36 pages of notes, almost all source citations. (This is by no means extreme; Lawrence Lessig's "The Future of Ideas: The Fate of the Commons in a Connected World" is a 386-page book with 66 pages of notes, meaning about 1/6th of the content is source citations).The notes are not indicated in the text at all and do not interfere with readability. Instead, the notes appear and the end in a massive block and each one begins with a page reference and a short description, e.g. '178: George realizes error: "Turf in Review," Morning Telegraph/Daily Racing Form, January 8, 1949."

This system won't work for Wikipedia because articles don't have page numbers, but it shouldn't require much effort to think of various creative solutions. Which are most readily implemented in Wikimedia is a question for the developers. One could, for example, imagine a system in which paragraphs are numbered, with the paragraph number appearing in the margin (and not interfering with readability), and the references are not marked inline, but, instead, the paragraph number links to the references for that paragraph.

I completely fail to understand how citing sources gets in the way of "providing context, synthesis, and balance."

Again taking Seabiscuit as an example, this book was a popular bestseller. Hard to see it as an "unreadable mess." And it is sourced to the hilt. Do you feel that her source citations prevented her from providng context, synthesis, or balance? How? Dpbsmith (talk) 20:19, 24 July 2006 (UTC)

I don't know what "universally sourced" means. Most good editors do cite their sources, Tony, for anything contentious, interesting, likely to be challenged, or where the editor wants to attribute the material to a source for reasons of intellectual honesty. That doesn't mean it has to be done mindlessly, but it is done a lot. Every indication I've seen from the Foundation shows they want more sourcing, not less. SlimVirgin ^(talk) 18:48, 22 July 2006 (UTC)

The information must be possible to check in reliable sources, but it is not necessary that there be specific citation to where to check. This policy currently blurs this a bit, but that is the encyclopedic principle. A chemistry textbook or Britannica doesn't have a footnote every sentence pointing to where it got the information, but a reader can still verify it by looking at any major source. We need citations on Wikipedia because, whereas Britannica makes sure that the articles are written by at least a nominal expert in the field, and are reviewed by other experts and fact-checkers, information added to a Wikipedia article could have just as well been added by a quack. As the principle, we don't accept the authority of the random editor, but this does not mean that obvious facts are not allowed, or that uncited implausibilities are not provisionally permitted. The Verifiability page currently has a summary of things that are in Wikipedia:Reliable sources or Wikipedia:Citing sources, but that is not essential to its meaning. —Centrx→talk • 07:12, 22 July 2006 (UTC)

We can't compare ourselves to the EB, where experienced encyclopedists or subject experts are writing. That's why we have to cite our sources and they don't. SlimVirgin ^(talk) 18:48, 22 July 2006 (UTC)

And I don't think Tony was suggesting that we have no footnotes. Rather the happy medium. A paragraph of 8 sentences with three footnotes is certainly fine. But if that same paragraph had 24 footnotes, it would be unreadably cluttered. I'm not sure we want to push for the situation where there are more footnotes than content, even on the most contentious articles like Bahaullah. Wjhonson 07:28, 22 July 2006 (UTC)

Yes, indeed. It boils down to common sense, which we can't legislate for. SlimVirgin ^(talk) 18:48, 22 July 2006 (UTC)

But can't we try? +whine and pounding fist+ Wjhonson 18:54, 22 July 2006 (UTC)

Common sense tells me we cannot. SlimVirgin ^(talk) 18:56, 22 July 2006 (UTC)

Meh, people should never be able to just talk their way out of having to provide a source when a claim is challenged. If we can include original research just because we happen to feel like it, that makes the project a real joke, if you ask me.

What the people who have this fear of WP:V don't seem to get is that only dubious claims get removed... if something's uncited, that's fine, so long as no one thinks the claim is incorrect. If people think the claim is wrong, and a source can't be found for it, then the claim should be removed... does anyone really think we shouldn't do it that way? But no one is going to run around deleting articles like Ice cream because they don't cite sources... acting like that is a logical result of "non-negotiable" is just silly. --W.marsh 19:45, 22 July 2006 (UTC)

But no one here, that I can see is disputing you. The WP:CIVIL way to flag a claim is with {{fact}} and allow the author or some other passer-by to find the cite. There was a recent, highly controversial dispute on the Hans Christian Andersen page that used this very method. And it was successful. All parties were satisfied. However on the flip-side, people remove claims, don't discuss why, don't ask for cites. I don't think people have a fear of WP:V, they have a fear that people will use it as a club to silence any view they don't like. We don't need wikicensorship any more than it already exists. And they will state that a source is not "verifiable" simply because they can't find it, or think it's hard to find. But WP:V doesn't set the bar that high. It merely states that it must be possible, not easy. Wjhonson 20:13, 22 July 2006 (UTC)

To clarify, I absolutely believe that challengable claims must be sourced. However I've recently encountered editors routinely removing unsourced content from list articles, in the belief that they were doing the right thing, even though the unsourced items they removed were common knowledge. It would have been more appropriate to ask on the talk page for citations, if they were thought necessary. So strict interpretations of verifiability are sometimes applied. If they were applied universally it would be silly, which is why I've said that "overly fastidious sourcing would reduce the quality of writing and make editing articles, if not impossible, very unpleasant". In general we're a sensible bunch so it's a non-issue, really. However for the above reasons I don't think it's appropriate to use terms like "non-negotiable" about this policy. It is in practice interpreted in the light of commonsense. --Tony Sidaway 20:23, 24 July 2006 (UTC)

Recently, I've run into the problems Tony is describing while I was editing a stub on the Enteric Nervous System. My content was deleted twice, each time. within hours of my edit. The stub already had a section of sources and I was merely summarizing. A cursory google search brings up a page of sites with essentially the same information. The requirement of citing every paragraph would mean attaching the same list to each paragraph. When I protested I was repeatedly referred to WP:V.

It seems to me that our primary concern should be our audience. We need to label every article that is inadequately. sourced and provide an index to the featured articles. h2g2 does that. They have levels of peer review and avoid the need for more than minimal censorship.

Being revert happy is not the only way to solve the references problem. I'd personally rather let unsourced information remain a while with a flag so other Wikipedians can track it down, Thats what collaborative publishing is all about but I was told that its against policy. But you could hear an important fact, on the radio for example and not have time to track it down. Why not delegate to someone who has the time? References are always a work in progress because authorities can publish mistakes that are later corrected. We can't expect the average Wikipedian to have the time and diligence of Lessig but we could put reference chasing on our todo list.

I'd like to share Tony's optimism but its not consistent with what I've seen of organizational behavior. From what I've seen they move to more rigidity and bureaucracy and ultimately ossification. I'm not saying it always happens but I haven't experienced an exception. Eventually the structure forks ( called a startup in the corporate world ). The parent may still be useful but like duck tape, handy but not cutting edge. Cayte 00:03, 25 July 2006 (UTC)Cayte

Clarification on self-published sources

Assume that a corporation has published to its web page a press release regarding a piece of proposed legislation. Clearly, the policy as phrased would permit the mention of this press release in an article discussing the corporation. Would this policy permit the mention of the press release in an article discussing the legislation itself? - O^O

It might be edging more into the WP:RS guideline's arena of address when it comes to particular sorts of websites. WP:V is giving you know, broad, generral statements of intent, policy to be followed. While WP:RS hammers out the differences in reputability between business websites and personal websites, between The New York Times and a college newspaper's reputability. Terryeo 11:56, 24 July 2006 (UTC)

If the press release is notable enough to be included in the article, I'm sure you could find mention of it in a reliable news source. If you want to include the business's stance on the legislation, again, if it is notable enough you should be able to find coverage in a reliable news source. --Dalbury 21:05, 20 July 2006 (UTC)

So-called "notability" is neither here nor there. A press release sourced directly to the corporation is preferable to a write-up in the press as a representation of that corporation's official view on the legislation. Whether that view should be included in any particular article is outside the purview of this policy. --Tony Sidaway 21:19, 20 July 2006 (UTC)

I know I've disagreed with Tony before, but in this I think he is spot on. Going directly to the source is preferable to a reporter's summary of the source. Do many share the opinion that a corporate press release is uncitable because it is self-published? - O^O

I agree with Tony to the extent that this is not covered by this policy. On the general issue, we don't want some flack for the WeSaySo Corporation including their official position on every issue we have an article on, although he will be citing their official press releases; not even if he labels all of them the company's PoV. Septentrionalis 16:27, 21 July 2006 (UTC)

The point I was trying to make is that we don't want to use a company's press release in an article about some piece of legislation (as was asked above), unless there has been press coverage that indicates that the company's position played a notable role in the drafting and/or passage of that legislation. As Pmanderson says, if we linked to every press release from companies that took a position on some subject, we could be giving a very unbalanced view of the subject of the article, and end up with far too many links from the article. -- Donald Albury^(Talk) 17:07, 21 July 2006 (UTC)

And then could a company representative themselves, be a wikipedian and edit the article on their company or some related relevant article and cite their own press release, even though technically it's not *secondary*? Wjhonson 18:52, 21 July 2006 (UTC)

I'm just having trouble thinking of when it would be appropriate to use a press release in an article. I think there will almost always be more reliable sources than press releases for information pertinant to an article. -- Donald Albury^(Talk) 23:58, 21 July 2006 (UTC)

Isn't the relative reliability of a press release, as compared to some other source, the domain of the WP:RS guideline(s)? WP:V's role, if I'm not mistaken, is to say that as a matter of policy, WP:RS should be taken into account when adding information to an article. WP:V itself should not be in the business of ruling in or out any particular types of sources across the board, so I really don't understand why there is so much momentum here to have it do exactly that. Am I crazy to think that the answer to the question "is X a reliable enough type of source for claims like Y in context Z?" should be debated in Wikipedia talk:Reliable sources and the decisions reflected in WP:RS? When was it decided that WP:V would make authoritative statements about the intrinsic unreliability of some types of sources, while leaving others in WP:RS? It seems to me that including such things in WP:V is elevating what are ostensibly guidelines to the level of policy. WP:V already does, by reference to WP:RS, require that editors apply due levels of scrutiny to sources, and that seems sufficient, assuming WP:RS is up to snuff (covering all types of sources), so I don't see why there needs to be this struggle to figure out what blanket statements WP:V can make about self-published sources and press releases. Let WP:RS do that. —mjb 03:54, 22 July 2006 (UTC)

Say that Congress were considering a major piece of automotive safety legislation. Personally, I feel that a press release by General Motors expressing support or opposition to that legislation is germane. But it appears that there are some here who would say that we cannot cite it, because it is self published. Personally, I feel that position is taking things to an extreme. - O^O

A publication by an automotive maker would be as an interested party. Press releases are published in the press, normally. But if a souce of a press release is small and unimportant and thier words are do not reach a newspaper but are only published on their own website, then that might constitute "self-published" as you are using it. But anything in a newspaper, well, a good newspaper, can be included in any article, I believe. Terryeo 11:51, 24 July 2006 (UTC)

How does one go about verifying the sources on this page?

Phi Alpha Literary Society Precis 12:21, 25 July 2006 (UTC)

That's an interesting question, which has come in terms of other materials of limited circulation. Much of the article, however, should be supported by the Phi Alpha Catalogue 1845-1890, of which the IC Library has four copies. Try interlibrary loan. Septentrionalis 20:48, 26 July 2006 (UTC)

Summary

I've noticed some work being carried out on this article. Is there a simple summary of the outstanding issues, or has everything been resolved now? Stephen B Streater 18:36, 25 July 2006 (UTC)

Suggested addition to "Burden of evidence" section

I do this suggestion as a result of these recent/ongoing discussions:

• http://en.wikipedia.org/w/index.php?title=Wikipedia:Deletion_review/Log/2006_July_13&oldid=64476061#Kitty_May_Ellis
• Talk:List of gay, lesbian or bisexual people#Is Rictor Norton's My Dear Boy a reliable source?

In both cases an editor contended (s)he had provided reliable sources, so the thus referenced assertions should be included in Wikipedia:

• In the first case the cited sources appeared quite out of reach of any average Wikipedian apart from the original editor. There was a practical near-to-impossibility to bring WP:V in practice: the content proposed for Wikipedia couldn't be verified by any wikipedian apart from the original poster. For clarity, these were the sources:
  • History of Snohomish County, J Daniels, 1962. Privately printed.
  • Clearview, founders and pioneers, Frances Smith, 1982. John Brown & Co. Everett.
  • Snohomish Tribune, various dates from 1956 to 1972.
• In the second case the proposed source was available/accessible enough, but not deemed representative enough (in other words: a "tiny minority view", which needn't be recorded in Wikipedia per WP:NPOV#Undue weight).

In both cases the original editor contended (s)he had done enough, while providing reliable sources (so conforming to Burden of evidence), so (s)he needn't do any additional effort to provide complementary sources that would satisfy the other editors.

So I propose to extend the WP:V#Burden of evidence section so that it makes clear that the Burden of evidence doesn't stop before reasonably accessible reliable sources can be given, which also demonstrate that the statement proposed for inclusion in Wikipedia is more than a tiny minority view.

This might be WP:BEANS (it would be for most of us), and a more handsome formulation of the idea is surely possible, but it might cut short several lingering discussions (for example I also think about "Hubbard's Lecture 7203C05" proposed as a source for the Suppressive Person article, but largely unavailable/unaccessible to anyone else than the single Wikipedian that appeared to have a copy of it - discussions have been going on for weeks). --Francis Schonken 07:52, 26 July 2006 (UTC)

I agree it's good to be able to have independent verification. I have many magazine articles, as well as company report and accounts, from the 1990s which are relevant to material covered in Wikipedia. Just because no one else on Wikipedia has a copy doesn't make them non-reliable sources. I suggest that providing copies of verification material should suffice if other people cannot find originals. In this modern day and age, it is easy to make a copy of the relevant section of an annual report and accounts from 1993 and put it on the Internet. Stephen B Streater 08:13, 26 July 2006 (UTC)

"Must allow independent verification" seems a good formulation of the principle too.

Re. "easy to put on the internet": not always: for example, that might be a copyright infringement in some cases (it would in the "Lecture 7203C05" case if I understand the context of that debate, and it would be for most journals and magazines published in the 20th century).

I was still thinking about another example, in Constructions of Subjectivity in Franz Schubert's Music four versions of an essay by Susan McClary are mentioned: (1) a 1990 lecture; (2) a 1992 lecture; (3) a publication in the Gay/Lesbian Study Group Newsletter; (4) publication in Queering the Pitch: The New Gay and Lesbian Musicology, ISBN 0415907535. The Wikipedia article contends that all four versions are different... Since only the fourth version is widely available, how could *independent verification* of the assertion that this version is the most "sanitized" of the four be conducted? --Francis Schonken 08:37, 26 July 2006 (UTC)

Usually fair use allows a percentage of a work to be copied. A few examples could suffice without having to copy the entire work. Of course, the real enthusiast would get permission for the entire source to be made available to Wikipedians. Although we all assume good faith, we are also sceptics as the author may have made a genuine mistake. It is often better to leave something out than to contain false or misleading information. I'm not sure of the legal difference between photocopying a sentence and quoting it though - I think the non-human copy would be required for real verification, but fair use may only apply to quotes. Does anyone know the rules here? Stephen B Streater 08:45, 26 July 2006 (UTC)

All in all not so relevant to the discussion here I suppose: quoting a few sentences of an inaccessible but otherwise reliable source on a webpage would only produce a websource that for Wikipedia would be refuted per Wikipedia:Reliable sources.

The topic I started is about the "Burden of evidence" section of WP:V. Not about the production of websources by Wikipedia editors. --Francis Schonken 09:26, 26 July 2006 (UTC)

I see your point. I'm looking into whether a reliable source can be made accessible, rather than finding new accessible reliable sources. Perhaps a photocopy wouldn't count, but I think it often would in practice. I agree with the general thrust that if proof is required, it must be accessible. As a matter of interest, would you count a book in the British Library as accessible? Stephen B Streater 09:38, 26 July 2006 (UTC)

No idea. Anyway, I don't think there is a wholesale answer to that question. I don't see the use of breaking my head over it. Mostly, for individual books, the question would answer itself: for instance a non-notable novel by a non-notable author whose major lifetime achievement was to have a single book enlisted in the British Library, for which every other copy of the book has gone lost... well you see what I'm getting at. --Francis Schonken 10:51, 26 July 2006 (UTC)

---

Look at the Policy section on the main page. Item 3 says

• 3. The obligation to provide a reputable source lies with the editors wishing to include the material, not on those seeking to remove it.

It would be helpful to replace this by

• 3. The obligation to provide a reasonably accessible reputable source lies with the editors wishing to include the material, not on those seeking to remove it.

In fact, it's rather surprising that the Policy section doesn't once stress the need for accessibility, which is obviously the cornerstone of Verifiability. If you'll forgive a little cynicism, this suggestion will never be implemented, because of the petty bickering that would ensue about the need to make precise the meaning of "reasonably accessible". As for SBS's comment, if editor Mr.X cites a reliable source in his possession that nobody else has, then I would think to myself "I don't know Mr.X, why should I trust that he quotes accurately from his source?" If Mr.X puts a copy of this reliable source on his personal website, I'd be much happier, but I'd still wonder about the reliability of the website purporting to reproduce the source. Precis 09:06, 26 July 2006 (UTC)

I'd be happy with your new wording. The level of verifiability will probably depend of the nature of the claim. Reasonable covers it well, as its meaning can adapt over time to reflect changes in Wikipedia culture. Stephen B Streater 09:15, 26 July 2006 (UTC)

Well put. Here is an example that I think could illustrate dependency on "the nature of the claim". Wharton School of Business subscribes to a database that costs hundreds of dollars to access. If a (wealthy) editor cites it to support a rather innocuous point, we, the proletariat, might be inclined to trust it, whereas if the editor cites it to support a highly controversial claim, we may be less inclined to consider the source as "reasonably accessible". Precis 09:44, 26 July 2006 (UTC)

It's an unworkable solution. There are millions of rare books which only exist in say 2, 10, 30, or 100 libraries worldwide. Where do you draw the line at "reasonable" ? It's so vague it's unusable. We are however, conducting discussion on some of these very points at WP:RS and I would suggest this move there, as "accessibility" is one of the issues we've been discussing. You are suggesting, that if I happen to have a rare book by, for example A.P. Sinnett which I do, an early copy with contains additional text, not in the final version, that I cannot cite these additional text, since I, the editor, would then have to *prove* that this particular edition, exists in more than 20 libraries say in the world. That puts an almost impossible burden upon editors who tend to work on obscure subject matter. I cannot cite the Journal of Contemporary Greek Archaeology unless I can also prove that this particular volume exists in more than 20 libraries? Wjhonson 20:43, 26 July 2006 (UTC)

Accessibility/availability is maybe rather a WP:V issue than a WP:RS issue. Whether a source is accessible does not say much on its reliability. But we need accessibility by at least one more Wikipedian than the original poster to bring wikipedia:verifiability in practice. Maybe this also explains why it is so difficult to come to a conclusion on the published/available debate on the WP:RS talk page. Anyway, that's how I see it now. --Francis Schonken 01:00, 27 July 2006 (UTC)

And it doesn't have to be obscure subject matter. None of the four or five standard lives of Alexander Hamilton is available on line, or will be for the foreseeable future, and the best of them is out of print. All of them should be in any decent research library in English - if not checked out; but what do editors who don't have access to one do? For the foreseeable future, what's on line will be

• Primary sources
• The nineteenth century lives.

The first should not be used alone; the second are largely non-factual; much work has been done cleaning up the old myths. (Much work has also been done manufacturing new myths, but that is a separate problem...) Septentrionalis 21:04, 26 July 2006 (UTC)

But are you suggesting that we can, or cannot, cite from these rare books? Wjhonson 21:10, 26 July 2006 (UTC)

The Hamilton bios are not rare; but they can be difficult to access. I think we must look at them to have a decent article. The Snohomish County history is at the other pole; If I found reason to check the assertion, I would demand evidence that there was such a book. (It's not in the RLG Union Catalog, which probably means that Washington State doesn't have a copy.)

As for Sinnett, is this a separate edition, or a hand-written note? And of what book? Septentrionalis 21:37, 26 July 2006 (UTC)

The book The Occult World went through several editions. The most noteworthy being that several long sections were excised on the grounds of plagarism. This is probably noted online somewhere, on Theosophy pages, but was well-stated in newspaper articles of the time, at any rate the most accessible online version does *not* contain the excised text. I believe they even state, that it is a fifth or sixth edition or something of that kind. At any rate, my own edition is an earlier one. As you know, at that time, they did not regularly state "xxx edition" in the book, so the only way we can know which editions are older or younger is by outside sources. I am slowly transcribing my edition into wikisource, so it can then be referenced by other scholars who may not be one of the hundred or thousand or ten thousand people to have their own copy of it. I have no idea how many copies were made, but to suggest that an article can't cite a rare book is fairly silly. The DNB cites thousands of books which we only know from perhaps 50 copies. And it's an "encyclopaedia" of the same sort as we're trying to build. Wjhonson 22:14, 26 July 2006 (UTC)

Does it have a date, either on the front or back of the title page? Septentrionalis 22:58, 26 July 2006 (UTC)

You can see for yourself here. The book was originally published, in London in 1881. The so-called 'American edition' that was the most widely bought was published several years later. However, this edition was an in-between edition that was published in Boston in 1882. This edition is known to bibliophiles, in fact its quoted on the topic of the plagarism charge, but its a rare book. The portions related to the charge were excised in the later edition. Wjhonson 06:44, 29 July 2006 (UTC)

I regard this as squabbling. It like arguing that "reasonable doubt" is an unworkable concept in law, because it is unclear where to draw the line. Precis 22:44, 26 July 2006 (UTC)

The effect of "reasonably accessible" will be either to make mandatory the use of online sources, which would be a disaster; or to make the acceptance of the Snohomish Co history a matter of editorial discretion. I would hope for the second, but I don't see that it would change much; such things are editorial discretion (as undue weight) now. Septentrionalis 22:58, 26 July 2006 (UTC)

Come on now, asking that there'd be at least a second Wikipedian (apart from the original poster) who has access to sources is not going to drive everybody to internet. oops. I'm on internet now. --Francis Schonken 01:00, 27 July 2006 (UTC)

Fine; that's a reasonable thought. How about exact phrasing of the form "sources that readers are able to consult"; although this may be redundant with what Precis quotes below? Septentrionalis 22:59, 28 July 2006 (UTC)

The WP:V page currently states: "Verifiable" in this context means that any reader must be able to check that material added to Wikipedia has already been published by a reliable source,, and surprise, surprise, nobody is using this to suggest mandatory use of online sources. Yet you say that the two lil words "reasonably accessible" may result in DISASTER. My oh my, thanks for scaring me straight. I'll say no more on the subject here. Precis 23:33, 26 July 2006 (UTC)

Actually, I have seen that argued. Please note that my objection is to the wording (policies should not be vague where precision is feasible), not the idea; so do continue. Septentrionalis 22:59, 28 July 2006 (UTC)

Thanks for accepting my sarcasm with good humor. I fully understand that you are not opposed to "reasonably accessibility"; rather you fear that what others deem reasonable may seem unreasonable to you. But despite your encouragement, I can't contribute further to the subject, since I don't know how to make the words "reasonably accessible" more precise. As SBS wisely pointed out, what is reasonable depends on the nature of the claim and on the nature of Wikipedia culture at the time--it's a dynamic concept. So I'll leave the task of fine-tuning to you and your cohorts. Good luck. Precis 23:47, 28 July 2006 (UTC)

You really can't. There are books in the US which *only* exist in 12 libraries, such as the New York Public Research library. They don't travel by I.L.L. and yet there they are. Is it unreasonable to cite them? In order to cite them, if I happen to have my own copy, do I have to prove where they are located? Wjhonson 06:49, 29 July 2006 (UTC)

Wikipedia is expanding. So having a source that any editor can verify is going to become impractical. However, having a source that a number of editors can verify will become easier. Many editors will have access to major libraries in big cities. So one option is for reasonably verifiable to mean several editors can verify. Stephen B Streater 07:04, 29 July 2006 (UTC)

Take for example, how much the Mayor of San Francisco paid for his townhouse. This information is published by the County government, in the lands records. Anyone can go to the courthouse and look it up. We are allowed to use primary documents, that are published by a reputable publisher. I don't think anyone would argue that the San Francisco land office is disreputable. The information is published, accessible, and by a reputable publisher. So I go look it up and cite it — $2.5 million dollars for a townhouse! Is it reasonable to expect, anyone who wants to know, to come to San Francisco to look it up. Remembering that the land office does not do research and won't answer requests by phone or mail. You have to come in-person or hire someone to come in-person. This isn't a hypothetical situation by the way. Wjhonson 07:14, 29 July 2006 (UTC)

Good example. SF is 8000 miles away from Wimbledon, but I've been there a few times. More importantly, there are hundreds if not thousands of independent local editors who can pop in to verify this information if required. Let's not forget that Wikipedia is a team effort. Stephen B Streater 07:56, 29 July 2006 (UTC)

WikiProject:Research

One solution might be to establish a project to check sources, not on line. This would be slow; I would be willing to look up some things, but I have my own editing. Septentrionalis 21:54, 26 July 2006 (UTC)

We had already had a very short discussion on trying to build a list of reliable sources. That went nowhere so far, but at least we're on-track I believe, on creating a list of what is considered "published" and "available" for the purposes of this wiki. There was a long discussion on when we can and cannot use blogs, not sure the state of that one, as I dropped out last week. Wjhonson 22:17, 26 July 2006 (UTC)

I don't believe that the suggestion was to build a list of reliable sources, rather to establish a project whose members have access to a decent research library and who are willing to spend the time needed to help verify that an off-line source that has been cited actually supports the Wikipedia text. Some content disputes revolve around whether sources have been used accurately or responsibly, rather than the mere existence of a source or the relative weight to give conflicting sources. Robert A.West (Talk) 12:04, 28 July 2006 (UTC)

It would also require that source-checkers are somewhat familiar with the content being discussed; My experience with source-checking (check out my edit history) is that sometimes sources are cited where one would need a thorough understanding of the subject before understanding what exactly is being cited.

For the record I fully support Septentrionalis's idea of a source-checking project, which is something that will do immeasurable good to Wikipedia's reputation in general. As for my personal involvement I go to a university with a heavy emphasis on humanities/social sciences which should come in handy for someone; I can help when I have time. Mr Bluefin 13:48, 28 July 2006 (UTC)

Suggestion concerning self-published sources

The current policy for when self-published sources can be used in articles other than about the author reads:

Exceptions may be when a well-known, professional researcher in a relevant field, or a well-known professional journalist, has produced self-published material.

My feeling is that this is overly restrictive. I understand that it isn't meant to be an all-inclusive list, but experience tells me that many editors treat it as if it were, and automatically rule out self-published sources that do not fit one of these two categories. My main concern is over the word "researcher"; while this is an excellent standard for articles about academic subjects, it makes it much more difficult to use self-published sources in subject areas that are non-academic in nature, and in which therefore there are few or no professional researchers to publish such articles.

I would suggest a change to make it read either "well-known, professional researcher or expert in a relevant field" or more simply "well-known expert in a relevant field". (I'm not sure whether or not "professional" should remain in this latter case) As an example of the kind of article that could be referenced under this new rule, but couldn't on the old one, consider any legal commentary published on Becker-Posner Blog by Richard Posner. As a US Appeal Court Judge, his opinion on such matters really ought to be reliable enough for citation here, but with the current situation, this would not normally be allowed.

Any comments? JulesH 21:50, 19 July 2006 (UTC)

What about...?

"Exceptions may apply if the author of the self-published material is a well-known professional researcher or notable expert in a relevant field, or a well-known professional journalist."

≈ jossi ≈ t • @ 22:00, 19 July 2006 (UTC)

We should make a distinction between expert opinion and scientific fact. If a recognized subject expert voices his opinion on his blog, then it's good and well and likely true. On the other hand, if someone publishes a preprint on arXiv or publishes a monograph on a specialized subject through BoD that that work has not undergone peer-review. It may well have received equivalent scrutiny after publication and is cited extensively within the field. Dr Zak 22:16, 19 July 2006 (UTC)

Notable expert sounds like good phrasing to me. Citations of the article is probably a good way of assessing, yes: if an article is cited in a positive way in a reliable source, then we can probably transfer at least some of that reliability onto the original article, however it was published. JulesH 23:47, 19 July 2006 (UTC)

Another modification would be to change the condition about self-published sources from related to his notability to related to his field of notability; which is I think what is meant (that is, we can't quote Joe Expert's blog on what he likes for breakfast, unless he is a cereal taster). Septentrionalis 17:59, 20 July 2006 (UTC)

I have a problem that the headline says "dubious sources". If they're dubious, they're by definition not reliable, and we should be clear we don't take information with dubious provenance. ~ trialsanderrors 19:41, 20 July 2006 (UTC)

I see the problem, but the intent is clear: we can use the FAQ of a blog as a source for that blog; even if we don't use the blog for anything else. How would you rephrase? Septentrionalis 14:26, 21 July 2006 (UTC)

I strongly disagree with broadening the exception, especially for areas like law where the quantity of intellectual output is enormous and there are many reliable sources available for any given subject. Posner is an immensely prolific writer and I'm sure almost all of his positions can be cited to one of his many books and articles (which can be found in practically every large public library in America).

As for obscure matters, there are cultural studies journals publishing articles on practically everything nowadays (even science-fiction universes), so there really is no need to rely on unreliable resources. These journals can be easily found through Google, Infotrac, JSTOR, LexisNexis, ProQuest, EBSCO, etc. If a matter is so obscure that not even one tiny journal or small-town newspaper or book publisher will give a few column inches to it, then it's probably not notable enough for Wikipedia. --Coolcaesar 19:05, 21 July 2006 (UTC)

But you would agree, if Posner himself had a blog and he wrote, "I'm filing a divorce from my wife", that that would be a WP:RS on the fact that he wrote that. And you would probably also agree that if a well-known reporter says "This may be printed later today, but Councilman Smith told me he's going to vote against the proposal". That that blog of a well-known reporter, on his/her own blog page, and without doubt that it's really that person, is a WP:RS on both what the reporter said, and what the Councilman said. Would you not agree with those two situations? Wjhonson 19:25, 21 July 2006 (UTC)

Let's say Posner was divorcing and it was a current event. Then I would agree that Posner's blog could be used as a source for his intent to divorce, and perhaps for his intent on a day-to-day basis as the divorce evolves. But if it is something in the past (several months or more), then someone would have published an article or book about it (which in turn would probably cite to Posner's blog and interviews and other sources and so on) and we could cite that article or book instead. Or if the divorce turns into a major case, then it would go up on appeal and we could cite the court opinion that eventually results (as I have done in the article on Richard C. Atkinson with regard to the alleged affair). And so on.

For certain short-term uses (within one or two months of a significant event, before one has the advantage of hindsight), a blog may be a reliable source because no sources of superior reliability may be available. But for stuff that happened several years ago, there is no reason to rely on inherently unreliable blogs when there are many alternative reliable sources available at any decent public library or community college library. That is, if an event is important enough, then a journalist or writer will go interview the key players, and we can quote the articles that result. Or as happened with the O.J. Simpson trial, many of the key figures involved may write books analyzing their own actions.

Anyway, most public libraries subscribe to Infotrac, EBSCO, ProQuest, or NewsBank, which can be accessed from one's home with a current library card number, so there really is no excuse for doing one's research.

As for the second situation, I do not think it is a reliable source because it is hearsay about another person's intent to commit a future act. Anyone can claim that someone else said they intended to do something; it is quite different when someone announces their intent and signs their announcement with their name. --Coolcaesar 07:31, 22 July 2006 (UTC)

I agree with your analysis except in the last part, the difference is, this is a well-known reporter, whose columns are cited as reliable sources generally. But they just happen to also have a blog where they post little bits now and then :) Wjhonson 07:35, 22 July 2006 (UTC)

OK; perhaps Posner was a bad choice -- he is well published, and as pointed out you can find out just about anything relevant about his opinions from well-known third party sources. Let's take a slightly different tack: C. E. Petit is a well-known and respected copyright lawyer. He's perhaps best known for representing Harlan Ellison is his case against Robertson et al (see bottom of Harlan Ellison#Controversy). He isn't widely published, except on his own web site. He has some interesting theories concerning the applicability of copyright law to fan fiction, that could be usefully incorporated into legal issues with fan fiction. But he doesn't appear to be a reliable source. At least, not the way most people read this page. Is he? I think he should be. JulesH 15:59, 25 July 2006 (UTC)

"Well-known expert in their field of study, or a professional researcher in their field of study, or a well-known journalist." (Journalists don't have a field of study.)Wjhonson 07:57, 2 August 2006 (UTC)

"Well-known expert or professional researcher in their field of study, or a well-known journalist" means the same thing and is less repetitive. My feeling is it's clearer. JulesH 08:07, 2 August 2006 (UTC)

I'd say journalists do have fields, although not of study. You wouldn't want to quote an IT journalist on interpretation of political events, say. JulesH 08:10, 2 August 2006 (UTC)

The issue with journalists is really a different one. The only time their self-published information can be cited, on any page other than their own, is where they are reporting additional news tidbits on a story already reported in a WP:RS. Reporters opinions, are never citable, unless they are widely-acknowledged as op-ed columnists. Op-ed columnists are secondary sources since they are analyzing primary material (alledgedly). But not every journalist would be a citable, op-ed writer. So the part about journalists, has to be framed in a succinct, but exact way to express all this ramblingWjhonson 19:43, 4 August 2006 (UTC)

Should this also apply to non-corporate sources?

Quote:

A press release sourced directly to the corporation is preferable to a write-up in the press as a representation of that corporation's official view on the legislation. Whether that view should be included in any particular article is outside the purview of this policy. --Tony Sidaway 21:19, 20 July 2006 (UTC)

Should this apply to non-corporate entities? For instance, if it were deemed relevant that a particular individual help some opinion, would it be reasonable to go to a self-published source for that opinion, even in an article not about that person (but, rather, on the subject of the opinion)? I'm assuming the person's opinion is deemed notable enough for inclusion. JulesH 15:33, 25 July 2006 (UTC)

Has this page been protected long enough?

Is there any present need for protection of this page? Can we try an experiment? Septentrionalis 17:59, 20 July 2006 (UTC)

I feel that this page has been protected long enough. It is time to unprotect it. Even if this backfires by unprotecting this page, an admin can always re-protect the page again. I see no harm in doing this as long as a couple of administrators include this page in their watchlist. --Siva1979^{Talk to me} 20:28, 20 July 2006 (UTC)

Mayor of SF's townhouse

Take for example, how much the Mayor of San Francisco paid for his townhouse. This information is published by the County government, in the lands records. Anyone can go to the courthouse and look it up. We are allowed to use primary documents, that are published by a reputable publisher. I don't think anyone would argue that the San Francisco land office is disreputable. The information is published, accessible, and by a reputable publisher. So I go look it up and cite it — $2.5 million dollars for a townhouse! Is it reasonable to expect, anyone who wants to know, to come to San Francisco to look it up. Remembering that the land office does not do research and won't answer requests by phone or mail. You have to come in-person or hire someone to come in-person. This isn't a hypothetical situation by the way. Wjhonson 07:14, 29 July 2006 (UTC)

Some lying prankster could say he looked it up and cite any figure he likes. If he cites $50 million, I'd probably revert. If he cites an undistinguished figure, I would let it go, thinking to myself, "Eventually somebody in SF might decide to waste time and money checking this figure. In the meantime, I'm not going to give it a whole lot of credence." Precis 08:08, 29 July 2006 (UTC)

Publish: 1. To prepare and issue (printed material) for public distribution or sale. 2. To bring to the public attention; announce. See synonyms at announce. [8]

Records that are available in a government office for inspection by the public have not been "published" in either sense of the word. "Publish" is not a synonym for "public."

The spirit of the verifiability requirement involves some implicit estimation of the amount of work that is required to verify a reference, and the implicit standard is "about as much work as it takes to look up a page reference in a print book or journal that exists in many copies and can be found in good research libraries or public library systems."

In other words, if verifying the reference involves a) ordering a book through interlibrary loan, b) travelling to my public library to pick it up, c) turning to the specified page, that's reasonable.

Travelling to San Francisco is not reasonable.

If the 2.5 million figure is all that significant and interesting, then you should suggest it as a news story to the local newspaper. Most papers publish contact addresses specifically for news items, and it is becoming increasingly frequent for reporters to include their email addresses. Then when the figure is published in the newspaper it certainly becomes fair game.

There is, however, another possibility that is worth discussing. What is the legal and copyright status of facsimiles (e.g. a digital camera image) of a public record? I know this sort of thing has been bruited about and I don't know the history here. (It is most often bruited about rhetorically, as when someone who does not agree with the verifiability policy says "What do you want, a photograph of his diploma" or whatever) but it seems worth discussing. If there are no legal/copyright issues, I'd personally accept a digital camera image of a public record (e.g Registry of Deeds records etc.). Of course such an image could be faked, Photoshopped, etc. but it doesn't require much "assume good faith" to discount that possibility. Dpbsmith (talk) 10:30, 29 July 2006 (UTC)

W.r.t. your last paragraph: this seems a non-issue to me. Wikipedia should not be the first publisher of previously unpublished material. That is a corollary (if not the core) of WP:NOR.

If the material is published elsewhere (that is outside Wikipedia), you've got a published source: then first assess its reliability & relevance (WP:NOR, WP:V, WP:RS, WP:NOT, WP:NPOV,...) before deciding whether the information contained in the source can be introduced in Wikipedia. --Francis Schonken 11:55, 29 July 2006 (UTC)

Not rhetorical questions, trying to understand where you draw the line. I think what you're saying is that the fact that the townhouse cost whatever has not been published before, and including good evidence of the truth of the fact does not affect that. OK, that makes sense. But, just to be clear:

1) When uploading images, one of the choices is "GFDL (self-made)." I used this when I took several pictures of the Boston Public Library and added them to the Boston Public Library article. None of these images has ever been published before. They are receiving their first publication in Wikipedia. Is this a problem? If not, why not?

2) The Boston Public LIbrary is accessible to the public. So I was able to take a photo of it. You could travel to Boston and look at it for yourself, or you can look at my picture of it, accepting my testimony that it really is a fair likeness of the BPL. How, exactly, is this different from taking a photo of a public record available to anyone who walks into the Register of Deeds office in San Francisco, then uploading it under the GFDL?

3) There are very likely images in print showing the Boston Public Library. It is, however, at least conceivable that nobody has previously published any image of the mural in the Post Office of Lancaster, Wisconsin. Although it is also quite conceivable that there may have been an image, pubished perhaps in a Lancaster newspaper at the time the mural was painted, or in some small-printing Lancaster Historical Society centennial booklet or something of the sort. Is this image a problem? Is it necessary to determine whether or not there was a published image before uploading and including this one? Dpbsmith (talk) 17:12, 29 July 2006 (UTC)

I agree with Dpbsmith. Slrubenstein | Talk 15:19, 29 July 2006 (UTC)

And I agree with Francis Schonken. If there is notable information in a public record, then someone will publish it in a reputable newspaper or magazine, and then we can cite to that. --Coolcaesar 16:52, 29 July 2006 (UTC)

Actually government offices are publishers. The documents they create, for public consumption, like copies of deeds, marriages, divorces, etc, are considered published documents. Wjhonson 17:16, 29 July 2006 (UTC)

First, if traveling to San Francisco is not reasonable, then where do you draw the line? Wherever you draw it, people will want to move it. I doubt we'd ever get consensus on what is or isn't reasonable going down that road. Wjhonson 16:56, 29 July 2006 (UTC)

Second, its true someone *could* lie and cite a source which doesn't exist, or doesn't say what they claim. That could be true of every source cited in this wiki. The number of claims, and the number of verifiers is vastly different. Out of 10 million claims, maybe one or five percent have been verified, by a truly independent editor, I'd say (just to make my point). Are we supposed to assume people are lieing? Or rather are we supposed to WP:AGF ? Shouldn't the burden of "you are lieing" rest on the accuser? If I cite a book and you don't believe me, shouldn't you be required to prove that I'm lieing? Because there is certainly no way for an editor to prove they are not lieing, if they have included a citation, except to post photoimages of the book which may violate copyright. Innocent until proven guilty? Or guilty by accusation alone without proof? Which do we want the standard to be on wikipedia?Wjhonson 16:56, 29 July 2006 (UTC)

Third, government documents are public. Any member of the public may do anything they wish with them, including posting images online or printing them in a book. And they *are* published. Government documents are printed, then microfilmed (generally) and made accessible to the general public. This is the act of publication. They do not have to appear in your local bookstore to be considered published. The government is often cited as the largest publisher in the country. That is how it is in the U.S. at any rate. Wjhonson 16:56, 29 July 2006 (UTC)

Fourth WP:RS states that we may cite primary documents that are published by a reputable publisher. If people don't like that guideline, they should discuss it there, but as it is, that's the way it currently stands. Wjhonson 16:56, 29 July 2006 (UTC)

'Public' does not mean 'published'. Making records available for public inspection at a designated place and time is NOT publishing in any sense that is meaningful for Wikipedia. - Donald Albury^(Talk) 15:25, 30 July 2006 (UTC)

So you don't accept that we can use court stenographer's transcripts of a trial? Because I'm a little confused about exactly *where* you think these transcripts are *published*. They are published in the office of the Clerk of the Court, in the County or State where the trial took place. You go there, and read them. Are you suggesting that perhaps the stenographer in his/her role as official recorder, is something more reliable than the county clerk in the next cubicle in *their* office as official recorder? I'm just failing to see the distinction you're trying to draw here. Unless of course you don't believe we can use court transcripts either. Wjhonson 18:21, 30 July 2006 (UTC)

WP:RS says that we can use the transcripts when they have been "published". To me, that means when they have been printed up in enough copies to be available in libraries and/or bookstores. I do not accept that 'published', as used in Wikipedia policies, means 'available for inspection in a public office'. -- Donald Albury^(Talk) 21:03, 30 July 2006 (UTC)

This is not about "published" but rather about "accessible". The items do meet the qualification of "published", you argument is that they are not "accessible". However there is, as of today, no consensus on just how accessible an item must be to met that threshold. Wjhonson 21:20, 30 July 2006 (UTC)

Another point, you're wrong about the meaning of "published". The government, when it "prepares, creates, binds, makes available material for public consumption", is indeed a publisher. That covers not just original work, but also primary material created by others.Wjhonson 18:29, 30 July 2006 (UTC) --

Hmm! 'Binds, makes available material for public consumption' implies multiple copies that the public can buy or otherwise take with them. I will agree that constitutes publishing for the purposes of Wikipedia. Whether that provides enough accessibility for a reasonable number of Wikipedians to verify the source is a case-by-case decision. Of course, if they government agency put the information up on a web site, that would be better. I regard government documents that can only be inspected by visiting the office where they are held as equivalent to documents held in an archive. An attempt to get such archived document accepted as reliable sources was turned down by the community in Wikipedia:Archives as sources. -- Donald Albury^(Talk) 21:03, 30 July 2006 (UTC)

That is reading too much into what I said. There is no requirement for multiple copies. An artist may "publish" their material to a website. Their website represents a single copy of that work, and yet is published, and accessible. (However being self-published, and not an authority, researcher, reporter acknowledged as an expert in their field, is not citable, except in the artist's own article.) "Take" and "Buy" is also incorrect. We do not need "take" or "buy" in order to get "distributed to the public". Distribution may be by means that you cannot take. For example a billboard "Cingular offers $10 phones!" may be put up in 100 cities, and is thus "published" but you cannot take or buy it. You are incorrect that "archives" represent published works by a publishing agent on primary materials. "Archives as sources" specially addressed primary material that had never been published, only held in a library, such as private letters, in their original form. In specific, it was noted that if those private letters were "published" by a reliable publisher, in a facsimile form (exact copies, images, photocopies) that *would* be a citable source. This example of a government agency publishing land records on microfilm, is identical to a publisher publishing a facsimile edition of the Dead Sea Scrolls without commentary. Wjhonson 21:17, 30 July 2006 (UTC)

I think the standard should be this: If I cannot order a copy on-line (at a reasonable price), or through inter-library loan, then it is not accessible enough to count. Slrubenstein | Talk 10:26, 30 July 2006 (UTC)

How about:

Without diminishing other requirements regarding reliability and publication, sources used for reference or quotation in Wikipedia need to be accessible or available to other Wikipedians, that is: other Wikipedians should be able to order or view a copy on-line (at a reasonable price) and/or a copy should be obtainable through a major inter-library loan network.

? --Francis Schonken 11:58, 30 July 2006 (UTC)

I think this getting somewhere. I would quibble about the 'reasonable price' to 'view a copy on-line', as I am in that crowd opposed to using on-line sources that require a paid subscription, or even simple registration, to view a source. But that's minor. We do want to find that ground between requiring that a source be available to every possible reader, and allowing sources that only a priviledged few can access. -- Donald Albury^(Talk) 15:25, 30 July 2006 (UTC)

What is the point of saying "major" ? Are there minor inter-library loan networks? By saying that the item must travel, you're basically removing every one of the 125,000 books in the New York Public Research library, none of which travel. You can't check them out, they don't go by ILL. Wjhonson 18:26, 30 July 2006 (UTC)

I think it all comes down to defining some kind of standard for the amount of work that needs to be done to actually verify that the contents of the reference actually match the description. If a book is not available through interlibrary loan, and photocopies can't be ordered from the library holding the item, and the book is, let us say, available in only one of the libraries on this list... well, regardless of what the policy says, most Wikipedia editors aren't going to be able to verify it. Regardless of how we may define "publish..." if a reasonable dedicated Wikipedian can't get his or her hands on it without engaging in plane travel... it's effectively unverifiable. Dpbsmith (talk) 18:38, 30 July 2006 (UTC)

Well what if 100 wikipedians live in San Francisco, and *could possibly* go to the land records office to verify the amount. Now an obstructive wikipedian lives in Guam and says "it's not verifiable!" How do you solve that issue? And remember we're not only talking about libraries on wikipedia, but all accessible sources. That could also include newspaper offices couldn't it ?Wjhonson 18:49, 30 July 2006 (UTC)

Again, and sorry if I have been lately the throwing cold water on these discussions, the reason for these discussions fail me. Let me ask some questions about Wjohnson initial example, with the hope that it may focus the discussion:

• "Take for example, how much the Mayor of San Francisco paid for his townhouse."

1. What is the context?
2. is this related to an article on the Mayor of San Francisco, an article on the price of town houses, or something else?
3. Is this being disputed by other sources, such as a book, a newspaper article, etc.?

• "This information is published by the County government, in the lands records. Anyone can go to the courthouse and look it up. We are allowed to use primary documents, that are published by a reputable publisher. I don't think anyone would argue that the San Francisco land office is disreputable."

Again, context is missing. Please provide some information about how this information will be used in the article and in which article.

≈ jossi ≈ t • @ 21:43, 30 July 2006 (UTC)

Jossi I'm not sure why the "reason for these discussions" fail you. As you can see, this area is not clear, otherwise, ten different wikipedians, who have each read the various policy and guideline pages wouldn't be disputing each other as to what they mean or don't mean. For my example, let's say on the Mayor's own article page, we see something like this: "He was criticized in an op-ed by John Brown of The Post for being a hypocrite on issues related to the poor, as this Mayor is 'the wealthiest Mayor San Francisco has ever had'. However Gail Thompson on the City Council feels that his program for the poor, "Is the best thing we've seen to date." (San Francisco Chronicle, July 12 2004, p 19). He paid $2.5 million for his townhouse. (Office of Land Records, San Francisco County, Deed of Trust, Vol 290, pg 18, 12 Aug 2005).Wjhonson 22:22, 30 July 2006 (UTC)

The first cited reference, to an report *about* an Op-Ed, of a reliable and verifiable source, which is an acceptable source, in the article about which it speaks, makes a claim, that can be further cemeted by the Deed showing what he paid for his dwelling. In this case, a published primary source, is "backing up" what a secondary, published, reliable source has stated. Note the snippet cut from the news story is not the source of the opinion, but is reporting the opinion as part of a larger story, and thus is secondary, not primary.Wjhonson 22:22, 30 July 2006 (UTC)

If I lived in Tuvalu, then I'd have to rely on other editors to verify facts only published in the list of US libraries. Similarly, if I lived in the US, I might have to rely on editors in the UK for published works only available in the UK. I don't think it makes sense to say every editor has to be able to verify everything cheaply. If we say this, we may also have to throw away lots of other articles like Gromov's theorem on groups of polynomial growth which the typical editor will not be able to understand, let alone verify. Stephen B Streater 21:48, 30 July 2006 (UTC)

Your example, Wjhonson, does not work as it fails WP:NOR. We cannot "connect the dots". If John Brown's op-ed refers to the deed of trust, we can cite it, if not we can't. ≈ jossi ≈ t • @ 23:50, 30 July 2006 (UTC)

How do you get that? The guideline page here WP:RS only states that we can use primary sources which have been published by a reputable publisher. It doesn't appear to imply that they must be referred to, by some other source, before we can cite them. Wjhonson 06:23, 31 July 2006 (UTC)

As said numerous time already, guidelines cannot bypass policy, and policies have to be abided by as a whole (all policies working together). Adding such a reference to the article, violates not only WP:NOR, but also WP:NOT. The editor adding that reference is forwarding the POV that the Major is very wealthy based on the value of his home. ≈ jossi ≈ t • @ 17:59, 31 July 2006 (UTC)

Or perhaps the view, somewhat distinct, that it was bought for him as a form of corruption. Neither is Wikipedia's business until secondary sources claim it; and even then WP:LIVING would apply. Septentrionalis 02:10, 1 August 2006 (UTC)

And what WP:LIVING says is quite clear:

Material from primary sources should be used with care. For example, public records that include personal details such as home value, traffic citations, vehicle registrations and home or business addresses should not generally be used. Where a fact has first been presented by a verifiable secondary source, it is acceptable to turn to open records as primary sources to augment the secondary source. Material that is related to their notability, such as court filings of someone notable in part for being involved in legal disputes, are allowable, as are public records such as graduation dates, dates of marriage licenses and the like, where they are publicly available and where that information has first been reported by a verifiable secondary source.

Undent. I don't see how the example violates WP:NOR, and the simple fact of what he paid is not an opinion. If a reader draws an inference from the fact, that's not stated, that burden isn't on the editor. WP:NOR does state that primary sources may be used, which is contrary to what you stated above. Wjhonson 06:24, 1 August 2006 (UTC)

And WP:RS states, "Wikipedia articles may use primary sources only if they have been published by a reliable publisher e.g. trial transcripts published by a court stenographer, or historic documents that appear in edited collections. We may not use primary sources whose information has not been made available by a reliable publisher. " The point is, material taken from primary sources may be misleading and confusing without proper context and interpretation. We, as editors, cannot provide the interpretation such material needs. That is why citations from secondary sources are so important. -- Donald Albury^(Talk) 10:24, 1 August 2006 (UTC)

Handel to Bach

In my London flat, I discover a letter that Handel wrote to Bach in 1749. In a Wikipedia article on Handel, I quote from this letter. How could I make this citation consonant with WP-policy? First I could scan the letter and place it on my website. Next, I could ask several well known experts on the history of baroque music to confirm (on their websites) the authenticity of the letter I have displayed. My website is now a reliable published source for Handel's letter, suitable for Wikipedia. Precis 12:39, 29 July 2006 (UTC)

Er, no. Placing an image of the letter on the Web may mean that it is published, but having 'experts' state opinions on their web sites does not make your website a reliable or reputable publisher. No 'expert' worth his salt is going to authenticate a letter based on a scanned image, so you will have to let them examine the original. If the letter appears to authentic, and has any significance at all, it will likely be written up in some journal. Then you will have a source that you can cite. -- Donald Albury^(Talk) 13:13, 29 July 2006 (UTC)

The experts based their assessments on the original letter, of course. Their public confirmation (that the original letter was authentic and that my website faithfully reproduces it) most certainly makes my display reliable. There is no need for me to wait until such a time that the letter is written up in some journal. Precis 13:52, 29 July 2006 (UTC)

I don't agree, and I would likely challenge any such attribution if I came across it. -- Donald Albury^(Talk) 14:46, 29 July 2006 (UTC)

Donald Albury is right. be content using your own website your own venue for your own concerns. Slrubenstein | Talk 15:17, 29 July 2006 (UTC)

I concur with Donald Albury and Slrubenstein. The only confirmation that I would accept is if a major expert actually wrote up their analysis of your letter in a professional journal or magazine, so that they have put their credibility on the line to back the authenticity of the letter. For example, if they wrote about it in National Geographic or Science (both of which have strict publishing standards), that would be sufficient confirmation for me. Then we could cite to the published article as a verifiable source, and link to your Web site for people who are interested in seeing a high-resolution image of the original letter.

Otherwise, it is just too easy for someone to make up fake experts, fake attributions, etc. As a one-time student of computer science, I have studied the history of social engineering, which overlaps with the history of hoaxes, and I have learned to be highly skeptical about any document. See Frank Abagnale for the ultimate example of why one must be skeptical about documents. --Coolcaesar 16:52, 29 July 2006 (UTC)

The first two entries above offer no specific objections, whereas Coolcaesar provides an interesting analysis which raises a question. First, the historians are internationally known scholars in the field, with long established websites. It stretches credulity to suggest that they are fake experts. These are the very experts that magazines like Science or National Geographic have come to in the past to verify documents. By placing their (independent) analyses of the Handel letter on their websites, they ARE putting their credibility on the line. So here is my question. If along with their letters confirming authenticity they placed scans of the Handel letter on their own websites (instead of just providing links to my scan), would you still challenge citations to these experts' websites? Precis 20:23, 29 July 2006 (UTC)

I feel that it's going to be hard to form a concensus to challenge citations to the websites of known experts in their field. As we're all aware, not all websites are equal. Anyone can create a website and say anything they like, but that should not stop us from validating that certain websites are in fact reliable sources. Wjhonson 00:41, 30 July 2006 (UTC)

A citation to a program on Knuth's website is unlikely to be challenged. But if the (demonstrably) identical program appeared on somebody's personal website, its reliability would be challenged, even if Knuth and other experts publish letters authenticating the program. Why view it as programa non grata despite the authentication? Perhaps the mantra "personal websites are unreliable" has made it difficult to think outside the box. Precis 06:20, 30 July 2006 (UTC)

Primary source material has to be published by a reliable source before we can use it. If I go to a court case and take my own notes, we can't use them as a source. If an official stenographer does the same thing, we can. So your letter from Handel would have to be published by someone reliable before we could cite it. SlimVirgin ^(talk) 10:43, 30 July 2006 (UTC)

The problem here is partly that personal websites can be updated. Just because they were accurate once, they may not be in future. There might be a technical fix with MD5, but most people cannot verify these. Stephen B Streater 10:58, 30 July 2006 (UTC)

I agree with SV's stenographer example, but that analogy doesn't apply here. The Handel letter on my site would ordinarily not be reliable, but it inherits reliability by virtue of the published letters of authenticity from independent experts. In short, my site becomes as reliable as the stenographer's. I have not heard a reason yet to indicate why the letter placed on an expert's sites is more reliable than the identical letter placed on my site, given that the independent experts' sites all point to my letter while vouching for the letter's authenticity. Their authentication is no different in principle than editorial oversight. We shouldn't lock in the notion that oversight can come only from the publishing industry. That's what I had in mind by "thinking outside the box". SBS is correct that I'd have to verify that the letter on my site isn't altered. (That's why I'd used the word "demonstrably" earlier.) This could be ensured in a number of ways. For example, the experts could all have low resolution versions of the letter on their sites. Another method might be to type the letter onto my User page so that months later someone could check the History for comparison. Comments: (1) Waiting for a journal to publish the letter is obviously the best course. But suppose the composers had been lesser known, important enough for WP but not important enough for a journal? (2) One reason I might want to keep the only high resolution version on my personal site is so I could charge a fee for downloading the handwritten copy. Precis 13:26, 30 July 2006 (UTC)

Stepping back a bit to the idea of publishing the letter on your own website after authentication - it depends on who you are and what the website is.

• First step - is there some way to verify that you are who you say you are? Are you a university professor publishing it on your webspace at your university? Are you publishing it on your website sponsored by a publisher where you are a featured blogger? Is it a website that has high enough traffic that if someone was fraudulently claiming to be you it would be likely to get back to you?
• Assuming you are, yourself, verifiable, then comes the issue of reputation. Do you have enough of a reputation that making fraudulent claims would seriously hurt you?
• Finally, is there some sort of independent verification of this, if not by a journal, then by your local newspaper? Guettarda 14:16, 30 July 2006 (UTC)

It's simple. Having 'experts' authenticate a letter you put up on your website does not make your website a 'reliable source'. Now, if a recognized expert posts the letter, or a commentary about the letter, on a website for which he clearly controls the content (say, his personal page at his university), then we can cite what he says about the letter. If a newspaper prints an article about the discovery of the letter, we can cite the newspaper account. But, we still cannot cite your personal webpage as a reference, because it is still not a reliable source, as defined in WP:RS. -- Donald Albury^(Talk) 15:34, 30 July 2006 (UTC)

The point to be understood here is that the actual contents of the original source that has been verified do not matter at all to what can be said in the wikipedia article; no matter how certain we are the source is authentic, anything that can be done with it is original research. The most that we can say in your situation is that the source exists, because this is all the experts have said. If the experts went on to comment on notable parts of the letter (which of course they would), then these comments could be referenced into an article, notwithstanding that the venue of publication is a personal web site, due to the fact that we are talking about well-known experts.

The interesting questions here, as far as I'm concerned, are (a) when is a primary source acceptable (e.g., if the conclusion to be drawn from it is blatantly obvious, then should it be acceptable to use such a source?) and (b) what are the criteria for determining if a personal web site is suitably reliable. In this case, the historians fall within the existing phrasing ("professional researcher in a relevant field") but what about fields where "researcher" isn't an appropriate word? Wikipedia covers a wide variety of topics, many of which are not academic (or even journalistic) in nature. This ties in to my comments above. JulesH 18:04, 30 July 2006 (UTC)

I would argue that making that distinction should be very easy for editors involved in a specific article. Trying to comment on a generic, hypothetical example, will not lead to clarity, IMO. ≈ jossi ≈ t • @ 18:11, 30 July 2006 (UTC)

It should be. At the moment it isn't, because of the number of editors who will remove without a second thought any source that isn't explicitly of a kind listed here or at WP:RS. JulesH 06:55, 31 July 2006 (UTC)

Thank you all for your input, but I'm afraid I've heard nothing convincing so far. All seem to agree that if the letter were published on the experts' site, then WP could quote from it. Yet if the experts point to a copy of the letter on some anonymous website and vouch for its authenticity (and the copies provably have identical wording), some say the copy on the anonymous site may NOT be quoted. Why not? Dalbury's literal interpretation of policy WP:RS seems to be splitting hairs. I would trust the authenticated copy on the anonymous site as much as an identical copy on the experts' site, and I've yet to hear a reason why I shouldn't. Precis 21:02, 30 July 2006 (UTC)

Precis, can you clarify why will you want to cite from an anonymous website when the material exist on another website that is authored by an expert in the field? I don't see the point of doing so. ≈ jossi ≈ t • @ 21:46, 30 July 2006 (UTC)

Sure, the owner of the anonymous site wants to make money from his find, and does not want to relinquish control of the high resolution copy of the letter. He's the kind of guy who finds a strad in his attic and won't let anyone play it. I'll close with a story.

A Wikipedia story, by Precis

A Wikipedian (W) and his date (D) are in my living room, having paid admission to see the Strad that I found in my attic.

W: I don't know why you dragged me here. That's not a strad.

D: What do you mean? The top appraisers in the US authenticated it and gave this address. And look, I downloaded this high resolution photo from Bein & Fushi. This violin looks exactly like it.

W: If this violin were in the offices of Bein & Fushi, then I'd say you have a point. But I don't know who lives here. Precis 21:52, 30 July 2006 (UTC)

Investor I and Musician M:

I: I've just bought the score of a newly discovered Beethoven manuscript. I've scanned it and put a certified copy on my website. Would you like to play it?

M: No - I'm a Wikipedian.

Stephen B Streater 22:02, 30 July 2006 (UTC)

You can play it, but you cannot use it as a reference to the article on Beethoven, as personal websites are not reliable sources, unless you are a renown collector of music scores and/or expert on Beethoven, that is. ≈ jossi ≈ t • @ 23:54, 30 July 2006 (UTC)

Actually I don't think that's enough. You can't be a mere expert, you must be a "widely acknowledged" expert. After all, anyone can *state* that they are an expert in a field of study. Wjhonson 06:26, 31 July 2006 (UTC)

I didn't know about the Beethoven, thanks. And here I thought my scenario was far-fetched :)

Precis 22:09, 30 July 2006 (UTC)

Real Life is always more amazing ;-) Stephen B Streater 22:21, 30 July 2006 (UTC)

Precis, these far fetched scenarios do not help. Let's focus on real examples, shall we? ≈ jossi ≈ t • @ 23:51, 30 July 2006 (UTC)

Sure. Would you say the two sources at Phi Alpha Literary Society are reliable published sources? Precis 06:19, 31 July 2006 (UTC)

OK, here's a real example. Look at this old version of the Barbara Bauer article, specifically the last paragraph of the Controversy section. Here we have an example of quoting a primary source (a list to a google groups search that shows results from 1997 onwards, proving without doubt that the "negative online discussions" mentioned in the previous sentence exist) that was later removed by an editor, along with the sentence that it backed up, as not providing verifiability. I'm personally pretty sure that is verifiable: anyone can look at it and immediately see that the sentence is true. But it falls outside of the letter of what was discussed here, so it has been removed. Does common sense not suggest that this should be allowed to stand? JulesH 06:55, 31 July 2006 (UTC)

It may be verifiable, but ading that reference violates WP:NOR. ≈ jossi ≈ t • @ 18:05, 31 July 2006 (UTC)

I tend to think that relatively inaccessible sources can be used as long as multiple independent people can verify them. The Google case is the opposite extreme - accessible but not traditional. I think that Google can now be considered a reliable source for what appeares on the web. Stephen B Streater 07:06, 31 July 2006 (UTC)

The website of the Phi Alpha Literary Society is a self-published source. It is reliable as to the stated purposes, location and current procedures of the society, largely because there is little to be gained by falsification. It is less reliable as to historical matters, such as founding date or the list of famous former members, because such claims often rest on doubtful grounds. For example, the American Whig-Cliosophic Society claims a 1765 founding date for Clio, and at one point claimed JFK as a former member. Clio claims continuity with the Plain Dealing Club, founded 1765 and forcibly dissolved (along with the Well Meaning Club) after a duel was suppressed in 1768. Both Clio and Whig were founded in 1769. Is this continuity? Since JFK attended Princeton only briefly, even most official society historians have regarded the claim as inflated, but some enthusiast could easily revive the claim on the website. I know of no similar controversies surrounding Phi Alpha, but a good Wikipedian should be careful. Robert A.West (Talk) 12:06, 31 July 2006 (UTC)

The website of that society is a reliable source for the article on the society, but not a reliable source for any other article. ≈ jossi ≈ t • @ 18:05, 31 July 2006 (UTC)

"Material from self-published sources, and other published sources of dubious reliability, may be used as sources of information about themselves in articles about themselves, so long as . . . It is not unduly self-serving;" [Emphasis mine]. Any article that accepts an organization's puffery without using outside sources needs improvement. Robert A.West (Talk) 19:03, 31 July 2006 (UTC)

Of course... ≈ jossi ≈ t • @ 19:06, 31 July 2006 (UTC)

As to the Barbara Bauer article, the objection to the Google search seems to be more that it was de minimis original research. Personally, I would not object, since I think the OR policy, like the law, should ignore trifles, but others would not agree. Robert A.West (Talk) 12:12, 31 July 2006 (UTC)

I that the problem with original research is that it is, by nature, non-verifiable (except by an expert in the subject). That is why including it in wikipedia is clearly wrong; wikipedia cannot verify its own contents, so we expect the readers to be able to do so, but if something is beyond their ability to verify it should not be here. In this case, the link might well be original research, but it is also verifiable: I feel that the verifiability of it should trump the original nature. Others seem to think the other way around. JulesH 14:17, 31 July 2006 (UTC)

Verifiability + NPOV + no original research. This formulation works... when all the elements are there. ≈ jossi ≈ t • @ 18:05, 31 July 2006 (UTC)

That may be so, but this article is only concerned with Verifiability. Stephen B Streater 19:07, 31 July 2006 (UTC)

I disagree... Each and every content policy of Wikipedia contains this wording. The lead of this policy reads:

"Wikipedia:Verifiability is one of Wikipedia's three content-guiding policies. The other two are Wikipedia:No original research and Wikipedia:Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in the main namespace. They should not be interpreted in isolation from one another, and editors should therefore try to familiarize themselves with all three. These three policies are non-negotiable and cannot be superseded by any other guidelines, or by editors' consensus. Their policy pages may be edited only to better reflect practical explanation and application of these principles." ≈ jossi ≈ t • @ 19:13, 31 July 2006 (UTC)

I don't disagree with that. I could have been clearer. This article is not the place to define the other two policies - it is the place to define verifiability. We have three articles, not one. Something can be verifiable but fail the other two. Stephen B Streater 19:37, 31 July 2006 (UTC)

The concern is that unless seen from that holistic perspective (i.e. not interpreted in isolation from other content policies) we may degrade the policy rather than enhance it. I base this assessment on the comments some editors are making on this page. ≈ jossi ≈ t • @ 19:49, 31 July 2006 (UTC)

Fair point. I'll bear this in mind on my subsequent posts. Stephen B Streater 20:15, 31 July 2006 (UTC)

Any article that accepts an organization's puffery without using outside sources needs improvement. Robert A.West (Talk) 19:03, 31 July 2006 (UTC) Good point. I made some reversions which should provide impetus for improvement. Precis 20:33, 31 July 2006 (UTC)

Let me get this straight...

When you say "verifiability, not truth", doesen't that mean we could be telling a lie to everyone, as long it is published in some kind of media? Pacific Coast Highway ^{(blah • I'm a hot toe picker)} 01:16, 1 August 2006 (UTC)

Yes. We simply report what reliable — stress, reliable — published sources say: "According to A, x, but according to B, y." SlimVirgin ^(talk) 01:31, 1 August 2006 (UTC)

Wow, there goes our credibility. Pacific Coast Highway ^{(blah • I'm a hot toe picker)} 01:37, 1 August 2006 (UTC)

As opposed to what happens to our credibility if we let editors add material that cannot be verified? -- Donald Albury^(Talk) 01:51, 1 August 2006 (UTC)

This is a policy I have always opposed. Google "not truth" (with the quotes). Once people see the second result, they'll give up on Wikipedia. --J.L.W.S. The Special One 01:54, 1 August 2006 (UTC)

To understand why this makes sense, you have to think it through a bit. If things are known to be untrue, we can present them as such. See Piltdown man. What this really means is that we depend on other reputable sources- and if they make a mistake, we might have a mistake too. When other sources show it to be a mistake, we'll have that in our article too. All it means is that we do not do our own independant research- we use other sources. Friday (talk) 01:55, 1 August 2006 (UTC)

Cranks often say that we are not competent to argue with their pet theory, and should therefore include it as The Truth, explaining why all other theories are wrong. They're right: We're not qualified; if anyone is, it is the whole body of scholarship on the subject. We are qualified to see whether the theory has been generally accepted (and the answer is almost always no).

It is almost certainly true that the consensus of scholarship will prove to be wrong on some things. But it is not Wikipedia's business to decide which. Septentrionalis 02:18, 1 August 2006 (UTC)

And that is why it is so important to find more than one source for a fact. The more reliable sources we can find supporting a fact, the more confident we can be that we are getting it right. If reliable sources don't agree, we note the disagreement by citing both sides. (oh, wait, that's the NPOV policy...) -- Donald Albury^(Talk) 02:23, 1 August 2006 (UTC)

Want to put in "Not three policies, but one policy...."? ;-> Septentrionalis 15:00, 2 August 2006 (UTC)

Responding to the original question, as phrased, I see virtually no way for a Wikipedia editor to lie within an article and still be within policy. Consider a false statement, Q. There are three cases:

1. No reliable source contains Q, in which case it violates verifiability.
2. All reliable and relevant sources contain Q and none contradict it, in which case there is virtually no way for the editor to know Q is false.
3. Some reliable sources say Q and some disagree, in which case making the unchallenged statement Q violates NPOV.

Yes, occasionally an editor feels that he or she has irrefutable personal knowledge that contradicts what the reliable sources have published. Most often, that "knowledge" will be doubtful or incomplete. Even eyewitnesses routinely misremember or misinterpret what they saw. There's not much we can do about the rare exception. Robert A.West (Talk) 22:29, 1 August 2006 (UTC)

For another example, if an organization describes itself using a self-published source that may be full of inaccuracies, there is not much we can do. Precis 23:07, 1 August 2006 (UTC)

Yes there is; we can

• observe that the source was published by the organization.
• include the facts which contradict it.

In the cases where nobody else has discussed the organization (Is Precis thinking of Phi Alpha again?), its notability is open to question. Septentrionalis 00:51, 2 August 2006 (UTC)

RfC Rah Crawford

An editor has posted a request here on how to indicate, on a living person, that you have permission from them, to post details from their biography, that although created by the subject, were published in an otherwise reliable source. The content was previously removed as a copyvio, but the source, is releasing copyright. How do you handle that, so other editors are aware on the page? Please respond to that talk page, not here. Wjhonson 02:35, 1 August 2006 (UTC)

First, how are they releasing it? I believe that we require it to be released under GFDL. See Wikipedia:Example requests for permission for more discussion. -- Donald Albury^(Talk) 02:46, 1 August 2006 (UTC)

Making discussed change

The change I made was reverted (here) with the comment "rv, I don't see a clear consensus to make the change, even if it does look benign".

In the discussion about the change at Wikipedia_talk:Verifiability#Suggestion_concerning_self-published_sources there were multiple responses that seemed in agreement that the change should be made, and nobody seemed to object strongly to it. Is this not enough consensus to make such a small change to this page? If it isn't, how should such consensus be achieved? JulesH 07:37, 2 August 2006 (UTC)

The distinction I'm trying to draw is that they aren't researchers, which is a term that (as most people understand it) applies only to academic disciplines. There are many wikipedia articles on subjects that are not academic in nature, and the phrasing as it is seems to make it harder for those to use a self-published source than an equivalent academic article (via the "researchers" part of the sentence), or article on a current event (via the "journalist" part of the sentence).

As a suggestion of the kind of expert who would be described, I'd give the suggestion of an expert in antiques -- perhaps for information including a verification of the authenticity of a particular item. It would certainly be a stretch to call such a person a researcher, but they might well be considered an expert, and I think that such a source should be allowed to provide verifiability. JulesH 08:04, 2 August 2006 (UTC)

Existence of opinions and/or rumors

Oftentimes a Wikipedia article will need to mention as a matter of course that rumors exist on a particular subject, or that certain outside opinions exist, etc. For instance, a lot of articles require a "controversy" section to really give complete coverage of a subject.

In that case, wouldn't it be appropriate at times to link to a blog or other unverifiable source, in order to prove "some people are saying this?" Or does it need to be to a 3rd party reliable news source that then quotes the unverifiable source?? heh... I suppose you could argue that linking to the 3rd party source avoids doing original research, but at the same time that also seems like a silly hoop to jump through...

An example I did recently was in MTV#Moral influence. Someone had asked for a citation for the sentence "Critics have said that MTV was like 'pornography for children,'" so I linked to a blog that did just that [9]. Is that appropriate, or is that a WP:VERIFY problem that I should revert? --Jaysweet 15:31, 2 August 2006 (UTC)

Is that blog, just a blog from an non-notable individual? If so, there is no using it as a source. There are millions of blogs out there with opinions on anything. I am sure, though, that if you dig a bit deeper you will find a reputable soure that descibes MTV's critics in these terms. ≈ jossi ≈ t • @ 15:48, 2 August 2006 (UTC)

I see, so unreliable sources can be acceptable as proof of the existence of certain opinions if they are prominent or prevalent enough? That kindof makes sense...

As a corollary to this, I have heard it said that a reliable news source citing what bloggers are saying is a WP:VERIFY problem as well. But that is not explicitly stated in this policy. Are those who claimed that was against policy incorrect, or is this article incomplete?

The article clearly cannot cover everything that is or isn't verifiable. And the policy doesn't make sense in all circumstances... for instance the MTV link above is disallowed, at least by a strict reading of the policy. It can make presenting minority opinions difficult, because those minorities often are not well-represented in mainstream publishing. My personal feeling is that citations of opinions like this should be restricted to quoting the opinions of people who are notable in their own right, at least in some fashion. Otherwise, it would be too easy for articles to descend into long lists of different minority beliefs about some subject or other. Any significant belief is likely to have adherents who are well known enough to qualify as notable. This isn't perfect: it doesn't account for viewpoints on subjects that only small groups are interested in, where those groups are quite likely not to have any notable individuals, for example. JulesH 17:15, 2 August 2006 (UTC)