Wikipedia talk:Verifiability/Archive 61

WP:Deletion policy states, "If the article's content severely fails the verifiability or neutral point of view policies, but when the topic is notable, the article may be reduced to a stub or completely deleted by consensus at WP:AfD."

King of Hearts, who I assume here is a reliable source for the statement, states, "It is very, very rare for a non-BLP article to be deleted because there are reliable sources but they are not included in the article."  I replied, "My experience at WT:V is that the editors there believe that WP:V is our most important policy."  In the specific AfD case, I recommended an Incubate, and as per my comment on the user talk page, there are two words in the article supported as per WP:V. 

I see here a disconnect between policy and practice that goes beyond this one AfD.  For example, at Wikipedia:Articles for deletion/Ensignbus an editor nominates an article that fails WP:V, doesn't mention the problem in the nomination, but then does so in the first response.  At Wikipedia:Articles for deletion/Helen Smith (New Zealand politician), my !vote is keep or incubate.

Meanwhile, in a comment today at Wikipedia talk:Article Incubator, an editor wants to shut down the incubator.

So, to focus the discussion, the question is, should closers of AfD discussion use the incubator for articles that pass WP:N but fail WP:V?  Unscintillating (talk) 00:24, 14 May 2013 (UTC)

No. Articles that fail to pass WP:V should simply be deleted. It's incredibly rare that an AFD closing that actually considers weight of arguments would keep an article that fails WP:V. That's not to say that such articles aren't frequently kept, but it's nearly always an error on the closing admin's part, and shows that he has given weight to weak arguments like "there must be sources out there somewhere, we just haven't found them yet."—Kww(talk) 00:36, 14 May 2013 (UTC)

What does "pass WP:V" mean?

WP:V requires every single bit of information to be verifiABLE. Except for these four types of information, absolutely zero sources need to be named in the article ("verifiED").

If the entire subject doesn't "pass WP:V", then it cannot, by definition, pass WP:N.

Let me give you an example:

A bone fracture is a broken place in a bone.

Does that "pass WP:V" in the sense that you're using? WhatamIdoing (talk) 00:59, 14 May 2013 (UTC)

WP:V requires that articles be based on material from independent, third-party sources with a reputation for fact-checking an accuracy, not simply that each individual statement be verifiable. It is impossible to demonstrate that an article is based on independent, third-party sources without naming those sources. The statement that an unsourced article can pass WP:V is a woefully common misunderstanding that results in keeping some terrible articles.—Kww(talk) 01:16, 14 May 2013 (UTC)

(edit conflict)Not necessarily so. The sources can be presented in, say, the AFD discussion itself and not be in the text of the article. There doesn't actually need to be any source in the text of the article per se to avoid deletion. That doesn't mean such articles are good in any measure, just that, as a subject matter, the article itself covers an appropriate topic. The actual article can be a piece of shit, but deletion is not a cleanup technique, and once it has been established that sources exist that demonstrate that the subject matter is worthwhile, there's no need to delete it since all that would need to happen is for an interested party to incorporate them properly. Of course, if no sources can be produced, the article should be deleted. But the threshold has always been the existence of the sources, not the quality of the actual writing and referencing in the article text. --Jayron32 01:29, 14 May 2013 (UTC)

I agree that that is an unfortunately common practice, but I've always viewed it as a misapplication of IAR.—Kww(talk) 01:35, 14 May 2013 (UTC)

(edit conflict) You might be thinking of pre-2007 versions of WP:N that required that sufficient sources exist to write an article.  As for the last question, WP:V is a content policy, but WP:Deletion policy applies to an article as a whole.  I have given three examples above of articles that IMO fail WP:V as a whole. Unscintillating (talk) 01:27, 14 May 2013 (UTC)

The sentence "Base articles on reliable, third-party, published sources with a reputation for fact-checking and accuracy." is in the current version of the policy, and has been present in some form for the last seven years. WP:V used to clearly state that if the topic was not covered in third-party sources, no article should exist, but I see that someone has removed that critical piece of guidance. Articles without sources cannot be demonstrated to be based on independent, third-party sources, and thus, are unacceptable per WP:V.—Kww(talk) 01:35, 14 May 2013 (UTC)

Indeed, "if the topic was not covered in third-party sources, no article should exist." The key is whether the topic has reliable sources, not whether the sources are actually in the article. As long as specific sources are presented in the AfD, that is sufficient. -- King of ♥ ♦ ♣ ♠ 02:35, 14 May 2013 (UTC)

No, because there's no reason to believe that the content reflects the material from the sources.—Kww(talk) 02:37, 14 May 2013 (UTC)

WP:V does not require that we have a "reason to believe that the content reflects the material from the sources". It does not require editors to "demonstrate that an article is based on independent, third-party sources" to prevent deletion. If we actually required that, then we'd actually say so, in plain language that could not be misunderstood. WhatamIdoing (talk) 02:57, 14 May 2013 (UTC)

Actually, there's no way to say an article meets any policy without having a reason to believe it does. It's just that people allow a lot of hand waving when it comes to WP:V. If a policy says "base articles on reliable, third-party, published sources with a reputation for fact-checking and accuracy" (which WP:V clearly does), then anyone that says an article passes WP:V needs to be able to demonstrate that the content is based on independent, third-party sources. It is in plain language. That you choose not to abide by the consequences is not the fault of the policy.—Kww(talk) 03:07, 14 May 2013 (UTC)

Look at the example of the substub I give above. Would you delete that? Do you believe that it fails to comply with WP:V? WhatamIdoing (talk) 05:22, 14 May 2013 (UTC)

Notice that the word "delete" (or "deletion") is never mentioned in WP:V, so failure to include sources in an article when they exist is not a basis for deletion. Please see WP:BEFORE D3: "If you find that adequate sources do appear to exist, the fact that they are not yet present in the article is not a proper basis for a nomination. Instead, you should consider citing the sources, using the advice in Wikipedia:How to cite sources, or at minimum apply an appropriate template to the page that flags the sourcing concern. Common templates include {{unreferenced}}, {{refimprove}}, {{third-party}}, {{primary sources}} and {{one source}}. For a more complete list see WP:CTT." -- King of ♥ ♦ ♣ ♠ 10:06, 14 May 2013 (UTC)

WP:BEFORE is not policy. Note points 13 and 14 under WP:Deletion policy#Reasons for deletion: articles not based on third-party sources clearly fall under those. As for your toy example, that's deletable under WP:NOT#DICDEF.—Kww(talk) 15:33, 14 May 2013 (UTC)

How long has it been since you actually read NOT#DICDEF? It says "articles that contain nothing more than a definition should be expanded with additional encyclopedic content." That's kind of the opposite of "deletable under NOT#DICDEF", isn't it? WhatamIdoing (talk) 17:01, 14 May 2013 (UTC)

Read the next sentence: "If they can not, Wikipedia is not the place for them". If, in the course of an AFD, the best anyone can do is that, there's no reason to retain it. It's not useful content, and it's not useful history.—Kww(talk) 17:17, 14 May 2013 (UTC)

Do you think a dicdef-style substub about bone fractures could "be expanded with additional encyclopedic content"? Oh, look, someone already did it. That tends to suggest that it can be done, and that therefore deleting it is not appropriate per DICDEF, doesn't it? WhatamIdoing (talk) 21:14, 16 May 2013 (UTC)

An essay like WP:BEFORE is better than a WP:VAGUEWAVE citation of policy. Nothing you've presented actually states that articles without citations should be deleted. -- King of ♥ ♦ ♣ ♠ 17:05, 14 May 2013 (UTC)

And nothing you've presented indicates that an article that fails content policies should be kept. My reasoning is firmly based in policy, and is far more than a handwave. There's no reason to retain content that doesn't conform to major content polices. As noted below, deleting unusable content is not a declaration that no article should ever be created on the topic, it's simply a reflection of the fact that the current contents don't represent a valid starting point.—Kww(talk) 17:13, 14 May 2013 (UTC)

I have: "the fact that [sources] are not yet present in the article is not a proper basis for a nomination." Simply put, if you have objections to an article's content, WP:SOFIXIT. If an article is unsourced but has ample coverage, and the material in the article looks mostly correct, then clearly it is salvageable just by adding citations to existing content. It would be a waste of editors' time to insist that the current version of an article always satisfy WP:V. -- King of ♥ ♦ ♣ ♠ 18:17, 14 May 2013 (UTC)

If you take the time to analyze the content, determine that it actually corresponds to the material present in reliable, third-party sources, and then choose to not make any indication that you've actually performed that analysis, it's not me that's wasting people's time. If you are keeping unsourced articles without actually performing that analysis, then you have no credible reason to believe that they are simply "lacking sources". They may well be false, or hoaxes, or possess any number of other problems.—Kww(talk) 18:24, 14 May 2013 (UTC)

Who says that this example is "an article that fails content policies"? Think about the content here:

A bone fracture is a broken place in a bone.

Do you actually believe that anyone would attempt to assert, with a straight face, that this content doesn't correspond to the content present in reliable, third-party sources? Exactly which sentence in the major content policies does this single sentence fail to comply with? I see no challenge. It's not a direct quotation. It's not about a BLP. I think it highly unlikely that anyone will challenge it in the future. Those are the only requirements for providing citations, and it complies with all four. There is no rule that says "If the sentence normally doesn't require a citation, but it happens to be the only sentence on the page at the moment, then it magically requires a citation now". Our requirement for a sentence like this is that it must be possible for a sufficiently determined and resourced person to find a published reliable source that verifies the claim made in this sentence. WhatamIdoing (talk) 21:14, 16 May 2013 (UTC)

So your point seems to be that articles can pass WP:V verifiability without having citations, which to my knowledge has never been a requirement.  How is this point relevant to the current discussion?  Unscintillating (talk) 23:14, 16 May 2013 (UTC)

It is relevant in two ways:

• because I don't understand what is meant by the initial question about a page failing WP:V (does the example I give fail WP:V? I don't think so), and
• because at least one participant in this discussion believes that supplying at least one citation on every page is actually a requirement of WP:V (I agree with you that it's not, but apparently not everyone agrees with us).

Your examples were not really enlightening to me. Wikipedia:Articles for deletion/Ensignbus was nominated because the nom felt that it failed WP:Notability due to a lack of significant coverage in independent reliable sources. Practically everything in the article can be verified at the company's own website, which means that it does meet WP:V. WhatamIdoing (talk) 23:59, 16 May 2013 (UTC)

• IMO, [1] severely fails WP:V.  Except for the pictures, the amount of effort that it takes to verify it is the same effort as re-writing it.  Unscintillating (talk) 00:29, 17 May 2013 (UTC)
  • WP:V doesn't care how much trouble it is. WP:V does not require that verifying the content be quick, easy, free, cheap, possible to do online, or anything else. It only requires that somewhere in the world, at some point in history, that some published, reliable source (including both non-independent and self-published sources) have already said whatever it is that the editor put in the article. WhatamIdoing (talk) 01:04, 17 May 2013 (UTC)
• IMO, [2] severely fails WP:V.  IMO, this is a credible article and sources have been listed in the AfD that indicate wp:notability.  But the research done during the AfD also suggests that WP:V verifiability for this material needs a local library in the capital of New Zealand.  Unscintillating (talk) 00:39, 17 May 2013 (UTC)
  • WP:V does not require that the sources be available in all countries. If there's a book in New Zealand that verifies this information, then the material complies with WP:V. WhatamIdoing (talk) 01:04, 17 May 2013 (UTC)
• As previously discussed, IMO, [3] severely fails WP:V, with only two words of the article being verifiable with current sourcing.  Unscintillating (talk) 00:50, 17 May 2013 (UTC)
  • WP:V does not care what the currently listed sources happen to cover. WP:V cares about what the possible sources say. If, somewhere in the world, there is at least one source that has previously published this material, then the material meets WP:V. WhatamIdoing (talk) 01:04, 17 May 2013 (UTC)

• Content that has been decided to be unverifiable (not to be confused with unverified) should be deleted. It should not be used for anything. It should not be used as an idea to start any new content building. --SmokeyJoe (talk) 01:48, 14 May 2013 (UTC)

  Well, unverifiable content can frequently be removed without anything being WP:DELETEd, but otherwise I agree with you. WhatamIdoing (talk) 02:57, 14 May 2013 (UTC)

  Of course. --SmokeyJoe (talk) 05:08, 14 May 2013 (UTC)

To get us back to the original question... Should closers of AfD discussions use the incubator for articles that pass WP:N but fail WP:V? I think the answer to that depends on whether incubator is actually achieving its purpose or not. I have my doubts. But, if not, we then have to ask whether there another system that would work better?

I have often thought that we should have two distinct forms of "delete" closure...

1) "Delete article" - to be used when the article has too many issues to merit keeping it. (And with no judgement on whether the topic is notable or not).

2) "Delete topic" - to be used when the topic is not considered notable enough for a stand alone article.

Just an idea. Blueboar (talk) 15:23, 14 May 2013 (UTC)

• I've seen admins specifically note that the deletion was without prejudice against an article being recreated, which is a very similar concept. I think you've hit on the real problem, and the reason for the mushiness in policy. People fear that if we deleted really bad articles because the current content was hopeless on a regular basis, that would be used as pretext to prevent good articles from being created on the same topics.—Kww(talk) 15:37, 14 May 2013 (UTC)

And, of course, we should not forget the option to Merge... which is used when the information is noteworthy enough to be mentioned somewhere in Wikipedia, but is not notable enough as a topic to merit a stand alone artilce.

AfD discussions too often are presented as a black and white, all or nothing, choice between "Save it" and "Kill it". Editors (both deletionists and inclusionists) need to remember that there are "middle ground" options available. Perhaps we need to explain these options in more detail in the policy? Blueboar (talk) 14:38, 15 May 2013 (UTC)

Coming from the WP:N side, there is no way for an article to pass WP:N without having some part pass WP:V. Notability based on the GNG requires secondary sources and ergo requires WP:V to be passed for those (those as noted, the sources don't have to be listed in the article but clearly identified somewhere in the working article space). Notability claims on one of the subject-specific notability guidelines like BIO require a statement that affirms the topic meets one of the guideline criteria, which again, requires a WP:V-meeting statement. So there is no case where WP:N can be meet but WP:V isn't. Now as to well the article meets citation MOS issues or the like, that's a different story. --MASEM (t) 00:09, 17 May 2013 (UTC)

That is exactly the point that confuses me. How the heck do you have significant coverage in multiple, independent, reliable sources and yet have an article that cannot comply with WP:V? From the comments above, the question seems largely to be based on a belief that WP:V requires far more than it actually does. WhatamIdoing (talk) 01:08, 17 May 2013 (UTC)

• WP:N is a guideline that requires nothing more than consensus that a topic be worthy of notice.  But the bigger picture here is that not a single editor here has supported WP:V.  One editor supports WP:Deletion policy, two say that admins know but that it is not something that is written, two are questioning whether we need to talk about WP:V since we have WP:N.  We've also got a discussion about unverified verifiability and a suggestion for the names "delete article" and "delete topic".  Likewise, the only comment about the incubator was that it wasn't working.  Unscintillating (talk) 01:20, 17 May 2013 (UTC)

• I'm glad we are having this discussion. In many cases, guidelines are interpretations or extensions of policy that explain what to do with the general principles of a policy. The reason I closed the AfD as I did is that when a guideline and a policy may both apply to a situation, a direct application of a guideline takes precedence over an editor's own interpretation of a policy. WP:N clearly states that the topic is notable since it has the required coverage, while WP:V is totally unclear on what threshold, if any, should be used to determine that an article is so unsalvageably unsourced that it's better off deleted. I'm not entirely against the idea of deleting notable articles with really bad sourcing which would require a fundamental rewrite, but the criteria for this must be clarified. -- King of ♥ ♦ ♣ ♠ 01:35, 17 May 2013 (UTC)

• Wikipedia:Deletion guidelines for administrators states,

Unscintillating (talk) 23:32, 23 May 2013 (UTC)

This entire section was started with a somewhat ulterior motive, in that Unscintillating seems to object to marking the incubation project historical. To that, I say, fine, but actions are what matter, not words, and the incubator has languished nearly unmaintained with an extremely low level of activity for quite a long time now. There are several editors who believe that it has outlived its usefulness, not just one as Unscintillating implies.

I was one of the editors who helped set up the incubator and define its mission, though I admit that I haven't really participated in improving incubated articles as much as I thought I would, as seems to be the case with pretty much everyone who set out to work on the incubator. The few articles that have made it out of incubation are hardly gems, and many of them would be borderline if they were renominated to AfD. At this point I think we can say "we gave incubation a chance, and it didn't really work". The situation seems analogous to a child objecting to the disposal of a broken toy they haven't played with in years, with them suddenly showing interest in it because someone wants to get rid of it. Gigs (talk) 15:11, 14 May 2013 (UTC)

This change, which was made by S Marshall back in January 4th of this year, defeats the purpose of WP:BURDEN; by telling editors that if they think the material is verifiable, then provide an inline citation for the material before tagging it as unsourced or removing it, WP:BURDEN is placing the burden on that editor to source the material. Before that change to the policy, the wording had been "it is better to try to provide" instead of "provide." The previous wording is not bloat, as it was called by the editor who changed the text; that's a different meaning that was there to avoid what I am now complaining about. And the changed text hasn't stopped editors from removing text when it's unsourced, whether they think it's verifiable or not. Because of these points, the original wording should be restored. I noticed the different wording in May, but am only just now commenting on it. Flyer22 (talk) 06:47, 7 June 2013 (UTC)

I don't think the change is necessarily that significant. I was worried the change was moving the sentence from a suggestion towards a requirement, but as the wording "try to" is still there I'm not overly concerned about this. That said, I wouldn't object to the original text being restored either. Doniago (talk) 13:34, 7 June 2013 (UTC)

I would. WP:PRESERVE is still policy in this encyclopaedia and it's not for WP:V to undermine it. Finding decent sources is everyone's responsibility.

There are good reasons why WP:BURDEN isn't an absolute----it's a powerful tool for dealing with unverifiable material, but it's also a useful tool for griefers and bad faith editors because they find it so much easier to challenge a sentence than to source it. One suspicious-minded WP:BURDEN fetishist can create an awful lot of work for subject-matter experts who've taken a lot of trouble to craft neutrally-worded phrases, and there has to be a fair balance between the BURDEN-type on the one hand and the good faith content creator on the other.—S Marshall T/C 14:07, 7 June 2013 (UTC)

I guess I'm a bit mystified as to why a subject-matter expert would be adding material without providing reliable sources (and how we can assume someone is a subject-matter expert) in the first place, but I suppose that's off-topic. Doniago (talk) 14:25, 7 June 2013 (UTC)

As much as I do not like the last two sentences of BURDEN (as I said in the last excruciating discussion about it that happened about that same time and following an RFC on it the previous October), I don't think that this does any additional harm when understood in context:

When tagging or removing material for not having an inline citation, make it clear that you have a concern that there may not be a published reliable source for the material, and therefore the material may not be verifiable. If instead you think the material is verifiable, it is better to try to provide an inline citation yourself before considering whether to remove or tag it.

I interpret that to say this:

Don't tag or remove material for not having a citation unless (a) you have a concern that it's not verifiable and (b) you express that concern. But you only have to have a concern, which only means that you have to have a doubt or suspicion that there may not be any and does not mean that you have to check to see if your doubt is valid. If, on the other hand, you believe that sources do exist — not "may" exist, but "do" exist — the better practice is to at least make an effort to provide a source, but that's just the best practice, not a requirement, and you can go on and just remove it or tag it if you care to do so.

About the only way an editor could get in trouble for deleting, rather than sourcing, unsourced material under this is if s/he routinely and habitually deletes material which is so unquestionably verifiable with easily-available impeccably–reliable sources that no one in their right mind could conclude otherwise or if he forgets to mechanically insert the word "verifiable" in his/her edit summary or on the talk page. Like Doniago, I wouldn't mind seeing it added back in (or, better, those last two sentences removed). Regards, TransporterMan (TALK) 14:28, 7 June 2013 (UTC)

While WP:PRESERVE is designated as policy, the vast majority of it is clearly stated in permissible permissive, not mandatory, terms and has little effect on what we're talking about here. Regards, TransporterMan (TALK) 14:30, 7 June 2013 (UTC)

TransporterMan, by "As much as I do not like the last two sentences of BURDEN," it doesn't seem that you are speaking of the following lines that are the last two lines of WP:BURDEN: "Sometimes editors will disagree on whether material is verifiable. The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material." You are referring to lines in the second paragraph instead, correct? You mean that S Marshall's change was discussed with the larger concern for that paragraph? Flyer22 (talk) 18:15, 7 June 2013 (UTC)

• Per S marshall, finding decent sources is everyone's responsibility, and the suggestion that people are allowed to find sources for unverified sentences does not undermine WP:BURDEN in the slightest. WP:BURDEN is about resolving conflicts where people disagree. One editor finding an unreferenced statement and then providing a reference for it is not a conflict, and we should be actively encouraging people to improve references all the time, not just when they themselves wrote the statement. Lots of unreferenced material is easily verifiable with a little digging, sometimes very little digging, and in many cases the original author is nowhere to be found, so WP:BURDEN does not provide good guidance in cases like that. Instead "do a cursory search through google and find a reliable source to cite" is best practice, better than "remove it all without bothering". I like the current wording just fine. --Jayron32 14:32, 7 June 2013 (UTC)

I agree: you see the glass as half full, I see it as half empty (or maybe it's the other way around), but we're both okay with the glass as it is. — TransporterMan (TALK) 14:36, 7 June 2013 (UTC)

• My main objection to the section is that it implies all material is worth keeping. If someone pops up with "The asphalt surface of Times Square weighs approximately 11.47 tons" or other similar trivia in an article about New York City, my first concern isn't whether the material is verifiable, it's whether the material is valuable. If I determine the material has no value, I will remove it before I consider searching for a source.

There's also a threshold of value. The real-life situation that occurs constantly is the addition of record sales for a particular country for a particular song. These are always, in theory, verifiable: I could send Nielsen Business Media $50 per song and they would tell me the number. Since I can't do that, I can't add a citation. Many WP:PRESERVE purists would argue that I should be tagging those figures, not deleting them, because they are verifiable, and they would shout about how paywalled sources and the like are acceptable. I can promise you that articles would quickly collapse under the weight of the tags, simply because people feel free to add the sales figures they find in online forums but no one can cite to them. Instead, I simply delete the unsourced sales figures when I see them.

The second thing I usual consider is the age of the material and the editor that added it. If a prolific editor adds unsourced material, it's not my role to be his errand boy and do the dirty work of source research. I will happily and quickly revert such additions. If I find a bit of unsourced material that has been sitting unchallenged in an article for a while, I will go through the decision making process about tagging and sourcing.

This source/tag/delete decision isn't one that lends itself to two sentences of pithy prose. There's a lot of reasonable approaches to a lot of situations.—Kww(talk) 15:04, 7 June 2013 (UTC)

Verifiability has nothing to say at all about suitability for an article, and so your objection is kind of irrelevant here. Other policies and guidelines cover what material should and should not be included in articles for reasons other than sourcing. I wrote a little essay here about the issue, but it has no bearing on whether or not people should be allowed to provide references for statements they find that need them. People can remove stuff from articles for reasons unrelated to verifiability, and the current wording makes no implication that they couldn't. --Jayron32 15:24, 7 June 2013 (UTC)

The underlying problem here is that Wikipedia always refers to its contributors as "editors"... but we also act as "authors". I think BURDEN make more sense if we separated these two roles.
In the real world, there is always a slight conflict between authors and editors... authors don't like having their work marked up by editors... but marking up an author's work is an editor's job.
Now, in the real world, when an editor comes across something that he thinks needs a citation, he will mark up the submission, and tell the author to go find one. This is appropriate... The author is the one who knows the applicable reference materials. The editor is not expected to try to source the material himself.
The author may not like having to do this extra work (grumble grumble, damn editor, grumble grumble)... but he accepts that if he wants to see his work published, the burden to find the requested source does indeed rest with him.
So... my question is... why should Wikipedia be any different? Blueboar (talk) 15:42, 7 June 2013 (UTC)

I still feel that S Marshall's text is placing the burden on the editor who did not add the unsorced material, and that absolutely does defeat the purpose of WP:BURDEN. Right now, the policy is saying that any material lacking a reliable source directly supporting it may be removed, and yet goes on to essentially say..."But if you think or know sources exist for the material, then keep the unsourced material and tag it for someone else to source it or source it yourself." That's often not how I or most editors, from what I have seen, work. When I see a bunch of unsourced text added to an article, especially when I'm only sure that reliable sources exist for parts of it, and when it's poorly formatted, I usually don't go out of my way to correctly format the material and provide reliable sources for it. I shouldn't have to. I shouldn't have to clean up after someone else. It should be a choice to do so, especially considering what WP:BURDEN starts off stating. And the choice to do so should be clear in WP:BURDEN, as it was before. As it stands, for example, the policy (with the exception of dubious unsourced or poorly sourced text about a living person, of course) is telling us to let unsourced material stand in articles, whether they are WP:GA, WP:FA or lower class articles, or to source it ourselves. Not good because, for example, every time unsourced material is added to an article someone watches, it is placing the sourcing work on editor who did not add it. And this somehow does not defeat the purpose of WP:BURDEN? I don't understand that. WP:BURDEN is contradictory in its current "try to" wording. It should state what it did before or "consider trying to." That stated, I am worried about editors who surf articles to remove large chunks of unsourced material from them, sometimes reducing the articles to nothing but stubs, when the unsourced text is clearly or likely verifiable. In those cases, they should clearly add the "This article does not cite any references or sources" tag or the "This article needs additional citations for verification" tag.

While tagging unsourced material with the "citation needed" tag is common, I usually do not see editors going out of their way to source newly added material that is clearly or likely to be sourced, which means that the "try to" part of the policy isn't, or largely isn't, working; I know this even more now from surfing for vandalism or other unconstructive edits while using WP:Huggle. S Marshall's text allows for an editor to state, "You shouldn't have removed my unsourced edits. If I should have added a source for them, so should you have...since you know that sources can be found for them." Therefore, to reiterate, it is placing the burden on the one who challenged the material. Yes, of course, finding sources is everyone's responsibility. In general, that is. The last part of WP:BURDEN, however, makes the following clear: "The burden of evidence lies with the editor who adds or restores material, and is satisfied by providing a reliable source that directly supports the material." If this policy is no longer going by that, then it needs to be revised more. S Marshall stated, WP:PRESERVE is still policy in this encyclopaedia and it's not for WP:V to undermine it." To that, I state, "WP:Verifiability, which WP:BURDEN is a part of, is policy as well." Furthermore, this discussion shows why changes to a policy page, like the one S Marshall made, no matter if they seem like a minor change, should not be made without WP:CONSENSUS first; the policy page states, "Changes made to it should reflect consensus." I suppose that since no one challenged S Marshall's change, we are supposed to assume that it was a consensus change; I don't consider it that. At the time, it came off as a minor copyedit change, from what I see. But since this discussion shows that the text that S Marshall changed is not likely to be revised back to how it was before or to something similar, I'm mostly done with this discussion or will start a WP:RfC about it later for wide input. Flyer22 (talk) 15:54, 7 June 2013 (UTC)

I think your concern could be addressed with a simple change that most editors would agree with "any edit which adds unsourced material can be reverted on that basis. It is the original editor's responsibility to address the sourcing concern. The valid issue that the WP:PRESERVE advocates have is with editors that go through articles removing old and stable material using sourcing as a pretext. That doesn't apply to a fresh edit that adds unsourced material.—Kww(talk) 15:59, 7 June 2013 (UTC)

(edit conflict) No, the act of removal is what institutes WP:BURDEN, and that's fine too. The current wording does not invalidate that. What the current wording does is to remind editors that Wikipedia is not worse off if they provide references rather than remove material. WP:BURDEN is only invoked when there is a genuine conflict over the inclusion of some material. The statement as written does not contradict that in any way. We should be encouraging editors, where possible, to expand the encyclopedia in appropriate ways. In cases where it is appropriate to remove material (for whatever reason, it can be sourcing, but there are other valid reasons to remove text as well), AND where someone objects to the removal, that is where WP:BURDEN comes into play. There are lots of other situations at Wikipedia, and it's not contradictory to recommend best practices to address those. --Jayron32 16:04, 7 June 2013 (UTC)

I stand by what I stated above. Truly take what I stated above into consideration. If S Marshall's aforementioned change is a recommendation, it should be worded clearly that it is, not come off as something we are required to do. And, yes, if editing here, we are required to follow the policies here...unless it's a WP:Ignore all rules matter. If S Marshall's "recommendation" doesn't come off to you or someone else as saying "do this" instead of "consider doing this," then there is absolutely no reason why the text shouldn't be changed back to how it was or to something similar. The previous wording is clearer that it is a recommendation and not a requirement. Flyer22 (talk) 16:54, 7 June 2013 (UTC)

I think there's a tendency on the part of some editors/authors to take the removal or challenge of their material as an implication that the material they added is untruthful (i.e. they're not operating in good faith). In some cases this may stem from an unfamiliarity with WP's policies regarding verifiability, while in others it may simply be that the author/editor is overreacting. Unfortunately it's difficult to resolve this disconnect. Contributing editors need to realize that tagging/removing material is a reflection on the material itself, not on them. Becoming confrontational ("WTF did you remove my material?") is only likely to polarize the situation, and I think in normal circumstances people would realize that...but they've allowed themselves to become invested in the presence or absence of the information and are no longer looking at the situation objectively.

Sometimes I wish every time material was deleted a template could be auto-inserted for a week or so that says something like, "Material was removed here due to verifiability concerns. You're encouraged and welcome to re-add the material, but please include a reliable source." I'd happily bypass any request that a proper citation be provided if at least the source itself was present. Doniago (talk) 16:01, 7 June 2013 (UTC)

• Whereas what I think is that there's a tendency on the part of some editors to remove material without even googling it, and it should be reasonable to administer a clue adjustment in those circumstances. Sure, there are more subtle and difficult situations as well, but hard cases make bad law. Well-written rules deal with the simple and obvious cases first.—S Marshall T/C 16:12, 7 June 2013 (UTC)

• One tendency doesn't preclude the other. Doniago (talk) 16:17, 7 June 2013 (UTC)
• If the material has just been added by an established editor that should know better, that's the correct and reasonable response. We have far too many editors that believe that it is their role to add material and other people's job to source it. That's an attitude that needs to be dealt with harshly. There's a big difference between taking issue with stable material that has been in an article for a long time and material freshly added by a sloppy editor.—Kww(talk) 16:19, 7 June 2013 (UTC)

What constitutes "old and stable"? In articles on more obscure topics, months may pass between edits. Should unsourced material be preserved, simply because it took several months for someone to notice that it was unsourced? Blueboar (talk) 16:23, 7 June 2013 (UTC)

I'd argue that material an editor considers to be "old and stable" but unsourced should be tagged, and after a "suitable" amount of time should be removed or relocated to the article's Talk page. Unfortunately I doubt we'll be able to form a consensus that would allow us to formally define either "old and stable" or "suitable amount of time". Doniago (talk) 16:27, 7 June 2013 (UTC)

• That's part of why this topic is hard to distill into two or three sentences. It can be really hard to define the difference between "old and stable" and "unnoticed crap".—Kww(talk) 17:04, 7 June 2013 (UTC)

I completely agree with Kww that "We have far too many editors that believe that it is their role to add material and other people's job to source it. That's an attitude that needs to be dealt with harshly." That is what WP:BURDEN tackles, but is currently also worded as to make sourcing the job of the editor who did not add the unsourced material. Well, like I stated above, it shouldn't be my job unless I'm the one who added the unsourced material. Therefore, WP:BURDEN will remain contradictory as worded. Nothing can convince me that the current wording is not contradictory. Flyer22 (talk) 17:09, 7 June 2013 (UTC)

Well then, I guess that there's no point in considering that you have anything worthwhile to add, since you just made it clear that you aren't willing to work with others in considering various points of view. Thanks for making it so easy to ignore you. We'll keep you're unwillingness to consider various points of view in mind when we decide how much credence to give yours! --Jayron32 17:27, 7 June 2013 (UTC)

Wow, because I stated "Nothing can convince me that the current wording is not contradictory.," it means that I don't have "anything worthwhile to add" to this discussion and that I made "clear that [I'm not] willing to work with others in considering various points of view."? Absurd. What others have stated in this section in agreement with me shows that I have something worthwhile to add to this topic. Sticking to a viewpoint does not mean that a person is not willing to hear other viewpoints or consider them; otherwise, I wouldn't have started this section and wouldn't be known to many editors of this site as someone who is "willing to work with others in considering various points of view." Hearing out editors does not mean that I have to agree with any of those other editors or should be open to changing my viewpoint. However, I am usually open to compromising, and compromising is a core aspect on Wikipedia when there is disagreement. Your rude reply further shows me why there are aspects of this site that I hate. And you criticize me for sticking to my viewpoint on this matter when you have made it clear that you are unwilling to change yours on this matter as well? Not good logic, but okay. Flyer22 (talk) 17:52, 7 June 2013 (UTC)

I have no real strong views one way or the other. --Jayron32 19:09, 7 June 2013 (UTC)

• I really don't get the concern, of course, "it's better" to keep truthful encyclopedic information rather than delete it for any reason. That's why our articles exist -- to present truthful encyclopedic information. Alanscottwalker (talk) 17:36, 7 June 2013 (UTC)

How do I know the information actually is "truthful"? For all I know, everything in the article may be completely inaccurate. I have no idea whether the author of the article knew what he/she was talking about. I demand sources! Blueboar (talk) 18:44, 7 June 2013 (UTC)

How do you know it's truthful? Probably because you are an educated, literate person with a modicum of reasoning and logic ability, as well as common sense. If your brain is a tabula rasa, it's doubtful whether you should be doing anything with articles here. Alanscottwalker (talk) 20:01, 7 June 2013 (UTC)

Alanscottwalker...let's say that you are "an educated, literate person with a modicum of reasoning and logic ability, as well as common sense." Does that mean that you know everything in the Circumcision article is truthful, without reliable sources validating that information? I don't think it does, so I don't understand your response to Blueboar...other than it being unnecessarily rude...unless editors are expected to know everything. If we should conclude, "Oh, my brain is hinting at me that all this material is truthful. Therefore, I'll let it stay.", then there is no point in even having the WP:Verifiability policy (which is not about truth). Still, as we know, even information that is supported by a source that passes as a WP:Reliable source can be non-truthful (inaccurate).

Side note: For those who don't know, Kww's 17:40 response is to Alanscottwalker; it was moved from its original placement due to other responses to Alanscottwalker. Flyer22 (talk) 20:33, 7 June 2013 (UTC)

Of course, there is point to it. We want it verified but every rule is to be applied with reason, and it is not a mechanical exercise, it is a matter that requires reasoned, and reasonably informed, judgment. Alanscottwalker (talk) 21:23, 7 June 2013 (UTC)

I'm pretty sure there's plenty of information on WP with truthfulness that cannot be determined based solely on the traits you just described. Anything regarding pop culture immediately springs to mind. If an editor claims that Brad Pitt's middle name is Horatio, the fact that I have a college degree doesn't seem relevant to my evaluating its truthfulness. Doniago (talk) 20:39, 7 June 2013 (UTC)

No one said those faculties are the sole basis for determining truthfulness, those are merely the faculties we think it "better" to employ in making editing decisions. In this context, the reason that is so, is that the removal of information is not without risk. Some information when it is removed may make an article positively misleading. Alanscottwalker (talk) 21:35, 7 June 2013 (UTC)

• How would you respond to an editor that persists in adding truthful statistics to thousands of articles while refusing to provide any sources? What policy would you point him at to help persuade him he was not being constructive?—Kww(talk) 17:40, 7 June 2013 (UTC)

• I would say that editor was displaying insufficient clue quotient and a need for support and direction. But I'd question whether that's a situation that needs to be addressed by our verifiability policy; surely we're dealing with a conduct issue there.—S Marshall T/C 17:45, 7 June 2013 (UTC)

• It's always nice to be able to point at something. I've certainly blocked editors for it, and those blocks have generally been upheld by other admins, but they always have kind of a "because I said so" feel to them.—Kww(talk) 18:00, 7 June 2013 (UTC)

Re: I'd question whether that's a situation that needs to be addressed by our verifiability policy;... are you kidding me?... In fact, it already is addressed by our verifiability policy. That sort of situation is exactly why we have a verifiability policy in the first place. How do we know the information actually is "truthful"? We are not going to simply take another editor's word for it - especially when it comes to statistics (lies, damn lies and statistics). Statistics certainly don't fall under the "Paris is the Capitol of France" type of obviously verifiable statement that don't need a citation. I absolutely would remove uncited statistics in any article, and would insist a source be cited in order to return them. Blueboar (talk) 18:25, 7 June 2013 (UTC)

Providing useful information is only half our mission. The other half is providing useful information which can be proven to be true. I would hope that Kww hasn't blocked editors for merely deleting unsourced information without other indicators of cluelessness. I've made this argument previously, but it bears repeating: We all have a responsibility, policy or no, to improve the encyclopedia. The best practice is to provide sources when you find unsourced information, but it's not a requirement. An editor who continuously, habitually, and systematically removes unsourced information without looking for sources needs to be made aware that practice is not acceptable, first by warnings, then by blocks, and then by a topic or site ban as needed. His or her sin is not, however, the fact that they've removed unsourced information but that they've shown themselves to have an agenda or interest which is not in keeping with the best interests of the encyclopedia. It's not because they're doing bad things or because they're not interested in improving the encyclopedia, but because they're an editor who isn't interested in improving the encyclopedia enough. Regards, TransporterMan (TALK) 18:41, 7 June 2013 (UTC)

You read me backwards, TransporterMan. I've blocked for repeatedly adding unsourced information.—Kww(talk) 18:59, 7 June 2013 (UTC)

I most certainly did. My apologies. Best regards, TransporterMan (TALK) 19:02, 7 June 2013 (UTC)

I'm not kidding you, Blueboar, I mean it. The key phrase in Kww's post was "persists in". An editor who adds uncited statistics is something WP:V should deal with (and does, by means of WP:BURDEN among other things). An editor who persists in adding uncited statistics is the kind of conduct issue that (rightly) leads to blocks, and it's something that WP:V should not deal with.—S Marshall T/C 18:52, 7 June 2013 (UTC)

Oh... I see what you are saying (phew)... yeah, we were talking about two different things... I was looking at it from the perspective of a regular editor (not an admin) who might come upon this behavior at an individual article ... where the appropriate action is to remove the unsourced information and insist on a citation per Burden ... but the persistent refusal across multiple articles is a conduct issue that goes beyond our Verifiability Policy. Got it. Blueboar (talk) 19:13, 7 June 2013 (UTC)

How would I respond to the editor that is adding statistics to articles without citation? I would point them to this policy and tell them to cite it. But if the statistics are relevant and easily verifiable, I would recognize that the article would be better with them then without, and I would plead and implore and beg (or provide the cite). But if the person persists, I would report them for not heeding either my good advice, or my pleas, and for being tendentious and unreasonable in not doing so. Alanscottwalker (talk) 22:04, 7 June 2013 (UTC)

The above discussion has reminded me of an issue that I have been meaning to raise for a while. Shouldn't we say something about when and how to use primary sources? A lot of new editors think that we are never allowed to use primary sources. I think we should include something to help correct that common misunderstanding. We are, after all, allowed to cite primary sources in certain circumstances. Indeed, sometimes a primary source is actually the most reliable and accurate source that exists... and thus should be used.
I fully understand that primary sources can be misused (this has ties to both NOR and NPOV), and so we would have to word what we say about using primary sources very carefully... but surely we can say something about them. Blueboar (talk) 21:46, 21 May 2013 (UTC)

How about

The use of primary sources inside of articles is permitted. However, given that articles should be based on independent, third-party sources with a reputation for fact-checking and accuracy, the amount of material in an article derived from primary sources (or secondary sources that are not independent of the subject) must always be substantially less than the amount derived from independent, third-party sources.

—Kww(talk) 21:56, 21 May 2013 (UTC)

Not perfect, but it gives us something to start with. I suspect that we will need a lot more discussion before we actually add anything to the policy. Blueboar (talk) 22:08, 21 May 2013 (UTC)

Flawed, actually. It is possible to write an article that is 100% based on independent, third-party sources and still 100% primary sources. WP:Secondary does not mean independent. Secondary sources are good for analytical and synthetic work, such as comparing and contrasting two artists' style. "Jef Mallett uses a variety of stroke widths" is a primary statement. "Jef Mallett's drawing style is reminiscent of Bill Watterson's" is a secondary statement. Both of these could be independent, third-party sources.

Sometimes it's possible to write an entire article without needing much, or possibly anything, in the way of secondary sources. We have a number of medium-quality stubs about experimental drugs that are sourced entirely to primary sources (peer-reviewed scientific papers reporting original experimental results), and the community has generally not complained about them.

If you want to work from this, then you need to decide whether your concern here is primary/secondary or independence/affiliation. Then you can say, "The amount of material derived from primary sources (or sources closely affiliated with the subject, if your concern is independence/affiliation) should be substantially less than the amount of material derived from secondary sources (or independent sources, if your concern is independence/affiliation)". WhatamIdoing (talk) 22:29, 21 May 2013 (UTC)

• My effort: You may use primary sources in articles. Primary sources are appropriate for (1) uncontroversial facts and technical details, and (2) views and opinions with in-text attribution.¹ The majority of article content still needs to be based on secondary sources.

  ¹When using primary sources for views and opinions, the in-text attribution is essential. Primary sources cannot be used to support statements in the form "X is Y". They can only be used to support statements in the form "Person Z believes that X is Y."

  This may be over-detailed?—S Marshall T/C 22:12, 21 May 2013 (UTC)

S Marshal... I get what you are trying to say, but what you actually said is wrong... Primary sources can (sometimes) be used to support statements in the form or "X is Y"... Consider the following statement and citation: "Thorin Okenshield is a character in JRR Tolkin's book The Hobbit <ref - The Hobbit>". This statement is an "X is Y" statement. It is verifiable by the book itself. We would not need to attribute this statement in the form "Person Z believes that Thorin is a character in The Hobbit". Blueboar (talk) 00:49, 22 May 2013 (UTC)

This is why I prefaced that section with When using primary sources for views and opinions. In a fictional context it's not an "opinion"; it comes under the first limb of "uncontroversial facts and technical details". If I used a primary source to say that "JRR Tolkein was the greatest writer of our age", then that would be unacceptable because it's a view or opinion, so any statement of that kind would have to be in the form "Source X believes that JRR Tolkein was the greatest writer of our age". I suspect the whole thing's moot anyway because of the emerging consensus below to keep primary source discussion within the realm of NOR.—S Marshall T/C 07:45, 22 May 2013 (UTC)

• I don't think that we do ourselves any favors by getting into the details about when to WP:USEPRIMARY sources and when to avoid them in this policy. PSTS is at NOR, and IMO should not be scattered around multiple policy pages. WhatamIdoing (talk) 22:25, 21 May 2013 (UTC)

  It neglects "independent" as being one of the criteria, and it's actually the most important one. The last section would need to be "..the majority of article content still needs to be based on independent, third-party sources with a reputation for fact-checking and accuracy." —Kww(talk) 22:29, 21 May 2013 (UTC)

  Independence has nothing to do with the primary/secondary/tertiary axis. The opposite of a primary source is a secondary source; the opposite of a primary source is not an independent source. You can have an independent, third-party primary source (most newspaper articles, for example) and you can have non-independent, affiliated secondary sources (e.g., books that summarize traditional uses for herbal medicines that the authors or publishers "just happen to" sell). WhatamIdoing (talk) 22:33, 21 May 2013 (UTC)

• Agree generally with WhatamIdoing. Let's keep that headache contained, although perhaps a more prominent pointer to where to find that could be placed somewhere here. I also do not think the "amount test" explicated above would be universally applicable to all articles or particularly understandable in practice.-- Alanscottwalker (talk) 22:38, 21 May 2013 (UTC)

• Agree with WhatamIdoing. An appropriate mix of primary sources with secondary sources is a WP:NOR issue, not a WP:V issue. If you want to intertwine WP:V and WP:NOR (a good idea), then see Wikipedia:Attribution. --SmokeyJoe (talk) 00:28, 22 May 2013 (UTC)
• Agree with WhatamIdoing. North8000 (talk) 00:52, 22 May 2013 (UTC)

I think that User:WhatamIdoing is using a different definition for “primary source” than what I’ve understood it to mean in WP context. She says that most newspaper articles are primary sources, but I always thought they were secondary sources because they publish information (e.g., quotes or accounts) from primary sources. Am I alone in this? —Frungi (talk) 02:21, 22 May 2013 (UTC)

See WP:PRIMARYNEWS. It's not enough for it to be barely second-hand. Secondary sources provide not only greater distance than interviewing the eyewitnesses while the police are still on the scene, but also analysis or what you might call other intellectual work. WhatamIdoing (talk) 02:31, 22 May 2013 (UTC)

Yes, Frungi, you are thinking of newspapers as secondary sources correctly--that is the common, understood meaning of secondary sourcing in the Wikipedia context. The fact that WhatamIdoing is using a different definition, one that I think is incorrect and unhelpful, doesn't take away from his point that independence is (sorry) independent of secondariness. That is, regardless of how we define 'primary' and 'secondary', it's possible to have both primary and secondary sources that are independent and non-independent. Jclemens (talk) 04:16, 22 May 2013 (UTC)

The editorial pages in the middle of a newspaper tend to be secondary sources, commenting on topical events. The front pages for news reports tend to be primary sources, accurate repeating of factual information without commentary. Newspapers sometimes contain "stories". Stories are likely to be secondary source material, while news reports are primary source material. Newspaper content spans the line between primary source and secondary source material, but it is not correct to say that most newspaper articles are secondary sources. Mere repetition or republication does not turn primary source material into secondary source material. See primary source and secondary source and the references contained in these articles. --SmokeyJoe (talk) 04:50, 22 May 2013 (UTC)

Editorial pages are opinion pieces, and not sources of factual information, and thus are not appropriate for sources of such for Wikipedia articles. In limited cases, editorials may be useful for direct quotes and attribution of specific statements to the author of the editorial itself, but little else. Newspaper reporting itself is a bad example of the difference between primary and secondary sources, as it doesn't really fit well on the primary-secondary-tertiary axis. At best, it's somewhere between a primary and secondary source.

• Primary sources are uncritical collections of data, basically. Census data, weather reports, experimental results as published in peer reviewed journals, etc. are all primary sources. Assuming reliability, primary sources are useful in reporting the actual data itself, but not useful in providing analysis, drawing conclusions, or granting significance or relevance to the data. Primary sources are data, and that's probably the most concise way to think of it.
• Secondary sources use the data from primary sources to draw conclusions, and provide context, significance, and relevance to that data. Books are great secondary sources, insofar as the books will collect the data, and present it in a form that gives it an overall narrative and shows how the data fits into the corpus of knowledge by providing it with context, though there are secondary sources in all sorts of media. Secondary sources provide context, which is why they are so important for Wikipedia because it allows us to incorporate the information without having to figure out for ourselves where it fits in.

The deal with newspaper reporting is that it can be either a primary source or a secondary source, or it can sometimes stradle the distinction between the two in the same article. The two concepts are not, of course, a binary option. For example, consider this news article about the recent tornado in Oklahoma: here While a lot of it is primary in nature (raw data about the number of people who died, wind speeds and the strength of the storm quotes from people who witnessed the event, etc.), there's stuff in there that falls under the "secondary source" type material, for example at several points it draws parallels to earlier tornadoes in the area. That's not just raw data, that's historical context, which is a "secondary" type thing. this article on the same storm is primarily of a secondary source tone. It discusses the storm in historical context and provides analysis as to the conditions that created the storm, and why it formed and behaved as it did. That's all secondary source roles. So we can't make a blanket statement like "If it's in a newspaper, its a primary source" Not necessarily. Each source needs to be looked at individually and assessed, with the understanding that the "primary-secondary" rubric is also not a binary choice, but a continuum.

The most important thing to remember, rather than trying to make the decision as to whether a source is primary or secondary, is that Wikipedia text cannot go beyond what sources say, and that we cannot make decisions about relevance and context and stuff like that, if the sources itself don't. So, if Wikipedia text is merely reporting data, then the source cited for that bit of text only needs do the same. If Wikipedia text is providing historical context, analysis, or drawing conclusions about something, the sources cited to support it better do the exact same thing. --Jayron32 05:15, 22 May 2013 (UTC)

• Yes The newspaper issue is nuanced. Part of that comes from the academic field of history, where newspaper accounts of a particular time in the past are generally referred to as "primary sources" because they are accumulated into secondary sources to make generalized statements about that time period. Moreover, a reporter who witnesses the immediate aftermath of an accident, for example, is in many ways a primary source about that accident. However, a reporter who reports about a set of government documents, for example, and provides context to them is not the primary source, the government documents are, and in some cases the government reports may be secondary sources about other things. Alanscottwalker (talk) 09:25, 22 May 2013 (UTC)

• Oppose addressing the primary source issue here at V, except to link to the proper policy. It's not WP:USEPRIMARY, but instead WP:PRIMARY, that sets the policy on that issue (and it does refer to the independence standard). By addressing it here all we're doing is taking the chance that those two policies will get out of sync and either appear to, or in fact, contradict one another. I basically agree with Jayron32's analysis of the newspaper issue, but think that it can be made even more simple than that: When newspapers report things which they observe themselves, then they're primary; when they get it from a third party, then they're secondary. If a newspaper reports a death count from a disaster, it's a primary source if the newspaper's reporters counted the bodies, it's a secondary source if they got the numbers from the medical examiner, first responders, or other third parties, even if they added up multiple sources to get to their total. If they report on a tornado, it's a primary source if they're only reporting what their reporter saw with his own eyes; it's a secondary source if they got the information from the weather bureau or from citizen eyewitnesses. (In many cases it won't be possible to determine how they got the information; in those cases, in my opinion, they should be presumed to be secondary, but that's another issue for another day.) Regards, TransporterMan (TALK) 14:25, 22 May 2013 (UTC)

What about something as simple as: "While in certain circumstances it can be appropriate to cite a primary source to verify information (see WP:PSTS for more), articles as a whole should primarily be based on reliable secondary sources." I think something like this would clarify the situation, and resolve the "primary always = bad" misinterpretation that I am concerned about. Blueboar (talk) 15:14, 22 May 2013 (UTC)

• Newspapers as primary sources: Quite shocked to see newspapers cited as a primary source, and relieved to see I'm not alone in disagreeing with that reading of then policy. One good reason we use primary sources with care is because they have no oversight. A newspaper is a forum for publishing what can be primary information, but the fact that information is published with oversight makes most newspapers not only secondary sources but often reliable secondary sources.(olive (talk) 19:24, 22 May 2013 (UTC))

  The prepare to be shocked some more, because that's what the published reliable sources say: "Primary sources may include newspaper articles, letters, diaries, interviews, laws, reports of government commissions, and many other types of documents."[4] See Wikipedia:Identifying and using primary and secondary sources#cite_note-1 for more. This is a common definition. There are minority definitions (e.g., Willie Thompson defines "secondary" as anything already published) but this is actually the common definition for news articles (NB: not analytical pieces that happen to appear in a newspaper). WhatamIdoing (talk) 23:23, 29 May 2013 (UTC)

It may sometimes in some situations be a useful rule of thumb -- but your comment minimizes the may you quote too much. Alanscottwalker (talk) 18:11, 8 June 2013 (UTC)

It is not necessarily the case that "primary sources...have no oversight", as Littleolive oil writes above. As an example, a report of an experiment reported in a scholarly journal is reviewed for importance by the editors, and for content, to some extent, by the peer reviewers. But of course, some regard everything that appears in a scholarly journal as a secondary source, which is still another reason not to discourage primary sources too much, because some reading the policy will think we are discouraging what the reader thinks as a primary source, but which the editor who inserted the discouraging words into the policy thinks is a secondary source. And since no one is in charge of the English language, there is no way to decide who is right. Jc3s5h (talk) 23:33, 29 May 2013 (UTC)

@Blueboar. I'd reverse your suggestion actually because as is, it still opens the door a crack for misunderstanding that primary sources under certain conditions can be acceptable sources. I revised your version a bit, below.

"While articles as a whole should be based on reliable secondary sources, there are circumstances where citing a primary source is an acceptable means to establish that content is verifiable (see WP:PSTS for more),(olive (talk) 19:24, 22 May 2013 (UTC))

That would be fine with me. Blueboar (talk) 16:22, 22 May 2013 (UTC)

Do we really need to use sources to define the terms “primary/secondary sources”, as User:WhatamIdoing did above? And is it relevant to use the academic meanings—is Wikipedia an academic institution? It’s my understanding that Wikipedia-namespace pages do not need sources, and that Wikipedia internally defines these terms on such pages. —Frungi (talk) 23:46, 29 May 2013 (UTC)

The terms "primary" and "secondary" are too widely used in policies, guidelines, and talk page discussions to expect editors to figure which page contains the official Wikipedia definition and refer back to make sure they understand the official Wikipedia definition before they ever use those terms in a discussion, or ever make an edit to an article while relying on the definition. The only practical thing to do is write policies and guidelines in a flexible way that takes into account the nebulous nature of "primary" and "secondary". Jc3s5h (talk) 00:09, 30 May 2013 (UTC)

Or to base our definitions on someone else’s, which is the practice I’m questioning. —Frungi (talk) 00:37, 30 May 2013 (UTC)

• There is some sentiment here that articles should be based on secondary sources.  WP:GNG states, "Availability of secondary sources covering the subject is a good test for notability."  But WP:GNG is only one path to defining wp:notability.  So while WP:GNG class articles might have secondary sources, what matters in writing articles is not primary/secondary but the concept of wp:verifiability.  Unscintillating (talk) 17:02, 8 June 2013 (UTC)

There seems to be a minor edit war over the phrase, "rumor and personal opinion":

• "rumor and personal opinion" => "gossip or personal opinion"
• "gossip or personal opinion" => "gossip, rumor or personal opinion"
• "gossip, rumor or personal opinion" => "unsubstantiated gossip or personal opinion"
• "unsubstantiated gossip or personal opinion" => "gossip, rumor or personal opinion"

I'm creating this section so we can discuss these changes. FWIW, I like the addition of "unsubstantiated". A Quest For Knowledge (talk) 17:31, 16 June 2013 (UTC)

Don't mind the addition of "gossip" or "unsubstantiated gossip"... my concern was the removal of the word "rumor" (which I think is subtly different than gossip). But I am done edit warring over it. Blueboar (talk) 18:22, 16 June 2013 (UTC)

In that case, I'm going to add "unsubstantiated" back in. If anyone disagrees, please post something on my talk page and I'll self-revert. A Quest For Knowledge (talk) 02:07, 18 June 2013 (UTC)

I agree that there is a subtle difference between "rumour" and "gossip". The Concise Oxford Dictionary has the following definitions:

• Rumour: "Report or hearsay of unsubstantiated accuracy"
• Gossip: "groundless rumours"

Given these definitions, the word "rumour" and "gossip" should be retained, but the word "unsubstantiated" is redundant.Martinvl (talk) 06:51, 18 June 2013 (UTC)

Just a side note (without being up to speed on the overall situation) "gossip" is generally pejorative, the other terms aren't. I think that dictionary.com says it better *Gossip: idle talk or rumor, especially about the personal or private affairs of others North8000 (talk) 12:05, 18 June 2013 (UTC)

Since April 2011, WP:V (at WP:SOURCES) has contained this:

Source material must have been published (made available to the public in some form).

This implies that our policy definition of the word published is "made available to the public in some form".

So, for instance, the inscription on a gravestone was "published" when the stone was erected. And every publicly-accessible document in an archive (such as The National Archives (United Kingdom), the County record offices and equivalent resources in other countries) was "published" when it appeared in the catalogue.

Is this the intention?  —SMALLJIM  21:04, 18 June 2013 (UTC)

I sincerely hope not. AndyTheGrump (talk) 21:10, 18 June 2013 (UTC)

I would say the part about records in accessible archives is correct. I'm not so sure about gravestones, but plaques in national monuments and the like would be published. But the WP:Identifying reliable sources guideline addresses whether a published source is reliable or not. Many sources in archives are effectively self-published, in that any record will be included if someone appears at the counter and pays the recording fee. Jc3s5h (talk) 21:50, 18 June 2013 (UTC)

I agree. While "published" contains an element of "made public" (don't these words share a common root?), historically it has implied more than that. The Web has confused matters, in that merely posting something is taken as a kind of publication, so perhaps a closer consideration would be useful. ~ J. Johnson (JJ) (talk) 21:57, 18 June 2013 (UTC)

Anything available openly on the internet is definitely "published" as far as WP is concerned – we have the concepts of "reliable", "independent/3rd-party" and "secondary source" to keep the bad stuff out. But I'd rather not get involved in those aspects here – let's keep it simple.

Anyone can go to the churchyard where Ivor Deadbody is buried and can read the inscription on his gravestone, and anyone can go to the government-funded archive where his diary is kept and can read what he wrote in it. Both items satisfy the condition of being "made available to the public in some form", so the way WP:V is currently worded, we can certainly make a case that they are "published", which allows us to make use of them in our articles. It's only limited use because they're primary sources, but if they're "unpublished" we can't use them at all, so this is of some significance. The policy should be clarified so that the intention (whatever it is – I don't know) is clear.  —SMALLJIM  23:42, 18 June 2013 (UTC)

The "made available to the public" was added specifically to allow publicly accessible archives. These are cateloged and thus "published". Blueboar (talk) 01:59, 19 June 2013 (UTC)

OK, thanks. I assume you're happy about the gravestone?  —SMALLJIM  11:51, 19 June 2013 (UTC)

One place that "publication" is defined formally is in copyright law. Interestingly, the US definition as presented in publication seems to include something that an archive makes available but exclude a gravestone. A gravestone, plaque, or similar, seems to fit the meaning of "public display", which is not publication. (I'm aware of the danger of interpreting a law without consulting case-law.) Zero^talk 13:03, 19 June 2013 (UTC)

Many web pages, lacking any of the traditional hallmarks of publication, would seem to come under merely "public display".

Does "publicly available" fail if a document is available only to members of a private group? ~ J. Johnson (JJ) (talk) 19:00, 19 June 2013 (UTC)

Obviously. AndyTheGrump (talk) 19:07, 19 June 2013 (UTC)

Re the gravestones... yes... while not published in the formal sense, a gravestone from a public cemetery would meet our definition of "reliably published source" (made available to the public). An "unpublished" gravestone (under our definition) would be one located in a private mausoleum that members of the public have no access to. Of course any public gravestone would be limited in how we could use it. We would consider any gravestone to be a primary source... and there are limitations that apply to all primary sources... but, keeping those limitations in mind they can be reliable.

Oh... One important caveat to the private group idea... having to pay for access does not make something "restricted to a private group"... as long as access is granted to any member of the general public who is willing to pay, it counts as "made available to the public". Just want that to be clear. Blueboar (talk) 19:56, 19 June 2013 (UTC)

If anyone can buy a membership, yes. But what about a whites-only mausoleum? ~ J. Johnson (JJ) (talk) 20:15, 19 June 2013 (UTC)

Thanks, Blueboar. I'm happy to go along with that, even though you've stuck that word "reliable" in there a couple of times :) Since those who expressed reservations here haven't followed them up, can we assume that there's a rough consensus that you're right, and therefore make the policy clearer by moving that text out of the parentheses, perhaps with a footnote. Something like:

Source material must have been published, the definition of which for our purposes is "made available to the public in some form".^[1]

1. This includes material such as documents in publicly-accessible archives, inscriptions on monuments, gravestones, etc., that are available for anyone to see.

 —SMALLJIM  12:15, 21 June 2013 (UTC)

No further comments, so I've made the above change. Happy to discuss further.  —SMALLJIM  13:26, 24 June 2013 (UTC)

This policy talks a lot about adding citations and removing unsourced text. Is there a policy or guideline about removing citations? I am specifically asking about citations, such as in-line citations, not links in an "external links" section. My understanding is that the Wikipedia:External links guideline does not apply to such citations, but this wouldn't be the first time I was confused by the WP:PAYWALL policy and the WP:ELREG guideline.

Often people say that a citation or URL was or should be removed for various reasons, such as:

• "subscription service so url removed" (Talk:Trevor Pinnock/GA1)
• "subscription only link, and therefore I would suggest should be replaced" (Talk:Electronic symbol)
• "The link to ... is subscription-only content. Replaced with a link to" (Talk:Ludwig van Beethoven/Archive 4)
• "The link is paywalled, but it's not clear to me that's disallowed in WP references." (Talk:Toyota Prius/Archive 1)
• "Whoever is using The Times online articles as sources will need to find something else. Their articles require subscription fees" (Talk:Afghan War documents leak/Archive 1)
• "I deleted the following link because it's subscription-only and the subscription costs $80. ... WP articles aren't supposed to link to subscription-only destinations" (Talk:New_Criticism)

Is there some policy or guideline that gives some clues as to what kinds of in-line citations definitely should be deleted, and what other kinds of in-line citations definitely should not be deleted? --DavidCary (talk) 03:11, 21 June 2013 (UTC)

If the citations that are behind paywalls are being replaced with more open-access citations that support exactly the same information. But removing citations without replacement just because they are paywall-sources is inappropriate. We should try to replace these but that's not always an option. --MASEM (t) 03:16, 21 June 2013 (UTC)

A source does not become unsuitable just because it is behind a paywall but it might be preferable to replace a paywall source with an equivalent free one, because that makes it easier for editors to verify that the source says what is claimed. I don't think it is right to remove a citation just because it is behind a paywall. Reyk _YO! 03:20, 21 June 2013 (UTC)

A source certainly should not be removed simply because the editor without access to the content has to pay for the content...unless a free, reliable alternative is provided in its place. If sources that require people to pay weren't allowed, that would generally place our WP:MED articles that require WP:MEDRS-compliant sources in a bad place. Flyer22 (talk) 03:36, 21 June 2013 (UTC)

Most certainly citations should not be removed solely because they are behind a paywall. While a "free" source (or link) might augment the paywalled version, I would not replace such a source as they are often the authoritative source. ~ J. Johnson (JJ) (talk) 00:01, 22 June 2013 (UTC)

In short, the people who provided all the quotes you listed, DavidCary, don't know what they are talking about. There is no requirement for the sources to an article to be free or even easy to acquire. The sources doesn't even need to be online at all. Anyone who states the contrary is simply mistaken or making things up. Now, if it turns out the exact same source is available from a paywalled website and a free website, then of course the free website is preferred unless it is a copyright violation, or virus-ridden, or some other trivial example of WP:ELNO. A paywalled source may also be removed if it is completely superfluous to the other sources given for the same statements. Someguy1221 (talk) 00:34, 22 June 2013 (UTC)

"Exact same source" can be problematical, as copies (such as PDFs of journal articles) can be altered. Even the source can subsequently corrected, and any discrepancies can be resolved only by reference to the authoritative source. If there is a doi link, no problem, and the convenience link can be to the more available source. But in no case should an authoritative source be deemed "completely superfluous". ~ J. Johnson (JJ) (talk) 20:04, 22 June 2013 (UTC)

I'd like to second the clarification made by Alanscottwalker earlier today in this edit. I think that's a useful clarification to the existing text which was a bit vague. Regards, TransporterMan (TALK) 14:53, 24 June 2013 (UTC)

• I could give multiple examples of references and citations removed before and during AfDs by editors !voting delete.  Unscintillating (talk) 15:18, 24 June 2013 (UTC)

Wikipedia_talk:Verifiability/Archive_44#Sourcing for summarizing articles seems to have the consensus that articles require sources, even when the sources are immediately available in blue links.  Yet the common practice seems to be that lists of notable people are maintained without sources.  This edit summary asks the question, "don't we need sources for this kind of thing? Especially if living people are involved"?  I'm not aware that "common practice" is a policy-based counter-argument.  Where is the policy-balance here?  Thanks, Unscintillating (talk) 12:11, 1 July 2013 (UTC)

This is where common sense applies. If all those individual articles meet Wikipedia's notability requirements, then obviously sources exist. Deleting valid content that you know can be sourced is a bit WP:POINTY. If someone wants the sources from the individual articles to be copied to the list, then they can do so. Deleting the list for not having sources when you have sources is just plain lazy. A Quest For Knowledge (talk) 12:32, 1 July 2013 (UTC)

The Notability requirements for lists of people are tricky... See WP:Stand alone lists, for more. In general, the consensus is that we can leave it to the bio article to source why each individual who is listed is notable (on his/her own)... however, at the list article we need to establish that the grouping is notable (so... for an article entitled List of Xs, we need to establish that being an X is notable).

We also need sources to establish that each person listed belongs in the list article.

So... In order to sustain a List of Foos article, we would not have to establish the Notability of each person listed (we can leave that to the linked bio article)... but the list article would have to establish (though sources) that a) being a Foo is considered notable, and then b) that each person listed was indeed a Foo.

To give a more specific example... consider our List of Freemasons (A - D) article ... in the intro paragraphs we establish (through sources) that being a Freemason is a notable thing to be. Then, after each entry, we cite a source to establish that the person listed was/is in fact a Freemason. Blueboar (talk) 13:17, 1 July 2013 (UTC)

I remember a similar discussion coming up before, and while in general, if the list has been pared to notable (blue-linked) people, and the categorizing is obvious/non-contentious (eg "List of people from X"), then there's no real need to reference that, as the references should be in the linked articles. But as soon as the subject categorize is something of contention (eg "List of LGBT supporters"), then even though there may be references, sourcing to justify their inclusion on the list should be included. --MASEM (t) 13:23, 1 July 2013 (UTC)

Blue links aren't citations. Blue links are the equivalent of directly citing the Wikipedia article, and we would reject that on its face. It's a reasonable expectation that at least one legitimate inline citation be provided for each element of a list, and there's no reason not to apply WP:BURDEN to such lists.—Kww(talk) 16:24, 1 July 2013 (UTC)

But, on the other hand, if you follow a tenant of WP:V where sources have had to be identified but not necessary in the article, a blue-link to a WP article that contains those sources is sufficient. Mind you, I do agree that one should have at least one inline cite for any such list, and put a lot of weight in having those cites when the list is based on something that is potentially controversial, but a list without references but only containing blue-links that include the proper references meets the spirit of what WP:V says. --MASEM (t) 16:39, 1 July 2013 (UTC)

Example: One often sees these blue-linked lists in "notable alumni" but there will be no cite that says alumni matter. Alanscottwalker (talk) 16:44, 1 July 2013 (UTC)

I can't see how including someone on a list based solely on a blue link can ever be justifiable - it is in essence citing Wikipedia as a source. It also makes checking eligibility much harder when new names are added (something that seems to be a favourite of new contributors, many of whom don't understand sourcing policy). If a list is worth having, it is worth sourcing properly. WP:BURDEN applies. AndyTheGrump (talk) 16:50, 1 July 2013 (UTC)

I never intended to catalyse such a serious debate!

Personally, I think that edit might have gone a little too far. I feel torn between Masem's position and Kww's position; on the one hand, WP:BURDEN is important, especially for BLPs, and simply linking to another article isn't a citation per se; but on the other hand we only have finite resources, so it may be unreasonable to demand citations for uncontroversial lists where target articles usually have their own sources which would support list membership.

Ultimately, we don't do policy for the sake of policy; deleting content that is "unsourced" is a way of reducing the risk that the encyclopædia might say something untrue. I tend to edit other topics where some editors like to add untrue things to lists, hence it's appropriate to be a little more aggressive with the delete key. On the other hand, for something like a list of notable people from a town - the list is uncontroversial, the target articles have their own sources (at one remove) which are likely to support list membership, and there's less reason to suspect other editors would be prone to making up their own list members - perhaps we have not yet reached WP:V nirvana where 100% inline sourcing is a realistic expectation. I'm happy to go along with whatever the community decides on this point.

I respect A Quest For Knowledge. When restoring the list, their edit summary said "sources should be in the linked articles". This is quite true, and I wouldn't revert their edit whilst citing a literal reading of WP:BURDEN - that would be supremely WP:POINTy. However, I have dealt with other articles in the past where editors said similar things whilst reinserting lists, but the target articles did not actually have sources supporting list membership. "But there's a source in the other article..." can be a red flag, not over some pedantic interpretation of WP:V, but warning that an editor is actually unable to provide sources which support the text they're adding. bobrayner (talk) 17:01, 1 July 2013 (UTC)

Sure burden applies to matters likely to be challanged but many of these things will not likely be per WP:EDITCONSENSUS. What Bobraynor says about deletion is true but one side of the coin, deleting relevant verifiable but not cited info leads to less informative articles, and may sometimes make them misleading (although probably not in these kinds of lists - ie. lists that cannot be reasonably taken as all inclusive or lists that include the header that it is possibly incomplete). Alanscottwalker (talk) 17:14, 1 July 2013 (UTC)

Just as a comment from dealing with non-BLP list articles that hvae similar limitations (eg List of Internet phenomena), I have found that editing the edit notice on the main page, and adding talk page headers to explain in both cases that any added entry must include a reliable source helps to minimize additions from newer editors for that page, or at least guide them to adding something with a source. You'll still get a few edit wars with users that don't bother to read, but that least helps. I would certainly say that if there is a list involving BLP where the inclusion requirement is something that is not immediately obvious, that similar steps can be done to inform editors what are the requirements for both inclusion of a person on the list, and for the entry about that person to have at least one reliable source.
Secondly, I would also note that for any of these types of lists, if they are to get to FEatured status, there will need to be references to justify every entry on the list, so it does become a not-a-deadline issue to add them over time. But there is a need to stress additions for any BLP list that can be a contentious matter. --MASEM (t) 17:26, 1 July 2013 (UTC)

• I'm seeing more about what is going on here.  The list in the example is called, "==Notable natives and residents==".  Wikipedia editors easily confuse the word "notable" with "wp:notable" and are likely to incorrectly conclude that a blue link proves that the topic is "notable".  This idea is reinforced in the convenience with which such a list can be verified.  But the policy basis for inclusion in an article is WP:DUE "prominence", for which "notable" is often loosely substituted.  Since "notable" is not equal to "wp:notable", and neither are equal to WP:DUE "prominent"; maybe we shouldn't be using the word "notable" in the section header.  Unscintillating (talk) 18:09, 1 July 2013 (UTC)
• Something else to consider.. articles can be edited and even substantially rewritten. Information can be reworded and even omitted. Sources can be replaced with other sources. One reason to repeat citations at a list article is that there is the real possibility that, over time, the bio article may be edited to such an extent that it no longer contains the citation that supports the information at the list article. Blueboar (talk) 19:12, 1 July 2013 (UTC)
  • Yes, but in this particular case, the editor in question could have added the cites, but didn't. A Quest For Knowledge (talk) 19:21, 1 July 2013 (UTC)
    • That may not a helpful approach. If we're going to be awkward about it, the policy says "The burden of evidence lies with the editor who adds or restores material". In this particular case, that's you. Blaming me for failing to provide sources for content that you reinserted is not the best way to nurture goodwill. The section that you restored still has the unreferenced-section tag dating back to March 2010. bobrayner (talk) 19:51, 1 July 2013 (UTC)
      • But the burden is met "once sources that an editor believes in good faith to be sufficient have been provided". BURDEN is written so that editors must collaborate to achieve consensus, not to provide an automatic "win" for people wanting to remove poorly sourced content. That's why, once a source supporting the content has been identified, BURDEN mandates that editors wanting to remove the material should instead include the reference themselves (or provide a specific reason why the provided reference isn't enough to support verifiability). Diego (talk) 22:35, 1 July 2013 (UTC)
        • Burden mandates no such thing, and all attempts to make it do so have been rejected. The burden falls on the editor including the material, and everything aimed at the removing editor is solely a suggestion.—Kww(talk) 05:42, 2 July 2013 (UTC)
          • That's a bizarre statement considering Diego is quoting directly from WP:V. A Quest For Knowledge (talk) 12:13, 2 July 2013 (UTC)
            • That's a footnote on the topic of in-line citations, not a section on vague hand-waving at blue-links and tags.—Kww(talk) 15:20, 2 July 2013 (UTC)

• An alternative to supporting WP:V is that we define in a guideline somewhere that lists of "notable" people are recognized as being lists of "wp:notable" people.  Unscintillating (talk) 20:10, 1 July 2013 (UTC)

• Note - concerning the list of notable people in the Elizabethtown, Kentucky article... What is needed are sources that establish that each of the people listed are indeed from Elizabethtown, Kentucky. Now, in some cases the needed source can be found at the person's bio article... and it would be POINTY to remove them from the town list when all it would take to fix the problem is a simple cut and paste. Let's not argue about who's responsibility it is to do this... it's everyone's responsibility. However, in the other cases, the information is not cited in either the town article's list or the bio article on the person. This means that Wikipedia has no source anywhere that establishes that the person is from Elizabethtown. I have marked those with a "citation needed" tag. If, after a reasonable search, no source can be found - it would not be POINTY to remove these people from the list. Blueboar (talk) 23:00, 1 July 2013 (UTC)

• Your research shows that there are legitimate sourcing concerns hidden behind the blue links.  Blue links cannot be used to verify such lists.  I've changed the name of the section from "Notable natives and residents" to "Prominent natives and residents".  Unscintillating (talk) 00:15, 2 July 2013 (UTC)

Such lists as "Prominent natives" (or notable) are frowned upon and I am unsure why this discussion even moved from list based articles to list based sections of articles. Those lists in articles are usually asked to be removed when going for GA status anyways. This is usually under the assumption that if the people in those lists cant be moved into the history section or some other section on economy or sports or landmarks then they probably aren't "notable" enough in that city's history in order to list them, for listing them would be for prideful reason such as "President Cleveland's father lived here!!!" along the lines of the (in)famous "George Washington slept here" tagline.Camelbinky (talk) 01:14, 2 July 2013 (UTC)

Adding additional citation-needed tags is a step forward, I suppose, but if the list as a whole was tagged as unsourced since 2010, at what point can we actually remove unsourced content (including claims about living people) rather than just tinker with tags and argue over reverts? If three years isn't long enough to wait for somebody else to provide sources, do we need four years or five? bobrayner (talk) 05:24, 2 July 2013 (UTC)

An important note... the the important thing isn't whether the content is sourced or unsourced... it's whether it is verifiable or unverifiable. If you think the unsourced information is probably verifiable... leave it in the article (and either find a source for it yourself, or tag it and leave it to someone else to find the source). If you think the unsourced information is probably unverifiable... take it out. Blueboar (talk) 12:07, 2 July 2013 (UTC)

Blueboar is correct. There's a huge difference between unverifiable and unverified. Unverifiable information should be removed. However, nobody has claimed that this content is unverifiable, and in fact, there doesn't seem to be any attempt to even examine whether this information is unverifiable. WP:V only requires sources for content that is challenged (or likely to be challenged). Since that isn't the case in this particular instance, see WP:PRESERVE. A Quest For Knowledge (talk) 12:46, 2 July 2013 (UTC)

Well, the flip side of WP:PRESERVE is WP:FIXTHEPROBLEM. In the case of the list at Elizabethtown, Kentucky, at least some of the material has been challenged. That's why I switched from a section tag to individual "citation needed" tags. We know that a source exists for the untagged entries ... those are verifiable and should be left in the article... But we don't know whether a source exists for those that now have the citation needed tags. By adding those tags, I have challenged whether their inclusion in the list is indeed verifiable ... so the clock is ticking on them.

That's where WP:BURDEN comes in. It is not the responsibility of those who question the verifiability of the material to prove the negative (ie to prove that the information isn't verifiable)... it's the responsibility of those who think the material is verifiable to prove the positive (by finding a source). Those who question the material should be understanding, and give others a reasonable amount of time to search for sources. How long we give them is a judgement call that depends on the nature of the material. In this case, the information is fairly innocuous, and so we can give people a goodly amount of time to search.

Those who think the material is verifiable, however, also need to be understanding, and now need to make finding the requested sources a priority. They also need to remember that removal is not permanent. They should not over-react if the tagged entries are removed. If the search for sources takes too long, they need to accept a temporary removal (and remember that removed material can easily be returned once the required source is located). Blueboar (talk) 12:07, 2 July 2013 (UTC)

OK... I have done my part, per WP:FIXTHEPROBLEM, by adding sources to the extent that I can... (it took me all of 15 minutes to cut and paste a few sources from the bio articles). It should be much clearer now which are actually being challenged and which are not. It is now the responsibility of those who wish to keep the "cn" tagged entries in the article to find sources. Blueboar (talk) 13:46, 2 July 2013 (UTC)

There has to be a reasonable expectation that the information isn't verifiable. We don't delete "Paris is in France" simply because it doesn't have a cite. Challenging for the sake of challenging isn't valid. In this particular case, I don't see anyone saying "Person A shouldn't be on the list because they're really from California, not Kentucky". As you know, a lot of Wikipedia isn't sourced (or properly sourced). If we went around deleting content for no other reason than not being sourced, half the encyclopedia would be gone. A Quest For Knowledge (talk) 14:09, 2 July 2013 (UTC)

I take exactly the opposite view... there has to be a reasonable expectation that the information is verifiable. For a statement such as "Paris is the capitol France" we have that expectation (we expect that it can be verified by looking at any up-to-date atlas)... but the same is not the case when claiming that a specific person lived in a specific town. That is something we expect to be sourced. If it isn't sourced, I think it is perfectly reasonable to challenge it. And, when challenged, it is up to the people who wish to add or keep it in the article to meet the challenge and find a source. That's what WP:BURDEN is all about. I don't have to prove the person isn't from the town... you have to prove that the person is from the town.

You are correct in saying that a lot of material in Wikipedia isn't properly sourced... It's a problem that needs to be fixed (not ignored). I think that most of that poorly sourced material should be removed. If it is worth returning, someone will eventually return it (properly sourced this time). That's called the editorial process - and in the long run, it makes for better articles, and a better encyclopedia. Blueboar (talk) 15:01, 2 July 2013 (UTC)

Having had to deal with a whole lot of lists of the type 'List of British Xians' (where 'X' is an ethnicity) in the past, I've seen far too many entries which were not only unsourced but just plain wrong to ever expect the entry to be verifiable. Fortunately, any reasonable interpretation of WP:BLPCAT suggests that such entries must be sourced, and consequently it is easy enough to spot new entries and source/remove them. With an unsourced list, there is no way to tell whether an entry has ever been checked, and it is an open invitation for intentionally misleading additions, along with entries based on vague recollections, assumptions and just plain guesswork. AndyTheGrump (talk) 15:24, 2 July 2013 (UTC)

@Blueboar:Considering I've never heard of this town or any of the people in this list, I have no reasonable expectation either way. But I do have to assume good faith in the editors who created the article. Had the deleting editor given a specific reason (beyond lack of cites) why they were deleting the list, that would have been a completely different situation. But we can't have editors going around deleting half the encyclopedia.

And we've had this discussion before. The way most editors deal with unsourced content is to draw a line in the sand between new and old content. When an editor adds new content to an article without a source, it usually gets reverted or at least a fact tag is added. For content that's been around for years, we don't delete it unless we have an actual concern about the content. A Quest For Knowledge (talk) 15:57, 2 July 2013 (UTC)

ah... I guess its time to play dueling acronyms... I like that game! I'll see your AGF and raise you one BLP... note that most of those now tagged as needing a citation are living sports figures. Adding the name of a favorite baseball or football star to the article on your home town strikes me as exactly the kind of silly prank that school kids would do. I suppose we could be more willing to assume good faith when it comes to the obscure historical residents, but for living people we demand sources to support the information. Blueboar (talk) 17:03, 2 July 2013 (UTC)

FYI - I've found sources for the remaining living people[5] except for one which I moved to the talk page.[6] A Quest For Knowledge (talk) 19:30, 2 July 2013 (UTC)

Is there not a case to create a single citation for everybody who has an article which explicitly states that they lived in the town (or were born there). For example, if I were to list Barack Obama as one of the notable people born in Honululu, it would be sufficient to use a citation which states "as per citation in Wikipedia biography". This citation would probably be shared by a number of people. In this way, every entry would have a citation, but the work done in creating those citations would be dramatically reduced. Martinvl (talk) 16:14, 2 July 2013 (UTC)

That is citing Wikipedia as a source - which we never do. And has already been pointed out, articles change, references can get deleted, sources rejected as unreliable - just because a biography once had a reference, there can be no assumption that it still does. AndyTheGrump (talk) 16:33, 2 July 2013 (UTC)

In this case, I actually checked the bio articles to see if there was a usable source... for the entries I have tagged and questioned there are no sources at the bio articles either. For all we know this is stuff the kids made up in school. Blueboar (talk) 17:03, 2 July 2013 (UTC)

The lists should not be existing anyways. Has anyone here actually put forth an article on a community for GA or FA status or even a peer review...? Combine the people into appropriate sections, sports figures in the sports section with a source, if they aren't notable to be where they would be put then they don't deserve mentioning at all. That is how Wikipedia articles on municipalities are done, why are bothering with how to source lists that aren't even supposed to exist? Discussions on these lists have been done ad nauseum else where and we've made that consensus many times.Camelbinky (talk) 17:28, 2 July 2013 (UTC)

• Although there was some initial confusion, this has not been a discussion about a stand-alone list.  For stand-alone lists, see Wikipedia:Manual of Style/Stand-alone lists#Lists of people, which has a shortcut WP:LISTPEOPLE.  FYI, Unscintillating (talk) 01:02, 5 July 2013 (UTC)

I will also note that in this case WP:BURDEN actually worked exactly as intended... Someone challenged the unsourced material... and editors who thought the material should be included went out, found sources to support inclusion, and added them (only one entry could not be sourced as living in the town, and that entry was removed). So now it is simply a matter of deciding whether having such a list in the article (at all) is a "good" or "bad" thing... is it useful information that should continue to be in the article, or is it mere trivia that should be omitted? That is an editorial decision that should be made on the article talk page with reference to other policies and guidelines... it isn't an issue for WP:V. I think we are done here. Blueboar (talk) 13:42, 5 July 2013 (UTC)

I draw people's attention to this discussion, where A Quest for Knowledge shows that he doesn't understand his obligations on this topic with regard to WP:BURDEN.—Kww(talk) 18:50, 2 July 2013 (UTC)

Above post is quite vindictive, I've read the linked discussion, and I agree with Quest's conclusion of Kww's baseless threats in that discussion. Given it has nothing to do with this section is it ok to strike or remove Kww's statement?Camelbinky (talk) 19:30, 2 July 2013 (UTC)

It's specifically about A Quest For Knowledge's behaviour with regards to this particular discussion, where he twice restored unsourced material after it was challenged. There's nothing baseless or vindictive in my comments.—Kww(talk) 19:38, 2 July 2013 (UTC)

It's a clear refusal to follow a core policy, but we need less drama here, not more. Given the choice between fixing the general case (ie. people disagree on the applicability of WP:BURDEN) that affects millions of pieces of content, or enforcement on one specific piece of content, I'd rather we focus on the former. Disagreements over policy and guidelines can become bitter and entrenched; let's try to mitigate that. bobrayner (talk) 13:56, 5 July 2013 (UTC)

Conduct issue, belongs on a conduct-related noticeboard rather than here. There's a relatively similar incident in my own recent editing history, where I felt (and feel) angry about being "forced" to provide a source before an editor would "permit" me to restore the obviously-verifiable consensus text (and the AfD closer was no bloody help either).

The general principle is that WP:BURDEN could potentially be used for griefing. In order to avoid that, where it's alleged that WP:BURDEN hasn't been used correctly, it's best to get a third opinion rather than for two users to lock horns in an unproductive way.—S Marshall T/C 15:54, 5 July 2013 (UTC)

I would say the best thing to do is not argue about it at all. If the material is really so verifiable, it should take little time and effort to find the requested source and slap it into the article. Calling in a third party simply continues the aggravation and debate... and will probably take a lot more time and energy than it would to simply supply the source. Let the idiot have his petty victory... just supply the source. Blueboar (talk) 02:12, 7 July 2013 (UTC)

There is at least one article I watch where I remove uncited entries in the list of examples on sight. They are inevitably drive-by edits by people who don't know what they're talking about. If there is any question as to whether the example is reasonable, I'll do a very cursory check; but usually it's not worth it to do so. If I didn't do this, the article would fill up with inaccurate examples. Now, for these residence examples I'm generally willing to take our own article as sufficient basis for inclusion; but if our article didn't mention such residence, I'd expect a citation and wouldn't feel bad about deleting the entry if it were lacking, since my assumption would be that the claim probably isn't true in that case. Mangoe (talk) 14:27, 8 July 2013 (UTC)

• I agree with the assessment that this discussion as initially stated does not belong on this page.  There is a general problem across Wikipedia that talk page themselves need talk pages, which is where this discussion IMO should be refactored.  Unscintillating (talk) 14:24, 6 July 2013 (UTC)
• On the one hand, we need more admins supporting our WP:V policy.  On the other, I think the policy issues here are not clear.  Unscintillating (talk) 14:24, 6 July 2013 (UTC)

(Ancient joke warning!) A guy goes into an old folk's home and goes into the sitting room to wait for his uncle. Twenty or so old guys are sitting around there and every now and then one of them calls out a number. "75" and every one laughs. "54" and everyone laughs. "8" and everyone laughs. The guy is perplexed, but his uncle comes in about that time. "What's going on with the numbers and the laughing?" the guy asks. "Oh," his uncle says, "all these guys have been telling the same jokes to one another for so long that they've given them numbers and just call out the number." The young guy thinks for a minute and yells out "65". Nobody laughs. "What's the matter," he asks, "isn't that a joke?" "Oh, it's a joke, but you told it wrong." Can't we just number the different arguments and responses which just keep getting endlessly made over, and over, and over again here and cut down on the discussion? Regards, TransporterMan (TALK) 17:33, 9 July 2013 (UTC)

I can see it now... "Number 42"^{[citation needed]} Blueboar (talk) 22:48, 9 July 2013 (UTC)

While looking for a preference to turn off the Visual Editor links, I found a gadget that allows you to add some dropdown boxes to the editing window to let you plug in quick edit summaries. One of them is "Removing unsourced content". In light of our new policies about what you have to say when you're deleting unsourced content, that ES is obviously inadequate. I've asked that it be changed to read, "Removing unsourced content, no published reliable source may exist, thus perhaps not verifiable" so that people don't step into a trap using the WP-provided tools. Regards, TransporterMan (TALK) 14:14, 10 July 2013 (UTC)

I fear that may be unhelpful. WP:V simply says "Any material that needs a source but does not have one may be removed"; let's not overcomplicate it. When you say "new policies", which new policy do you have in mind? bobrayner (talk) 02:22, 11 July 2013 (UTC)

Per WP:BURDEN: "When tagging or removing material for not having an inline citation, please state your concern that there may not be a published reliable source for the content, and therefore it may not be verifiable.^[1]"

It was added pursuant to this RFC which ended in October, 2012. Please don't interpret my request as an indication that I like this provision. To the absolute contrary. But if we've got it, we don't need to be leading people into violating it inadvertently. If someone were dragged before ANI for violating it, I'd be willing to argue that the "please" in the main section and the "advisable" in the footnote make this advisory/best practice rather than mandatory, but I rather suspect that there are others who might not feel that way. Regards, TransporterMan (TALK) 17:07, 11 July 2013 (UTC)

I don't think that we want to have an automatic summary making claims that the editors may not actually believe. People sometimes remove unsourced material even though they fully believe that sources exist. WhatamIdoing (talk) 00:55, 12 July 2013 (UTC)

If they do, then they have probably violated policy since the sentence just after the one quoted just above is, "If instead you think the material is verifiable, try to provide an inline citation yourself before considering whether to remove or tag it." Now, that is admittedly a violation which would be hard to prove at ANI since it would require proof of what was inside the editor's head. Moreover, since the "try to provide" links to WP:PRESERVE and WP:PRESERVE is advisory / best practice rather than mandatory (see the argument I just made a couple of minutes ago a couple of sections above), then arguably this provision is either not mandatory or it only requires you to do something that is mostly optional, so that would make a violation of this sentence even more difficult to sanction but hey, a violation is a violation and we don't want to promote it, do we? Regards, TransporterMan (TALK) 14:17, 12 July 2013 (UTC)

• Another point in the Elizabethtown, KY discussion is, what constitutes a challenge, [WP:CHALLENGE], for a list of prominent people?  This issue started with a section tag {{Unreferenced section|date=March 2010}}.  But looking at the diff just before the material was removed shows that the tag was falsifiable, the section had a reference.  That one reference would be a reason to remove the tag if the paragraph were prose.  Yet, everyone realizes almost without thinking that that one reference doesn't suffice, so why then does the one section tag suffice?  I would say that the answer is that it doesn't, that the remedy in this case is to remove the section tag as too vague, and note that individual {{citation needed}} tags provide clarity.  IMO we have seen in the subsequent example the clear benefit of using individual {{citation needed}} tags.  Unscintillating (talk) 15:14, 6 July 2013 (UTC)

Um... is there a tag for "this section needs more sources?". My guess is that the person who added the "unreferenced section" tag simply chose one that came close to explaining the problem (which was that a lot of the information was not properly referenced).

But let's not get hung up on trivial technicalities ... what ever the original tag should have been... the one that actually was placed in the article did what it was supposed to do. It alerted other editors to a problem that needed fixing, and we worked together as a community to fix it. All the information in the tagged section is now properly sourced. The system actually works. Blueboar (talk) 02:31, 7 July 2013 (UTC)

If you think this is a "trivial technicality" then I think that you are completely missing the big picture here.  Unscintillating (talk) 02:43, 7 July 2013 (UTC)

On other articles I've come across various problems with lists (ie. some entries are sourced and some aren't; some entries may be OR or pointy or misleading; and other variations on the theme) and none of our existing warning templates was a perfect fit. I have often toyed with the idea of creating a couple of more specific tags for lists with problems...

We also need to bear in mind that articles are dynamic. If a list is tagged as unsourced and somebody subsequently fixes that problem for some entries, that's great news for those entries but the rest of the list still has the original problem. Also, other editors may visit in the meantime and add or remove individual entries, with or without sources. If a list as a whole suffers from chronic sourcing problems, simply adding {{citation needed}} tags to each new entry is like using a bucket chain to fix a burst dam. If the community is unable or unwilling to enforce WP:V, the least we could do is have an appropriate warning template for the list as the whole. bobrayner (talk) 11:43, 7 July 2013 (UTC)

I don't see what the issue is here... we started with a section that was almost completely unreferenced, so someone added an "unreferenced section" tag. Now, technically, the tag was wrong (one of the entries did have a source)... however, since the rest of the section was completely unreferenced, and since we don't have a tag for "Almost completely unreferenced section" the tag was close enough to being correct that I find the difference trivial. More importantly, the tag did want it was supposed to do... it alerted editors to a problem with the section... so we could WP:FIXTHEPROBLEM.

Which we did... I was able to initiate a partial fix (I added sources to a few of the entries, per WP:BURDEN)... and that partial fix made the broad "section" based tag obsolete ... so I removed it, and narrowed the challenge by leaving individual "cn" tags on the entries that still needed sources. This change allowed AQFK to focus on finding and adding sources for the remainder... which he did (also per WP:BURDEN). He was not able to find a source for one entry, and he removed it (Again per WP:BURDEN]].

All of this is how the system is supposed to work. A problem was highlighted, discussed, identified and fixed. No more problem. So why are we continuing to discuss it? You are obviously upset about something that occurred during the process, but I don't understand what you are upset about. Blueboar (talk) 16:59, 7 July 2013 (UTC)

My problems with this issue is that:

1. There was apparently no good faith attempt at finding sources. Anyone can use Google to find sources, but apparently nobody bothered trying. WP:PRESERVE says that you're supposed to try to find sources yourself. WP:PRESERVE is a Wikipedia:Editing policy and it cannot be ignored without a good reason.
2. No specific reason for challenging the content was ever given. Did anyone say that the content was incorrect? No. Did anyone say that the content was unverifiable? No. Did the content end up being correct and verifiable? Yes. So what was the reason for the challenge?
3. WP:BRD says that anyone can be bold in changing content, but if someone reverts, you need to seek consensus on the article talk page. While WP:BRD is not official policy, WP:CONSENSUS is. Why was this policy ignored?
4. Half the encyclopedia contains unsourced content. Do we really want editors to go around deleting content without a good reason? A Quest For Knowledge (talk) 17:57, 7 July 2013 (UTC)

It was challenged as being unsourced. Right here. Colloquial phrasing, perhaps, but that is a removal for being unsourced. Whether or not that was a great removal, it kicked WP:BURDEN into effect, and anyone restoring the material was obligated to provide an inline citation for the material. An inline citation. Not a blue-link. Not a tag. Not a hand-wave. An inline citation to a reliable source. It was removed a second time because the restoration was a blatant and willful violation of that policy. That you twice restored the material without making any effort to provide citations beforehand shows willful disregard for policy. You certainly did eventually help provide a few citations, but you did so only after you had caused other editors to shoulder the work that was yours to do. Finding sources for these things is good. I'm glad you did so. But do it before you restore material to articles, not afterwards.—Kww(talk) 18:07, 7 July 2013 (UTC)

Can you cite Wikipedia policy where it specifically says that WP:BURDEN overrides WP:PRESERVE? Can you cite Wikipedia policy where it specifically states edit-warring to win content disputes overrides WP:CONSENSUS? A Quest For Knowledge (talk) 18:17, 7 July 2013 (UTC)

They are both policies. However, Bobrayner made a change, which you reverted without fulfilling your obligation under WP:BURDEN, I fixed your error, and than you reverted a second time without fulfilling your obligation under WP:BURDEN. I only see one editor in that exchange that an accusation of "edit warring" could be leveled against.—Kww(talk) 18:34, 7 July 2013 (UTC)

OK, I acknowledge your admission that WP:BURDEN does not override WP:PRESERVE and that edit-warring to win content disputes does not override WP:CONSENSUS. Please do not do so again otherwise you may be blocked. A Quest For Knowledge (talk) 19:00, 7 July 2013 (UTC)

It truly is disheartening to see how a generally productive editor can be so intent on misreading policy to support the addition of unsourced material to articles, including unsourced material about living people.—Kww(talk) 19:18, 7 July 2013 (UTC)

Technically it is not against any policy to add material without a source attached to it. I can create an entire article without a single source and I will have not violated any written policy nor can one get blocked or otherwise sanctioned for it. We do not EVER discourage people from adding legitimate new information that helps the encyclopedia grow. Now if someone is new and perhaps does not know how to add an inline citation using our mark-up language then we would HELP them to learn and be more productive. Policy does state that lack following correct procedures does not EVER delegitimize an edit or an attempt at a bureaucratic function (writing a DYK as an example). Burden does not state you have to cite something BEFORE it being challenged, only if you think it could be challenged. AGF however forces everyone to assume that unless there is a huge amount of evidence that this is intentional, that the person who added must have assumed it would not be challenged (because of a blue link to the subject's article where there was indeed a source for example). Honest editing should not be disparaged such as Kww seems to be doing, and doing constantly. I suggest this all be collapsed and all the recent talk (3 threads now?) be allowed to quickly go into archives.Camelbinky (talk) 19:46, 7 July 2013 (UTC)

Read the discussion more closely: AQFK is defending the practice of restoring material that has already been challenged, basically by trying to set hurdles for the quality of the challenge. It is required to cite information that has already been challenged, and there's no provision for saying that you reject some challenges because you think the editor that made them is lazy or misguided. I'm not arguing that people should eradicate all unsourced material on sight, simply that once it has been eradicated, it's disruptive to edit-war it back in without following WP:BURDEN.—Kww(talk) 20:05, 7 July 2013 (UTC)

Replying to Blueboar, way up there: the problem was the sequence, not the end result. The material should have been restored from history as sources were found. Reverting unsourced material en masse while making accusations of bad faith and then searching for sources is completely inappropriate.—Kww(talk) 20:09, 7 July 2013 (UTC)

Kww and Quest... you are both essentially complaining about a technical procedural foul... not a major infraction. Seriously, this entire argument is sounding lamer and lamer.

Quest... how would you interpret WP:CANTFIX (the section immediately after WP:PRESERVE... where it says: Several of our core policies discuss situations when it might be more appropriate to remove information from an article rather than to preserve it. WP:Verifiability discusses handling unsourced and contentious material;... I interpret that as saying WP:V (which includes WP:BURDEN) can take precedence over WP:PRESERVE.

As for making a good faith attempt at finding sources before issuing a challenge... I think it perfectly reasonable to assume that when something has been tagged as being unsourced for over three years there probably isn't one (surely someone should have found a source in all those years.) OK... it turns out that this assumption was incorrect, but I find it a very reasonable assumption to make, nevertheless... and because the assumption is reasonable, the challenge that resulted from it was made in good faith. More important... the challenge resulted in us actually FIXING a long standing problem. This was not a case of someone being POINTY and barging around removing unsourced material just because they can... this was a specific challenge enticing us to fix a specific long standing problem. I certainly did not mind taking a moment to look for sources to keep this material in the article per WP:BURDEN... and I don't really understand why you resented it so much. Blueboar (talk) 20:21, 7 July 2013 (UTC)

The reason I get grumpy is that the reversions make it nearly impossible to remove the data. This particular instance, it came to a decent result, and in that sense my complaint looks like a procedural whine. But the normal place for such a thing to end is that it gets removed and someone restores it with some tag, so it languishes in the article forever. I don't see any reason to believe that AQFK was ever going to find a source for any of that data, as he did not do so after either of his reversions. They appeared to be simply reflexive reversions, and he did not undertake sourcing anything until after you had done so. It's not the biggest of BLP violations, but there was even a BLP violation in there, as no one can find a source for one baseball player's hometown. If Wikipedia is ever to be cleansed and repaired of unsourced material, we have to draw a line, and WP:BURDEN draws it quite clearly: the person restoring material has to provide an inline citation when he does so. No exceptions.—Kww(talk) 20:47, 7 July 2013 (UTC)

There have to be exceptions. Imagine if editor A has some kind of grudge, gripe, or even a legitimate cause for concern, against editor B. Editor A goes through editor B's contribution history and removes the last 100 content additions citing WP:BURDEN. (We're probably all aware of similar incidents having actually taken place.) Editor A is always going to be in the wrong because they're using BURDEN for griefing. And, demanding a citation for "Paris is the capital of France" isn't exactly a brilliant use of WP:V either. In fact, I'm going to go further and say that I think using BURDEN against people you clash with on-wiki is only to be done with forethought and restraint.—S Marshall T/C 08:25, 8 July 2013 (UTC)

Not so fast... Are you saying that Editor B has made 100 content additions without adding a single citation? That would be a red flag for me ... indicating that there is a serious issue with Editor B that needs to be addressed. I would at least want to look deeper into the conflict between A and B before I condemn Editor A's actions. Blueboar (talk) 11:41, 8 July 2013 (UTC)

Well, let's not overanalyse the randomly-selected example. What I am saying is: Looking for sources is everyone's job. WP:BURDEN is not a licence to revert an addition without looking for sources yourself. It's not meant to be used as heavy artillery against an editor you dislike; it's to be used with sound editorial judgment and good sense. I'm not saying KWW was wrong in this specific case. What I'm saying is that BURDEN is not exception-free.—S Marshall T/C 12:06, 8 July 2013 (UTC)

Actually... if you go back and examine the history of how WP:BURDEN came to be, the intent was explicitly to say that a challenger can revert an unsourced addition without looking for sources himself (for one thing, the challenger may not have the resources to check himself). The entire point of BURDEN is to say that when unsourced information is challenged, it is up to those who wish to keep the information in the article to do the work of finding the sources. That's what the word "Burden" means. It is not the job of the challenger to prove the negative (I looked and there are no sources)... it is the job of the adder/keeper to prove the positive (Here is a source).

That said... We all agree that going on a mass removal campaign is wrong... the thing is, the rational given for the removal campaign does not really matter... it would be just as wrong to go on a mass removal campaign that used WP:NPOV or WP:NOR as the rational. The problem you are trying to address is mass removals, not the rational given for the mass removals. I don't know if this already exists, but if not perhaps you should write a WP:Don't go on a mass removal campaign behavioral guideline. I would be happy to help. Blueboar (talk) 12:48, 8 July 2013 (UTC)

As usual I agree with Blueboar. The only obligation on the remover is to say that s/he has doubts about verifiability and, perhaps, to preserve the material on the talk page. (Since this is winding down, I won't churn it by giving my screed on why I believe WP:PRESERVE to be advisory rather than mandatory, especially in comparison to WP:BURDEN. Also, I am of the opinion that a failure to express doubt about verifiability may deserve a reminder plus a slap on the hand if repeated, but does not affect the burden of any restorer.) Mass removals are wrong, but the thing we are lacking is a good definition of "mass removal", so they have to be enforced via a Potter Stewart test. Regards, TransporterMan (TALK) 13:36, 8 July 2013 (UTC)

Re Tagging: A bit off topic but one of the issues about section and article tags, is that in many circumstances, it is unclear why exacty they are there. It is incumbant upon the tagger to make it clear by saying something on the talk page when the tag is placed, and it is perfectly approraite that such unclear and undiscussed tags are summarily removed, and not left there for years. -- Alanscottwalker (talk) 14:05, 8 July 2013 (UTC)

I agree with the spirit of your comment, but would quibble ... surely an "unreferenced section" tag left on on a section that has no references is clear as to what the problem is without a talk page comment to explain. "Not discussed on talk page" is not a valid reason to remove the tag. If a tag is unclear, it is incumbent on the responders to ask for clarification before they remove it. Blueboar (talk) 14:36, 8 July 2013 (UTC)

As to your first sentence, I did not say all circumstances. As to your second, a tag is subject to challenge also -- an unclear tag is not an article improvement. Alanscottwalker (talk) 20:29, 8 July 2013 (UTC)

• Well, it's not just mass removals that I'm objecting to. I see the removal of easily-verifiable content as problematic behaviour. I don't mean the obscure or complicated stuff; if I went to solar twin and added, "The most sun-like star yet found is HIP 56948", then it's reasonable to challenge that under WP:BURDEN. But if I went to a less technical article and added content you could verify with thirty seconds' googling or by reaching for a standard dictionary, then if you revert it under WP:BURDEN then that behaviour would help me to form an opinion about you. It may not be required but it's certainly good practice to perform a perfunctory search of your own, and an editor who persistently fails, or refuses, to do that is not helping to write an encyclopaedia.—S Marshall T/C 14:20, 8 July 2013 (UTC)

• Sorry, I don't agree. As I mentioned above, I watch several articles to which people like to add uncited examples. These are the people who aren't helping write an encyclopedia— or rather, the encyclopedia they're "helping" to write is the kind of junk source which plagues the fringe theory noticeboard. There are many, many articles where the number of junk contributors is high and where the quality of their insertions is low; the only way for the few watchers of these articles to keep up is to enforce the need for citations rigorously. A vote for tolerance in these articles is a vote for error. The differential between someone putting in their personal opinion and someone having to chase down possible citations puts the former at the advantage in getting their falsehoods kept. Mangoe (talk) 14:36, 8 July 2013 (UTC)

The problem I see the discussion dissolving into is that WP:Editing policy is schizophrenic when it comes to subsections like Burden, Preserve, and FIXTHEPROBLEM. While the wording has become closer in agreement the core reason behind each section is in opposition between Burden and Preserve, and the reason is that there are two philosophical camps in Wikipedia when it comes to uncited material and the requirements of editors; lots of "policy" decisions are compromises between people and groups who disagree and come to the best conclusion they can on leading us to make decisions in the future on issues such as this. Those who cite Burden want to put the burden on those who add information to cite it themselves and for the most part do not want to be burdened themselves with adding a citation. Those who cite Preserve want those who go around tagging everything they don't like, to go ahead and make an honest attempt to find a citation first and remove only as a last resort. Fixtheproblem is more compromise between the two groups. We need to recognize our policies aren't clear cut demands on how to do things and understand if someone means to help the encyclopedia that we are here to help them in return. Let's all just take some time this week and if you see something that needs a citation, to go ahead and find one, if we all fixed 5 a day for the next 5 days the encyclopedia would be all the better. Tagging and removing information is most definitely not as helpful as finding a citation, and don't we all want to do what is best for Wikipedia? BTW, removing information and assuming someone will come along and know that it is in history and it needs a source to come back is unreasonable, a tag at least lets the occasional reader know not to take it for definite fact; and it serves to let advanced editors know to take time to find a source. Anyone who removes content without taking a good faith effort to find a source should be ashamed and there should be consequences for continual behavior of that sort (eg- constantly being proved by others that a 30-second google search solved the problem results in sanctions).Camelbinky (talk) 15:25, 8 July 2013 (UTC)

• There's no schizophrenia, it's simple. The stable wording of WP:PRESERVE is "Preserve appropriate content" (which, incidentally, is my phrasing). If content is appropriate for the encyclopaedia then it's up to you, editors, not to remove it. If it's inappropriate then there's obviously no obstacle to its removal. If you're not sure whether it's appropriate then there's an onus on you to search for sources and make up your mind about its appropriateness before removing it. If you do remove it then there's an onus on me to source it before I add it back. When editors do what they're supposed to do and co-operate in looking at the sources together, where's the conflict? There's none.

  When editors demand that other people do all the work, frayed tempers are only to be expected, really.—S Marshall T/C 16:15, 8 July 2013 (UTC)

OK... time for an airing of one of my pet peeves... I am getting tired of people saying that the only way to "contribute" and "improve" wikipedia is to add material (be it information or the sources that support it). The fact is removing material is also be a way to contribute, and improve wikipedia ... removing material can be just as helpful as adding information. It all depends on what is being added or removed. Blueboar (talk) 16:18, 8 July 2013 (UTC)

I'm sorry, but I couldn't disagree with Camelbinky more. Let's take this deletion I made as an example. I came onto it and The Ford House doing {{coord missing}} patrolling. And I knew from editing The Ford House that both that article and the Chertsey Abbey article were plagued with unsourced material about being spiritually haunted. I prodded the house article, which was subsequently deleted, and {{cn}}–tagged the unsourced assertion referenced above, which fit into the haunting nonsense with the house article. After a month, I deleted that unsourced assertion. Did I search for a source? Nope, not for a minute. Do I feel ashamed for it? No, not in the slightest. (I didn't preserve it to the talk page because, to tell you the unvarnished truth, I wasn't aware of PRESERVE at that point in time, but I probably would today — and may go back and do so now.) I very rarely delete unsourced information, but I generally have a good reason for doing so when I do other than the fact that it is unsourced, but when I occasionally come across an unsourced and highly–improbable assertion inserted by a fringecruft–lover such as this while doing something else, I'm probably not going to take the time to look for sources, nor am (or should) I or anyone else be rule-required to do so. Regards, TransporterMan (TALK) 16:28, 8 July 2013 (UTC)

First of all there are no rules requiring editors to edit in a certain way. The ONLY thing, and I swear to G-d this is not up for debate, that matters is that you are trying to improve Wikipedia (and Blueboar is right, that does not mean just adding, it can mean removing). It does not matter about bureaucracy, policy, or even citations; as long as the information you are adding is, in your best faith knowledge correct and factual, it does not matter one bit how you put it in. If it creates "work" for someone else to fix it up, too fucking bad, seriously, this all boils down to people not feeling they should clean up other's messes. Well, we are a work in progress, some new editors add good things but don't know how, or add things that prod others to make it better. This crap of harping on others because they didn't add the information the way we want it and that they aren't following rules has got to stop. This is the kind of harping crap that has driven away potential editors, the only new editors that stay are ones that ignore the rudeness and learn the !rules, but then of course that causes a feedback loop because then they think they need to act the same way to newer editors. You see something that needs a citation, if you have reason to believe it is not true, tag it, if it is tagged and still not fixed, remove it; if you have some reason to believe it is true then YES YOU MUST DO SOME SEARCHING!!!!!, it is common courtesy, common sense, and it will go a long way from solving your laziness problem.Camelbinky (talk) 18:33, 8 July 2013 (UTC)

"[A]s long as the information you are adding is, in your best faith knowledge correct and factual, it does not matter one bit how you put it in." That statement completely invalidates WP:V. When information is not cited to a reliable source, the average reader has no way to confirm whether or not it is real-world-reliable. And the more unsourced information that's here, the more worthless this encyclopedia becomes (and FSM knows we're accused of being worthless enough as it is) and arguments such as this just make it that much harder to purge the encyclopedia of that kind of information because for every good-faith editor here who does care about sourcing and the quality and reputation of the encyclopedia, there are ten fanboys, POV-pushers, and Randy's from Boise packing in the unsourced information. That's the reason that it is, and should be, easy for editors of good faith to remove unsourced information if they merely suspect that it might be unverifiable: if it's really valuable and reliably-sourceable someone who cares enough to do it right will add it back in. Regards, TransporterMan (TALK) 18:59, 8 July 2013 (UTC)

I'm surprised this kind of discussion is still ongoing. If Wikipedia aspires to be taken seriously then the burden of providing references has to be on an editor who wants to add or include material. If an editor, in good faith, believes unsourced material is not true or is implausible, or just considers that it shouldn't be included without references, they should be able to remove it. Then another editor can restore it if they provide a good quality reference. There should be no expectation on editors to look for sources for this kind of material rather than remove it. For various good reasons, they may not have the time/resources/inclination to do so. Removing questionable text is improving articles. As a reader, I would have a lot more confidence in an encyclopedia if I knew it had been ruthlessly edited and purged of questionable or unsourced material, than if I knew the editorial policy was to keep stuff in as a default. --hippo43 (talk) 19:41, 8 July 2013 (UTC)

• A practical example will probably help here. I've just created a fresh biographical article, Marianne von Willemer. It cites absolutely no sources at all. It is nevertheless fully compliant with WP:V, because everything in it is verifiable and nothing in it is challenged or likely to be challenged.—S Marshall T/C 19:51, 8 July 2013 (UTC)

I agree that sort of article may well be accurate (and I'm sure this one is) but the point is about when something is challenged. I could remove any of the unreferenced facts that you have asserted, and the onus would rightly be on any editor wanting to restore them to provide references. If I can't/won't look for sources myself, that shouldn't remove the burden from editors who want to restore it. --hippo43 (talk) 22:03, 8 July 2013 (UTC)

True; looking for sources is best practice, not a requirement. And I see you wisely chose to tag the article rather than remove any content. But if you won't look for sources yourself, then how does that affect your credibility as an editor? And if you really are sure the article was accurate, then why have you tagged it?—S Marshall T/C 22:44, 8 July 2013 (UTC)

Who says looking for sources is best practice? Removing unsourced, questionable material seems like best practice to me. Then looking for sources if you have time/resources/knowledge etc would be nice. Truly best practice is probably not creating an entirely unreferenced article.

My credibility as an editor is in the eye of the beholder. Some editors think looking for sources in this kind of situation is a big deal. I don't. IMO creating a completely unreferenced article is more harmful to an editor's credibility.

I wrote "I'm sure this one is (accurate)" as a kind of shorthand. Obviously I'm not sure at all. I don't know anything about this woman, but I'm giving you the benefit of the doubt because you're using it as an example here. If I had just stumbled across that article, I would probably tag a lot more. Deleting the unreferenced parts of it would seem pretty brutal as there'd be nothing left. --hippo43 (talk) 08:41, 9 July 2013 (UTC)

Two points: first, a whole article is a different kettle of fish. Dealing with it would in general require pushing through a merger or a whack at AFD. To do the latter one has to do some basic research— or at least someone is bound to do it. Second, the one claim which cries out for verification is the last, and it is also the easiest to verify. At the same time, I don't feel the obligation to go back and do your work for you. If you can write this article at all, it's because you have a source for it; there's no reason not to expect you to admit what that source was. I've written articles (and especially not-ready-for-publication articles) where I didn't lay out the in-line citations, but just put them in a list at the bottom. A lot of my older articles did get published that way, and I haven't felt too compelled to go back and clean them all up because as a rule it's not hard to find the claims in the cited works (both they and the articles tend to be short). But especially for someone who actually knows what they're doing, there's really no excuse at the time for publishing an article without any references at all. We wouldn't let anything through AFC in that state. You're using "verifiable" in a sense which would apply to any string of declarative sentences; I cannot verify that your sources are being represented accurately, because I don't know what they are. What I'm seeing is that this is a situation where you can get away with doing a bad job. Mangoe (talk) 22:46, 8 July 2013 (UTC)

Anyone else want to give me attitude about my failure to add a source they can't check because it's in a language they don't speak? I may collect all this stuff up for posterity.  :)—S Marshall T/C 23:08, 8 July 2013 (UTC)

Not sure about the attitude part, but I think you should have provided a source or not created the article.  Tell me, do you agree that we need more admins supporting WP:V and the need for sources?  Unscintillating (talk) 01:40, 9 July 2013 (UTC)

With all due respect, Unscintillating, you're completely missing the point. That article fully complies with WP:V as-is, with no sources at all. Because every word in it is sourceable and there's nothing in it that's challenged or likely to be challenged. There's no academic dispute about Marianne von Willemer or her role in Goethe's writings. There's no nationalist partisanship involved. She's not part of a special interest group or involved with a political affiliation.

The purpose of writing the article is of course that someone might want to know about Marianne von Willemer, but the purpose of omitting the sources is to facilitate a discussion about WP:V that's more practical than the theoretical-level discussions above. I'm finding it illuminating: faced with this practical example, nobody's behaving in an extreme way. There's grumbling that I should have listed my sources, and someone's added a tag, but despite the misinterpretations of WP:V we saw in the discussion that preceded the article's creation, in practice nobody feels comfortable with removing anything.

I also see that nobody has lifted a finger to source it themselves, so my failure of best practice in not listing my sources is mirrored by other peoples' in not doing any work either.—S Marshall T/C 08:36, 9 July 2013 (UTC)

In fact I could not find a good citation for the last sentence, so I removed it. A quick look didn't turn up a source making that exact claim, and while I found some testimony that she did write some of the work, I need to get in the shower in a moment, so I don't have time to dig through that and sort out a passage that I can cite. So I took the whole thing out, as it is better to have a lack of claims rather than claims that I cannot be sure of. I really don't care whether you may have used a German source, or for that matter a source in Urdu or Tlingit; all unreferenced sources are equally unreadable. I see no reason to believe that you can read German, if you don't cough up a source. I take a hard line on exclusivity and priority claims, and delete them all if they don't have a good citation, because they are usually incorrect if they are not cited. I would not be surprised that there are good German sources, but I disbelieve that you are using them unless you cite them; I don't see anything in what is written that couldn't be taken from a misinterpretation of the English-language sources I turned up. If you feel you must stand on principle, I am willing to do likewise; however, when push comes to shove, it is my (and the world's) higher standards that need be met. Mangoe (talk) 10:29, 9 July 2013 (UTC)

And now let's throw that open to discussion. Is Mangoe being POINTY?—S Marshall T/C 13:29, 9 July 2013 (UTC)

No more than the rest of us. Seriously... we are all getting a bit POINTY here. And a bit too personal in our comments. Time to back off, folks. Blueboar (talk) 13:55, 9 July 2013 (UTC)

The question of whether it can be POINTY to remove easily-verifiable content is germane.—S Marshall T/C 14:05, 9 July 2013 (UTC)

Of course it can be POINTY to remove easily-verifiable content... We all know that. However, it is not always POINTY to do so. Sometimes it is absolutely appropriate. Note that it can also be POINTY to return such material without adding a citation. People can get POINTY about all our policy and guideline provisions, including both WP:BURDEN and WP:PRESERVE. . Blueboar (talk) 14:21, 9 July 2013 (UTC)

If I cannot easily verify it, then it's not "easily verifiable". And I would say I'm a bit better at this than average. You set up a test case; I say it doesn't pass, and there's nothing POINTY in the police showing up when you climb the Reichstag. The thing is that you, as the writer, cannot be the arbiter of what is "easily verifiable"; it's too ready a means of avoiding ever having to supply proof for possibly fallacious claims. As I said, a quick search didn't turn up a statement that she was the only one to ghostwrite some of Goethe's material, so I took that assertion out. The rest of the sentence died to the cause of accuracy because at 6:30 in the morning I didn't have time to do the research needed to rewrite the rest of the sentence around the excision. And speaking as a reader who also edits, and particularly in the latter case as someone who deals with fringe theory material, I don't have a lot of confidence in Wikipedia articles which lack citation. We have had far too many hoaxes in which plausibility and a lack of scrutiny protected false information; therefore the mere claim that some material is "easily verifiable" tends to make me suspect that it is not only not verifiable, but should be assumed false until someone demonstrates otherwise. Someone without my experience in researching and fact-checking, who is doing a quick Google search to get a rough idea about something, is surely not in a position to easily verify anything, and shouldn't trust someone's easy assurance that the information is easily verified; the refusal to pony up I translate into the assumption that the material shouldn't be trusted. Mangoe (talk) 16:40, 9 July 2013 (UTC)

Well, I hear all that. It's understandable that someone who hangs around the fringe theory parts of Wikipedia would get suspicious. To you, BURDEN is vital because it's a stick to beat the POV-pushers with. I get that.

But not all of Wikipedia is fringe, or full of POV-pushers. Most articles are completely uncontroversial----the individual species of spider; the religious articles that have been untouched since they got copy/pasted from the 1913 Catholic encyclopaedia; and that one you always get when you click "random article" which is about Where the Fuck, Idaho (pop. 312). BURDEN needs to be about these general cases. It doesn't need to be specially tailored for the Fringe Theory Noticeboard or the Palestine-Israel controversy or whichever other topic area's plastered all over AN/I right now (I don't keep track). And the reason why BURDEN needs to deal with general articles, not controversial ones, is because hard cases make bad law.

With these general articles, wandering around randomly removing unsourced facts from material you haven't researched is not a good idea. For example: A reader who looked up "Marianne von Willemer" on Wikipedia is probably doing that because they've heard of someone who's won the Marianne von Willemer Prize for Literature. But I haven't written the section on the Marianne von Willemer Prize yet. That section would need to be set in context about her significance as a writer. But when some random editor goes to the recently-written article and removes the only sentence that says why she's important as a writer, the subsequent reader loses all context. They'd probably think the Marianne von Willemer Prize was some kind of acting award, or something. Do you see?

Part of the reason why I see this as so important is because randomly removing sentences in this way could so easily create a WP:DUE or other WP:NPOV issue. I think editors should research all changes they make to the encyclopaedia, and that includes subtractions.—S Marshall T/C 17:54, 9 July 2013 (UTC)

The problem is that (as the list of hoaxes shows) there isn't any place in the encyclopedia which is above suspicion anymore. Most successful hoaxes work by incorporating innocent-looking material in articles that aren't heavily watched; I suspect that the quantity of deliberate falsehoods which have gone undetected because they weren't seen by an expert and weren't outrageous enough to tip off the passer-by is vastly larger than what has thus far been detected. From what I can tell there are a lot of "Sodding Chipwich" articles out there which document utterly fictitious communities, often enough because the name is copied from some source which has made some stupid mistake along the way. But if I cannot tell what that source is, how can I tell? We've had some big arguments in the lighthouse project about fictitious stations created out of mistakes made by the people who wrote up the USCG history pages. The NRHP project is continually plagued by mistakes in he official database; they make up phony towns all the time, to the point where there is a place for keeping track of and reporting the errors back to the NPS.

Back seven or eight years ago it was tolerable to throw a few sentences together on a subject and call it a day, knowing (meaning hoping) that later on someone would come back and to the work right. The project is too mature for that kind of shortcut any more. Readers should be able to trust the material, and the apparatus of that trust is, in part, the provision of the sources used. Really, at this point the standard should be raised from "verifiability" (which I interpret to mean that I can verify that the article reflects its sources) to "verification". I may well put in a proposal to do that. But even so, the insistence that I must assume good faith about unreferenced material, even on what is commonplace, is now out of line. There is simply too much falsehood being slipped into the work through that door. Mangoe (talk) 18:19, 9 July 2013 (UTC)

I don't actually understand your most recent edit to the article we're talking about; you seem to have changed it from a claim that she co-authored part of the book, to a claim that she co-authored part of the book. Did you see what I said about removing sentences actually creating NPOV and DUE issues, and do you have a response to that?—S Marshall T/C 19:08, 9 July 2013 (UTC)

• We need both WP:PRESERVE and WP:BURDEN.  In this particular case, the section tag was added by an editor who soon left Wikipedia using vulgarity.  The next few years show a consensus with dozens of edits that the tag was not actionable for the purpose of adding sources.  It is also not actionable under WP:BURDEN because it is too vague to know what is being challenged.  It is like a tag that says, "This article has problems that need to be fixed."  There seems to be only one proper next edit here, which is to replace the single tag with individualized cn tags.  Unscintillating (talk) 02:09, 9 July 2013 (UTC)

The content in Elizabethtown, KY was challenged repeatedly, so it's rather different from S Marshall's example which is, dare I say it, rather atypical. Articles in the wild with zero sources are generally good candidates for gutting, rewriting with new sourced content, redirecting, or AfD (or CSD) because they usually have problems with accuracy, neutrality, or notability (the latter is based on sources, of course). I have no interest in making WP:POINTY edits to Marianne von Willemer! :-) bobrayner (talk) 10:57, 9 July 2013 (UTC)

What was the challenge? A Quest For Knowledge (talk) 11:00, 9 July 2013 (UTC)

One good example would be this edit. I'm surprised that you were unaware of that challenge, since you reverted it, and your revert failed this policy. bobrayner (talk) 11:10, 9 July 2013 (UTC)

No, I'm asking what the reason was for the challenge. AQFK (talk) 11:14, 9 July 2013 (UTC)

The reason for Bob's challenge was stated in the edit summary: "don't we need sources for this kind of thing? Especially if living people are involved". Blueboar (talk) 12:07, 9 July 2013 (UTC)

Taking a look at that edit summary, the answer to that question is: No, we don't need inline citations. This goes back to the difference between verifi'able and verified. A Quest For Knowledge (talk) 14:29, 10 July 2013 (UTC)

The material included unverifiable information about living people, AQFK. Don't get so obsessed with the specific language of the challenge: it wasn't sourced. It was challenged on the basis of it not being sourced. The challenger removed it on the grounds of it being unsourced, and hence unverified. No challenge has to meet your personal standards for quality before WP:BURDEN kicks in. Any and all restoration of that material without having bothered to find a source prior to restoring the material was disruptive.—Kww(talk) 00:55, 12 July 2013 (UTC)

No, it appears that the question was based on honest misunderstanding between the difference between verifiable and verified. Since no challenge was offered, WP:BURDEN doesn't apply (but if it did), WP:PRESERVE also applies (and no argument why WP:PRESERVE should be ignored has been presented so far). But even still, the information was proven to be verifiable by the simple fact of all the citations that were easily added.[7] So even if we were to ignore policy as well as common sense, how does one explain the fact that the information was BOTH verifiable yet verified? In any case, I'll let you get in the last word. A Quest For Knowledge (talk) 02:51, 12 July 2013 (UTC)

AQFK, PRESERVE states that facts "should be retained if they meet the requirements of the three core content policies: Neutral point of view (which doesn't mean No point of view), Verifiability and No original research." This material was clearly challenged on the basis of verifiability, therefore PRESERVE does not support its retention. In any case, BURDEN does not state that material has to be challenged before it can be removed (even less that it has to meet your definition of challenged). It clearly states "Any material lacking a reliable source directly supporting it may be removed." --hippo43 (talk) 03:22, 12 July 2013 (UTC)

Moreover, note that "should". As I have alluded to before, though PRESERVE is part of a larger policy it is a part which is only a best practices or advisory provision, not a mandatory provision. If it were mandatory, that sentence would read "retain them if they meet the requirements of the three core content policies" or "they must be retained if they meet the requirements of the three core content policies". For the same reason, the intro line to the laundry list at the bottom of PRESERVE reads "Instead of deleting text, consider:" (emphasis added). Thus an editor must only give consideration to doing those things, but is free to do as s/he likes. The entire tenor of the subsection is advisory / best practices, not mandatory. Like most other best practices provisions in policy, that's not to say that you can't get in trouble if you ignore it often enough, especially if you're engaged in a POV or other self-serving program, but for an isolated instance or occasional flouting, not so much. Regards, TransporterMan (TALK) 14:06, 12 July 2013 (UTC)

• PRESERVE is mandatory. Content that's appropriate for an online encyclopaedia should be preserved. This is not in any sense optional and there's no room for discretion there at all: if the content belongs in an article somewhere then you may not delete it. There is, of course, plenty of room for reasonable people to disagree about whether a particular piece of content is "appropriate".—S Marshall T/C 18:54, 14 July 2013 (UTC)
• In my OP to this section, I mentioned the outright removal of unclear section tags.  However, I want to reiterate alanscottwalker's point above, "a tag is subject to challenge also".  For other approaches, Blueboar has demonstrated that the section tag can be removed and replaced with inline tags.  It is also possible to remove the section tag and just add a perfunctory number of inline tags to get the ball rolling to zero in on the actual problems.  If the entire article has one giant tag, it may be necessary to first replace the article tag with section tags before working on the individual section tags.  Unscintillating (talk) 21:35, 14 July 2013 (UTC)

I'm reading a lot of biographies (of deceased people) and claims are often made that are supported by books written about other people. Although the book itself might be well-researched (let's take that as a given to save time), it's about Person A, not Person B. So, if Author X says Person A mentioned that Person B was gay, doesn't that still make it hearsay? I should also mention that neither Person A nor Person B are alive to dispute what Author X has written.

The claim could very well be true but I don't think this meets the requirements of a reliable source on the life of Person B...by the way, Person A was not claiming to have had a relationship with Person B, if that matters, just that they went to the same kind of parties where gay men associated with each other (this all happened 60-70 years ago).

This happens a lot in books about people in the entertainment industry and I'd welcome hearing something definitive about this from more experienced editors, specifically what WP policies I can read and point to for guidance. 69.125.134.86 (talk) 20:00, 16 July 2013 (UTC)

Pending the arrival of the resident experts I'd say that subject to considerations of appropriate weight, we would attribute what Author X says, per WP:SUBSTANTIATE. Something like "In X's book My Life in Films (1979), he wrote that in 1920 A mentioned that B was gay."[full reference] or – better – 'In My Life in Films (1979) X wrote: "At that party in 1920, A told me that B was not the marrying type."'[full reference].  —SMALLJIM  21:10, 16 July 2013 (UTC)

A lot depends on what article you are discussing all this in, and what the context is. Person A's opinion of B might be relevant to mention in the article on A, and yet not relevant to mention in the article about B. DUE WEIGHT is also an issue... Does mentioning A's opinion of B (as passed on to us through X) give A's opinion too much weight? Blueboar (talk) 00:13, 17 July 2013 (UTC)

Agree with the other responses. Also I'd like to note that two elements of your question complicate it a bit. You said that your example was about a deceased person, yet the wording of your question did not include that as a condition (exclude living persons) and so it could include wp:blp situations, a special case where policy imposes a higher standard. Second, the example includes a major and potentially controversial statement (that someone was/is gay) which could also raise the bar, doubly so if a wp:blp situation. North8000 (talk) 14:41, 17 July 2013 (UTC)

This is a more common occurrence than one might think, and I hope everyone will excuse me if I don't use actual examples here. Generally speaking our BLP policy about living persons still applies for recently dead and that can be up to 2 years. See WP:BDP. However, there is also the consideration of whether or not the source is from a reputable publisher (not a vanity press) and if the material can be considered creating a situation of/or prolonging victimization. See WP:AVOIDVICTIM. I believe the best policy we have to cover this is: "If an allegation or incident is noteworthy, relevant, and well-documented, it belongs in the article – even if it is negative and the subject dislikes all mention of it. If you cannot find multiple reliable third-party sources documenting the allegation or incident, leave it out.". See WP:PUBLICFIGURE. If the belief, accusation or understanding about the figure is well documented, it may be included, think Rock Hudson(OK, there's an example I can use). If, on the other, this is the only source or all sources are not reliable, tabloid journalism, vanity press or self published etc., then it should simply not be used.--Amadscientist (talk) 00:23, 20 July 2013 (UTC)

I think that that is good advice . North8000 (talk) 00:33, 20 July 2013 (UTC)

I have reverted CamelBinky's edit to the 'Offline sources' section [8] as it simply doesn't make sense:

Articles may be based entirely (or in part) on offline reliable sources; there is no general requirement that all or any sources for an article be available online. Offline sources may include books, maps, audio, video, carvings, monuments, and other physical material that is used in a manner that would be consistent with any other editor viewing or listening to the same material. Be careful not to violate WP:OR by using specialist knowledge to expand on the source, best to be cautious and ask for help before going too far in using such a primary source.

Which are the 'primary sources' referred to here? Some offline sources may be primary, others secondary - there seems to be no reason to go off on a tangent and raise this here at all. AndyTheGrump (talk) 23:34, 19 July 2013 (UTC)

Removed the offending statement. Good to know all I have to do is edit something in order to get Andy to discuss something.Camelbinky (talk) 23:41, 19 July 2013 (UTC)

A polite request on my talk page might have worked. AndyTheGrump (talk) 23:44, 19 July 2013 (UTC)

http://en.wikipedia.org/w/index.php?title=Wikipedia:Verifiability&diff=563792693&oldid=563777619

I'm trying to understand the revert. It wasn't on the basis of the sentence being necessary that I included it. I cited a recent Village Pump (policy) discussion (Wikipedia:Village_pump_(policy)#Are_online_references_mandatory_for_new_articles.3F), and further experience of the problem as the reason for inclusion. I agree that it is not necessary that the sentence be included, but I don't think that is sufficient reason for exclusion, because most sentences in the policy are unnecessary; the reason they are included is because they usefully respond to problems that occur. It's not necessary that we correct misconceptions, but it is useful. --Atethnekos (Discussion, Contributions) 16:57, 11 July 2013 (UTC)

I agree the revert was probably just a knee-jerk reaction to someone daring to change a policy. I believe Atethnekos' contribution should be returned. This is a perennial topic that I have seen come up for the last five years or so every so often. It needs to be stated clearly and with no wiggly room- offline sources are perfectly fine to cite and online sources are not MANDATORY.Camelbinky (talk) 20:02, 11 July 2013 (UTC)

I agree with User:Bbb23 that it ought to be unnecessary. Then again, it ought to be unnecessary to have warnings that external links are not reliable sources scattered throughout WP:EL, and putting it six or eight such warnings has proven to be quite useful. Perhaps this is another case in which redundancy will help some people grasp the point. WhatamIdoing (talk) 00:40, 12 July 2013 (UTC)

I don't like adding language to policy unless it's necessary. Just because somone or even more than one person fails to grasp a certain point doesn't mean we have to address it. If we did, the policy would be even longer and harder to read than it already is. And, @Camelbinky, I'm not interested in your speculation as to why I do what I do. In any event, you're wrong.--Bbb23 (talk) 00:52, 12 July 2013 (UTC)

If people are failing to grasp a certain point, then it is necessary to improve the way that we address their confusion. That's the only reason we have written down any of these policies in the first place. WhatamIdoing (talk) 01:58, 12 July 2013 (UTC)

• Support inclusion of the point, but it should be stated in the positive and with helpful advice. For example: Printed sources are acceptable, but include quotations, brief summary, etc in the reference data to assist others in reviewing the use of the source. --SmokeyJoe (talk) 02:51, 12 July 2013 (UTC)

  That will naturally be interpreted as "Printed sources are unacceptable unless you have included quotations, brief summary, etc in the reference data". Providing access to sources that you have and someone else doesn't is a courtesy; it is never required. WhatamIdoing (talk) 22:23, 12 July 2013 (UTC)

• Describing better practice in this basic guideline (intended to be read by new editors) is a good thing. --SmokeyJoe (talk) 03:03, 13 July 2013 (UTC)

I think it was already well stated that printed sources were acceptable, the problem is that some people fall under the impression that they by themselves are not adequate. I think a reminder that one can include summary and quotations in some circrumstances can be appropriate, but an unqualified command to include such is potentially very damaging. This is because when quoting copyrighted works, the reproduction of such material must always be discretionary and carefully included only toward the purpose of scholarship. If editors end up just including quotations from copyrighted works by rote, then the risk of copyright infringement would be quite high. The section does well in directing people to the WP:REX, which can help with accessing sources somewhat. There are other useful source-access pages on this project; I've started to put these together and other links at User:Atethnekos/Source_Access, and I am still interested in expanding that page with the help of others and eventually having a genuinely helpful page to which we can direct those seeking sources. --Atethnekos (Discussion, Contributions) 23:28, 12 July 2013 (UTC)

It can also lead to stupid summaries: "President Joe opposes killing baby seals.^{ref Joe, President (2013) "I Oppose Killing Baby Seals", News. Summary: President Joe opposes killing baby seals.} We explicitly state elsewhere that quotations are normally provided upon request and as a courtesy. We don't normally add these, and if you don't or can't provide quotations, that still doesn't invalidate the source or your use of it. WhatamIdoing (talk) 14:45, 14 July 2013 (UTC)

Agree. It is already a task to provide the source itself. A needed task but real work nonetheless. Let's do nothing to make it all the harder. Alanscottwalker (talk) 15:28, 14 July 2013 (UTC)

I think the problem in this instance was an Article for Creation reviewer was under the impression that although some of the sources for an article could be print-only, it was necessary that at least some of the sources be online (which is a bogus interpretation of the policy). So the question is whether this particular misconception is widespread enough that the acceptability of an article based only on print sources be spelled out. Jc3s5h (talk) 13:07, 13 July 2013 (UTC)

• Based on this discussion I have been bold, and I have separated the sentence about online vs. paper sources out into its own sub-section. (see this diff). That should probably be enough to highlight it. Blueboar (talk) 13:57, 13 July 2013 (UTC)

OK... I made an additional tweak which Bbb23 reverted here with the comment: "Please don't make it worse." No problem, but could you explain why you think my edit made it worse? I thought it made it better. Blueboar (talk) 16:48, 14 July 2013 (UTC)

The second part of your change just added unneeded verbosity. The first sentence ("The medium in which a source is published does not matter.") is worse than unnecessary. I assume you're trying to hit the reader over the head that a reliable source may be used whether it on an online medium or on a print medium, but because it's unnecessary, it may cause the reader to think there's yet another medium you're referring to. What medium is that? Water? Glass? Stone? Sarcasm aside, there are other media that might be sources, and your statement is simply overbroad and ambiguous at the same time.--Bbb23 (talk) 17:00, 14 July 2013 (UTC)

I agree with Bbb23 - the change adds no more information, and does nothing for clarity. Conciseness is always preferable. AndyTheGrump (talk) 17:04, 14 July 2013 (UTC)

As others have stated there is a misunderstanding with how the wording has been currently stated. There have been many discussions on noticeboards, discussions that would have been unnecessary if this policy wording was indeed hitting the editors over the head that "ONLINE SOURCES ARE NOT PREFERRED OR MANDATORY OR REQUIRED OR NECESSARY". Blueboar, more than any other editor in this discussion has been most active in having to inform and educate editors over and over that verifiable does not mean "every editor, immediately, from their computer can verify every source" and this is one of those policy statements that causes the confusion that Blueboar and many others at places like WP:RS/N have to deal with on a weekly and monthly basis. If there is a problem, then we have a duty to fix it, conciseness is not mandatory nor is it preferred when it is standing in the way of improving editor's understanding of fundamental policy.Camelbinky (talk) 17:13, 14 July 2013 (UTC)

And an example of regarding medium being more than print or online is from several years ago a discussion at RS/N that Blueboar was involved with solving regarding whether a source that could only be found at the USS Missouri museum in Texas was allowed since an editor would have to physically go there to see and verify the source. It was correctly stated that it didn't matter if the editor in question had no way to verify it, as long as in theory AN editor could verify it at some point is the policy, not that every editor can.Camelbinky (talk) 17:18, 14 July 2013 (UTC)

I think that we need this clarification. If people understood this point, then we wouldn't be getting the same question over and over and over again. It's easier to add a sentence that directly addresses this point than it is for us to go fetch Bbb23 and Andy and make them personally explain it to each inexperienced user until they, too, have enough personal experience with this ongoing source of confusion to think that it merits a more comprehensive explanation. On the other hand, if the two of them aren't willing to trust the rest of us, including RSN regulars, when we say that this is a real problem, then we could make it be their problem. WhatamIdoing (talk) 20:38, 14 July 2013 (UTC)

Above, I was asked: "...it may cause the reader to think there's yet another medium you're referring to. What medium is that?"... actually there are multiple other mediums besides paper and electronic format that sources can come in. There's audio tape and vinyl recordings, several mediums for video ... hell, there's even carvings on rocks and impressions left in clay tablets. Sure, the vast majority of our sources will consist of written words in either paper or electronic form... but there are other mediums in which sources can come. What matters (in the context of this section) is whether the source is accessible to members of the general public ... in whatever form it comes. Blueboar (talk) 18:11, 14 July 2013 (UTC)

Despite possibly targeting myself to calls of "ownership", I have to say that those wanting to make a change to wording of this policy have amply shown cause, whereas those who would keep wording at the status quo have shown no reasons beyond "it has always been this way" and "what problem?"... Those who would know what the problem is and work the RS/N as What and Blueboar do on a daily basis (and I used to years ago, and had the same problem even back then) have an overriding interest in changing policy. Consensus does not require all no-sayers to be placated nor gives them a veto through filibustering. The talk has gone its course.Camelbinky (talk) 16:32, 16 July 2013 (UTC)

Could we discuss my proposed new opening sentence for the section... "The medium in which a source is published does not matter." I understand that the primary point is to say that sources that are not online are acceptable, but I do think we need to clarify that offline sources can come in other forms than just paper. For example, I could see an article citing only audio or video recordings, which would be fine as long as those audio/video recordings are reliable, and located in a publicly accessible archive. Blueboar (talk) 16:26, 16 July 2013 (UTC)

Hm? You want a theoretical policy for a theoretical article that does not exist? Why? Alanscottwalker (talk) 01:33, 17 July 2013 (UTC)

I'm not convinced that a whole new sub-section is warranted – I think Atethnekos's original addition to 'Access to sources' may be enough. But if we do have such a section, its purpose would be to alert editors before they make the wrong assumption that an article must include at least one online source: which is the problem being reported here. To start the sub-section with an emphasis that anything that has been published counts as a source would obfuscate the main point making it less likely to sink in. I suspect that this emphasis may also have unwanted consequences, around the area of Notability for instance.

I would start the sub-section with Atethnekos's: "There is no requirement that all or any sources for an article be available online" and then go on to other points, as consensus deems appropriate.  —SMALLJIM  11:00, 17 July 2013 (UTC)

Hmmm... I don't mind starting off with that... it's the second part that caused my concern. The opposite of Online isn't "Paper" ... the opposite of Online is Offline - ie Physical (which includes but is not limited to paper books). Blueboar (talk) 13:42, 17 July 2013 (UTC)

"There is no requirement that all or any sources for an article be available online. Offline sources may include books, maps, audio, video, carvings, monuments, and other physical material that is used in a manner that would be consistent with any other editor viewing or listening to the same material. Be careful not to violate WP:OR by using specialist knowledge to expand on the source, best to be cautious and ask for help before going too far in using such a primary source."

I believe that covers just about everything that recent threads have mentioned and concerns that have been brought up? Please feel free to comment and/or change that around.Camelbinky (talk) 15:05, 17 July 2013 (UTC)

• Concerns: this policy should not generate arguments, that assume any source in particular is "reliable," nor arguments that contravene the universal consensus that articles should be based primarily on "secondary sources." Moreover, it is not clear to me whether our guidance on "Citing" or on "Identifying reliable sources" have developed consensus conventions on how to identify, for example, the "authentic" recording (compare, with a book we have the ISBN number). Citing is a problem because it does not seem good enough to just say watch that two hour movie (or even that ten minute video), or listen to that thirteen hours of tape (an official transcript would help as would official time-stamps, but again that goes back to identifying the "official" copy). Sometimes it's just better to say less per WP:BEANS, when we are thinking theoretically, and on the high plain of general policy. Also, as noted above, this section policy should be written in the positive, because the positive covers more than the particular instance and guides better. but if we need the negative for emphasis something like: "Articles may be based entirely (or in part) on offline reliable sources; there is no general requirement that all or any sources for an article be available online." Alanscottwalker (talk) 15:53, 17 July 2013 (UTC)

I find it ridiculous that there is a serious problem that has been identified, and because a few don't want policy to ever be changed we are stuck with a conservative non-changing policy. There is a problem with the way the policy has been worded all this time, as identified by What and Blueboar, and yet we are back to having the policy state almost exactly what it was saying, despite Alanscottwalker's claim "Blueboar's was better"... Blueboar's last edit put it back to the original with barely any changes, so in effect Alan is stating there is no problem. Ridiculous. I guess we can add Alan to the list of users we can contact to come to the RS/N to explain over and over to new users every month that offline sources are ok.Camelbinky (talk) 02:56, 20 July 2013 (UTC)

Your attacks on other editors as in one of your recent edit summaries [9] suggests that you should reconsider being anywhere near a policy page or a help page. Of course, it's the consensus of everyone that changes to policy need to be conservative, because of unintended consequences (read the "stability is paramount" instruction at the top of this page). Blueboar's edit is better and clearer -- much better and has taken in the concerns of multiple editors, including Blueboar. That's how we do things, here. Alanscottwalker (talk) 11:17, 20 July 2013 (UTC)

I've been on Wikipedia a year longer than you and you're personally attacking me and being a WP:DICK. Don't tell me how we do things around here, I've been working on policy pages for years and years. Policy is not supposed to be conservative, EVER, policy is supposed to reflect how we do things, and not proscribe how we will do things in the future. This policy wording has caused confusion for years and you are saying Blueboar has made it better when his hand was pretty much forced to put it back to what it was. Here's another "attack"- you're entire comment is assholish and you're a dick.Camelbinky (talk) 16:13, 20 July 2013 (UTC)

Well, as your comments suggest that you don't actually understand how we work here (or understand personal attacks, which I did not do but you have), my suggestion that you reconsider seems even more apt. Moreover, no one forced Blueboar to do anything as one can see below. As this is your second comment that does not substantively respond to to my policy "Concerns" above but shows a desire on your part to put people on lists and engage in attacks, we should do that discussion at the AN/I that was opened. Alanscottwalker (talk) 18:06, 20 July 2013 (UTC)

Let me explain why I did a complete about face and proposed my most recent version ... I realized that the section in question is about the ACCESSIBILITY. It isn't about the reliability of sources... it isn't about avoiding Original Research... it isn't about the appropriate use of primary vs secondary sources... it isn't even about paper vs stone. It's about accessibility.

The stated concern was that some editors have a mistaken belief that only online sources are accessible. THAT misconception is what we are trying to correct. The fact is, all other concerns are irrelevant to this section. In order to correct the misconception, all we really need to say is "There is no requirement that sources be online. Articles may be based entirely or in part on reliable offline sources." We don't need to define what a reliable offline source is in this section... nor do we need to give examples of the different types of reliable offline sources... because this section isn't actually about types of sources... its about accessibility. Blueboar (talk) 16:15, 20 July 2013 (UTC)

That's a good point – the simple wording as it is now, that you have quoted, seems to be quite sufficient. See Rewording 'Access to sources' below for some more, though.  —SMALLJIM  21:11, 20 July 2013 (UTC)

Following on from "Does the medium matter?" above, I think that the policy subsection 'Access to sources', could be made clearer without changing its meaning. What about changing this (as it is now):

===Access to sources===
Other people should in principle be able to check that material in a Wikipedia article has been published by a reliable source. This implies nothing about ease of access to sources: some online sources may require payment, while some print sources may only be available in university libraries, or in off-line sources. Nonetheless, difficult to obtain reliable sources may still serve as sources for articles, and should not be rejected solely on the basis that it is difficult or costly to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

to this:

===Access to sources===
Although people should in principle be able to check that all material in a Wikipedia article is based on published reliable sources, this implies nothing about the ease of access to these sources. There is no requirement that sources be online, and articles may be based entirely or in part on offline sources. Some online sources may require payment and some offline sources may only be available in university libraries or archives, but sources such as these may still be used in articles: they should not be rejected solely because it is costly or difficult to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

I've changed the first sentence because it's just recapping what we've already been told, but also amended it because "...material in a Wikipedia article has been published by a reliable source" is simply wrong (unless it's a quote or a copyvio etc). I've then added in Blueboar's bit about there being no requirement for any online sources, and the rest then follows with minor changes for clarity ("offline" instead of "print" and "archives" instead of "off-line sources") and flow (such as removing that "Nonetheless" and the duplicated "difficult to obtain" and "sources", and replacing "on the basis that" with "because"). No problem if opinion is that 'Offline sources' should still be split out: in that case we'd have:

===Access to sources===
Although people should in principle be able to check that all material in a Wikipedia article is based on published reliable sources, this implies nothing about the ease of access to these sources. Some online sources may require payment and some offline sources may only be available in university libraries or archives, but sources such as these may still be used in articles: they should not be rejected solely because it is costly or difficult to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

===Offline sources===
There is no requirement that sources be online. Articles may be based entirely or in part on reliable offline sources.

All civil comments welcome :)  —SMALLJIM  21:26, 20 July 2013 (UTC)

You know, I actually like your first suggestion... it flows nicely: first we present the idea that sources may not be easy to access... then highlight this by addressing the most common complaint about ease of access (that the source is not on line), then go on to talk about a few more specific issues with accessibility of both online and offline sources. I would add just one more sentence (in bold below):

===Access to sources===
Although our readers should in principle be able to check that all material in a Wikipedia article is based on published reliable sources, this implies nothing about the ease of access to these sources. Nor does it imply that any specific person (ie you personally) be able to access the source.. There is no requirement that sources be online, and articles may be based entirely or in part on offline sources. Some online sources may require payment and some offline sources may only be available in university libraries or archives, but sources such as these may still be used in articles: they should not be rejected solely because it is costly or difficult to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

I do think it is important to tell people that a source can still be verifiable, even if you personally can not access it. Blueboar (talk) 02:34, 21 July 2013 (UTC)

If we are reworking this section, I do think that we should include something to indicate that we actually do prefer readily accessible sources for purely pragmatic reasons. I could, for example, source the fact that "Running Bear" by Johnny Preston was the best selling record in the UK to a hardcopy issue of Record Retailer, a source that can only be verified by people with access to an exhaustive collection of old British periodicals, or I can source it to http://www.officialcharts.com/archive-chart/_/1/1960-03-12/ , a source that anybody that can access Wikipeida can verify. Are both technically acceptable? Certainly. Is there a strong preference for the online source? Certainly. I'd actually be suspicious of claims sourced to ancient periodicals when the same material is available online, and it would prompt me to do extra investigation.—Kww(talk) 02:57, 21 July 2013 (UTC)

I disagree... What we really want is the most reliable sources given the specific topic... whether they are web based or dead tree (or in some other medium). Now, for pop culture topics, the most reliable sources might well be online... but this is not the case when it comes to more academic subject areas. In those cases, the most reliable sources tend to be dead tree sources (I'm not talking about out of date periodicals, but recent scholarship). Web based sources in these areas tend not to be all that reliable... and so we actually want to encourage editors to do their article research in the library and not online. I suppose it really depends on the topic area. Blueboar (talk) 03:32, 21 July 2013 (UTC)

I support Smalljim's proposal (first version). It is much better than the existing text. Smalljim's proposal (first version) is very good but it has a problem that Amadscientist identified. The rule "based on published reliable sources" is too lenient, since it could be argued to allow novel conclusions made about facts published in reliable sources and also to allow synthesis of several published sources. What we want is that all the facts, analysis of the facts, and opinions about the facts, have been published. Regarding Blueboar's amendment, I agree in principle but would simplify it to "Nor does it imply that you personally must be able to access the source." Ok? About Kww's proposal, I think matters like this belong in a guideline document rather than a policy document. (Policies state what must be and guidelines state what we prefer.) Zero^talk 04:46, 21 July 2013 (UTC)

• (edit conflict) I wish this was not just clearer, but worded better as well. I only wonder KWW if adding that specific wording of preference, would require a discussion to decide if that really were our preference. I understand your thought here, but that begins to ad weight to the argument that articles "should" have online sources or sources that are only easily accessible, free and obtainable without cost or fee. It may not really sound like much of a change, but there are many editors that have access to reliable sources that others do not. But, more importantly is the reason why and how we source and it is the same as, for example, a university level paper and sources would not be required to be obtainable on the web. As a tertiary source that might begin to limit our sources with a small amount of instruction creep. I would contend that such a proposal, might garner some interest in discussing before adding but that discussion can take place here and now. Why not. We are talking about it. I sorta presented an argument against in some manner anyway. But really...I can't help but think that the very existence of such a detailed explanation of why we accept sources that are not easily accessible, is because it is already more common, but perhaps not necessarily preferred.

I also think we need to address the main concern brought up by Smalljim. Yes, that section is badly written, but the change is slightly off as the material itself must still have been previously published. The "material" that we write is a summery of the published source. It isn't technically "based on" the source but is summarizing the information or the material. So, in that way it is actually a little more accurate to just add the word "summarized" behind the word "material". I think that saying "based on" might be too ambiguous.

Access to sources

People should, in principle, be able to check that material summarized in a Wikipedia article has been published by a reliable source. Ease of access to sources is not a requirement. Some online sources may require payment, while some print sources may only be available in university libraries, or in off-line sources. Difficult to obtain reliable sources may still serve as sources for articles, and should not be rejected solely on the basis that it is difficult or costly to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

Should be "published in" not "published by". Otherwise it refers to the publishing company, not the book or journal. Zero^talk 04:06, 21 July 2013 (UTC)

I can't help disagree with Blueboars suggestion. Sorry Blueboar. It sounds to repetitive but also a bit like it says no one need be able to access the source at all. That just isn't verifiable, although I believe what you are stating is that the source need not be accessible to a specific editor to be used. Would it sound better if we said some thing like that --Amadscientist (talk) 03:51, 21 July 2013 (UTC)

What about the wording I suggested just above your comment? Namely, "Nor does it imply that you personally must be able to access the source." Zero^talk 04:36, 21 July 2013 (UTC)

I'm also not sure the sentence about WikiProject Resource Exchange should be there. It is not part of the policy. Could we move it to a footnote? Zero^talk 04:49, 21 July 2013 (UTC)

I hesitate to remove such a useful tool for someone seeking or reading this guideline. It is directing them to a place where they may be able to obtain an easily accessible source.--Amadscientist (talk) 05:02, 21 July 2013 (UTC)

(edit conflict)You know, I think my issue is with the term "you personally". It almost makes the implication, that you are personally being overlooked or bypassed in a formal manner, but it really is about the "Not easily accessible sources" and the editor that places it there, not really about the editor that finds it. We are already guiding the editor that finds it not to rejects it just because it is difficult to obtain. We need not personalize the guideline. How about this:

Access to sources

People should, in principle, be able to check that material summarized in a Wikipedia article has been published by a reliable source. Ease of access to sources for anyone is not a requirement. Some online sources may require payment, while some print sources may only be available in university libraries, or in off-line sources. Difficult to obtain reliable sources may still serve as sources for articles, and should not be rejected solely on the basis that it is difficult or costly to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

Also, I didn't address why I changed back to people instead of reader. I tend to attempt not to inject the reader into our policies when referring to sources. In this case we cannot begin to assume what source the reader may have, any more than the editor. So in this manner we just say "people", and not editor or contributor.--Amadscientist (talk) 05:00, 21 July 2013 (UTC)

"published by" should be "published in": sources don't publish anything, it is the producers of the sources that publish them. Also, I would not understand what "for anyone" is doing there if I hadn't previously read the discussion here. It doesn't solve the stated problem. Another small suggestion is to replace "some online sources may require payment" by "some sources may require payment" since that can apply to paper sources too. Finally, "off-line" is spelt "offline" in the article as well as in its own article. Zero^talk 05:19, 21 July 2013 (UTC)

A publisher is actually part of what makes a reliable source. Part of the source is reliable publication. I think its micromanaging in a manner not needed as the publisher is, in fact, the source. If the author were the publisher, we would have a problem. Self publication is not RS. See what I mean?--Amadscientist (talk) 06:21, 21 July 2013 (UTC)

No. You are discussing the question of when a source is reliable, which is not the question being addressed in this paragraph. It is your wording that I disagree with. Books, journal articles, etc, are "sources" for us. That's the main way that the word "source" is used in Wikipedia. But books don't publish anything; you can't say something is published by a book; it is meaningless. The reliability of the book depends on the author, the publisher, the nature of the book (eg fiction vs non-fiction) and many other considerations, not just the publisher. In the final analysis, it is the book that is reliable or not, and the material we can use if the book is reliable is in the book. Zero^talk 06:59, 21 July 2013 (UTC)

I am only speaking about the suggestion of changing "published by" to "published in". For this discussion the question is, can a source publish and is that wording accurate. The fact is, all our sources are published regardless of ease of access. Books, journals etc. are all "publications". Saying something is "published by a reliable source" is accurate as a publisher is the source as much as the author. That is why I feel the wording need not be changed. However, it could just as easily be changed to..."referenced from a reliable source".--Amadscientist (talk) 07:38, 21 July 2013 (UTC)

You write "can a source publish", then you answer yourself in the negative with "our sources are published". In the question the source is the subject and in the answer it is the object. (Incidentally until now I didn't realize the wording you used was the existing wording. I thought it was your innovation, sorry about that. I still think it is wrong though.) Zero^talk 08:04, 21 July 2013 (UTC)

Maybe this is getting confusing, so let me recap. The phrase "published by a reliable source" makes sense in reference to the publisher of a work. However, I don't believe the author of that sentence was intending to refer only to the reliability of publishers. I'm sure that the reference was intended to be to the wikipedia concept of "reliable source" which involves much more than the publisher. In my opinion it is simply bad English to say that something was "published by" a book; that is misuse of the word "publish". Zero^talk 08:24, 21 July 2013 (UTC)

I did not answer in the negative. I answered in the positive. Our sources are published. That is not a negative. I think this is being overly complicated. What is a source? It is where the information comes from. Is the information published? The question for us is only whether or not the statement is accurate to be considered proper English. While the difference may be measurable in a real life discussion, I am not so sure it is here. The issue is the source. "Published in" means that the information was found in a published form. "Published by" means that the information was made available from a publisher. I truly believe this is a matter of semantics and that the two terms, while meaning separate things, are still accurate and properly phrased.--Amadscientist (talk) 08:50, 21 July 2013 (UTC)

Having said that, I could still support a re-wording of that particular portion as long as it was in full agreement in discussion.--Amadscientist (talk) 09:13, 21 July 2013 (UTC)

1. As far as I can see, the first sentence (or clause) does not provide any new information, it just provides context for the rest of the section. It should in fact reiterate or summarise the very first sentence of the policy: "In Wikipedia, verifiability means that people reading and editing the encyclopedia can check that the information comes from a reliable source" because that is what this section is qualifying. So something like: "Although this policy states that people should be able to check that the information in our articles comes from reliable sources," is then followed by the qualifier: "this implies nothing about the ease of access to these sources."

2. The above change aside, I strongly feel that we should separate the process of rephrasing the section to make its current meaning clearer from the process of adding new content or provisos to it. As soon as a new way of phrasing it is agreed and settled in (or if we decide not to change it after all), we can then go on to consider whether it should contain any new information (like adding "you personally" or "prefer readily accessible"). These are two separate processes and to mix them will make coming to a consensus much harder – as innumerable other discussions have proven.

3. If the above is accepted, then we should all be very careful about the exact wording used, because that is precisely what we're discussing. The latest proposed version has changed much of what I (carefully) rephrased without any discussion at all (for instance "offline sources" has been reverted to "print sources", "archives" is back to "off-line sources", and "on the basis that" is back too). I certainly don't expect my words to be immutable, but if they are rejected, some reason should be given.

So, with all the above in mind, may I present my latest suggested revision:

Access to sources

Although this policy states that people should be able to check that the information in our articles comes from reliable sources, this implies nothing about the ease of access to these sources. For example, some online sources may require payment and some offline sources may only be available in university libraries or archives, but sources such as these may still be used in articles: they should not be rejected solely because it is costly or difficult to obtain them. WikiProject Resource Exchange may be able to help obtain source material.

Offline sources

Sources need not be accessible online. Articles may be based entirely or in part on reliable offline sources.

 —SMALLJIM  10:06, 21 July 2013 (UTC)

I have addressed many of the things I did not agree with in good faith. However I certainly can explain areas I have not discussed already. Opening with "Although this policy states that people should..." seems indecisive and I don't feel the change is an improvement over the old language while just cutting out the superfluous text makes the policy as clear as possible. I feel if we are going to re-write the section, the changes should be concise. I don't know about others, but I feel you are almost saying that you want us to go back to your "version" and then want us to agree to a specific form of deciding this. If you wish such structure I recommend a formal RFC.--Amadscientist (talk) 10:43, 21 July 2013 (UTC)

Well I'm definitely not doubting your good faith, but I do think we should go back to the parts of my wording that haven't yet been discussed, because I really don't know whether the changes you made to the second sentence (in my version just above, 3rd & 4th in yours) were because you disagreed with my wording or simply because you based your version (which was mostly rooted in a concern about the first sentence only, I think) on a copypaste from an earlier version. The whole point of this is to discuss the precise wording of the section to ensure that it's as clear as possible, and there certainly seems to be some agreement that there is room for improvement.

Regarding that first sentence, don't you agree with my 1st point: that because this section qualifies the general principle, it should start with a clear restatement of that principle to set the context?  —SMALLJIM  12:26, 21 July 2013 (UTC)

First sentence "although": It shouldn't qualify the thought "information in our articles comes from reliable sources" and we don't want to qualify that. Arguably it qualifies "people should be able to check" but what this section really is meant to do is state what that means in practice (no matter where the source is, or its cost). Here, "although" leaves a connotation of 'but not really' and we do not want to leave that impression, even in the slightest, about any of this. A better prefatory phrase would be "It is true that" or "Under this policy it is true that" but no prefatory word or phrase would be less complex. Alanscottwalker (talk) 14:14, 21 July 2013 (UTC)

That's useful, thanks. Although I don't get those connotations from that word, I agree that we mustn't leave any doubt. What about:

Access to sources
Verifiability means that people should be able to check that the information in our articles comes from reliable sources, but this implies nothing about the ease of access to these sources. For example ...

Is that safer?  —SMALLJIM  14:41, 21 July 2013 (UTC)

Safer, yes. Alanscottwalker (talk) 14:49, 21 July 2013 (UTC)

Smalljim, you removed the words "in principle" from the first sentence. Can you argue for that? It seems like a substantive change. Thanks. Zero^talk 14:41, 21 July 2013 (UTC)'

That phrase has always seemed too vague to me, because in fact people should be able to check, if they are in the right place or have paid. Alanscottwalker (talk) 14:49, 21 July 2013 (UTC)

(ec) I think so, Zero. All the first clause is doing is setting the context for the rest of the section, and it's doing that by paraphrasing the first sentence of the whole policy. It seems to me that that's the best way to do it instead of trying to recreate the context by using very different words. This helps hammer that primary concept home again, too. It's the first of the three ideas I had this morning :)  —SMALLJIM  14:55, 21 July 2013 (UTC)

I don't want to argue against this as I think your latest edition is pretty good and we should adopt it before discussing any finer points. But for the record I think the idea of "in principle" was to add an allowance of common sense to the statement. Saying "people should be able to check" is well and good, but what if the source is in Mongolian? We don't exclude that, but "in practice" as opposed to "in principle" most people cannot verify anything in that source. But, as I said, I can leave this issue aside at least until after we adopt your current edition. Blueboar, do you agree? Zero^talk 15:11, 21 July 2013 (UTC)

Thanks! Being in Mongolian is surely just another lack of ease of access issue, fixable with money? Maybe "written in an obscure language" could be added as another example, in round 2 – we could actually add as many varied examples as would be considered helpful.  —SMALLJIM  15:44, 21 July 2013 (UTC)

You have the idea now of what I was saying but we do need to continue to discuss what the text shall read.--Amadscientist (talk) 22:36, 21 July 2013 (UTC)

"...continue to discuss..." Please do – I'm still here. Would continuing from my message above in reply to yours be a good place to start?  —SMALLJIM  10:07, 23 July 2013 (UTC)

I suggested something on this above, but it seems to have gotten lost in the shuffle... I think this is an important point to add to the section... but it is a distinct issue from other concerns being discussed, and so perhaps we should discuss it separately. Blueboar (talk) 14:33, 21 July 2013 (UTC)

Good thought but isn't it true that we want to convey that you personally can gain access if you're willing to look in the right place (and deal with the authorities where it is) or pay enough? Alanscottwalker (talk) 14:39, 21 July 2013 (UTC)

I hope we can tackle this in a second round after agreeing on a text without it. Otherwise we might never agree on anything at all. Zero^talk 14:45, 21 July 2013 (UTC)

I really get the impression a great deal was lost Blueboar as it appears Smalljim is insisting that his written versions be what goes into the article and that the discussion center on his versions over those recommended by others. I strongly disagree with that, but at least another editor got the point across about the wording of the first sentence. I had felt that the suggestion you made was sound but the idea used seemed to personalize the issue more than needed and added a small mention that stated that the source need not be easily accessible by any one in particular. I am willing to continue the discussion and re-work that to your satisfaction, but I would like this discussion to be open to all recommendations and not owned by a single editor with good intentions.--Amadscientist (talk) 22:35, 21 July 2013 (UTC)

That's why I hived this off into its own sub-discussion. I think it would be helpful to say something along the lines of what I suggested... but I'm not insisting on any specific language. All suggestions on how to phrase it are very welcome. (heck, I am not even insisting that we have to address it at all, I think it would be helpful, but if others don't agree that's fine. I just want to discuss addressing it). Blueboar (talk) 02:06, 22 July 2013 (UTC)

Blueboar, it would help if you identify the problem you are addressing. Is this it: someone might come along and say "this source is inadmissible because I can't access it"? Is it really a problem? After editing for 11 years, I'm not sure I can remember anyone trying that except for a few newbies who thought everything should be online, and they are taken care of by the online/offline sentences. Zero^talk 03:00, 22 July 2013 (UTC)

It's not just newbies. All the time the RS/N has to handle that question of "online/offline" and other accessability, and not just there. The question that set off this entire discussion several threads ago was regarding a new article that was wrongly reviewed and told that because all the sources were offline it wasn't validly done. This has been an ongoing problem that both Blueboar and User:Whatamidoing have addressed and mentioned earlier is a problem. We've had problems of geography as in a source that could only be accessed at the USS Missouri, we've had hardback books (including one article I put up for GA where the original reviewer said I couldn't have ANY sources that were offline in a GA), we've had questions regarding non-English language sources (even online), we've had questions regarding pay-to-see sources. That's just a few examples I remember off the top of my head.Camelbinky (talk) 03:10, 22 July 2013 (UTC)

Understood, but do any of these call for text in the policy that is not there now? There is already an explicit statement that offline sources are ok, and nobody is proposing to remove it. What misunderstandings are occurring that are not easily disposed of by referring to the present policy? Zero^talk 08:19, 22 July 2013 (UTC)

I think it is something that should be addressed to clarify that the inaccessibility to anyone in particular is never an issue. One of the main places this clarity may be needed is in our road and highway articles (and while this is but one example, we have a lot of them ) where information is, many times, referenced from sources requiring payment from city or county services to access maps and documentation used to properly reference an article and editors may be attempting to exclude such referencing believing that if they can't verify it themselves they may exclude it on that basis. But another place where this does come up (and where I discovered both, our policy here and the fact that so many articles do depend on this type of referencing) is in the review process such as GA, FA and all other rating or peer reviews. It is when these editors, who may be unfamiliar with this policy, raise the issue that this specific wording (in some manner) clarifies that, on top of the accessibility issue in general, those wishing to review and check these sources cannot drop the rating or exclude the article from GA or FA status based on that alone.--Amadscientist (talk) 17:38, 22 July 2013 (UTC)

OK. But that last seems like a GA/FA guideline issue. That is also a process where one is actually asking other editors to review the work and say they Approve. Think of it from the approvers perspective when he/she passes an article with a totally bogus source, and it turns out to be embarrassingly bogus for the Pedia, with news articles, and blog posts are written all over the web, etc. etc. If "fact checker" is one of the things they think they need to do, how do they do that? Alanscottwalker (talk) 19:21, 22 July 2013 (UTC)

It's been awhile since I've put an article up for GA, FA, or FL status (when I had a reviewer like I mentioned tell me all sources had to be online I quit bothering to get such a designation though many of list articles I have done would pass), but I do believe that for GA there is no requirement that they be "fact checkers", I had one GA go so thorough that by the time I had gotten everything they wanted fixed that I had begun to think they were using the FA requirements instead of GA! I agree with Alan that a lot of times these reviewers do not take the time to go through the proper checklist of what they are and are not supposed to be looking at. A big problem is that people want to do things around here without first reading what our policies are on how to actually do it. I'm starting to think- even if we rewrite policy more to hit people over the head, would it really help User:Blueboar in not having to tell people every month the same spiel?Camelbinky (talk) 19:40, 22 July 2013 (UTC)

I think I see what Alan is saying. That because it was an issue of a GA reviewer, would that not make it a GA criteria issue. No, not really as what is at issue is the actual verification of the summary not so much fact checking. Although we do want to make sure the source is reliable and that the summary used is supported. The specific GA criteria where this would be an issue is GAC#2 and it is simply a verification criteria: WP:GACR which is a three point criteria:

Verifiable with no original research:

it contains a list of all references (sources of information), presented in accordance with the layout style guideline;

reliable sources are cited inline. All content that could reasonably be challenged, except for plot summaries and that which summarizes cited content elsewhere in the article, must be cited no later than the end of the paragraph (or line if the content is not in prose);^[1] and

it contains no original research.

So this is basically an example of what may happen if someone is attempting to verify no original research and the feel stonewalled by a source that is not easily accessible by either cost, location etc., or anything that makes an otherwise reliable source, simply not available for that particular editor who feels they may be justified in using verification and/or issues as the excuse because they haven't been able to see the source.

This isn't such an issue I think, that it is essential to the wording, but could be an improvement by clarifying that no individual person must have easy access to a source to verify our policies etc.. It is a matter that the information is "verifiable" not that it is easy to do so.--Amadscientist (talk) 20:06, 22 July 2013 (UTC)

Just to be clear, by totally bogus source, I meant a source that does not actually exist, although some editor maliciously, lazily, or negligently, created a cite for it and text that it is suppose to support ("fact checker" was in quotes meaning not exactly a fact checker as in publishing but like that, and the facts I am thing of are: 1) there is this source, and 2) it says this; and the mixed question of fact/policy/prose: 3) it is reliable). Alanscottwalker (talk) 08:59, 23 July 2013 (UTC)

I had a long run-in over such a source— not that the person in question had made it up, but it was (as it turned out) passed along as a bogus citation to justify a bit of Catholic George Washington legendarium. The supposed ultimate source was difficult to get to and I was helped out by someone who was able to visit a fairly distant library and determine that the supposed source didn't exist (in this case a journal which wasn't published on the date question and which did not contain the kind of material claimed). Right now the system is gameable this way, by demanding people accept citations of works that they cannot tell exist and where we cannot tell that the editor actually read them. So in some respects the issue isn't so much verifying what the source said as it is verifying that it is a source at all. It seems to me that there needs to be some confidence that the cited source was actually used, if nothing else. Mangoe (talk) 13:06, 23 July 2013 (UTC)

Oh no, I see that above as a slippery slope of taking WP:BURDEN and having it now apply to citations and that, in my own opinion would be unacceptable and cause so many headaches and problems beyond those problems we already have with people abusing BURDEN to remove material without giving editors a chance to truly do anything. I see the slippery slope quickly leading to people saying- "It's not on the internet, I cant tell this source was really read by the editor who added it. That editor hasn't responded to that tag I left two days ago. I'm removing the information and the source". If people want to talk about changing this discussion into about verifying editor's due diligence then we should start a new thread for that, as it's getting off topic.Camelbinky (talk) 14:33, 23 July 2013 (UTC)

• This whole area is bursting with Dunning-Kruger effect. I'd boil it down to this:- Anyone who'd falsify a source is unfit to edit Wikipedia and we should be looking at lengthy blocks for a first offence. "Falsify a source" includes culpable errors, such as relying on a machine translation to source a fact based on a foreign-language source, or (in an article such as Differentiation under the integral sign) making a mistake because you're doing a calculation you're not actually competent to do correctly. Anyone who objects to a source purely because it's written down on paper in a proper book published by a proper book publisher is also unfit to edit.

  But, there's another breed of editors who doubt things that are perfectly true and widely-accepted among subject matter experts. That's a good thing if it leads them to check sources carefully. It's not a good thing if it leads them to disrupt and annoy more competent editors with persistent asshattery. Misuse of BURDEN should also be blockable.—S Marshall T/C 16:17, 23 July 2013 (UTC)

• Dunning–Kruger effect. Well OK then. But I would say I don't think that we are actually discussing blocks as that is an administrative issue. The issue before us is a re-writing of this particular part of the guideline...but I may understand what your point is. I will say that I love that link. Never heard of that before and it is interesting.--Amadscientist (talk) 00:49, 24 July 2013 (UTC)

We seem to have gotten way off track... all I am talking about is including some sort of language in the Accessibility section, to make it clearer to editors what "Accessibility" means... to clarify that accessibility does not mean that any specific person has to be able to gain access to the source. As long as a generic member of the general public is able to access it, then it is accessible. Blueboar (talk) 01:45, 24 July 2013 (UTC)

I do agree but the above was a very interesting take.--Amadscientist (talk) 02:00, 24 July 2013 (UTC)

There was a discussion on an article page Talk:Annery, Monkleigh about whether or not it's proper for an editor to use gravestones that they have observed as a source.

It seemed to me that from the WP:Verifiability guideline: information must be from "previously published information" - and excludes original research.

I was told that the footnote: "This includes material such as documents in publicly-accessible archives, inscriptions on monuments, gravestones, etc., that are available for anyone to see." means that contributors may use monuments or gravestones that they've observed.

However, the footnote refers to the sentence: "Base articles on reliable, third-party, published sources with a reputation for fact-checking and accuracy. Source material must have been published, the definition of which for our purposes is "made available to the public in some form""

Can you help me out with this? This has come up as an issue in another article (John Dennys - Funerary Monuments in St. Thomas a Becket Church, Pucklechurch, Glocestershire} and I'm just wanting to check that I've got the right interpretation. Can contributors cite the physical monuments or gravestones as a source for the article? Thanks!--CaroleHenson (talk) 18:28, 10 July 2013 (UTC)

I'm not sure exactly what your question is, but if it's about the publication bit there was a good discussion here about that just a few days back. Regards, TransporterMan (TALK) 18:50, 10 July 2013 (UTC)

Thanks, you're right that did help! My question was: Can contributors cite physical monuments or gravestones as a source for the (an) article?

It seems from the material that you provided that they can be, because they are available to the public, but are considered a primary source. I take this to mean that if there are reliable secondary sources, that would be better - but information from a gravestone or monument could be cited. Is that right?--CaroleHenson (talk) 21:48, 10 July 2013 (UTC)

I think in these cases, the best thing to do is to cite the gravestone or marker as a source for what the gravestone or marker says. That is, when in doubt, use overt in-text attribution of information, and say things like "According to his gravestone, blah blah blah" with a footnote to the marker in question. We should only speak in Wikipedia's voice where it is backed up scrupulously by reliable sources, and in the case of using something like this, it's best to make it clear and overt to the reader where the information comes from. --Jayron32 02:33, 11 July 2013 (UTC)

Absolutely. And to clarify (or more likely, having read this, complicate!) further, Carole, it doesn't apply to descriptions of monuments, only to the inscriptions that they bear: words and characters that if, say, half a dozen people were to carefully transcribe, all of them would agree on.

That clarification to the policy simply confirms that those inscriptions are "published" – the very first step in deciding if we can make use of them in our articles. We then have to consider what that use can be, by considering further questions:

• are they reliable? – usually they will be (though WP:IRS isn't written with this type of material in mind – it seems to me to make more sense to understand "reliable" for this kind of material as determining if it's genuine or a fake)
• what kind of a source are they? – primary, almost every time; and
• are they independent? – gravestones etc. usually aren't because they're written as a eulogy by someone closely involved with the deceased (a bit like a press release).

So as Jayron32 says, in-text attribution is usually necessary. And we mustn't make any non-obvious inferences from them either (that's OR).  —SMALLJIM  10:46, 11 July 2013 (UTC)

And regarding the understandable confusion that you express in your 4th sentence: that the footnote appears in a para that starts with the apparently irrelevant "Base articles on reliable, third-party, published sources with a reputation for fact-checking and accuracy." I think anyone who regularly consults the policy (any of our policies actually) would agree that after ten years of tinkering by hundreds of people it's basically a jumbled mess that ought to be completely rewritten in a logical order. But there's no chance at all of that ever happening. WP:A was a noble attempt that failed.  —SMALLJIM  11:23, 11 July 2013 (UTC)

Ah, thanks very much for the clarifications (source for inscription, not description of the gravestone), clarification of footnote intended meaning in WP:Verifiability, in text attribution, etc. That helps and I can work on edits for the funerary monument citation/notes to the best of my ability, having not seen them myself.--CaroleHenson (talk) 17:12, 11 July 2013 (UTC)

I updated footnote #6 on Wikipedia:Verifiability based upon this discussion.--CaroleHenson (talk) 18:09, 11 July 2013 (UTC)

Can we get the grammar hammered out first please? Change made resulted in:

Publicly available information, including documents in publicly-accessible archives, inscriptions on monuments, gravestones, etc., that are available for anyone to see may be cited according to WP:Primary sources. It does not include descriptions of gravestones or monuments.

Needs some work to make it more clear and less ambiguous before it goes into the policy. Zad68 18:22, 11 July 2013 (UTC)

First attempt:

Publicly-available information - including documents in publicly-accessible archives, inscriptions on monuments and gravestones, and other items that are available for anyone to see - may be cited in accordance with WP:PRIMARY. Publicly available information does not include descriptions of gravestones or monuments.

Is that what is meant? Zad68 18:24, 11 July 2013 (UTC)

That looks really good to me. Thanks!--CaroleHenson (talk) 18:30, 11 July 2013 (UTC)

Great! Updated... Zad68 18:33, 11 July 2013 (UTC)

What about when this publicly-available information is not being used a primary source? Doesn't this wording imply that it still has to be cited in according with WP:Primary when it is not in fact primary? --Atethnekos (Discussion, Contributions) 19:07, 11 July 2013 (UTC)

Well, I'm sympathetic, but this is actually wrong. You can cite a monument, or any other visual artwork, as a primary source for what the artwork looks like. See WP:USEPRIMARY's examples. I admit that, unless the article is actually about said artwork, then you're almost never going to need to do this. But it is actually a "legal" use of a primary source. Saying "There is a statue of Stalin pointing towards the clock at the end of the square.{{cite sign}}" is actually acceptable. (You can't, however, cite the sign to say that it's called the дать часы statue.) WhatamIdoing (talk) 00:49, 12 July 2013 (UTC)

There was once a similar discussion on whether you could say "The house had a red door" and cite the house as the source, obviously someone could drive by the house, verify it was in fact red. I was actually against that, but consensus said, yes you can cite anything that is verifiable, regardless of whether it was something printed or visual. A monument would be the same.Camelbinky (talk) 20:45, 15 July 2013 (UTC)

Re the last sentence: Publicly available information does not include descriptions of gravestones or monuments. It sounds like you are saying that a published description of a gravestone or monument (say in a book) would not be considered "publicly available"... which is patently absurd. If you mean that a gravestone would not be a reliable source for a description of itself, that is wrong too. Or do you mean something else entirely? Blueboar (talk) 21:18, 15 July 2013 (UTC)

Since it seems to have been the cause of some confusion, I should clarify what I meant when I said above (11 July) "...it doesn't apply to descriptions of monuments...". I wasn't very clear there, but it was in the context of Carole's questions. I was thinking of descriptions potentially made by us WP editors that go beyond the obvious Statue of Stalin example that WhatamIdoing mentioned. After going to see a statue or monument for ourselves, we shouldn't say, for instance (unless a secondary source has done so) that a statue stands upon a Carrera marble plinth supported by pillars in the Baroque style, or that a certain church monument bears the arms of Courtenay quartering Bohun, because "any educated person with access to the source but without further, specialized knowledge" (as mentioned in WP:PSTS > Primary sources > Policy) would probably not be able to verify those qualifiers. There's obviously a grey area here, but we should err on the side of caution to ensure that we don't stray into OR land.  —SMALLJIM  21:48, 15 July 2013 (UTC)

Ah... in other words, you are concerned about WP:No original research... specifically in regards to gravestones, monuments and other physical objects. Blueboar (talk) 22:09, 15 July 2013 (UTC)

Yes, though it all relates back to that minor change that we discussed and I made to the policy last month. Carole and I had been dealing with some pretty extensive OR in this area, and I felt that clarification on this point would be useful – in fact it ended up with the original editor quoting that new footnote back to us, which was quite ... satisfying.  —SMALLJIM  22:39, 15 July 2013 (UTC)

My I suggest that if clarification is needed, the place for such a clarification would be at WP:NOR and not here. From a WP:Verifiability perspective, a gravestone or monument would be treated no differently than any other primary source... limited in use, but OK to use within those limitations. Blueboar (talk) 16:05, 16 July 2013 (UTC)

You're right in general terms except, of course, that a comment posted to WT:NOR wouldn't have been much help in trying to clear up the confusion expressed above. This is actually a fine example of how interrelated the V and NOR policies are and how (IMHO) they are artificially and unhelpfully separated.  —SMALLJIM  20:45, 16 July 2013 (UTC)

I have had this ongoing debate with Wiki about eyewitness testimony, which is a primary source, and so generally considered to be acceptable to Wikipedia articles, with certain precautions. Yet what if the primary source is the editor, or reviser, himself? I feel this should be allowed, if and again, the additions are carefully and scrupulously made. I feel this is how history itself often will grow. After all, one of the finest historical, secular, testaments left to us, The Peloponnesian War by Thucydides, is to a fair degree an accumulation of witnesses... by the Author himself! John G. Lewis (talk) 07:30, 24 July 2013 (UTC)

History grows on the basis of primary sources, but it is not the role of an encyclopedia to grow history. ~ J. Johnson (JJ) (talk) 20:37, 28 July 2013 (UTC)

Please see Talk:Prince George of Cambridge#Unique and unprecedented format - disputed bullet - ricochet and Talk:Prince George of Cambridge#RfC: Bullet point in the "title and style" section. Prince George was born on 22 July and named on 24 July. Is it alright for us to claim that he was referred to as "Prince George" before 24 July? That is obviously untrue, but some claim that since the name is "backdated", it is okay to disregard the fact that no source, reliable or not, referred to this person as "Prince George" before 24 July. None of those users has attempted to cite a pre-24 July source that refers to him as "Prince George" and which would prove the claim; indeed, one of them has admitted that no such source exists. In fact, the Internet is full of sources that confirm exactly the opposite. Therefore, I must ask whether that article is an exception to the verifiability policy? Surtsicna (talk) 18:06, 2 August 2013 (UTC)

I'm not sure this is a verifiability issue, as everyone knows when he was named. It is quite normal to employ common present names for entities when referring to times before those names existed, and if there is a reasonable excuse and nobody is going to be misled, I don't see the problem. Zero^talk 02:39, 3 August 2013 (UTC)

I'll add that since a large fraction of babies are named a day or two after birth, hundreds (maybe thousands) of sentences in Wikipedia of the form "SoAndSo was born on SomeDate" would be wrong if we got pedantic about this point. Let's use common sense instead. Zero^talk 02:42, 3 August 2013 (UTC)

Agree with Zero. This is a bruhaha over nothing. This sort of thing happens all the time, definitely a situation to use some common sense. FurrySings (talk) 14:29, 3 August 2013 (UTC)

Zero, nobody's saying that it's wrong to refer to George as "George" when discussing events that took place before he was named. It is, however, wrong to explicitly state that he was formally styled as "HRH Prince George of Cambridge" before anyone even knew his name. Nobody styled him as "HRH Prince George of Cambridge" before the name George was announced. That is a simple, undisputable historical fact. The source used to support the claim does not support it; instead, it proves exactly the opposite by saying that George will be styled as "HRH Prince George of Cambridge". Surtsicna (talk) 22:35, 3 August 2013 (UTC)

I agree with Zero and FurrySings. This is silly. A Quest For Knowledge (talk) 23:44, 3 August 2013 (UTC)

I also agree that this is not an issue worth major attention. Appropriate wording can be worked out in the article or on the talkpage, but too much time spent focusing on this nuance is probably not productive. Newyorkbrad (talk) 23:48, 3 August 2013 (UTC)

All attempts to finesse this with a different wording (such as not including a date where the name-and-style are introduced, or conversely, not including the name where the date and style are given) are being gang-reverted. Addition of appropriate tags (notably {{not in source}} for claims that are... not in the given source) meet a similar fate. Discussion on the talkpage proceeds in a circular manner: 'It doesn't matter that this isn't verifiable; it doesn't matter that it's verifiably not correct; it's the way we like it, and have always done it.'

While I agree that a disproportionate amount of time has been expended on this this already, it would seem to set a bad precedent that unverified -- and inherently unverifiable, due to being very publicly demonstrably incorrect -- can be maintained in an article by sheer pressure of "local" editors filibustering on the topic. (If that precedent's not already been set some time ago.) 84.203.39.131 (talk) 04:07, 4 August 2013 (UTC)

Come now, we're splitting hairs here. The child is verifiably named Prince George, and the fact that for about 48 hours there was no announced name is already noted in the article. It contorts the language within the article in question to scrub all references to the child's life during that ~48 hour period to not say George. It's really not that big of a deal. --Jayron32 04:11, 4 August 2013 (UTC)

(Though the fact it was publicly announced that no name had at that point been decided is not, though I supplied a reference for that fact on request on the talk page, a line of discussion which seems to have since been "quietly dropped".) That's not what's being suggested here. No-one is proposing the he be referred to as "nameless royal sprog" for the period July 22-24, and "George" only thereafter. This is specifically about the line:

• 22 July 2013 – present: His Royal Highness Prince George of Cambridge"

Apparently it's not acceptable to omit the "22". Nor is it acceptable to omit the first name for that period. No-one is willing or able to produce a source for this claim. It's not acceptable to add a tag noting that this information is unsourced. Replacing the bullet points with a prose (near-)equivalent has also been rejected and reverted. It's difficult to see what "appropriate wording" can be arrived at, if a group have editors have predetermined that only one possible form of words and format of expression can ever be used due to "royal article convention". 84.203.35.99 (talk) 05:48, 4 August 2013 (UTC)

And why is it so critical to get the right version up now, instead of waiting for dispute resolution to run its course? Someguy1221 (talk) 05:58, 4 August 2013 (UTC)

It's not, though the sooner the righter the better, no? (Note that I didn't start this discussion here, haven't been editing the article at all recently, not least because it's semi-protected due to other silliness, and that I wasn't aware of this article being in "dispute resolution" as such.) 84.203.35.31 (talk) 02:18, 8 August 2013 (UTC)

I removed this section (before any comments had been posted) as it was a duplicate of RSN, and I still think the discussion should occur there, not here because this is not a noticeboard to discuss individual cases. Johnuniq (talk) 06:06, 4 August 2013 (UTC)

Actually, this article should be deleted as it fails WP:BLP1E. :) A Quest For Knowledge (talk) 12:32, 5 August 2013 (UTC)

But an article failing WP:BLP1E should not be deleted, it should be turned into an article about the event. :-P Diego Moya (talk) 16:14, 5 August 2013 (UTC)

I agree, this isn't the optimal location, but felt it was appropriate to raise some counters to some arguments raised here, IMO erroneously. 84.203.35.31 (talk) 02:18, 8 August 2013 (UTC)

Do we have a template half-way between {{Failed verification}} and {{Dead link}}? For use, for example, when a statement is sourced to a web page, with an access date, and the fact was on that web page at the given date, but the web page has been updated and the information is no longer there, even though the link is not "dead"? {{Failed verification}} seems to imply that it was never there. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:26, 19 August 2013 (UTC)

I don't think so, but you might consider {{verify source}} as an option. That would suggest at least that some work needs to be done to chase down the information. WhatamIdoing (talk) 03:08, 20 August 2013 (UTC)

Currently in the section Access to sources, the first sentence is,

"Other people should in principle be able to check that material in a Wikipedia article has been published by a reliable source."

Suggest changing "material" to "information" since "material" suggests the verbatim text, when what is actually meant is the information contained in the text. The result would be,

"Other people should in principle be able to check that information in a Wikipedia article has been published by a reliable source."

Please comment and indicate Support or Oppose. --Bob K31416 (talk) 16:40, 27 July 2013 (UTC)

Comments:

• Nuetral not much material difference, but it would seem that material is arguably broader or just as broad as sometime we do use verbatim quotes. Alanscottwalker (talk) 16:49, 27 July 2013 (UTC)
• Oppose change  I've seen this issue before.  From my viewpoint, "material" continues to be the preferred word.  Unscintillating (talk) 18:42, 27 July 2013 (UTC)
• Prefer a wider-ranging rephrase. What we have is a ponderous, bloated mass full of needless words and complex phrasing. At the moment that one paragraph repeats the word "source" eight times in five sequential sentences. It also says "obtain" three times. It only uses "material" twice, but in a different sense each time (once for material in an article, once for source material). It's full of wordy circumlocutions, such as the passive voice that I'm apparently not allowed to fix, or poncey verbose phrasing like "solely on the basis that" when we could say "just because".—S Marshall T/C 11:30, 28 July 2013 (UTC)

I could see much of that. Why not propose the wording you think is decent (points will be ironed out and done). Alanscottwalker (talk) 12:10, 28 July 2013 (UTC)

S Marshall (sidebar) you are brilliant and do excellent work. But you should realize that when you are talking about active vs. passive, it usually isn't really about active vs. passive, it's usually about:

1. What you call "passive" is usually something that specifies what the "end result" should be
2. What you call "active" is usually specifying/commanding a step that is (only)usually applicable towards that end result.

Sincerely, North8000 (talk) 12:18, 28 July 2013 (UTC)

• It's about grammar, mate. "...material in a Wikipedia article has been published by a reliable source" is a passive voice construction. "... a reliable source has published material in a Wikipedia article" is the active voice equivalent. Style guides generally prefer the active voice, all else being equal, because it's less wordy and easier for most people who don't use English as their first language.—S Marshall T/C 12:57, 28 July 2013 (UTC)

I was more reflecting on your general comment (which covered past proposals). My comment is really not applicable to / useful for the specific case that you just described. Sincerely, North8000 (talk) 13:15, 28 July 2013 (UTC)

S Marshall, North8000, Would you care to comment on the topic of this section, which is a proposal to change one word in one sentence? Support, Oppose?--Bob K31416 (talk) 13:53, 28 July 2013 (UTC)

What I said was, I think it doesn't go far enough. Fraid that doesn't distill down into a single word in bold, sorry about that.—S Marshall T/C 18:51, 28 July 2013 (UTC)

It wasn't meant to cure everything that you or anyone else wanted to improve. It was only meant to be an improvement. So, in your opinion, would the present proposal be an improvement over what is currently there? --Bob K31416 (talk) 20:44, 28 July 2013 (UTC)

• It wouldn't make it any worse, but I'd prefer a proper re-edit of the whole paragraph.—S Marshall T/C 21:28, 28 July 2013 (UTC)

Would the proposal be an improvement? --Bob K31416 (talk) 21:46, 28 July 2013 (UTC)

In answer to Alanscottwalker above, an example of what I think is better phrasing might be:-

Does that seem simpler and clearer?—S Marshall T/C 21:28, 28 July 2013 (UTC)

• Yes Good . . . but -- you knew that was coming -- put a parenthetical "(and its use conforms to other policies)" after reliable in the 4th sentence. Alanscottwalker (talk) 21:56, 28 July 2013 (UTC)
• Just looking at the first sentence, your change in order left a clumsy phrase that suggests a reliable source is publishing in Wikipedia, i.e. "a reliable source has already published the information in a Wikipedia article". However, I did like your change from "material" to "information". Trying to work with your first sentence, you might want to change it to, "In principle, people should be able to check that the information in a Wikipedia article has already published in a reliable source." --Bob K31416 (talk) 22:09, 28 July 2013 (UTC)

• Oppose. Sorry Bob. Your word change avoids confusion over the meaning of the word "material", but there's no point tinkering with broken text. As I proposed above, that first sentence should follow the wording of the first sentence of the policy lead – that's what this section is qualifying so why continue to try to say the same thing in different words? That policy lead sentence has been stable for one year now [10] following the RfC, and this section should have been amended to match at that time – it still uses wording based on the rejected "verifiability, not truth" lead (here, for example).  —SMALLJIM  22:18, 28 July 2013 (UTC)

Actually, you seem to support "information" over "material" when you wrote "avoids confusion over the meaning of the word 'material' ". I don't mind if the change from "material" to "information" is eventually incorporated in a big change, rather than in an incremental change like this proposal. However, there may be less chance of it ever being incorporated if agreement can't be reached on a larger change. --Bob K31416 (talk) 23:02, 28 July 2013 (UTC)

P.S. I notice that there are proposals below that avoid the issue by not using the first sentence, which seems OK to do. However, if they are not accepted, we are still stuck with "material" instead of "information". --Bob K31416 (talk) 23:56, 28 July 2013 (UTC)

How about a complete rewrite that avoids the issue all together:

That also cuts the paragraph down to its basics Blueboar (talk) 22:28, 28 July 2013 (UTC)

• Ooh, yes, that's lovely. I can make it even more basic, though!

Is that being too greedy?—S Marshall T/C 22:39, 28 July 2013 (UTC)

I could live with that. Blueboar (talk) 23:10, 28 July 2013 (UTC)

Me too - something along these lines neatly sidesteps my main objection to the first sentence.  —SMALLJIM  16:44, 29 July 2013 (UTC)

You might want to change the first two sentences in your latest version to "Some reliable sources may not be easily accessible. For example, they may charge a fee or may only be available in a specific library or other archive."
Also, you might want to reword "they may charge a fee" since it's not the article that is the reliable source that charges a fee.
However, I'm uncomfortable with an idea of this section. Suppose an editor supports material with a reliable source that the editor says is available in a library in Tuva, and no one can find it anywhere else. If it looks like OR or just made up, what do you do then? --Bob K31416 (talk) 23:32, 28 July 2013 (UTC)

Similar concerns but mine involve "primary sources." A "new" letter of 'Famous person' that so and so says exists but is unconfirmed by secondary sources could dramatically alter scholarship. I know we all want clarity but we have the cases where a "fundamentalist" reading of policy is just not right. Alanscottwalker (talk) 00:34, 29 July 2013 (UTC)

@Bob... That's the point of the section... you don't have to ever personally see the book... someone else can check it for you. If you question the accuracy of something cited to a source found in an obscure library in Tuva, find someone in Tuva who can check the citation on your behalf. It may take a while, it may take a lot of effort... but if you put enough time and effort into it, the information can be verified.

@Alan... you do realize that we are actually allowed to cite primary sources... albeit in limited circumstances. This might include citing a "new" letter just discovered in a public archive somewhere. Besides, the primary/secondary nature of the source has nothing to do with its accessibility. Blueboar (talk) 01:23, 29 July 2013 (UTC)

Blueboar: you do realize that authenticating letters is not something we do here. Of course, accessibility has everything to do with whether a new document is accepted in scholarship, anyone who says there is a new document but refuses to produce it is scorned in scholarship (see eg, the Kramer documents discussed in Sexuality of Abraham Lincoln#Historical scholarship and debate or The Protocols of the Elders of Zion, etc.). Alanscottwalker (talk) 01:36, 29 July 2013 (UTC)

I agree and we have to be careful to not write rules so black and white that they allow absurdities. If someone pops up and claims to be quoting from a letter they found in a public archive in Tuva, which letter has never even been mentioned in a secondary source, I for one won't allow it. It would be an invitation for malicious editors to fake sources — exactly the sort of thing that verifiability is supposed to prevent. Zero^talk 01:43, 29 July 2013 (UTC)

Perhaps this issue can be addressed by adding to the section, "If questionable material is supported only by a source that is not easily accessible, note that verifiability does not guarantee inclusion." --Bob K31416 (talk) 15:17, 29 July 2013 (UTC)

• No, please don't do that. The last thing Wikipedia needs is another way for subject matter experts with access to extensive specialist libraries to be challenged by Stuart from Where the Hick, Nebraska, age 14.—S Marshall T/C 15:29, 29 July 2013 (UTC)

The so-called expert with alleged access to a unique place where the reliable source is kept could be "Stuart from Where the Hick, Nebraska, age 14", or could be Joe the OR pusher, etc. The section Verifiability does not guarantee inclusion was put in policy for good reason and if questionable material is supported only by a source that is not easily accessible to check, that section is very appropriate. For reference, here is what that section says.

"While information must be verifiable in order to be included in an article, this does not mean that all verifiable information must be included in an article. Consensus may determine that certain information does not improve an article, and that it should be omitted or presented instead in a different article."

So it comes down to consensus, not just Stuart or Joe. --Bob K31416 (talk) 16:08, 29 July 2013 (UTC)

Not sure it's needed now that the bolding and absolutist caste is not in the current policy. In reply to SMarshall agreed, but the "last thing Wikipedia needs" is many things (no personal pov pushers, no hoaxers, included) but general policy is written for all these. Alanscottwalker (talk) 16:18, 29 July 2013 (UTC)

The content of an alleged letter in Tuva is covered by WP:EXCEPTIONAL anyway, isn't it?  —SMALLJIM  16:42, 29 July 2013 (UTC)

In part, but ...Amelia Earhart "loved flying anywhere in the world," and she could "think of nothing better than taking off into the boundless blue." [Earhart Tuva Letter 14.369] ... may not seem exceptional but it is a total lie and completely unvarifiable because I just made it up. You do touch on another point, a cite in Wikipdeia is itself "information in a Wikipedia article" that asserts that the source exists. Alanscottwalker (talk) 17:10, 29 July 2013 (UTC)

Which is why the source not only has to exist... but exist in a location that is open to the public: A library, archive, museum, etc. Let us assume that the "Earhart Tuva Letter" did exist... if it was located in the Tuva Public Library... then it would be verifiable. Anyone who is willing to fly to Tuva is able to verify it. In fact, you don't even have to be willing to fly to Tuva... because in our modern age it isn't that difficult to contact someone in Tuva to verify on your behalf). Blueboar (talk) 00:13, 8 August 2013 (UTC)

Well and good, but we would still not to want to write this small part of policy, so as to cut off the original research inquiry. There are serious OR issues(not to mention other issues) should Wikipedia announce to the world that a new Amelia Earhart letter exists. Beginning with, is it really written by Amelia Earhart? Alanscottwalker (talk) 20:10, 21 August 2013 (UTC)

If a user don't provide a source is that means he/she is lying? 174.89.25.75 (talk) 14:34, 22 August 2013 (UTC)

It's possible, but you should always assume good faith. Perhaps they didn't realize they should provide a source, perhaps they didn't realize they didn't provide one, etc... DonIago (talk) 14:50, 22 August 2013 (UTC)

What is the role of verifiability in determining the gender of the article subject? For example, suppose that a notable subject spent the majority of their life (including the period of time when they had the most notability) being identified as a male, by reliable sources and in self-identification. Then the notable subject announces that they have started self-identifying as a female. At this point, multiple reliable sources start identifying the subject as a female, while multiple reliable sources also continue identifying the subject as a male. How does verifiablity play into this? From my understanding, the normal resolution to a disagreement between reliable sources is to use in-text attribution, but this does not seem possible when the male/female disagreement affects almost every pronoun in the article which refers to the subject. There are also WP:BLP issues to consider, so these changes may be applied before discussion and consensus can be formed. Does anybody have any thoughts about this? Kind regards, Matt (talk) 07:09, 23 August 2013 (UTC)

Note: Matt Heard's post is the result of the Bradley Manning/Chelsea Manning matter. It's a topic that is popping up all over Wikipedia. Flyer22 (talk) 07:19, 23 August 2013 (UTC)

Thank you, Flyer22. Looking back at what I wrote, I should have been clearer about that. I was purposely being as general as possible so that the questions could be answered about the Verifiability process rather than about the particular article about Manning. Kind regards, Matt (talk) 04:13, 26 August 2013 (UTC)

• Use the gender pronoun the reliable sources use. Always follow the sources, don't try to be ahead of them.—S Marshall T/C 17:13, 23 August 2013 (UTC)

The Manning situation is an oddball one in that I don't believe we have any actual source, reliable or not to demonstrate that Manning himself is stating the self identification. Lawyers are not the subject and we cannot know that it is their actual wish, or some legal tactic, or even true altogether. Always follow the source. If the subject contacts someone (which is doubtful in this case) it could be amended at a later date, but for now we do summarize just what the sources state.--Mark 18:26, 23 August 2013 (UTC)

The answer to all these questions is simple... have a some patience. Let a little time pass... so we know whether the sources accept or reject Manning's name change. Blueboar (talk) 01:43, 24 August 2013 (UTC)

@Mark: Should we not consider for the sake of self-publication that a lawyer is a reasonable representative for a subject? Kind regards, Matt (talk) 04:13, 26 August 2013 (UTC)

@S Marshall: What should be done in situations where some reliable sources use "he/him/his" and other reliable sources use "she/her"? Kind regards, Matt (talk) 04:13, 26 August 2013 (UTC)

Current consensus on how to deal with such pronouns is at MOS:IDENTITY. Verifiability applies to direct quotes, where the exact pronoun used by the source is copied. Sentences made in Wikipedia's voice use the pronoun of the latest gender by which the BLP subject self-identifies - provided the self-identification is itself verifiable. Diego Moya (talk) 09:50, 26 August 2013 (UTC)

Matt Heard pretty sure we can consider the statement to have come from Manning. My point was badly made and was simply pointing out that we have to follow the source to verify that this is the subject making these identifications. As long as that is satisfied (and from everything I am seeing, I assume it is) then MOS:IDENTITY should apply and BLP policy does apply across the project broadly, including article titles.--Mark 10:05, 26 August 2013 (UTC)

WP:DOB (Privacy of personal information and using primary sources. Bolding for emphasis): Wikipedia includes full names and dates of birth that have been widely published by reliable sources, or by sources linked to the subject such that it may reasonably be inferred that the subject does not object.--Mark 10:17, 26 August 2013 (UTC)

• You know, it is amazing how waiting even just a few days will clarify issues over breaking news... despite all the angst and knee jerk contention, it turns out that there is no need to change this policy in any way. We now actually have reliable sources that discuss Manning's desire to shift from "Bradley" to "Chelsea" (and sources that discuss the shift of pronoun from "he" to "she"). It is not only verifiable, but verified. WP:VERIFIABILITY is no longer an issue (there may be other policies and guidelines that need to be discussed... but not WP:V). Blueboar (talk) 14:59, 26 August 2013 (UTC)

One should ask at what level the relevant information resides in the real world that we should focus on. I explain this here and here. Count Iblis (talk) 15:07, 26 August 2013 (UTC)

On several occasions in the past, and in a certain recent discussion, some people have made the argument that using a person's new name when discussing periods of their life during which they were known by another name violates wp:Verifiability, wp:No original research and/or wp:Synthesis. (If need be, I can dig up examples of people making this argument.) I'd like to start a discussion of whether or not that is the case. I appreciate your help in ensuring that this remains a discussion of that specific question and does not become a discussion of whether or not retroactive use of a new name violates any other policy and/or is a good idea. I would also appreciate your suggestions of a better forum for this discussion, if you can think of one.

Say we're writing an article about someone who was born 'Pat Xyz', discovered foobarium as 'Pat Xyz', then started going by 'Lou Zyx'. Say we have reliable sources confirming each of those things. Say, for purposes of this discussion, that our article on this person is at [[Lou Zyx]]. If we have sources saying "Xyz discovered foobarium", is it a violation of VERIFY, NOR and/or SYNTH for us to write "Zyx discovered foobarium"?

For my part, I contend that it is not, because we have sources verifying that Pat Xyz = Lou Zyx. I don't think anyone would ever argue that it was a violation of any of those policies to write "foobarium is water-soluble" in our article on foobarium, even if the only available references said "foobarium is soluble in H2O", so I admit that I can't even discern what the argument to the contrary is.

I've posted notices of this discussion to WT:NOR and WT:MOS; if this discussion is moved, those notices should be updated. -sche (talk) 17:34, 30 August 2013 (UTC)

None of WP:V, WP.NOR or WP:SYNTH apply to a case where the change of name is documented by reliable sources and not challenged by any. It is only a matter of writing style. If someone is very well known under one name, the "principle of least surprise" may mean using that name for a longer period (together with a brief explanation). For example, former Israeli prime-minister Yitzhak Shamir was called Icchak Jeziernicky for his first 35 years at least, yet the article uses "Shamir" throughout. That is fairly typical. Perhaps the correct place for this discussion is some talk page associated with WP:MOS. Zero^talk 00:04, 31 August 2013 (UTC)

Could you clarify what you mean by "the change of name is [...] not challenged by any [reliable sources]"? If one source says Pat started using the name Lou, but another source calls that "a persistent myth", then it's clearly necessary to verify whether or not the name change occurred. But say all sources acknowledge that Pat started going by Lou (and thus acknowledge that references to "Lou" are references to Pat), but some sources dispute that it was (legally/morally/etc) appropriate/valid for Pat/Lou to change names. (This is the scenario in which people typically make the arguments I'm asking about the validity of.) If enough reliable sources accept the name change as valid that Wikipedia has decided to use "Lou" in reference to the person's post-name-change actions, does there need to be a separate, additional discussion of whether or not using "Lou" in reference to the person's pre-name-change actions is verifiable? -sche (talk) 07:58, 31 August 2013 (UTC)

To me, the WP:V question is, "can we verify they changed their name," not "can we verify everyone else follows our style about the retroactive use of names" or whatever, for the reason that as you said the two names mean the same thing. Really, you could ask the same thing about anything in the MOS. For instance, we had a whole super duper long discussion about whether changing a hyphen to a dash, per MOS dash rules, in the comet hale–bopp article raised verifiability or OR concerns because seemingly all astronomers use a hyphen, but then really the outcome was that we follow the MOS over styling even if on a particular matter our styling is out of synch with RSs. I don't think it's that different here, really. There is no substantive change being made in this situation because the two names refer to the same person, so it's a styling matter not a substance matter. Ymmv. AgnosticAphid talk 14:37, 31 August 2013 (UTC)

To give an example of when we use a person's old name (and all but ignore a name change)... we have William Pitt, 1st Earl of Chatham... in this article we consistently refer to the subject as "Pitt" (and down play "Chatham") throughout the text. It really does depend on the specific person and what name readers will expect to see presented. Blueboar (talk) 15:31, 31 August 2013 (UTC)

I agree that whether to make retroactive use of a later name, continue using an earlier name, or use both names is something to be decided on a case-by-case basis. It occurred to me earlier that I hadn't noted (so I will now note) that if Wikipedia decided to use "Pat" throughout the article on my hypothetical "Pat Xyz"/"Lou Zyx", even when discussing periods of their life after which they'd come to be known as "Lou", that would also not violate WP:V, IMO. (And one would logically expect that the people who do think using "Lou" when discussing "Pat"'s childhood violates WP:V must also think that using "Pat" when discussing "Lou"'s adulthood violates WP:V. But what I'm suggesting is that any anachronistic use of a personal name, while it might be good or bad for other reasons, does not violate WP:V. And everyone here seems to agree on that point.) -sche (talk) 22:49, 31 August 2013 (UTC)

This reminds me of the movie Stripes. "My name is Francis but everyone calls me Psycho. Call me Francis and I'll kill you." There are two issues that conflate this. One is a simple name change and discerning its scope, notability, verifiability, etc. For example, if someone has a nickname and it becomes well-known, it may be appropriate to use it in the lead along with the proper legal name. Other organizations where names are part of a uniform, they require legal name changes (i.e. Metta World Peace was Ron Artest and had to legally change his name, Chad Johnson wanted his uniform to reflect "Ochocinco" and it was denied by the NFL until he legally changed his name.) I think it serves the encyclopedia to use the legal name, then common name and nickname as part of the name. It is still a case by case basis though because some pet names like "sweetypie" might not be appropriate whereas Bess Truman is an example of the pet name becoming the common name, article title and lede. Case by case is appropriate in less contentious articles. But the 2nd issue which is Wikidrama of the week is a name and gender change when neither has legally or anatomically been changed and all the notability came under the previous identity. Even if we accept they are the same person and there is no confusion (H2O and water analogy), there are very real differences in pronouns and the equivalence breaks down. The use of one name or the other is perceived as the declaration of gender, rather than the person. Wikipedia has to balance the preference of the person with the legal status of that person (both name and gender) in regard to past, present and future. Our sources generally don't redact their articles to reflect a change. Even if they are all electronic (i.e. Slate), the article doesn't search/replace every reference and neither should WP especially if the change isn't particularly relevant to the notability of the subject and the change has not been recognized legally or anatomically. In the present drama, Manning is the topic of debate. His notability was for an illegal act (personally, I don't even know why he has an article as his life is not particularly notable outside the single event). Manning recently said he was female and his name was now "Chelsea" instead of "Bradley." All fine and dandy except no body of authority recognizes it. Nor will he use his new name and gender in future litigation or incarceration. The court is the authority over his name, the Army is the authority of his rank and the examining doctor will determine his gender. It is acceptable to note Manning's preference or declaration but the wholesale change of every reference to Manning to reflect his preference is not supported by reliable sources. If Michael Phelps came out and said he was female, the Olympic committee would not suddenly acquiesce and allow him to compete as a woman nor would the record book redact all references to his accomplishments in Men's swimming and change it to Women's swimming. There is an authoritative physical exam to determine gender (and yes, there have been intersexual athletes that have been barred from participating as the gender they identify with) and the name engraved on the trophies is the name at the time. In the case of Manning, all of his notable acts occurred when he was "Bradley." That may be very relevant to his username for the systems he compromised, it's certainly on all the charging documents and his enlistment in the Army. He was certainly notable under the name "Bradley." The next phase of his life will be prison. A doctor will examine him and determine his gender to assign him to the appropriate barracks including whether he should start a gender reassignment medical regimen. That is the authoritative source of his gender. Manning's self-identity as female is currently not supported by any medical exam and is simply his own belief. There is a difference in recognizing his personal identity while also recognizing it is not presently supported by professionals that would be considered reliable sources. We have seen many people claim an identity (i.e. Native American, African American, Hispanic) that is challenged by reliable sources (or not proven by outside sources). Only a doctor with transgender training would be able to prescribe or perform any reassignment surgeries and I personally doubt that doctor would make that determination while Manning is under a tremendous amount of stress. Gay, transsexual and transvestite are not the same and only one would be a condition diagnosed and treated by a physician. Therefore, self-identification as gay, bisexual or transvestite has a much lower threshold than transsexual where a medical diagnosis would be required to start a gender reassignment regimen. Manning may very well be female and eligible for gender reassignment treatment but he isn't going to attain that until a doctor says so. That's the reliable source standard. When that happens, the gender of the sourced reference should be maintained either chronologically or for notability. If the real world operated solely on the basis of self-identification of gender, there would be no male criminals. They would just declare themselves female and expect society to accept it and house them in female prisons. If they declared they were gay, transvestite or bisexual, no one would care and they would still go to a male prison. The bar for claiming a medical condition has a necessarily higher standard for sources. --DHeyward (talk) 23:56, 31 August 2013 (UTC)

I apologise if my original post wasn't clear... if you re-read it, I think you'll see that you haven't addressed it. Your argument for why Manning's name and gender change isn't 'valid' is well thought-out (though I disagree with it); you clearly think that Manning shouldn't be referred to as Chelsea at all. However, this is a discussion of something different, something specific: several people, including you, suggested that using a name or pronoun retroactively — or using a source that said "(name A) did X" or "he did X" to support a sentence in an article that "(name B) did X" or "she did X" — raised verifiability concerns; other people suggested it raised synthesis concerns. Those are the concerns I'm addressing at the moment. -sche (talk) 06:38, 1 September 2013 (UTC)

I understood it, and for the record my preferred name in the lede is Bradley Edward "Chelsea Elizabeth" Manning. I also note he used "Breanna" previously as well. My concern is that editors believe that the pronouns used in an article become hostile when someone self-adopts a gender change but apparently only hostile when Wikipedia uses them. No one has "Slate" redo all their articles to reflect the latest incarnation. We didn't do it when Manning used his "Breanna" persona, nor did our sources. I don't consider the pronouns hostile, however, as they accurately reflect his identity when it was sourced. To avoid battleground issues, it would be best to leave the pronouns and gender as they are found in sources. WP isn't in the gender identity business so pronouns should generally reflect the sources unless there is high bar (i.e. reliable source other than just the persons wishes). Manning, so far, doesn't appear to have done anything notable as "Chelsea". There is a stronger argument that s/he became notable when s/he was using the name "Breanna." If we look at Manning through his Gender Identity Disorder defense arguments for mitigation, he blamed GID for emotional instability, assaulting a female superior officer and leaking classified documents. The wholesale acceptance of that, without 3rd party reliable sources, stigmatizes people that have GID as being emotionally unstable, potentially violent and untrustworthy for a security clearance. I don't have a particular problem with revising an article that reflects a reliably sourced gender change (i.e. medical, legal, etc). But prior to that, I am inclined to mention the gender claim while still reflecting the gender status in the sources because otherwise it implies we accept GID on the claim by the person at face value. I don't think that is fair to people that have been diagnosed, treated, and taken medical and legal steps to become who they are. Transexual issues have a higher bar than other LGBT issues simply because it is classified as a medical condition. No one blinks an eye at pronouns like "His husband" for two males that marry. Gay is not a medical condition. Manning however said a DSM psychological disorder (GID) caused him to do illegal things. Understandably, for those that also have GID, the bar should be pretty high for that claim. Imagine a hypothetical case 40 years ago, a married man was caught molesting children and used a DSM classification of homosexuality as a medical disorder that caused him to commit those acts. Would we rush to make sure that every reference to him reflected his sexual preference? It would not be tolerated to replace "The married father molested three children..." with "The gay father molested three children..." if he tried to mitigate his crime with the now outdated DSM disorder. Today, I hope we would see that it's not particularly relevant to the bio or the crime. Manning mitigated his crime with the GID defense and we should be very reluctant to simply accept it. It's worth mentioning what he wishes to be called and what he considers himself to be, but a wholesale retroactive change based on his press release, seems premature and unenlightened until/unless he takes steps that others take (including diagnoses, medication, surgery, name change, etc). If the Army doctor for Leavenworth, for example, said he can't be housed there because Manning is female, I'd have no problem with changing pronoun references. --DHeyward (talk) 22:25, 1 September 2013 (UTC)

The name issue cannot possibly be a problem. Take, for example Metta World Peace. He lived most of his life as "Ron Artest" and most of the sources for information on his page are to publications that pre-date his name change, and thus refer to him as "Ron" and "Artest" with no mention of his current name. The section on the Pacers–Pistons brawl, for example, uses sources from 2004, 7 years before the name change. But if we are going to take seriously the worry about name changes and verifiability, then we have to ask why is information about a brawl that a source says involved "Ron Artest" on a page for a person whose name is "Metta World Peace". The answer is because we have some reliable sources that Artest was involved in the brawl and other reliable sources that Ron Artest is the same person as Metta World Peace. We do not need to find some new source that says "Metta World Peace was involved in a brawl in 2004". The idea that there is a verifiability issue here is absurd.

For gender, take the following real example: 11 year old future singer Robin Thicke was being babysat by hockey legend Wayne Gretzky on the day that Gretzky was traded from Edmonton to Los Angeles. This quirky fact has been deemed interesting enough to be included on Thicke's page. The source that is cited for the information uses a male pronoun when mentioning Robin Thicke twice in a throwaway comment. The only other indication in that source that the source believes that Thicke is male are two uses of the word "son" (to define his relationship to Alan Thicke). But it is easy to imagine the same article without these mentions of gender. If it had been written with no male pronouns and no uses of the word "son" for Robin Thicke, someone who took the gender verifiability question as seriously as some have would be forced to say that we cannot conclude that the source believed that Robin Thicke was male and so the Wikipedia page cannot say, as it does right now, "Wayne Gretzky had been babysitting him while his father was on vacation...." That's also absurd. 99.192.79.173 (talk) 15:59, 1 September 2013 (UTC) (99.192....)

I've seen two comments made elsewhere which I think inform discussion of this issue: this opinion about why names and pronouns are used, and this opinion that retroactive use of names violates WP:SYN. I have asked the author of the latter post to copy the post over here, so it can be better discussed in this central place. -sche (talk) 20:20, 1 September 2013 (UTC)

Right now, WP:SOURCEACCESS holds that

"Do not reject sources just because they are hard or costly to access."

However, since the WP:NOT holds that Wikipedia Encyclopedia is an 'online community of individuals,' suggest that a priority should be given to sources that are more easily accessible to the majority of editors; e.g. online articles, famous journals, instead of obscure books. This will solve any problems that are caused by subjective interpretations of such sources.

In the same way, while WP:RSUE prefers English-language sources more than non-English ones

"English-language sources are preferred over non-English ones"

It is not too explicit and there are instances of non-English sources being used when English ones could be used instead. With such sources, English-only editors have trouble editing materials based on such sources, even if they can contribute to the topic in question.

So, in short - propose that "Sources more easily accessible and written in English should be valued over their counterparts, and editors are encouraged to replace inaccessible and non-English sources if they can, with more accessible ones." kkj11210 (talk) 03:54, 31 August 2013 (UTC)

That would privilege bad sources over good ones. The policy already says that English sources should be preferred when "of equal quality and relevance". Your suggestion would give preference to an article in The Daily Mail over an article in a highly-respected German history journal. --Hegvald (talk) 16:13, 31 August 2013 (UTC)

• Other editors' ignorance of foreign languages is not, and has never been, the article writer's problem. If you can't check a source because it's in a language you never learned, ask an editor who can for help.—S Marshall T/C 18:48, 31 August 2013 (UTC)

• I would definitely oppose favoring on-line sources over hard copy "dead tree" sources. Quality is more important than ease of access when it comes to sources. Blueboar (talk) 22:13, 31 August 2013 (UTC)

Like Blueboard says: quality (and reliability?) should be the paramount consideration. This should not be weakened simply for convenience. ~ J. Johnson (JJ) (talk) 22:55, 31 August 2013 (UTC)

I agree entirely, a weak easily accessible English source should not be preferred over a high quality foreign language source that is harder to access simply because the English source is ease of access.--64.229.165.126 (talk) 23:07, 4 September 2013 (UTC)

• Nope, I oppose, too. While this would make our lives easier in some ways, we're not trying to be the easiest encyclopedia to write for in the world, but the best. Regards, TransporterMan (TALK) 13:21, 5 September 2013 (UTC)

Time to time discover english readers the sources of the articles in languages ​​other than English. What is the point of these? You read articles on ENWP and would love to have sources written in English, it is natural. Many times I have come across articles where most sources, written in a language other than English. If English readers do not understand what is written in the source then the source is meaningless. All parties would benefit if all sources on ENWP were in English. Lifeglider (talk) 17:46, 28 August 2013 (UTC)

No. AndyTheGrump (talk) 17:47, 28 August 2013 (UTC)

That may be the ideal, but should not be a rule that would exclude sources. There are many things that would make a source more difficult to "access" (off line, behind a paywall, obscure, a different language) but they should not be grounds for excluding them. North8000 (talk) 17:52, 28 August 2013 (UTC)

On the other hand, since English is one of the major world languages, the English language Wikipedia opens up the subject to readers whose home language is neither English nor the language of the original source. For example, assume that most of the sources regarding a particular subject that is related to country A are written in the language A. A reader from country B wishes to know about the subject, but can only read his own language, language B and English. In this case the English Wikipedia will open up the subject to him. Of course, Wikipedians who read both English and language A will ensure that the article accurately represents the sources. Martinvl (talk) 23:17, 28 August 2013 (UTC)

When there are quality English language sources to support the information in our articles, our policy favors using them over non-english sources. However, many of topic don't have quality sources in English. We still want the articles, and we still want the articles to be well sourced... even if the sources aren't in English. Blueboar (talk) 01:16, 29 August 2013 (UTC)

• Absolutely oppose. If I'm writing about Germany or a German person, it's utterly insane to use out-of-date English sources when there are quality, up-to-date German-language sources that are easily accessible and, usually, more comprehensive.—S Marshall T/C 10:00, 29 August 2013 (UTC)

  I agree with S Marhsall. In the other direction, I've cited English-language sources on de.wikipedia, when writing articles about North American plants. Of course, I think it's helpful to both quote and translate the relevant excerpts of the sources. -sche (talk) 17:45, 30 August 2013 (UTC)

Agree with S Marshall. One option, which I don't think is currently sufficiently utilized, is to translate articles into English on WikiSource, which would make it easier for editors in any language to use sources which are primarily if not exclusively non-English. I think that this would apply most strongly to material in foreign language reference books, particularly on topics which seem to more directly relate to that language than English. Here I'm thinking of, for instance, encyclopedias in the Russian language which give wonderful articles about significant Russian historical developments or culture, but which aren't as regularly discussed in foreign language sources. If anyone knows of any way to stimulate development of such translations, I and I think others would love to see them. John Carter (talk) 16:57, 6 September 2013 (UTC)

While that may not be wildly known, even native English speakers are neither structurally unable to nor legally forbidden from learning other languages. Or from commissioning a translation. Restricting ourselves to English sources would increase systemic bias, decrease quality, and leave us overall much poorer. Also, arguably it's easier to learn (say) Persian as a language than to learn enough math or physics or computer science to understand many academic sources in English. Just because one can read the words does not mean one understands the source. --Stephan Schulz (talk) 17:15, 6 September 2013 (UTC)

Je suis très opposé à cette proposition. Sławomir Biały (talk) 18:11, 6 September 2013 (UTC)

• Absolutely oppose. That material must be verifiable does not mean that it must be verifiable by any given reader no matter how limited his or her competency. ~ J. Johnson (JJ) (talk) 22:42, 6 September 2013 (UTC)

(This proposal was originally posted at Wikipedia talk:Policies and guidelines#Limitations of RS+NOR, with a solution, but I was asked to post it at WT:RS and WT:NOR, and, by implication, here.)

Limitations of RS+NOR+VERIFY, with a solution (Rev. 1):

The motivation for this discussion is my frequent annoyance over the years with articles that don't explain things well, or don't give enough information, because experts can't simply write what they know, due to NOR and the need for RS.

An example is Sailing faster than the wind, where section "BOLD EDIT NOTICE" of its Talk page presents an excellent explanation (the analogy of a geared transmission) that couldn't be included in the article due to NOR and lack of RS. I personally believe that this explanation is actually correct, making this a very nice example, but it doesn't actually matter if this explanation is correct or not, just as we can rarely be sure that any information in any article is complete and correct, even when lots of RS have been cited. And, while I'm providing this as a motivational example, I want to discuss WP policies and guidelines here, not details of this example.

Now to get to the proposal: reliance on the availability of good RSs leads to a good encyclopedia, but this technique has limitations. Once in a while a Talk page provides some OR explanation that is clearer that the explanation given in the article. This example, I believe, is such a case. While this isn't the purpose of a Talk page, it is a very valuable service for WP readers who read the Talk pages as well the articles, as I do.

It also shows an inherent limitation of the RS+NOR policies, as applied to articles. An improved WP policy, and the solution I'm offering for discussion here, would be to allow OR explanations or knowledge in articles and Talk pages, without a reference, until someone provides a good OR reason to object to them, or an RS is found that supports a good replacement. Note that this implies that the information added without RS must be uncontroversial. Any unresolved argument among editors cancels this new policy and falls back to requiring RSs.

This new policy also applies to the large percentage of (mostly old) articles that contain no references or insufficient references. That means that presumably correct information in these many articles cannot simply be deleted by any editor. There must be an objection to it, with the opportunity to discuss the deletion on the Talk page. This allows some protection for old yet correct information.

This would be an additional policy, modifying the RS, NOR, and VERIFY policies, or would be a modification of the RS, NOR, and VERIFY policies themselves (I'm not specifying which here).

In addition to the above, this proposal affects the VERIFY policy by requiring modification to explain the situations in which OR without RS will be allowed. These situations are explained above. I will be happy to provide explicit revisions to all affected policies if this discussion eventually results in agreement. David Spector (talk) 13:26, 18 August 2013 (UTC)

Oppose: WP is intended for a general readership. Editors should not be placed in a position where they can be told that they can't remove unsourced information because someone allegedly versed in the technical areas covered by the article knows better than they do. I find it antithetical to existing WP policy to espouse a policy that renders unsourced and potentially unverifiable information inviolate. Editors have already faced difficult situations in some cases where they removed unsourced information, even information that had been challenged for months before they did anything more serious about it. If editors cannot or will not source information, especially information that has been challenged long-term, than they should not be able to prevent its removal by claiming nothing more than, "We know this is true and we don't think it's controversial." In fact, I think once information has been challenged it should effectively be considered controversial. Given that you say any unresolved argument among editors cancels this policy, does this not basically boil down to, "You must challenge the information via a method that does not involve its removal before you may proceed with removing it?" DonIago (talk) 16:51, 18 August 2013 (UTC)

Oppose. Wikipedia is not an appropriate place for the publication of original research, regardless of whether other contributors think that it is 'uncontroversial'. This proposal radically alters the whole ethos of Wikipedia, and could only ever lead to a reduction in article quality. AndyTheGrump (talk) 17:12, 18 August 2013 (UTC)

If a piece of OR really is absolutely uncontroversial (i.e. not "challenged or likely to be challenged") then in fact policy already allows this. What triggers the effect of policy is the act of disputing something (e.g. by changing or removing it). At that point we resolve the dispute by reference to sources. But if nobody challenges it and nobody is likely to challenge it, then whatever you've written can stand unchallenged forever. I would be opposed to altering the wording of the policy because it's unnecessary to achieve what you want.—S Marshall T/C 17:30, 18 August 2013 (UTC)

• This idea has also gotten a less than welcoming reception at the long discussion at Wikipedia talk:Identifying reliable sources. Nothing you have written changes the gist of my response: if you cannot find a cite to support adding content, then sorry -- out of luck on the Pedia -- or not enough work has been done to find the cite. The reason a source is important beyond the information itself, is because they serve an important function in vetting the information for its acceptance and reliability. Alanscottwalker (talk) 18:12, 18 August 2013 (UTC)

Support - I understand the proposer to be introducing a new class of material - "Tutorial material" - material which is useful in Wikipedia, but comes under class (B) in [[this essay. Supposing that I was writing an article and I developed an example showing how a debt of £100 mounts up if interest is charged at 5% per annum simple interest vis-à-vis interest at 5% compounded annually. (My mother explained this to me when I was about 10). Do I really need to get the example from a text book (and possibly run into copyright problems), or can knock out the figures on a spread-sheet and quote them? There are many such examples in Wikipedia. Come on, lets be sensible and add a new class of information called WP:TUTORIAL_MATTER. User:S Marshall identifies a sore point - what happens if somebody challenges some decent tutorial matter that you have written just to get at you personally. I would like to be able to quote a policy and so get the community support to tell the person to "desist". Martinvl (talk) 19:12, 18 August 2013 (UTC)

Oppose – This is a recipe for disaster. The ban on original research is the main reason Wikipedia is such a useful resource and weakening it has to be done only with very great care. But no such care is visible here and no suitable proposed wording has been presented. Zero^talk 23:34, 18 August 2013 (UTC)

Oppose per Zero and Andy. This should not have been posted at the other talk page or even at this talk page. The suggestion on the other talk page was to post it here or at WP:VPP. For a change of this kind, VPP would be a better place. Why David thinks it's "too informal" is beyond me.--Bbb23 (talk) 23:47, 18 August 2013 (UTC)

Oppose per BBB23. Wikipedia is not Citizendium. Indeed, WP would not even be as good as Citizendium because it would not have CD's equivalent of peer review. People who use Wikipedia need to be able to confirm what they read here and not have to depend on self-appointed experts. — TransporterMan (TALK) 00:05, 19 August 2013 (UTC)

Lets look at an example - the editors the article Complex number introduced the subject with this example. I don't know if this example was the creation of the author in question or if (s)he copied it from a text book. In either respect, it is hardly cutting-edge science - any maths undergraduate can confirm that it is correct. Where does this stand in terms of WP:VER and of WP:OR? BTW, the page in question gets between 1000 and 2000 hits a day and is ranked at about 7000 in terms of numbers of visitors. Martinvl (talk) 00:27, 19 August 2013 (UTC)

Without looking at the article, if anybody who understands the subject can tell you that the example is accurate, then it's fine. What we might need is a statement directly about the acceptability of giving obviously correct examples, comparisons, etc. We do have a small but dedicated number of editors who apparently believe that if it's not what the rest of us call plagiarism, then it's a NOR violation. I assume that this proposal is born out of long experience with such editors. WhatamIdoing (talk) 04:56, 19 August 2013 (UTC)

I was going to provide a tldr reply, but I can't say it better than what Doniago said at 12:31, 19 August 2013, and Monty845 said at 13:42, 19 August 2013, both just below. — TransporterMan (TALK) 14:23, 19 August 2013 (UTC)

Oppose, but with some sympathy for the types of problems that I think are causing this proposal. The thing is that I think editing suggestions which put things nicely are either blocked by other editors because their is a controversial amount of non-obvious originality or else because of a misunderstanding of the policies we already have. As per WhatamIdoing, I can perhaps imagine that some wording tweaks in policy might be possible, but I think openly calling for originality will not work.--Andrew Lancaster (talk) 09:35, 19 August 2013 (UTC)

A good deal of the problem is caused by the sentence "must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material being presented". Editors who might not understand the subject often look for in-line citations when in fact they are not necessary, or when "being able to cite" material is a matter of the expert saying "Here is a list of text books, all of which discuss the subject. Which one do you want?". Martinvl (talk) 10:06, 19 August 2013 (UTC)

This seems a bit misleading to me. In my experience the easiest way to resolve these situations, not to mention satisfy WP:BURDEN is to provide a source, not to get into a debate over the issue. An expert on the subject should be able to provide a source, and ideally should recognize that arguing over it wastes far more time and effort on the part of all parties. Why ask which source should be provided rather than simply choosing one and providing it? If the expert provides a source and the source itself is then challenged, then there's grounds for a more serious discussion of the matter.

That said, I question the use of the term "expert" in any case. I certainly don't know who here is an expert on what subject matter, and unless there are procedures for official recognition of "expert" status on WP...if there are I've never seen it come up...then nobody should wield the "expert" stick because it should be abundantly obvious that anyone can claim they're an expert on anything without that actually being the case. DonIago (talk) 12:31, 19 August 2013 (UTC)

Moral Support/Practical Oppose I wish there was a way for this to work, but as a practical matter, I just don't seen a standard to apply that gives more leeway then WP:Burden that wouldn't do more collateral damage then good. Requiring the opponent of inclusion to come up with a source to disprove something is simply unreasonable, if its hard to find a reference that provides a specific example, it will be far harder/impossible to find a reference that disproves it. What we really would need is editors to only invoke burden if they really do have a strong, good faith belief that the material is not accurate, and otherwise leave the material in. But the problem is that basing a decision on the subjective motive of an editor, rather an objective standard of what RS provides is just asking for Drama and baseless accusations of bad faith. I just don't see a way for the idea to work in practice. Monty₈₄₅ 13:42, 19 August 2013 (UTC)

Support (sort of) This is an important problem which has caused Wikipedia to plateau out in several areas. But it really doesn't need a new policy. It just needs:

• An important tweak in current policy to bring the mechanics of the policy in line with (rather than conflicting with) the spirit defined by "challenged or likely to be challenged". So it would remain just as easy to knock out unsourced questioned material, but harder to knock out unquestioned material for bad reasons.
• Explanation (maybe a guideline) to kill some immensely rampant-in-Wikipedia destructive urban legends which have no basis in policy. Namely that verifiability is a force for inclusion rather than just a requirement for inclusion.

Together these two would enable expertise in editing without weakening wp:ver North8000 (talk) 13:52, 19 August 2013 (UTC)

Oppose At least, not here on WP. We have other projects where this might make sense and be more welcome, such as Wikiversity. WP cannot have a mechanism that could make this work without abandoning the principle of "anyone can edit", without which WP would die a slow and painful death. LeadSongDog come howl! 18:30, 19 August 2013 (UTC)

I don't see where their proposal (vague as it is) would affect that. North8000 (talk) 11:48, 20 August 2013 (UTC)

Oppose. An expert's unsourced opinion might (but not always!) be more reliable than some non-expert's citation of random sources, but the proposal says nothing about expert qualifications; this proposal would allow all kinds of junk from anyone that "knows" something to be true. If there is a problem regarding the scope an editor has in formulating examples (etc.) that should be discussed at NOR. ~ J. Johnson (JJ) (talk) 18:42, 22 August 2013 (UTC)

Thank you. I did not see any insurmountable objections to my proposal. Here are my responses to these responses. Interspersing them would interfere with their flow, so I've grouped them all here. Please feel free to continue the discussion. David Spector (talk) 17:16, 23 August 2013 (UTC)

• DonIago, "WP is intended for a general readership. Editors should not be placed in a position where they can be told that they can't remove unsourced information because someone allegedly versed in the technical areas covered by the article knows better than they do." WP is not all intended for a general readership, by your implied definition. Many of the mathematics articles simply cannot be understood by your general readers, because they may not be educated in that technical area. The same is true of physics, and many other topics. You may have misinterpreted my frequent use of the word expert. I don't mean it to refer to someone who is highly educated in some specialty. A person who has played with toys called Beyblades may only be a teenager, yet may be an expert in that subject. She may meaningfully contribute correct information about the toy's operation, even though no one can find a single RS corroborating that information. Assuming that the toy itself is NOTABLE, the new information improves WP coverage of it. My proposal would protect that information until either a source can be found, or someone else corrects it. My proposal helps WP to grow organically, as it is meant to do. "I find it antithetical to existing WP policy to espouse a policy that renders unsourced and potentially unverifiable information inviolate." My proposal shows confidence in the editors. This is a good thing, and follows WP principles. Also, you are actually incorrect. My proposal only protects information until it is challenged. As soon as it is challenged and the challenge seems to hold, my proposal no longer applies. "Editors have already faced difficult situations in some cases where they removed unsourced information, even information that had been challenged for months before they did anything more serious about it." I'm not sure what your point is here. Please explain. "If editors cannot or will not source information, especially information that has been challenged long-term, than they should not be able to prevent its removal by claiming nothing more than, "We know this is true and we don't think it's controversial." If you mean that one editor claims the change is controversial and another claims it isn't, then it's controversial and the proposed policy no longer applies. If one of the editors is unhappy with the resulting situation, they are free to make use of the existing mechanisms for resolving disputes about policy. "In fact, I think once information has been challenged it should effectively be considered controversial." Exactly, that's correct. "does this not basically boil down to, 'You must challenge the information via a method that does not involve its removal before you may proceed with removing it?'" Yes, that is part of the implication. The proposal delays removal until there is a reasonable challenge (a wild claim with no rationale by an IP editor would not be reasonable). The proposed policy provides explicit protection for information currently at the mercy of anyone to delete.
• AndyTheGrump, "Wikipedia is not an appropriate place for the publication of original research..." I agree. The proposal should not be interpreted as establishing WP as a place for first publication of academic research. However, if someone knows something valuable about a subject, that will improve understanding of it, something that ought to have an RS but for some reason does not yet seem to have one, this proposed policy would allow that "expert" to share her knowledge with us. The safety in doing so is that another "expert" can claim that this alleged knowledge is highly unlikely, and explain why. Such a reasoned response cancels "OR without RS" instantly. "This proposal radically alters the whole ethos of Wikipedia, and could only ever lead to a reduction in article quality." I'm not sure what "ethos" means, but could you explain how my proposal, in allowing experts to contribute, will always reduce article quality? That would seem to be a contradiction, since WP has always benefitted from experts in all fields.
• S Marshall, "If a piece of OR really is absolutely uncontroversial (i.e. not "challenged or likely to be challenged") then in fact policy already allows this." Sorry, no. WP:NOR forbids adding information known by the editor, no matter how sure the editor is. WP:RS forbids adding anything for which a RS cannot be found. "What triggers the effect of policy is the act of disputing something (e.g. by changing or removing it)." It may look that way, but it is not so. An editor is allowed to remove text when it violates policy--the editor is not permitted to "dispute" policy by doing removals of text anywhere she pleases. On the other hand, she is permitted to dispute any text on the Talk page for the article. "But if nobody challenges it and nobody is likely to challenge it, then whatever you've written can stand unchallenged forever." In practice, this is exactly what happens, especially in older articles. My proposal modifies policy to allow this common case. "I would be opposed to altering the wording of the policy because it's unnecessary to achieve what you want." Currently, anyone can delete up to 25% of WP, because is consists of statements having no references. And many of those statements are true, but there are no RSs that will support them. My proposal will achieve protection and a rationale for keeping all this valuable information in WP until corroboration (or dis-corroboration) is found.
• Alanscottwalker, "This idea has also gotten a less than welcoming reception at..." I have responded to all comments there, as here. So far, I have seen little true understanding of my proposal and no objection that could not be easily countered. I think my proposal is doing very well so far, considering the experience of the long-term editors who are contributing to this discusssion. "...if you cannot find a cite to support adding content, then sorry -- out of luck on the Pedia -- or not enough work has been done to find the cite." Yes, that is the current policy, you're simply repeating it. Repeating it doesn't argue for it. "The reason a source is important beyond the information itself, is because they serve an important function in vetting the information for its acceptance and reliability." I fully agree with you. RSs are the cornerstone of the integrity and reliability of WP, such as it has. However, the reality is that 25% (or whatever) of WP consists of old articles that have no RSs at all. You can ignore them, but they're there. They are tolerated because deleting them would empoverish WP. Full enforcement of your statement would ruin WP. And, besides that, allowing better and more intuitive explanations in articles (my primary example) would clearly improve WP.
• Martinvl, I appreciate your proposal, which weakens mine so that it only applies to certain articles. I reject that. My proposal should apply to all articles, allowing people who know to contribute what they know, even when RSs cannot easily be found.
• Zero, I agree that WP should not be used to present major new research or new ideas. Yet I want to protect probably-true information and useful explanations. I believe I have done so using language that respects the importance of the NOR policy. If you can offer "suitable wording" to clarify this distinction, I would undoubtedly accept it.
• Bbb23, Please explain why a proposed change in three policies should not be presented on their Talk pages, or on the Talk page for all the policies and guidelines, for that matter. It's not obvious to me. WT pages are for discussion of WP articles. I felt that Village Pump would not be the place for this discussion, but I'm willing to move it there if others agree with you.
• TransporterMan, WP has a very different strategy from Citizendium. WP's strategy is more successful. The difference is that WP gives up some authority and reliability in return for allowing anyone to edit its articles. WP believes almost everyone has something valuable to contribute. As a result, our older articles are sometimes entirely lacking in references. We have so many articles that peer-review is haphazard, and many articles "fall through the cracks" year after year, remaining without RSs. As WP editors we accept our strategy and the policies that govern it. But many editors see flaws in those policies. One flaw is that one cannot always find a RS for a good and true statement that one would like to add to an article. My example was an attempt to illustrate this. I'm not asking WP to move in the direction of Citizendium. Their policies have not proven the right ones for growth, only the right ones for providing good information (and maybe not even that, I don't know).
• WhatamIdoing, "We do have a small but dedicated number of editors who apparently believe that if it's not what the rest of us call plagiarism, then it's a NOR violation. I assume that this proposal is born out of long experience with such editors." Well, er, no. I did not have that in mind. However, that looks like another area that would benefit from my proposed modification of policies. Of course, you're employing hyperbole here--plagiarism is forbidden by WP policy.
• Andrew Lancaster, I don't understand your points. You seem to be supporting the proposal.
• Martinvl, I'm not following your response to Andrew. I don't see how this relates to the proposal.
• DonIago, Your arguments make sense only when a RS can be found. But my proposal deals with the many situations in which a RS cannot be found, such as in the example I gave.
• Monty845, "I just don't see a way for the idea to work in practice." Fine, but you don't seem to be exploring the proposal, but instead the application of BURDEN. I'm not seeing why you find the proposal impractical.
• North8000, Your first suggestion is intruiging. I'd like to see more about it. Your second appears to be incorrect. Any statement in an article can be removed when it has no RS or even inadquate RS. Please point me to policy if I'm wrong. I don't see any real objection to my proposal, though.
• LeadSongDog, "WP cannot have a mechanism that could make this work without abandoning the principle of 'anyone can edit', without which WP would die a slow and painful death." My proposal allows anyone with certain knowledge to add it to WP without having the bad experience of seeing it instantly be deleted before they can even add RSs. This supports the principle of anyone can edit in a very direct way. You must provide proof if you seriously wish to claim that my proposal will make WP "die a slow and painful death." That is a sweeping and dramatic claim, one which even other opponents of my proposal have not made so far in this discussion.
• J. Johnson (JJ), I have deliberately not brought in the need to evidence the expertise of an editor. Sometimes an editor will know a lot (which they can contribute to make WP better), sometimes an editor will know little or nothing about the article. In the spirit of WP's strategy of allowing edits to converge to something good, I worded the proposal so that editing and editor discussion would resolve most issues, as it does under the current policies. My proposal does not require anyone to be an expert in some academic way, just to have some information that they wish to contribute to the article. If it is false, other editors will remove it right away. If it is true, other editors will respect that, and allow it to stand. But my proposed policy explicitly allows such contributions to stand even though they have no references yet. It is a way to allow anyone to contribute what they know, and at the same time conserve the 25% (or whatever) of existing unsourced information in WP.

David Spector (talk) 17:16, 23 August 2013 (UTC)

While your response improves my understanding of your proposal, I still feel it ultimately makes it more difficult for editors to remove unsourced information that does not belong here without sourcing. It's my experience that a desire/willingness to remove unsourced information is already not something found in great abundance among editors here (indeed, I've been commended on occasion for my willingness to get in the trenches to defend the removal of unsourced material...which I don't think should have needed to happen in the first place), and I believe there are sufficient protections governing the removal of such material. I don't feel we need more pressure placed on editors to leave added material alone. Bluntly - We have WP:BURDEN for a reason, and if an editor objects to material being removed then they should be able to satisfy said policy and consequently reinsert their material. If they can't, or worse won't satisfy said policy, then I don't believe the material should be published here. DonIago (talk) 18:10, 23 August 2013 (UTC)

David we agree that "RSs are the cornerstone of the integrity and reliability of WP, such as it has". But you've missed my point because you've misdiagnosed the full reason why that '25% or whatever,' unsourced exists. Sourcing can be hard and requires work. Policy should continue to encourage that which we all would rather not do but need to for integrity and reliability -- more work. The rather-not-do impulse does not need reenforcement; it does well enough on its own. The insistence on sources also has important systemic benefits for collaboration in this anyone can edit model. It encourages discussion about content not persons, and increases trust and communication in the know what we're all talking about vein -- sources literally serve as the common point of reference (internally, as well as externally). Alanscottwalker (talk) 23:34, 23 August 2013 (UTC)

Thank you, all. Again, I did not see any insurmountable objections to my proposal. Here are my responses to these new responses. Interspersing them would interfere with their flow, so I've grouped them all here. Please feel free to continue the discussion here. David Spector (talk) 19:39, 28 August 2013 (UTC)

• DonIago, "I still feel it ultimately makes it more difficult for editors to remove unsourced information that does not belong here without sourcing." Henry, the idea that unsourced information does not belong here is part of the current policy, not my proposal. My proposal is orthogonal to the question of providing references to sources. My proposal allows uncontroversial unsourced information until such time as RSs can be found. The great benefit of removing unsourced information is that, especially in the early days of WP or of most editors, there was a tendency to add incorrect information (for example, information that was remembered incorrectly). I do not oppose eliminating incorrect information. My proposal may add incorrect information, but it will also add correct information, benefiting WP by encouraging editors to add what they know to be correct. They will sometimes also make mistakes. My proposal does not prevent mistakes, but it does allow correct information to be added and does encourage editors to contribute to WP. "I don't feel we need more pressure placed on editors to leave added material alone." I'm not clear on how my proposal increases pressure on editors. It merely allows situations where RSs have not yet been found to stand instead of be deleted. We have entirely too much deletion on WP--frequently good information is deleted just as quickly as bad. BURDEN states, "Any material lacking a reliable source directly supporting it may be removed." My policy modifies this; it preserves useful material until RSs can be found. To the extent that we trust editors to grow and to do a good job, this will improve the quality of articles. Editors will be encouraged to contribute information they know to be true. Not guess, but know for sure. Maybe this should be added to the proposal. But, while RS and BURDEN certainly improve WP in the absence of true experts, my proposal would further improve WP. (Encyclopedia Brittanica is an example of an encyclopedia whose philosophy is to collate the contributions of recognized experts. WP has a different philosophy.) "Bluntly - We have WP:BURDEN for a reason, and if an editor objects to material being removed then they should be able to satisfy said policy and consequently reinsert their material." Blunt or not, this is simply a statement of current policy. Do you believe that current policy has evolved to the point that it cannot be improved? "If they can't, or worse won't satisfy said policy, then I don't believe the material should be published here." Okay, so it seems that your arguments pivot on this final statement, which is simply a belief in the current policy. Belief is not a good enough rationale. Yes, I agree that the current policies are excellent; the quality of many WP articles, developed through the work of dozens of "non-expert" editors, is excellent. But pay attention, too, to the many articles that have no references at all. Pay attention to the editors who leave WP because they find it difficult to face continual arguments over policy. And pay attention to the many new editors who leave after editing just one or a few articles, either because they were criticized by other editors who know the policies inside and out, or because they found it difficult to spend so much time on research when they actually know the few pieces of information they contributed.
• Alanscottwalker, You make a good point: "Sourcing can be hard and requires work. Policy should continue to encourage that which we all would rather not do but need to for integrity and reliability -- more work." My proposal is not meant to eliminate the need for sourcing. I wouldn't want to open floodgates of adding garbage that then could never be removed. The proposal would merely make WP more lively, providing a way for editors who have good knowledge to contribute it. If even one editor, reading a new addition, sees that it is garbage, they can give a concrete objection to it and later delete it for lack of RS. The point is that at WP we trust "ordinary folks" to edit. We don't assume that they are idiots who can only contribute garbage. We welcome them, and my policy encourages this. Besides, there are many cases where RSs either haven't been found (older articles), or where RSs are difficult or impossible to find (my example of explaining how it is possible to sail faster than the wind). It is in these cases where my proposal really helps. It doesn't avoid work, but it does allow spreading it out instead of insisting that anyone can remove material lacking a RS. Finally, as to avoiding work, all editors should be encouraged to add references to older articles that have none. It may not be glamorous, but it can be fun, and satisfying to confirm correct information.

David Spector (talk) 19:39, 28 August 2013 (UTC)

• "My proposal allows uncontroversial unsourced information until such time as RSs can be found."

And this is why I'm opposed to your proposal; you're prohibiting the removal of information indefinitely, even if it is unsourced. As I stated previously, my feeling is that if the information has been challenged then it is no longer uncontroversial.

• "in the early days of WP or of most editors, there was a tendency to add incorrect information"

Some editors might argue that this is an ongoing tendency.

• " I do not oppose eliminating incorrect information."

But your policy would make it more difficult to remove such information because an editor could claim they had the right to retain it until an RS is located. Or are you saying it should be incumbent on the challenging editor to prove that the information is untrue? If so, how would they do that?

• "My proposal may add incorrect information, but it will also add correct information, benefiting WP by encouraging editors to add what they know to be correct."

I don't believe we need to make it easier for editors to add correct information, especially if it means they're adding information that may be correct but also is unverifiable.

• " My proposal does not prevent mistakes, but it does allow correct information to be added and does encourage editors to contribute to WP"

I'm not aware of issues preventing editors to contribute to WP. IMO the requirement that information be sourced or face potential deletion is not a significant challenge to contributing.

• " I'm not clear on how my proposal increases pressure on editors. It merely allows situations where RSs have not yet been found to stand instead of be deleted."

Which increases pressure on editors to let information stand instead of deleting it, because an editor who takes issue with the deletion can obstruct the process by claiming they need an indefinite amount of time to locate sources. I've already played the "Just give me another week to locate a source" game, and it doesn't work. Either the editor doesn't provide a source, or I move on to other things and the unsourced and potentially erroneous information remains. Where's the harm in saying, "Please re-add this information when you can provide a source"?

• "We have entirely too much deletion on WP--frequently good information is deleted just as quickly as bad."

I don't believe that unsourced information should be called "good" information. It should be called "information that should be sourced". As for your opinion that there's too much deletion on WP - that's your opinion, and I haven't seen anything to support it.

• " My policy modifies this; it preserves useful material until RSs can be found."

At what point are we allowed to conclude that they can't be found?

• "Editors will be encouraged to contribute information they know to be true. Not guess, but know for sure. "

How are we supposed to know whether an editor knows information is true, is guessing that it's true, or is deliberately adding incorrect information knowing they'll later be able to claim they just need time to locate a reliable source?

• "Do you believe that current policy has evolved to the point that it cannot be improved?"

There may be room for improvement within the current policy; I simply don't see how your proposal improves the current policy; rather I see it as weakening current policy.

• "Pay attention to the editors who leave WP because they find it difficult to face continual arguments over policy."

In my experience most arguments over this particular policy could be averted if the editor desiring to include information would spend more energy locating a source than arguing for the inclusion of unsourced material. Also I'd like evidence that this is why editors are leaving...have you conducted exit interviews, or is this a guess on your part?

• "And pay attention to the many new editors who leave after editing just one or a few articles, either because they were criticized by other editors who know the policies inside and out, or because they found it difficult to spend so much time on research when they actually know the few pieces of information they contributed."

It's my opinion that contributing to WP shouldn't just be about adding what you know to be true; it should be about what you know to be true and can provide a reference for such that other readers can determine from whence your information originated.

In short, I don't doubt that you're passionate about this, but I don't see how your proposal ultimately does anything to improve the project. As near as I can determine, at best it restates information that is already true ("Editors can add unsourced information") while at worst it undermines verifiability and gives obstructionist editors the defense of "I know my information is true and I promise to add a source...sometime."

If you opt to respond I would appreciate it if you did so directly after my comment to make it easier to follow the conversation, rather than your current technique,which makees the flow of conversation less clear. Thank you.

Who's Henry? DonIago (talk) 20:36, 28 August 2013 (UTC)

Oppose. (I voted likewise at WT:RS.) The problems are probably fatal but it's worth considering whether they can be solved.

In answer mainly to editor David Spector at WT:RS (the editor suggested replying here) and at this topic/section:

I agree that experts usually agree on a given statement in a field in which they're expert. But most of those statements would be sourceable. When they're not, disagreement becomes likelier, because being in print or on the Web attracts scrutiny about the certainty of the statement, so that the lack of print or bytes reduces scrutiny and lessens weight for certainty. While knowledge development tends to reinforce the core of a given body of knowledge, because the development of new edges challenges the core so that if a challenge fails the core is reinforced (e.g., perhaps gravity does not exist but the known data is so vast it's likely to overwhelm any attempt to come up with a contradictory explanation that accounts for all the known data), if a core statement is not sourceable then developers at the edges are less likely to know of the core statement and therefore may not either agree or disagree with it, thus may not either reinforce or discredit it.

Active bodies of knowledge that are neither very recent nor very old are more likely sourceable; inactive bodies of knowledge (like on how to repair a Macintosh 128K) are those that have not lately been much developed (as far as most of would know) and thus have not lately faced known challenges to their validity, so if they're not sourceable there's also probably less agreement among its experts as to a given statement.

Any editor adding unsourceable content is presumably implicitly claiming to be an expert. (Exceptions include vandals and the mistaken (e.g., grammar correctors who inadvertently alter substantive information).) By claiming x is true without a source, one is presumably asking to be believed without citing an authority other than oneself, and that probably defines a claim of expertise.

Regarding the statement by David Spector, "my proposal would rapidly have had no effect in this example [on how dangerous is lead in house paint], since experts differed. WP has no way to prevent erroneous information from being added, but its virtue is that such errors are rapidly eliminated, as soon as someone with better knowledge comes along." Yes, but that means either that a second expert's unsourced view could replace a first expert's unsourced view or that replacement would not be allowed unless sourced which would mean that the first view would stay although unsourced while a second expert searched in vain for a source. So we'd either have battling experts (and we don't have a system for designating or recognizing experts so we'd really be dealing with battling editors who may or may not be experts), a pretender to expertise getting in first who stymies any expert trying to clear a matter up but has no source, or a relative success (an expert able to stop editing by a nonexpert who came afterwards); two out of three of those results are especially undesirable.

My other examples that may not have been clear have these properties: On string theory and the disagreement with it, on an unsourced disagreement, whether a string theorist or a nonstring theorist should be the editor permitted to add unsourced content is the question; both can't be that editor. Between archaeology and Egyptology as separate disciplines, each could be the dissenter from the other, so that a point of disagreement regarding archaeology in Egypt can't be resolved be turning to the mainstream, since there are two mainstreams; the same is true of political scientists and sociologists (I think sociologists) with respect to mass movements; the same is true of scholars debating who discovered America, with even more mainstreams. The theological case had that property and also the property of applicability to a great many nontheological articles in Wikipedia, which risks that battles between experts would not occur as often but mainly because the second experts would probably not know that a first expert had posted somewhere that was secular (e.g., on voting) and so probably wouldn't edit or counter-edit there.

I may have misunderstood in thinking that your proposal was for experts in particular (not just any editors) to gain a degree of preservation of their unsourced edits. If it is for nonexperts too and therefore anyone to edit so that other editors should assume that any unsourced edit is the product of expertise, making reverting harder, that amounts to weakening the RS and NOR requirements without, in my opinion, enough compensating gain. I have a thought about when a baseball batter has to decide whether to swing relative to a pitch's start but I have no source and I'm not a sports expert; if I add it to Wikipedia my view might be read by thousands of readers because no one could delete it without first gaining consensus that my view is controversial or finding a contrary source, if there is one. Maybe I should post it at a baseball discussion forum where reliable baseball experts could weigh in or possibly unreliable commenters could refer me to a reliable source one way or the other, since most readers of Wikipedia are not expert in the subjects they're reading about. Wikipedia is heavily used by nonexperts, most of whom don't edit it at all (I tried to persuade a public librarian who found an error and had a source to edit WP herself but she wasn't willing). I frequently edit unsourced edits on the basis of my sourceable expertise; if I had to bring all of them to the talk page to gain consensus instead of being bold I'd probably soon give up for most of them, as would most editors for reasons of time and productivity, and the result would be lower editorial quality across WP (unless those with the initiative to edit first tend to be inherently more expert than those who correct them, which I doubt).

What I was talking about regarding adding OR to other media is that OR should be submitted to subject-relevant websites or other subject-relevant media which have experts in the relevant subjects and that are RS and then, after the other media have vetted the OR, a WP editor should paraphrase from the other media into WP. It's not your proposal but it may be a solution to the problem you outline, a solution that preserves existing policies and avoids the problem of awaiting consensus before deleting wrong information because perhaps it was added on the basis of unsourceable expertise.

When new non-BLP content is rapidly deleted for want of sourcing when sourcing could have been found, the solution now is to re-add the content but this time with sourcing.

The plagiarism-vs.-NOR issue is not hyperbole, because it was describing what some editors call plagiarism and other editors don't when the latter editors demand what the former would call plagiarism because they don't trust the quality of the paraphrasing they see and so think they see OR. Arguments of that kind come up now and then.

I was not concerned that you have a hidden agenda. Something like this is not much affected by hidden agendas.

WP:BLP requires immediate deletion of contentious content. I don't know if you propose that that policy be changed, but changing it could expose the Wikimedia Foundation to legal risks.

Wikiversity is not a site I know well. Wikiversity:Reliable sources, Wikiversity:Original research (a proposal), and Wikiversity:Adding content seem relevant. At first glance, I'm not sure that OR is allowed there.

Crafting specific language is advantageous at some point in the discussion (and I think that's now because of already-open disagreements between you and some other editors about what your proposal actually is), posted in a way that lets it be updated before finalization as debaters raise issues you can resolve by fine-tuning the specific proposal. Crafting what a template would say is also helpful but it's not important to program one (unless you have a fancy one in mind and one could question whether it technically can be created, but what's been so far mentioned is not technologically fancy). Your proposal that if one of these cases of unsourceable expertise goes to arbitration the tradition of requiring RS and deleting OR would take over for the case is a simplifying approach; if this includes not just arbitration (a term normally used only for ArbCom proceedings) but any dispute resolution mechanism, your proposal essentially would end when an editor requested dispute resolution, but that means that consensus that an edit is controversial would be unnecessary because a single editor could seek dispute resolution. This would lead to more disputes at the dispute resolution level and less posting to article talk pages seeking consensus. If you want to preserve the tradition that dispute resolution only follow an attempt at talk page discussion, then you should craft language so that as a result each relevant dispute resolution forum requires (or continues to require) that a consensus have been arrived at on the article's talk page (assuming it's not a sham consensus) or that a failure to achieve consensus be invalidated.

Nick Levinson (talk) 16:53, 30 August 2013 (UTC) (Edited generally: 17:14, 30 August 2013 (UTC))

DonIago and Nick Levinson, thank you for your many points. I thought of replying to them individually as I have been doing here so far (many of them have simple answers based on simply understanding the proposal with an open mind), but realized that at this point, with mostly Oppose votes, that my continued defense is not worthwhile. If I'm seeing value in the proposal, but no one else is, the probability is that I'm wrong and everyone else is right. I therefore yield to the vast majority and sincerely thank everyone for considering my ideas. I know that present policies are not quite good enough to encourage some with good personal knowledge to offer it, so maybe I'll be back here again someday with a better and less objectionable proposal. David Spector (talk) 21:50, 9 September 2013 (UTC)

VisualEditor's reference dialog, especially the way it handles citation templates, is being redesigned. Suggestions and opinions are wanted at mw:VisualEditor/Design/Reference Dialog. (It's over at the sister project, Mediawiki.org, but you should be able to use your Wikipedia username/password to login there.) This is a practical, focused workshop to improve the design. Views from people who regularly edit at other Wikipedias are also needed, since some Wikipedias do not use the same citation templates (or don't use citation templates at all).

If you aren't using VisualEditor regularly, it may be helpful to turn on VisualEditor (at the bottom of Special:Preferences#mw-prefsection-editing) and try to add or edit a couple of refs in your sandbox, or to read Wikipedia:VisualEditor/User guide first. Whatamidoing (WMF) (talk) 19:02, 16 October 2013 (UTC)

Pass... I still use the "old fashioned" <ref>xxx</ref> format... which I find much more flexible and easier to amend when necessary. It isn't that I dislike the VisualEditor's citation template... I dislike the whole concept of citation templates to begin with. Blueboar (talk) 19:42, 16 October 2013 (UTC)

I think it would be beneficial for someone like you to go over to MediaWiki and remind the devs that manually formatted citations are important and desirable. I haven't heard of any plans to remove that option, but it would be unfortunate if only the proponents of citation templates had any say at all in the design. Whatamidoing (WMF) (talk) 18:39, 17 October 2013 (UTC)

I also think that citation templates are a bad idea. They make it so that only a fraction of editors (those who have learned the syntax for the particular template) can edit them. Unlike visual editor, getting rid of them would actually do some good in lowering the entry bar to editing. North8000 (talk) 17:03, 21 October 2013 (UTC)

I think that citation templates will be less intimidating in VisualEditor compared to in wikitext, since it's looking like a fill-in-the-blank system like several of the tools on the English Wikipedia. But it would still create the same templates for people who are using wikitext. If you've got opinions, though, I really think you should leave a note at the other discussion. Whatamidoing (WMF) (talk) 17:19, 22 October 2013 (UTC)

This page does not help with the case where there is an issue on which the editors can find no reliable source.

For example, take the case of whether Rooibos tea is a stimulant or not. The question is highly relevant to the article, but there are no reliable sources either way. How should this be handled?

To ignore the issue is to silently imply that the tea is not a stimulant, especially since the article states that it contains no caffeine. Willbown (talk) 17:31, 21 October 2013 (UTC)

Hi there. Please note that new threads should generally be posted at the bottom of Talk pages. Thanks!

Given your example you can look at it in one of two ways - either the information shouldn't be added because no reliable source can be provided...or the information can be added, but any editor who questions it will have the right to tag and/or remove it for lacking sources.

Personally I prefer not to see unsourced information added to Wikipedia articles, but YMMV. DonIago (talk) 18:50, 21 October 2013 (UTC)

• If you can't find a source for whether or not rooibos is a stimulant then the issue should not be mentioned in the article. I suggest that we should wait for a decent study to be published (not a study funded by people who want to sell rooibos tea, but proper science done by a proper scientist) before we make any claims about that.—S Marshall T/C 00:42, 22 October 2013 (UTC)

S Marshall is correct. If we can not find any reliable sources to support a statement, then we should not say it. Blueboar (talk) 01:06, 22 October 2013 (UTC)

Agree with S Marshall. And I agree with Blueboar's statement if the qualifier "if the material has been challenged" is added. Sincerely, North8000 (talk) 01:21, 22 October 2013 (UTC)

North... No matter how much we may want to say something in an article... if we can not find a reliable source to support the statement, we should not include it. To put it another way... we should self-challenge the statement (on both WP:V and WP:NOR grounds) before we add it. Blueboar (talk) 13:03, 2 November 2013 (UTC)

Follow the reliable sources. If they say that it contains no caffeine but do not say whether or not it is a stimulant, then that's how Wikipedia should treat the subject. If you think that this silently implies that it is not a stimulant, then you would similarly think that the reliable sources silently imply that it is not a stimulant. This is just following the reliable sources. To add to the article something that implies that Rooibus tea might be a stimulant would be questionable speculation not supported by a reliable source and thus not desirable for Wikipedia, in my opinion. --Bob K31416 (talk) 02:37, 3 November 2013 (UTC)

• Thank you all for your kind responses. I wish to disagree with all of you.

Donlago, you say, "I prefer not to see unsourced information added to Wikipedia articles". At no point have I suggested doing such a thing. Blueboar, you I think have misunderstood me in the same way when you say, "If we can not find any reliable sources to support a statement, then we should not say it."

S Marshall, I think you can only take your position because because you don't think the question of whether Rooibos is a stimulant or not is important.

Bob K31316, your argument does not hold as many of the publications that count as reliable sources do not have the same remit or goals or editorial guidelines or audience as Wikipedia.

I am not proposing stating as fact things for which there is no reliable source. What I am saying is that there are times when an issue needs to be addressed and there is no reliable source. And I am asking for guidance on how to do that, and more precisely saying that this page should address the issue.

This is a much more general issue than Rooibos tea. For example, suppose we have news of a bomb exploding in London but have no reliable source on whether there are any casualties. Then, by the reasoning of the respondents here, Wikipedia would not mention the question of casualties. What I am saying is that it would be better to say to the reader, "We have no reliable information on whether there are any casualties or not."

To repeat myself, I find it odd that this page does not address what seems to me a very common issue. Willbown (talk) 23:00, 3 November 2013 (UTC)

You cannot prove a negative. Who is the "we" supposed to be? If, on the other hand, you can find a reliable source stating that the question is open, you can include it. --Stephan Schulz (talk) 23:19, 3 November 2013 (UTC)

Willbown: You say "there are times when an issue needs to be addressed...." Yes, but not necessarily on Wikipedia. You should keep in mind WP:what Wikipedia is not. E.g., it is not newspaper, nor a publisher of original reporting, nor a forum. Wikipedia is an encyclopedia, which is a summary of existing knowledge. In your specific case, it appears there is no knowledge of whether Rooibos is a stimulant. At best you could only say "there is no knowledge that ...." And it is questionable whether even that is sufficiently notable to warrant mention. ~ J. Johnson (JJ) (talk) 23:22, 3 November 2013 (UTC)

What makes you, Willbown, think that this tea is or might be a stimulant? Is it personal experience? Then it can't enter Wikipedia. But if it is a widely held belief, even if not yet substantiated, then you may be able to find a eliabel source that says that such a view is widely held. If so, that can be described in the article. Or if ther have been published claimsthat it is a stimulant, even if not accepted by a scientific consensus, it might be stated that "Soandso claims that it is a stimulant, but..." provided that this doesn't give WP:UNDUE weight to a minority or fringe claim. DES ^(talk) 11:09, 4 November 2013 (UTC)

Is Obama secretly a alien lizard overlord from outer space? I cannot find reliable sources that confirm or deny this. I think it's incredibly important to state my suspicion, and the suspicion of many others on the internet, that Obama may be a secret alien lizard overlord from outer space. Hyperbole aside, the verifiability policy represents a project-wide consensus that Wikipedia should only host content that is at least hypothetically verifiable by reliable sources. As others have mentioned, we as editors are permitted to simply mention the existence of rumors/questions/uncertainties, without taking a stance on their validity. This is assuming that reliable sources have mentioned those rumors/questions/uncertainties. The historical consensus is very clearly against simply throwing in whatever a random editor thinks is important in the absence of reliable sources. Wikipedia is not a soapbox for what you personally think is important. Someguy1221 (talk) 11:25, 4 November 2013 (UTC)

Please add the following important reference book about Lee Harvey Oswald and Marina Oswald Porter: MARINA AND LEE By Priscilla Johnson McMillan (Harper and Row,1977.) Easily the best biography about these two significant historical individuals.

Richard Polese [Personal information redacted.] — Preceding unsigned comment added by 71.222.230.19 (talk) 20:52, 18 November 2013 (UTC)

Please note: As stated at the top: "This is the talk page for discussing improvements to the Verifiability page", which is about a Wikipedia principle, not Wikipedia content. If you think this book could lead to an improvement on some particular Wikipedia article you should mention it on the article's "Talk" page.

Also: you probably don't want your personal information posted here, so I have removed it. ~ J. Johnson (JJ) (talk) 23:02, 19 November 2013 (UTC)

Within the posts at Talk:Li_(surname)#RFC_regarding_multiple_Chinese_surnames_transliterated_to_the_same_surname_in_English is a proposal and some support for using Chinese characters in titles to disambiguate between different Chinese surnames that transliterate to the same word (Li in this case). Reliable English language sources do use Chinese characters to disambiguate, so there appears to be a conflict between the polices WP:V and WP:UE. Please provide input on how to resolve this there.
(After reading the top of the talk page, I'm not sure it's appropriate to post this request here. Is there a better place than the policy page to ask for input on potential conflicts between policies?)--Wikimedes (talk) 06:59, 2 December 2013 (UTC)

• If there was a conflict between policies then I would think WP:V is the one to follow, but I'm not convinced this is, strictly speaking, a verifiability issue. I don't think it's "challenged or likely to be challenged" that there are several Chinese-language surnames all transliterated into English as "Li", and I think the translations themselves are unchallenged as well. It seems to me that the issue is simply whether we use the Chinese character to disambiguate or an English translation such as "Li (surname meaning this)", "Li (surname meaning that)" or "Li (surname meaning the other)".

  If you want my opinion, it's that the English language wikipedia is designed to be read and searched by people using latin-alphabet characters so I'd go for the latter on an ease-of-use based approach, but that's all it is: my opinion.—S Marshall T/C 08:38, 2 December 2013 (UTC)

I concur with S Marshall, because the policies aren't intended to conflict, but to help readers. --Lexein (talk) 13:18, 2 December 2013 (UTC)

This debate (over how to disambiguate the various "Li" articles and lists) has been going on for a while. There have been multiple AfD discussions, and several RfCs. I have suggested several times that BEST the solution is to create an article that does not need any disambiguation at all... to have one single article explaining how there are various Chinese names that are presented in English as "Li"... instead of separate articles for each of the various Chinese language surnames that are transliterated into English as "Li". To my mind, one article is the best way to explain all the complexities of the name/names. And if we discuss it at all in one single article, there is no need for disambiguation in the title... the disambiguation can be left to the body of the text and presented through a few very to read charts. I actually created a draft of what this unified page might would look like... see User talk:Blueboar/drafts - Li (surname). Blueboar (talk) 19:59, 2 December 2013 (UTC)

It is one thing to have "one single article" explaining how Chinese names are to be disambiguated, and quite another thing to actually disambiguate them in a title (where an id number would suffice). But this problem is hardly unique to Wikipedia: scientific journals are also wrestling with how to distinguish authors. E.g., see "One Wei or Another" and "Are You Ready to Become a Number?" in Science. ~ J. Johnson (JJ) (talk) 23:18, 2 December 2013 (UTC)

But if we go with the "discuss them all in one article" approach, then we wouldn't need to "actually" disambiguate... We would not have multiple article titles that need disambiguation... there would just be one article title. Blueboar (talk) 02:36, 3 December 2013 (UTC)

But, Blueboar, that's the wrong way to approach the subject. These are three separate surnames, with three separate meanings, denoted by three separate phonemes. Although they're all transliterated into English as "Li", that's only because of language tonality. They're entirely separate, unconnected subjects. The reason we have separate articles on Lee (English surname), Lea (surname) and Leigh (surname) is because they're genuinely distinct names, with separate etymologies, that happen to sound the same. Would you lump all those in with your one combined article on "Li"? as well?—S Marshall T/C 17:03, 3 December 2013 (UTC)

Actually, The problem with the various "Li"s is actually the exact opposite of the issue with Lee, Lea and Leigh. With those we have three distinct names that are homophones... they sound the same but are spelled differently. With Li, on the other hand, we have the opposite... multiple names that sound different (at least in Chinese) but are spelled the same (in English). Wikipedia is a visual medium. Which means we base articless on spelling rather than sound sound. Nothing says we can't cover three separate names that are spelled the same in one single article... especially if doing so will clarify for our readers the complicated fact that they are not actually the same. If need be, you can consider this a WP:Ignore all rules situation... its an exceptional situation that I think needs a somewhat creative solution. Blueboar (talk) 20:24, 3 December 2013 (UTC)

Two questions. First: is the concern here about names as names, or names of particular persons? Second: if there was some language where the names Jones and Johnson were transliterated in the same string of characters, should a Wikipedia in that language cover those names (whether as names, or of particular persons) in a single article? Or separate articles? ~ J. Johnson (JJ) (talk) 23:14, 3 December 2013 (UTC)

I think "Should it?" is the wrong question. This isn't a matter of "should" or "should not" (which sounds too legalistic to my ear)... but one of "what makes the most sense in this situation". If covering two names in one article is deemed the best way to explain the how one string of characters can stand for both names... why not do so? Blueboar (talk) 12:34, 6 December 2013 (UTC)

The approach to explain all of the various variations of the English surname Li in a single article appears to be the best approach to me. However, if there end up being separate articles on the various Chinese flavors of the English surname Li, then I think the parenthetical clarification in the title ought to remain completely in the English alphabet using a Latin keyboard. N2e (talk) 16:03, 23 December 2013 (UTC)

Content farms "[employ] large numbers of often freelance writers to generate large amounts of textual content which is specifically designed to satisfy algorithms for maximal retrieval by automated search engines. Their main goal is to generate advertising revenue through attracting reader page views as first exposed in the context of social spam." Surprisingly, none of Wikipedia's policy pages ever suggest that content farms should not be considered reliable sources, as they are rarely edited, they are essentially blogs that users are being paid to write, and we should not be providing traffic to websites as sources whose only real purpose are to game search results for profit rather than journalistic integrity. As such, I think we need to add examples of them to relevant areas of the page.

In fact, I know a few content farms (i.e. Examiner.com and Associated Content) are already blocked by the spam filter, and I personally think some of Huffington Post's articles are designed for a similar purpose given this article. ViperSnake151  Talk  02:54, 23 December 2013 (UTC)

I reverted you; it is best that you get WP:Consensus for those changes here on the talk page first. Like the policy page states, "Changes made to it should reflect consensus." Flyer22 (talk) 03:17, 23 December 2013 (UTC)

This isn't the right policy page for that question anyhow. Discussions on wp:RS/N will show what the community consensus is. That may then be added to wp:RS if needed. LeadSongDog come howl! 04:04, 23 December 2013 (UTC)

• I Support the addition. - Or, at least, I think this is the right page for this discussion... after all, this policy does have a section entitled WP:V#Sources that are usually not reliable... and what Viper wants to add seems to fit with everything else in that section. RS/N would be the right place to discuss whether a specific site qualifies as a "Content farm" or not... but I think including a generalized caution about Content farms is appropriate for this policy. Blueboar (talk) 14:29, 23 December 2013 (UTC)

• I suspect that we should not conflate policy and guidelines any more than they already are. This suggestion sound like it should be part of a guideline. --Bejnar (talk) 05:09, 24 December 2013 (UTC)

The section on non-English sources in this policy suggests to me that there should be a guideline saying something such as: When citing a non-English source, it is appropriate to provide at least a translation of the title into English. Similarly, if the source is in a non-Latin alphabet, a romanisation or transliteration of the author's name, and a romanisation/transliteration/translation of any journal title, newspaper name or publishing house is appropriate. Is there any such guidance? If not, where would be the best place for it? Wikipedia:Citing sources#What information to include? --Bejnar (talk) 05:09, 24 December 2013 (UTC)

No... citations do not need to be be in English... as long as someone (in this case someone who knows the language) can check the source, then the information is verifiable. What needs to be in English is the text of the article... so if the article quotes from a non-English source, or if it mentions the work/author in the body of the text... that would need to be translated/transliterated (for the benefit of our English-only readers). Blueboar (talk) 13:59, 24 December 2013 (UTC)

Wikipedia now seems to have a policy about conflict-carrying sources: 'Questionable sources are those that ... have an apparent conflict of interest.[fn8, with extensive text, from e.g. the NY Times] Such sources ... should only be used as sources of material on themselves.' That's from here.

That quotation is a generally salutary addition to Wikipedia, but it surprised me, because about four years ago all of the many Verifiability-page denizens who spoke up were opposed to the proposal that there be some type of policy on conflict-carrying sources (as distinct from the policy on conflict-carrying editors; that latter policy already existed). Such opposition is visible here (about 28 screenfulls).

The current conflict-carrying sources policy arrived in the text without discussion/thinking by those many Verifiability habitués. (That's my conclusion from reviewing the history.) So they now have a chance to exclude the policy, as they did before. Bo99 (talk) 17:48, 26 December 2013 (UTC)

• I think that you have a point. WP:Ver, wp:npov and most policies generally say that if the "floor" of wp:RS is met, that nothing else (e.g. objectivity & expertise with respect to the item which cited it) matters. (But in practice, those considerations are assessed and applied at wp:RSN) But this particular sentence goes to the other extreme, basically saying that apparent-COI sources should be used in only very restricted situations. COI isn't just financial interest, it includes other advocacy. Perhaps we should find some more general and middle ground. Basically that the strength of sourcing is defined by expertise and objectivity with respect to the item which it is used as a cite for, as well as the other wp:rs criteria. And stronger claims need stronger sourcing and vica-versa. This "continuum" would be workable. Sincerely, North8000 (talk) 18:08, 26 December 2013 (UTC)

I think there are also other issues with this. The citation for COI mentions 'professional considerations' and 'the trust that people have in professionals' - but these are just words written within the citation in WP here. Sourced to Columbia University is: 'A conflict of interest is a situation in which financial or other personal considerations have the potential to compromise or bias professional judgment and objectivity'. However this ignores a large reliable literature critiquing career 'professionals' and related institutions and bureaucracies as inherently conflicted by financial and other motives, and that there is a widespread distrust of them especially in some quarters.

Furthermore, questionability is exampled as 'widely considered by other sources to be extremist', yet this stands out as a non sequitur from the preceding definition of 'poor reputation for checking the facts, lack meaningful editorial oversight, or have an apparent conflict of interest.' How does the latter relate to the former?

The above together suggest to me that Wikipedia is assuming a position that established professional bureaucracies are inherently reliable and free of conflict, and any extreme opposition to them - especially if involved in revolutionary political or other actions - is inherently unreliable. Sighola2 (talk) 09:07, 28 December 2013 (UTC)

Wikipedia articles should represent what mainstream sources say. That does not mean the mainstream is correct or unbiased, but it would be impossible to write an encyclopedia without any standard for sourcing. The policy is not overly onerous. Columbia refers to "reasonable" belief that a conflict may compromise a writer. So a website funded by the tobacco industry would be questionable, while the peer-reviewed work of a professor who received tobacco industry funding would not be. The reasoning is that a conflict may cause someone to falsify the facts. But there are good sources written by people from across the political spectrum, and their views are not something I would consider in selecting a source. TFD (talk) 09:44, 28 December 2013 (UTC)

I appreciate your last point and I would prefer the policy to clarify that it's not suggesting otherwise. If, as the policy itself says, a source has sufficient fact-checking, editorial oversight and is not overly affected by COI, then it is acceptable in terms of reliability? Whether 'mainstream' or 'alternative'? (of course policies like due weight & 'exceptional claims require exceptional evidence' may apply).

The example of tobacco is interesting because several reputable medical journals now completely refuse to accept research (by anyone) with such funding, recently the BMJ, so they clearly consider it questionable ('The tobacco industry, far from advancing knowledge, has used research to deliberately produce ignorance and to advance its ultimate goal of selling its deadly products while shoring up its damaged legitimacy.'). Sighola2 (talk) 12:49, 28 December 2013 (UTC)

Another notable thing about the current conflict-carrying sources policy is that its stated remedy is clear and severe (which has a positive side and a negative side). The current policy requires that a conflict-carrying source be disallowed, period -- 'Such sources ... should only be used as sources of material on themselves.' (That's from here.) That's more severe than the standard conflict of interest approach, which was proposed and explained in detail here, in the 2009-Jan-22 posting, 'Disallow, Or Disclose'. Such 'disclose' option means sometimes allowing the conflict-carrying source but disclosing the conflict (and the resulting risk of untruthfulness), and thereby discounting the proposed source to some extent. Bo99 (talk) 15:11, 29 December 2013 (UTC)

I think that's a good point. But it seems to raise issues of overlap with WP:NPOV which talks about assessing bias and either discarding the source or putting it in context. As anyway do the existing mentions here of whether a source is 'promotional' or 'extremist' or has too much 'conflict of interest'. I do wonder whether questionability in terms of factual reliability, as in this policy, shouldn't just restrict itself to whether sources seems to lack sufficient fact-checking and editorial oversight ('rely heavily on unsubstantiated gossip, rumor or personal opinion'). But if some assessment of bias is included, I do think COI is the most appropriate concept. Sighola2 (talk) 12:02, 30 December 2013 (UTC)

I agree with you that, someday, the conflict-carrying sources policy belongs logically not within the word 'Verifiability' but rather within the word 'Neutrality', someday. Bo99 (talk) 19:09, 31 December 2013 (UTC)

Another notable thing about the current conflict-carrying sources policy is that it is easily triggered (which is standard): 'A conflict of interest is a situation in which financial or other personal considerations have the potential to ... bias professional judgment ... [, in the view of a hypothetical] reasonable person ... . [So for example] conflicts of interest include ... [a professional] having financial interests in the [subject] being reported ... .' (That's from here.) Therefore a professor who receives cigarette-industry funding would have a conflict of interest, and the professor's text about cigarettes would be barred as a Wikipedia source under the current policy (even if the professor's work is reviewed by peers with no conflict of interest). I mention that example because it was raised earlier. But the above-quoted wording is much more broad (which is standard): a COI (in a source for a given proposition) is any situation in which any consideration has any potential to bias the judgment of the source's author (regarding the given proposition). Bo99 (talk) 16:01, 3 January 2014 (UTC)

So, in sum:

The conflict-carrying sources policy is a substantial improvement, over having no such policy at all as the Verifiability denizens demanded.

However, the current policy needs work: it deviates from the norm in that there is only one, severe remedy: disallowance; with no possibility of allowing-but-disclosing. (For background reading on such norm, see e.g. the article.)

And such disallowance makes the current policy contradict at least two other Wikipedia principles:

1. 'Biased sources are not inherently disallowed based on bias alone ... .' Neutrality Policy

2. 'While a source may be biased, it may be reliable in the specific context.' Source-Reliability Guideline

Anyone who wants to make the correction is encouraged to do so. However, i'm not going to spend the time to do the correction myself again, because of that prior attack by the Verifiability denizens on the proposal that there be any such policy at all.

No need to respond. Bo99 (talk) 16:36, 6 January 2014 (UTC)

Don't mean to jump in again on the thread, but I do find this odd. Looking back Bo99 was explicitly accused on his talk page of being a troll by apparently senior Wikipedia denizen(s), apparently for pursuing a logical analysis of this issue in a certain way, yet later down the line it's inserted into the policy by someone else without objection, and now is not even commented on. Sighola2 (talk) 17:33, 8 January 2014 (UTC)

Why isn't there a direct definition of a SPS in this policy? Right now -- as I read it -- there is no defintion at all.

Definitions generally have a format that reads something like "A self-published source is a source whose author and publisher are the same person or entity."

The reason that some definition is needed is because I have heard editors argue that a SPS is:

• "one where there is COI between the author and the content's subject matter of the publication"
• "one where any user can add content"
• "one where the media is the Internet hosted by a free service" (Examples: GeoCities, Flickr, WebCite, Yahoo)
• Even a few other editor definitions that now escape my mind

It seems an actual complete definition can avoid at least some of these various interpretations. A common manifestation of the problem is that every editor arguing SPS is always claiming that the other SPS editor "does not understand the SPS policy" and the other editor arguing SPS is always claiming "No, you are the one who has the wrong interpretation of the policy". So the argument never ends except thru a long debate and the addition of more editors to break the stalemate thru the mandated consensus. It seems that if we had a better definition in place, a good portion of these debates and consensus-building sessions wouldn't be necessary.

To elaborate a bit, the problem, IMO, is that when we read WP:SPS it starts by explaining the dilemma and not by giving the definition (which is never given anyway). It says, "Anyone can create a personal web page or publish their own book, and also claim to be an expert in a certain field." But that's not a definition. Worse yet, it then continues that "self-published media, such as books ... are largely not acceptable as sources", which would seem to imply that anything that's a book is a SPS, and therefore not valid as sources for Wikipedia! (I know the current "definition" intended to say "SPS books" instead of just "books", but that is not what it says.) In any event, the real problem is the lack of an actual definition, not so much the "book" travesty. Can we add a rock-solid definition of what is meant by SPS? Giving several examples could also aid in giving a robust definition, as would stating what a SPS is not. Thanks, Mercy11 (talk) 06:09, 11 January 2014 (UTC)

I agree that a definition is needed, rather than just examples as now. Could the already wikilinked terms such as 'publish their own book' provide the relevant text? The ambiguous wording re SPS books seems like it could be easily edited to clarify. Sighola2 (talk) 10:37, 12 January 2014 (UTC)

Discussions on the related policies and guidelines, such as Identifying reliable sources, have always been based on the idea that self-published sources are those created by one or a few individuals, and published through a channel that does not exert any meaningful editorial control, such as a vanity publisher or internet service provider. Adopting a more expansive definition would require a complete overhaul of all the related policies and guidelines.

One point that might be considered is whether meaningful editorial control must occur before publication, or may take the form of punishment after publication. For example, an elected official may make a truly asinine statement, but is likely to lose the next election if he/she does so. Jc3s5h (talk) 12:39, 12 January 2014 (UTC)

On that last example... after the fact "punishment" like that is not a form of editorial control. If we said that, someone would be sure to argue that re-election after making an asinine statement qualifies as "peer review" and the asinine statement is thus verified.

That said... meaningful editorial control can take place both before and after publication... for example reliable newspapers have an editorial staff that (presumably) reviews the stories before publication... they also have editorial control after publication... in the form of corrections/errata columns. Blueboar (talk) 14:05, 12 January 2014 (UTC)

Can we get some more uninvolved editors to participate is this RSN discussion? A Quest For Knowledge (talk) 13:31, 15 January 2014 (UTC)

Assuming that you mean the "Brad DeLong blog RS?" section? North8000 (talk) 18:59, 15 January 2014 (UTC)

Yes. I've reformated my original post to make the link more obvious. A Quest For Knowledge (talk) 21:24, 15 January 2014 (UTC)

Can any One Please tell me the actual place of the Eldorado so i can find more artifacts for it and help it, to prove it true — Preceding unsigned comment added by 117.196.9.145 (talk) 10:36, 16 January 2014 (UTC)

I'm afraid that this is not the proper place for that kind of request. Try remaking it at Reference desk/Humanities. Regards, TransporterMan (TALK) 14:31, 16 January 2014 (UTC)

According to this, the Eldorado is located at 140 Water Street, Henderson, NV. Blueboar (talk) 01:50, 20 January 2014 (UTC)

This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.

OMS - which was the Oriental Missionary Society - then OMS International is now One Mission Society. I am a member of this organization

72.45.60.102 (talk) 02:17, 20 January 2014 (UTC) Gwen Pinkerton

  Not done: as you have not requested a change. OMS does not appear on this page at all - I am not sure what you are asking to change, or where.
If you want to suggest a change, please request this in the form "Please replace XXX with YYY" or "Please add ZZZ between PPP and QQQ".
Please also cite reliable sources to back up your request, without which no information should be added to any article. - Arjayay (talk) 19:49, 20 January 2014 (UTC)

The reason for the ban on using SPS as sources for living people is clear: to avoid spreading dubious or false statements about a living person. We want to use sources like newspapers because they fact-check in a way that SPS don't (and often can't).

However, I see no reason for this ban to apply to statements of opinion, which establish no facts about an individual. For instance, suppose Neil deGrasse Tyson (on his blog) were to praise 'Cosmologist Katy' as "the most significant physicist of our generation", citing her recognized achievements (as established by other sources). I see no BLP problem with citing this remark. Is there any difference substantive between quoting this remark from his blog and quoting it from a newspaper? I submit there is not. I propose that the ban on using expert SPS as sources for living persons be explicitly confined to uses of SPS which establish statements of fact. Steeletrap (talk) 01:04, 26 January 2014 (UTC)

Contrary to the statement,...if the information in question is really worth reporting, someone else will probably have done so there are circumstances in which RS do not directly take up every utterance or position that is relevant to a given topic. The case of fringish and narrowly focused ideological statements and entrenched ideological positions are a case in point.

The specifics that generated this thread, which can be seen at the current RS/N thread [[11]]. In short, the statements of Murphy, a popularizer of fringish libertarian economics "theories" ('predictions' in the case at hand) are not deemed legitimate enough to merit a more thorough treatment. When Murphy challenges Krugman to a debate (with Murphy's 'backers' offering $100,000 to a charity of Krugman's choice), Krugman declines, stating that he will not provide a "platform" for Murphy's predictions, which get everything wrong.

In other words, it is not necessary for Krugman (and Brad DeLong) to spend much time and text to debunk the utterly errant posturing of an ideologue seeking publicity for his political positions. Krugman mentions DeLong's refutations of Murphy and then addresses Murphy's response, but it does not seem reasonable that the brief statements made by DeLong in his blog should be excluded as RS in advance. Here are links to those statements[12][13]. And here are the links t Krugman's columns Is Our Austerians Learning? On Not Learning, Continued

The policy statement excluding the use of personal blogs of experts commenting with respect to matters in their respective field of expertise is being used as a rationale to enable Murphy to be passed off as some sort of mainstream economist, because the fact that he is so marginal (or fringe) that his ideas do not merit more thorough treatment by the mainstream. It is only his political associations and profile that make any response necessary. If Wikipedia doesn't cover the limited response made by prominent mainstream academics to the fringe, Wikipedia could end up serving as the platform to promote the fringe insofar as the minimal criticism of fringe statements by prominent mainstream is not aired due to this policy.--Ubikwit ^連絡 _{見学/迷惑} 06:06, 26 January 2014 (UTC)

I don't think this is really a Verifiability (SPS) issue... it seems more of a WP:NPOV (and specifically a WP:DUE weight) issue. It is verifiable that Krugman and other experts hold certain views of Murphy and his theories... the question is how much weight to give these expert viewpoints. That depends on context. Reporting an experts SPS opinion may be absolutely appropriate in one article, and may be completely undue in the context of another article. Certainly in the context of a bio article about Murphy, we would give his views more weight than that of others. He is the subject of the article after all. However, that does not mean we necessarily should give his detractor's no weight at all. The key is report on the expert's views of his theories... not their views on him personally. Blueboar (talk) 14:08, 26 January 2014 (UTC)

The question here was in regard to the SPS issue, but if you think that it is not an issue, then we should work on a revised text for the statement at issue.

Regarding the other points--probably article talk page issues, but--the Krugman and DeLong have addressed Murphy's "predictions" as well as his continued adherence to those in the face of countervailing evidence. Accordingly, they have, to a certain degree, called into question his competence as an economist, pointing to contradictions in his stances as well as dogmatic intransigence regarding the inflation prediction.

Krugman's second installation ends with the following passage

Being willing to learn matters. Unfortunately, that willingness seems absent from many people who consider themselves economic experts.

I don't know if you would read that as a comment on Murphy personally or on his competence as an economist, but that is from a NYT opinion piece, not SPS, in any case. DeLong says, on his blog,

The most terrifying thing of all is that being completely, comprehensively, unmistakably, fundamentally, fatally, totally wrong has not led Robert Murphy to rethink or modify any of his analytical positions or ideological beliefs by even one iota.

I don't know if that is addressing Murphy's theories or his refusal to reconcile his prediction with extant theories in a theoretical manner. He does ascribe both "analytical positions" and "ideological beliefs" to Murphy.

With regard to weight, the statements of DeLong and Krugman are brief and to the point, as the "theories" at issue are not really theories per se, but a simple prediction that Murphy refuses to admit was wrong, and a contradiction with respect to other theories in the rationale he offered.

Meanwhile, though DeLong and Krugman's statements are brief, there is Murphy's lengthy reply. Note that no one appears to have ventured an explication of any of that, but the fact that he has even ventured such a reply is demonstrative of further notability of the exchange. He also uses a cartoon and various political images and the like to add a sensational air of theater to the post.--Ubikwit ^連絡 _{見学/迷惑} 14:34, 14:55, 15:26 26 January 2014 (UTC)

That is indeed the example which prompted this discussion. But I don't want a one-time exception; I want a general policy change. It makes a lot of sense to ban SPS from being sources for facts about living people, since blogs etc don't tend to fact check as well as newspapers. It doesn't make any sense to ban expert SPS as sources of opinions about, e.g., the work or theories of a scholar in their field; there is no substantive difference between quoting from their blog and a comment they made to a newspaper. Such statements should be allowed, so long as they are strictly used to provide the opinion of the cited expert, and not to establish any facts about the living person. Steeletrap (talk) 19:34, 26 January 2014 (UTC)

SPS ban is especially needed on opinion statements since on a self-published blog individuals are more likely to let lose with negative, derisive, exaggerated statements than they would in edited WP:RS publications. Carolmooredc (Talkie-Talkie) 23:11, 26 January 2014 (UTC)

I disagree. The ban clearly intends to prevent the dissemination of false or dubious information (i.e. facts) about a person, not to squelch expert criticisms of his or her work. Scathingly negative statements about living persons appear both in newspapers and blogs. Naked personal attacks should of course be excluded. But a categorical ban on SPS has little to do with preventing those. Steeletrap (talk) 23:26, 26 January 2014 (UTC)

Expanding the definition of acceptable sources to make it easier to include "scathingly negative statements about living persons" seems like a very bad idea and a step in the wrong direction. One of the problems with self-published sources is that they lack independent reviewers (those without a conflict of interest) validating the reliability of the material. Most books and journal articles undergo strict editorial review before publication. Self-published sources do not have impartial reviewers to evaluate the accuracy or fairness of the material. That's why the use of self-published sources is strongly discouraged and in the case of third-party sources regarding living people, are strictly forbidden. What's more, individual policies should not be considered in isolation with one other. There's also the issue of WP:WEIGHT. If the information in question is really worth reporting, someone else will probably have done so. Writing encyclopedia articles should be easy. If there are no third-party reliable sources to support some content, that's a strong indication that the content does not belong in an encyclopedia. A Quest For Knowledge (talk) 01:41, 28 January 2014 (UTC)

The editor who initiated this discussion is part of an Arbitration where her use of SPS in BLPs was a major motivating issue for ArbCom to take the case. So it seems inappropriate that she even bring it up here at this time. I've been hassled even for merely trying to get clarification at policy pages during talk page disputes, so I'm sure trying to change policy in the middle of an Arbitration is questionable and the editor should stop trying to do it. Carolmooredc (Talkie-Talkie) 11:02, 28 January 2014 (UTC)

I think that attacks and smears on people nearly always use methods other than making clearly factually false statements. And so I think that the comment above essentially that the prohibition is only intended for or applicable to the latter is a mis-statement. North8000 (talk) 11:48, 28 January 2014 (UTC)

Not clear what comment you are replying to or what you mean :-( Feel free to rephrase and erase my question. Carolmooredc (Talkie-Talkie) 12:48, 28 January 2014 (UTC)

I suggested at the RS/N board that the present policy as worded did seem to preclude use of the source, but that as there seemed to be a lack of consensus regarding the interpretation and application of the policy, the policy issue per say should be brought up here, perhaps accompanied with proposals for changing the wording.

Comments have continued at the RS/N discussion and both have been somewhat productive, in my opinion. As a result, I don't see a way to reword the policy to ensure that loosening the constraint doesn't backfire. The intent of the policy is debatable (and has been debated in a productive manner), but for the present case, it has sufficed thanks to the columns by Krugman.

I think that it has to be borne in mind that Krugman just expanded on DeLong's commentary in many regards. They both consider Murphy to be an ideologue and called his competence (and/or sincerity) into question. The policy seems to play an important role in preventing Wikipedia from citing a cascade of blog-based insults, and that function has to be maintained if the policy were to be relaxed to allow comments such as those made by DeLong to be cited directly.

This thread was not opened directly in relation to the Arbcom case, however.--Ubikwit ^連絡 _{見学/迷惑} 17:27, 28 January 2014 (UTC)

The section on non-English sources in this policy suggests to me that there should be a guideline saying something such as: When citing a non-English source, it is appropriate to provide at least a translation of the title into English. Similarly, if the source is in a non-Latin alphabet, a romanisation or transliteration of the author's name, and a romanisation/transliteration/translation of any journal title, newspaper name or publishing house is appropriate. Is there any such guidance? If not, where would be the best place for it? Wikipedia:Citing sources#What information to include? --Bejnar (talk) 05:09, 24 December 2013 (UTC)

No... citations do not need to be be in English... as long as someone (in this case someone who knows the language) can check the source, then the information is verifiable. What needs to be in English is the text of the article... so if the article quotes from a non-English source, or if it mentions the work/author in the body of the text... that would need to be translated/transliterated (for the benefit of our English-only readers). Blueboar (talk) 13:59, 24 December 2013 (UTC)

Blueboar seems to assume that English speaking readers are not interested in sources, and that sources are only for Wiki gnomes. That is not the case. I am proposing new guideline language, not a new concept. Please discuss. Is this the wrong place, should I be in Wikipedia:Citing sources? --Bejnar (talk) 18:02, 13 January 2014 (UTC)

You misunderstand my point. When an English speaking editor/reader is interested in a source that is written in a language he/she does not understand, the editor/reader has several options... 1) the reader can go take a language class so he/she can read the original... 2) the reader can ask someone who does understand the language for a translation. Either way... the information is verifiable. This is no different than (for example) a source that requires payment to access. A specific editor may not be able to access it (or may not want to go through the expense of accessing it)... but as long as someone else can do so on his/her behalf, then the source is verifiable. Blueboar (talk) 14:11, 18 January 2014 (UTC)

I feel that there are two important considerations here. First: translating a source, or substantial portions of a source, and posting the translation on Wikipedia could constitute a copyright violation. Second: we have to be careful not to create a tyranny of ignorance. I mean that multi-lingual editors are valuable members of our community and we don't want to discourage them from using foreign-language sources by forcing them to do extra work transliterating the source into something doubting monoglots can understand... Wikipedia does far too much pandering to people who don't understand the subject area they're editing in. In my view the current language of the policy is correct and editors who want to check something, but don't speak the language, should ask someone who does.—S Marshall T/C 21:19, 15 January 2014 (UTC)

I am not talking about translating the source, I am talking about the metadata.

• Example:

Original:  Борхсениус, Н. С. (1966). Каталог щитовок (Диаспидоидеа) мировой фауны (in Russian). Москва: Академия наук СССР – Зоологический институт.

Proposed:  Борхсениус, Н. С. (Borchsenius, N. S.) (1966). Каталог щитовок (Диаспидоидеа) мировой фауны (A catalogue of the armoured scale insects (Diaspidoidea) of the world) (in Russian). Moscow: Академия наук СССР – Зоологический институт (Zoological Institute of the USSR Academy of Sciences).{{cite book}}: CS1 maint: multiple names: authors list (link)

Which is more helpful to the reader? --Bejnar (talk) 05:25, 18 January 2014 (UTC)

• To the reader who speaks Russian, they're equally helpful. To the reader who doesn't, and will therefore never be able to evaluate the source anyway, they're also equally helpful.—S Marshall T/C 08:54, 18 January 2014 (UTC)

Actually there are English language sources that will help a non-Russian speaking person evaluate the Zoological Institute of the USSR Academy of Sciences as a publisher. Second, Borchsenius (as he is known in English language scientific publications) is well known as "Borchsenius" among those who study scale insects, and his works have been evaluated in numerous English language scholarly publications. If you put the Russian title into Google scholar, you get one hit, the document itself. If you put in the English language title, you get 52 hits. That is the reason for the proposed change. --Bejnar (talk) 00:50, 20 January 2014 (UTC)

• It seems to me that an editor can't, in any meaningful way, evaluate a source that they can't read. I'd be interested to see what others think about this.—S Marshall T/C 16:41, 20 January 2014 (UTC)

S Marshall: I think you're right (of course). A slight exception is that you can read what has been written in your own language about the source, and estimate its credibility on that basis. Sometimes a title alone can indicate unreliability: a blog post from the Buddhist–Nazi Togetherness League is probably not reliable. That's a long way from checking that the facts in the article faithfully summarize what's in the source, of course, of course. I think we basically have a trade-off here: between bias toward English sources and allowing facts cited from non-English sources to get checked less rigorously than English sources. I don't think any policy could be fully satisfactory here, but our current one deals well with the practical reality. The practical reality is that we have a huge bias toward English sources that are available on-line, because that's what most of our editors can find and can check, but we allow non-English, offline sources. No doubt the latter get abused sometime, but no doubt the former get abused a lot more. —Ben Kovitz (talk) 16:50, 20 January 2014 (UTC)

Certainly providing the English translation of the title and romanization of the author's name is more helpful to an English-speaking reader than leaving them out, unless it creates a lot of clutter. Let's not forget that most readers are not editors, and many readers who look at sources are looking for more information, not to fact-check. An English translation of a title can provide a valuable lead to a reader who is using the article as a start point for research. I've used foreign-language sources in this way myself, though I was not put off by the lack of an English translation. —Ben Kovitz (talk) 17:05, 20 January 2014 (UTC)

This has been discussed before, and while I like the idea of providing a translation when that's convenient, attempts to require it have been rejected because translator time is precious, because there are sometimes disputes about the One True™ Translation, and because sometimes it isn't appropriate (i.e., the non-English title is the name that English speakers recognize, especially for cultural works). My advice is to add translations when you can, and to not tell others that they're doing it wrong if they don't or can't. WhatamIdoing (talk) 19:45, 28 January 2014 (UTC)

OK... User:A Quest For Knowledge recently edited the third paragraph of the lede to conform it to the standardized language used in our other core policies (see this diff). This edit has been reverted by User:S Marshall (see: this diff). Personally, I don't have a strong opinion on either version... as both versions do say essentially the same thing. However, I lean towards AQFK's version, because I think using standardized language in all these policies helps to avoid the potential for confusion and misunderstanding. Please discuss. Blueboar (talk) 18:36, 28 January 2014 (UTC)

Like you, I don't have a strong opinion. Either is okay. Is the matchy-matchy version slightly better? Maybe. Better enough to matter? Maybe not. WhatamIdoing (talk) 19:54, 28 January 2014 (UTC)

• I suggest we standardise on the concise version rather than the verbose one.—S Marshall T/C 19:55, 28 January 2014 (UTC)

I don't have any specific preference of which version is better. I was simply attempting to standardize the wording. The other two policies followed a standard wording and WP:V was the odd man out. Whichever version we go with, let's make it standard across all three policies. A Quest For Knowledge (talk) 00:20, 30 January 2014 (UTC)

That would work for me as well... Perhaps a centralized discussion is in order? Blueboar (talk) 16:35, 30 January 2014 (UTC)

I think that a substantive discussion here would be the best start. Clarity on the proposed changes is important because right now everyone is just essentially pointing to a series of edits on a bundle of changes. North8000 (talk) 12:04, 20 February 2014 (UTC)

Should a list of links to Wikipedia pages have sources? This has come up at Sports in Alaska where at talk:Sports in Alaska we are discussing if this list needs sources, or if it should be something else other than a list article. -- 70.50.151.11 (talk) 01:32, 22 February 2014 (UTC)

I've offered my two cents. DonIago (talk) 15:19, 24 February 2014 (UTC)

Is it safe to tell everyone that all information in Wikipedia ought to have a citation, even if it does not?

I have been thinking about the practice of citing Wikipedia itself as a reference. If one goes to any article and looks at the menu on the right, there is a link called "Cite this page". That link (Special:Cite and for example "an article") is populated with text from MediaWiki:Cite text, which is hardly touched or discussed. Right now it says, "Most educators and professionals do not consider it appropriate to use tertiary sources such as encyclopedias as a sole source for any information—citing an encyclopedia as an important reference in footnotes or bibliographies may result in censure or a failing grade. Wikipedia articles should be used for background information, as a reference for correct terminology and search terms, and as a starting point for further research. As with any community-built reference, there is a possibility for error in Wikipedia's content—please check your facts against multiple sources and read our disclaimers for more information."

I was thinking of changing this to reflect that all information in Wikipedia should have a citation, and that instead of citing Wikipedia it is ideal to cite the sources which Wikipedia itself is citing. Right now, this recommendation is not part of this.

What are others' thoughts? Should visitors to Wikipedia be directed to cite what Wikipedia cites rather than Wikipedia itself? Blue Rasberry (talk) 02:00, 20 February 2014 (UTC)

It is not required to provide citations for all information. Only direct quotations and information that has been challenged, or is likely to be challenged, requires a citation to support it. Jc3s5h (talk) 02:16, 20 February 2014 (UTC)

Yes, reinforcing, that is the answer. North8000 (talk) 03:14, 20 February 2014 (UTC)

I view that as a bad practice useful only for tolerating content retained from our earlier days. At some point, we need to switch to a mandate that all additions be cited, and, some time after that, purge ourselves of all uncited information. It is, however, not really relevant to the original question. Certainly, people should be advised to consult and cite the source used in Wikipedia, not Wikipedia itself, and not to rely on any uncited information in a Wikipedia article. We are not a reliable source of information.—Kww(talk) 13:51, 20 February 2014 (UTC)

Oh God. So we all spend ten years providing citations for "Berlin is a city in Germany".—S Marshall T/C 15:24, 20 February 2014 (UTC)

Certainly obvious facts need not be cited, as per WP:BLUE. But moreover, if we choose to insist that all additions be cited, we will probably significantly reduce editor retention much further, which we do NOT need. Would we then require all or most new articles to go through wp:AFC or a some similar process? DES ^(talk) 17:20, 20 February 2014 (UTC)

Certainly all facts should be cited, including ones that you find obvious, DES, per WP:NOTBLUE, an essay well-grounded in policy as opposed to WP:BLUE's shaky basis.—Kww(talk) 18:38, 20 February 2014 (UTC)

I'd characterise this as the view of someone who spends most of their time editing in fringe or contentious areas and has a great deal of experience of untruthful editors. Verifiability is a very positive thing for the encyclopaedia, but taken to extremes like this, it will (a) become utterly impractical and (b) cross the line into copyright violation. If we have to put a citation after every single fact, then not only will our ever-decreasing number of productive editors find it ever-increasingly hard to expand articles, but our entire encyclopaedia will come to consist of close paraphrases of copyrighted sources.—S Marshall T/C 18:50, 20 February 2014 (UTC)

We disagree there, Kww. I point out that few if any academic works cite/footnote every fact mentioned, even those that seem obvious. Certainly no print encyclopedia does. Indeed I would be interested to hear of any academic work that cites a statement such as "Paris is the capital of France." Moreover, it is my view that citing at the level of density you seem to suggest would make Wikipedia articles quite awkward to use, much less to edit, and would hide the significant source citations among a forest of unneeded and pointless cites. I do not think such an approach has or is every likely to get consensus. I hope it doesn't. DES ^(talk) 18:51, 20 February 2014 (UTC)

Apples and oranges. Wikipedia cannot be compared to academic works because, first, it's written by non-experts and, second, academic works get vetted by professional editors whereas WP is vetted by no one other than other non-experts. V, supported by the citation of information, is the only bulwark Wikipedia has against living up to the charges of unreliability with which we're being constantly, but incorrectly, tagged. Live up to those charges and see what it does to editor retention. Regards, TransporterMan (TALK) 19:35, 20 February 2014 (UTC)

My problem with this is that you are following the idiotic "may be deleted" policy for unreferenced documents. For the most part, these documents are obvious to people in the field. The ones that are not, because they are incorrect or biased, (whether or not they are referenced!) are the ones that will be deleted. The problem with unreferenced documents isn't that they are unverifiable, it's that they don't supply additional reading for genuinely interesting readers. It would be nice to see this mentioned in the verification reference. Don't be scolding; be encouraging! — Preceding unsigned comment added by 67.100.124.159 (talk) 00:50, 23 February 2014 (UTC)

MediaWiki:Cite text also says: "For more detailed advice, see Citing Wikipedia."

I think this is sufficient. The latter includes: "However, much of the content on Wikipedia is itself referenced, so an alternative is to cite the reliable source rather than the article itself." PrimeHunter (talk) 03:31, 20 February 2014 (UTC)

What we require is that everything be verifiable, we don't require that everything actually be verified. What this means is that it should be possible to cite everything in Wikipedia... it does not mean everything in Wikipedia needs to be cited. Some things (the usual example is the statement: "Paris is the capital of France") have so many sources that could be cited (and are thus so obviously "verifiable") that it is silly to actually provide a citation for it.

That said... it is never "wrong" to give a citation... so... when in doubt, provide a citation. And, if someone requests a citation for something you don't think really needs one... it is almost always much quicker, and a lot less of a hassle, to simply provide one than it is to spend days arguing about why one isn't needed. Blueboar (talk) 13:45, 20 February 2014 (UTC)

Indeed. While it may not exactly be assuming good faith, the more an editor insists that a source isn't required, the more dubious I tend to become that one actually exists. I kind of wish we could include in policy something to the effect of, "Consider that providing a source is generally a more productive and simpler resolution to verifiability disputes than arguing that a source is not required." DonIago (talk) 14:48, 20 February 2014 (UTC)

I think we already, in effect, have the requirement as a practical matter. To focus on the "challenged or likely to be challenged rule" without considering what it means or why it's there is myopic. The purpose of V is to make sure that ordinary users can prove to themselves that what Wikipedia says is reliable. With a paper encyclopedia they can rely on the judgment of the professional editors (and, ultimately, the marketplace: unreliable encyclopedias fail), but we don't have that. What we have is V. So, in that context, "likely to be challenged" means anything the accuracy of which a prudent reader is likely to, or should be likely to, question. And that's pretty much everything except the most innocuous and universally-accepted information. Best regards, TransporterMan (TALK) 16:12, 20 February 2014 (UTC)

• PrimeHunter - You are more concise than me and using existing language is best. Just doing what you proposed and nothing else would satisfy me, but if people want to go further then that would be cool too.
• Kww I do not want to disrupt the existing community but yes, after some amount of time passes, I would like for all information in Wikipedia to be verified and for people to feel that the standard practice is to use citations. Right now that is not a standard practice, but just something that people do for other reasons.
• Jc3s5h North8000 I do not want to change current practices, but I think already citations are being used much more than is required and community customs are more strict than the rules actually say.
• Doniago I agree with you that the policy should say something like that and seems not to do so.
• Blueboar I agree with you but I do not want to discourage anyone who fails to use a citation. If you have ideas for doing this with a light touch then share. Regarding your Paris example - people at Wikidata have plans to host obvious factual references and propagate these out to Wikipedias because many of these things are not obvious cross-culturally. For example, how many English speakers know all the capital cities in Africa? There could be central verification of these things propagated out at once with citations to 200+ language Wikipedias.
• TransporterMan It certainly seems like a cultural requirement to me and your description is apt.

Blue Rasberry (talk) 16:17, 20 February 2014 (UTC)