Wikipedia talk:Verifiability

"Warning" signs on bio. pages and nomination for deletion

Have added a large number of {{Unreferenced|date=August 2008}} and {{BLPsources|date=August 2008}} to a large groupe of players in the swedish top league, aka Allsvenskan, i myself is a major fan of Halmstads BK, also a team in Allsvenskan, and have made major contributions to that teams and players articles, i am trying to improve all articles regarding to Swedish top fotball, Allsvenskan and Superettan mainly, and was hopping that some user and IP numbers would help if i added this tags, this however havent happend and i feel that i know to little about some players to writte theire bios on my own. I also added this signs to the articles since i dont feel like they reach the demands on Wikipedia regarding bios. So now i wonder how long i should wait before i put up a bio. page for deletion or if there is something else i should do instead? --> Halmstad, Talk to me 21:17, 17 August 2008 (UTC)

Context

I want to raise an issue that has been bothering me for some time. It is mentioned in RS but I think it needs to be spelled out, at least succinctly, here.

Many times an editor adds a statement with a complete citation. The statement and citation are accurate. But upon investigation I discover one of two things: (1) I can find the source - it is verifiable - and I discover that the quote, while accurate, is taken out of context and in one just reads a little more, discovers that the quote is then contradicted, or revealed to be a fringe POV. My point is that part ov "verification" is the availability of context. But this leads me to my bigger worry, (2) I don't have the book in my local library and when I try Google Books I find just the page that is quoted - quel surprise - and Google Books tells me that the next page is unavailable. This means I cannot check the context.

I think we have to have a statement that the reliability of a quote depends largely on the context, and verifiability means that one must be able to check not just that the quote itself is accurate, but must be able to check the context.

To be clear: I am NOT saying quotes from Google books are not verifiable and violate policy - my example is only meant to illustrate how I have come to realize the importance of this issue. I am sure I can get the full book through inter-library loan, and verify the context.

I think the main value of adding something to the policy, and it need not be long, just a few sentences as most, is to do what our policies largely do: help educate newbies about how to do quality research. We need to communicate that research is not just hunting for quotes taken out of context, but for evaluating quotes in their context ... that this is essential to "reliability" and the ability of others to do this is essential to verifiability."

If others agree, does someone want to propose an elegant way to word this?

Slrubenstein | Talk 14:59, 16 January 2009 (UTC)

How about:

Be very careful not to quote sources out of context. For example, suppose author A says that author B says the moon is made of green cheese, and author A goes on to dispute this view. In this case you cannot use author A to support the statement "The moon is made of green cheese." --Philcha (talk) 15:21, 16 January 2009 (UTC)

Well, that certainly makes the point, but perhaps a bit too broadly. How about:

Be very careful not to quote sources out of context. For example, a quote that may at first glance have a straightforward meaning may actually be using a particular word in an unusual or idiosyncratic way; you may need to read an entire article, chapter, or even book to understand how the author defines and uses an important term. Or, for example, a quote may seem to be making a strong assertion, when in fact the author is being ironic; you may need to read the entire article or chapter for this to be clear. Or, the author may go into some detail laying out an argument only to explain, later in the article or chapter, or even book, how and why the argument is wrong and rejected by most other scholars. In these cases, it is not enough to provide a quote and simple citation. You must explain the context, and an inclusive citation, or multiple citations, directing Wikipedia readers where to look to verify that the quote is being used properly.

Or something like this? Slrubenstein | Talk 15:45, 16 January 2009 (UTC)

"You will often need to read an entire article, chapter, or even book to understand the author's true opinion" would be a useful addition to what I previously suggested.

However the rest is too legalistic in phrasing. I think "an inclusive citation, or multiple citations" is not going to be clear to inexperienced editors. How about:

Be very careful not to quote sources out of context. For example, suppose author A says that author B says the moon is made of green cheese, and author A goes on to dispute this view. In this case you cannot use author A to support the statement "The moon is made of green cheese."

You will often need to read an entire article, chapter, or even book to understand the author's true opinion. In cases like the "green chesse" example, you should cite both the chapters / pages where author A describes the theory he / she opposes (if that theory is relevant), and those where he / she explicitly states his / her own opinion.

(Ain't political correctness a drag?) --Philcha (talk) 16:09, 16 January 2009 (UTC)

Well, I'd like something less condescending, and more useful, than "the moon is made of green chsses." We have some ideas out, how about we see what other editors think, and maybe some others can offer proposals or help improve what we have laid out? Slrubenstein | Talk 16:30, 16 January 2009 (UTC)

WP:UNDUE also applies to the content of individual sources. Gwen Gale (talk) 16:39, 16 January 2009 (UTC)

I prefer Slrubenstein formulation of 15:45 rather than blue cheese.Mccready (talk) 17:22, 16 January 2009 (UTC)

<ud> This is an important issue, but to me it's an original research matter. Maybe something like "Remember that citations must accurately reflect the context of the source, and take care to avoid the original research of introducing a meaning not intended in the original source." . . dave souza, talk 17:56, 16 January 2009 (UTC)

I think that has a loophole: selective quotation doesn't add info, it misleadingly withs the additional info that the context gives. --Philcha (talk) 18:12, 16 January 2009 (UTC)

Good point, it would be better phrased as "presenting a meaning not intended..." . . dave souza, talk 18:55, 16 January 2009 (UTC)

introducing the concept of OR at this point will be confusing. NPOV and OR and V are inextricably linked, but it is clearer to use only one or another at a time when discussing something in particular. The actual motivation for problems of this sort is most likely to be carelessness---and, as originally mentioned, in practice unavoidable given the availability of good print library resources to most editors, together with their general unwillingness to use what they do have access to. Constructing one's own summary of an argument or the selection of just what to quote is inherently research, and necessarily involves one's own understanding of the matter and thus is to a certain degree unavoidable OR. Even the misquotation because of lack of NPOV is not necessarily conscious--we all have the tendency to find first what will support our position--I know that when browsing my own eyes will land to anything on a page I am hoping to see there--this is probably psychologically inherent in the operation of browsing. DGG (talk) 18:16, 16 January 2009 (UTC)

Well to put a slightly different complexion on it, I think it's a general problem with quotes. It's always best to read a source and then include the info. A simply statement to the effect that we should avoid over using quotations and that they should always "be put in context and given any necessary explanation. As an editor, it is your responsibility to read the source of the quotation thoroughly, in order to prevent misrepresentation." That's from WP:QUOTE. I agree context is important, maybe we should link to WP:QUOTE from here, and have a brief summary of what it says there? Alun (talk) 18:28, 16 January 2009 (UTC)

I agree it is a problem with quotes. With respect to DGG, most Wikipedia editors seem not to have access to good libraries. From what I have seen at the articles on my watchlist, many people rely on google books or other on-line resources that often do not provide the entire text. This leads to out-of-context quoting.

I think it becomes a verifiability problem when the source provided is a link to an online text. Of course that link enables us to verify that the quote exists outside of Wikipedia. But such links often take us to incomplete sources where it is impossible to verify what the quote means. For example,k if a link takes me to a page at google books, and the preceding page is not available through google books and the subsequent page is unavailable through google books, it can be very difficult to verify that the quote has been used appropriately. I do not think this is a matter of original research, it is a matter of appropriate source-based research which must be the basis of all wikipedia articles. Right now, this matter is broached in WP:RS which is a content guideline; I think it needs to be addressed in a core policy. I agree that this problem has NPOV and NOR dimensions - our three core policies often intersect or have moments of overlap. I certainly would not object to someone raising this issue at NOR or NPOV!!! But I do think there is a verifiability issue. Put most simply: quotes are often being to made claims that are verifiable, but require more information, either other quotes or multiple page numbers, in order to verify the claim. Slrubenstein | Talk 19:00, 16 January 2009 (UTC)

DGG makes 2 good points:

• NPOV and OR and V are inextricably linked. However rather than try to narrow any particular discussion artificially to one of these, I'd be quite happy for each of these policies to say in the simplest, plainest terms that selective use of sources is unacceptable.
• I also agree that perfect observance of NPOV and NOR and V is impossible, becuase an encyclopedia inevitably summarises and selects. So what really matters is that we don't distort the source's meaning. So terms like "distort" and / or "misrepresent" must appear in the policy wording. --Philcha (talk) 19:01, 16 January 2009 (UTC)

I think Slrubenstein is looking at the Google Books issue the wrong way. If the citation said nothing about Google Books and your local library can't get you a "dead tree" copy within a couple of months, you can't challenge the citation. In fact a lot of reviews WP:AGF on books. So Google Books gives sceptics such as the Slrubenstein presents a chance to see whether there is a real risk of midleadingly selective use. It's not Google Books that's the problem (in fact its' a help), it's selective use of sources. --Philcha (talk) 19:07, 16 January 2009 (UTC)

What's happened to me is that I'll do a google search, and google books will pop up. The sections of google books that are available will only give a taste of the subject. But if I want to quote from google books it's easy to provide a link to the very page from the book I quote from. It's easily verified. Now if the quote is out of context even I don't know. BTW I never quote from google books, I don't see the point. If it's so easy for someone to check the quote, then they don't need it to be reproduced in Wikipdia, they can just follow the link to google books and read it there. I agree that we can use words like distort and misrepresent when it comes to quotes, but I also think we need to say "unintentionally" distort or misrepresent. If the policy acknowledges that many online sources are inclmplete, and that due to their incompleteness it may not be possible to get the full meaning of the author, then at least we're being fair and WP:AGF. Alun (talk) 19:17, 16 January 2009 (UTC)

(ec) My point was only an example. But Google Books is often a problem since it often provides only portions of a source, in those cases it invites misleading selective use. And providing the citation to Google Books does not help because we get the citation for the page the quotation came from, but not the citation to those pages that make clear what the quoted passage means (because those pages are often not available through Google Books). But this is only an example. I have never suggested specifying "Google Books" in any edit to this policy. My point is that in adding a quote and a citation of the source, we need to be able to verify not only that the quote appears in the source, but that the quote is being used consistent with that of the source. To verify this, we may need to know not only the page the quote comes from, but other pages that provide the necessary context for the quote. Slrubenstein | Talk 19:20, 16 January 2009 (UTC)

If Google books - or any text available to an editor does not provide enough text so that the editor can understand the authors intention in the quote ... that is, if a good-faith editor cannot be sure s/he is not taking the quote out of context, s/he just shouldn't do it ... to do so would be to make what is in effect a hard to verify claim. Slrubenstein | Talk 19:22, 16 January 2009 (UTC)

No harder than citing a book that's not avaiable online. --Philcha (talk) 19:40, 16 January 2009 (UTC)

You are wrong. If you have the entire book in front of you, rather than a fragment (it doesn't matter whether the whole book is on-line or paper, and it doesn't matter whether the fragment is on-line or paper) it is easier to see the context for the quote - and provide a complete citation. That is my main point. That Google Books usually provides people with fragments, and lending-libraries usually provide people with entire books, is a secondary point. Slrubenstein | Talk 20:26, 16 January 2009 (UTC)

Context of a source is more than just relevant, it's central to whether the source actually supports what it's being provided for. A change along the lines Slrubenstein suggests make a lot of sense. FeloniousMonk (talk) 03:14, 17 January 2009 (UTC)

I'd support that change too, and Slrubenstein's wording, except perhaps a little tighter, and also probably minus the point about idiosyncratic word use, because that's more to do with basic comprehension than context. SlimVirgin ^{talk|contribs} 03:36, 17 January 2009 (UTC)

I'd suggest our eventual wording be positive--we should, for example, talk about the need to make sure that quotes are in context, not that we forbid their being out of context. this is meant to be a guide o good editing, not a weapon. DGG (talk) 04:59, 17 January 2009 (UTC)

Yes, let's say what WP:QUOTE says about context, and link there. Alun (talk) 13:09, 17 January 2009 (UTC)

I would be very happy if Slim Virgin, DGG, and Alun crafted wording they felt appropriate and useful; I appreciate their suggstions and would like to see a concrete proposal .... Slrubenstein | Talk 17:13, 17 January 2009 (UTC)

I also support Slrubenstein's proposed change, along with SlimVirgin, DGG and Alun's suggested modifications; context is everything. Dreadstar † 17:22, 19 January 2009 (UTC)

So here's a suggestion. Currently the Burden of evidence section of the policy states:
"The source cited must clearly support the information as it is presented in the article." Then there's a footnote <ref>"When there is dispute about whether the article text is fully supported by the given source, direct quotes from the source and any other details requested should be provided as a courtesy to substantiate the reference."</ref>
I suggest that we move the footnote to the main body of the text, and add a further two sentences from WP:QUOTE:
"Quotations should be put in context and given any necessary explanation. As an editor, it is your responsibility to read the source of the quotation thoroughly, in order to prevent misrepresentation." We also link to WP:QUOTE at the head of the section. So we end up with:

For how to write citations, see Wikipedia:Citing sources

See also: Wikipedia:Quotations

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged must be attributed to a reliable, published source using an inline citation. The source cited must clearly support the information as it is presented in the article. When there is dispute about whether the article text is fully supported by the given source, direct quotes from the source and any other details requested should be provided as a courtesy to substantiate the reference. Quotations should be put in context and given any necessary explanation. As an editor, it is your responsibility to read the source of the quotation thoroughly, in order to prevent misrepresentation. The source should be cited clearly and precisely to enable readers to find the text that supports the article content in question. Editors should cite sources fully, providing as much publication information as possible, including page numbers when citing books.

How does that sound? On a different note I also noticed that WP:QUOTE is an essay, I wonder if it should be a guideline? It makes sense to me that we have a proper guideline for quotations. Alun (talk) 06:25, 20 January 2009 (UTC)

I do not think that "Burden of proof" should be change to place the footnotes into the text. The section needs to by as succinct as possible. This problem should be addressed elsewhere. --PBS (talk) 11:10, 28 January 2009 (UTC)

OK, how about adding the sentence "Quotations should be put in context and given any necessary explanation" to either the body of article in the section "Burden of proof" or to the footnote about quotations? It's quite a short sentence, and it addresses the problem of quoting out of context, which is what we want to address. Alun (talk) 12:52, 28 January 2009 (UTC)

biased sources (more precisely: conflict-carrying sources): proposed discounting as a "reliable source"

I recommend adding the following bolded word/link to the following sentence:

"Articles should rely on reliable, unbiased, third-party published sources with a reputation for fact-checking and accuracy." Comments? Bo99 (talk) 00:29, 19 January 2009 (UTC)

(update, 2009-Jan-20: Such word "unbiased" incorporates the given link, and therefore is clear shorthand for the more cumbersome phrasing:

"Articles should rely on reliable, third-party published sources that have a reputation for fact-checking and accuracy, and that do not have a conflict of interest on the proposition for which the source is being cited." - Bo99, 2009-Jan-20)

Then you violate NPOV, which assumes that no one view is "unbiased" and requires us to add all significant views. Even if you think they are biased. And biased views often come from biased sources. What is important is that the source is verifiable. If it has a bias, add other views from other sources. Slrubenstein | Talk 01:08, 19 January 2009 (UTC)

I think you are saying that a biased source is a Wikipedia-permitted reliable source. Is that right?

aside: My posting above probably did not talk about neutral-point-of-view as you suggest, merely about sources. Assume a Wikipedia article properly and neutrally presents majority view A, and minority view B, and excludes fringe view F. Under view A, the article cites source A1, A2, etc. Under view B, the article cites source B1, B2, etc. The issue i probably focused on in my posting above was: if there is a source B3 who has a personal motivation to support view B, then is B3 a Wiki-prohibited unreliable source? Then, if B3 is disqualified, that would not diminish the fact that the article neutrally presents view A (and source A1 and A2) and view B (and source B1 and B2), so there would seem to be no violation of neutral-point-of-view, as you suggest, in excluding the biased source B3. Bo99 (talk) 03:14, 19 January 2009 (UTC)

I understand your point, it is worth raising. I think it is a more complex issue. If we just want facts on say the percentage of blacks in the US, we want a third party source that is unbiased. But if we want a source on American Nazi views, an American Nazi web-site, clearly biased, is a reliable source on what American Nazi's claim. See? A third example: a psychology journal has an obvious bias against mathematics or astronomy, thy will almost certainly reject submissions on those topics. But they are important sources on psychology. We ust need to make sure the policy is worded in a way that accommodates these very diferent cases. And, finally - remember than NPOV is the one non-negotiable policy. No other policy can conflict ith NPOV. Slrubenstein | Talk 03:31, 19 January 2009 (UTC)

Thank you for your reply.

I agree that wording is important. We are not making up anything new. It seems to be a generally accepted principle that a person who has e.g. a significant stake on a particular view has a conflicting interest on that topic. In your example, the Nazi has no bias or unreliability on what Nazis claim. In your next example, a psych journal has no bias against reasonable use of math in a psych article.

On your NPOV point, i am not talking about precluding a view A or a view B, but about a generally accepted principle of recognizing when a source A3 or B3 has a conflicting interest. Bo99 (talk) 04:04, 19 January 2009 (UTC)

I do not understand what you mean about a conflicting interest on a topic. There is no reason a Nazi can't edit the article on nazis, as long as he complies with NPOV, V, and NOR - indeed i am sure it has happened in the past. Are you saying Jews cannot edit the article Judaism because of a conflict of interest? I think the key thing is that editors comply with our policies. Slrubenstein | Talk 04:11, 19 January 2009 (UTC)

I mean that it seems to be a GAP (generally accepted principle) that COI applies to every topic. A nazi can edit a wiki article on nazis, properly, if s/he complies with all relevant Wiki principles, e.g., as you say, NPOV, V, and NOR, but also e.g. Conflict Of Interest Held By A Wikipedia Editor, e.g. "Do not edit Wikipedia to promote your own interests ... unless you are certain that the interests of Wikipedia remain paramount." So far so good? Bo99 (talk) 04:32, 19 January 2009 (UTC)

Yes, and this is already covered by NPOV, NOR, and COI. Why bring it up when it is well covered? Slrubenstein | Talk 04:50, 19 January 2009 (UTC)

I brought up Conflict Of Interest Held By A Wikipedia Editor because it is a Wiki guideline. So far so good? Bo99 (talk) 05:04, 19 January 2009 (UTC)

It is not related to this policy. Slrubenstein | Talk 05:43, 19 January 2009 (UTC)

Regarding this Verifiability policy, i'm not yet returning to talk about that, as you are in your single sentence just above. But probably we agree that an editor should follow the following: not just NPOV, V, and NOR, but also for example Conflict Of Interest Held By A Wikipedia Editor, which roughly says that if you are editing to further your own aims, you should not (subject to the usual caveats, exceptions, etc). Bo99 (talk) 14:09, 19 January 2009 (UTC)

No. The Conflict of Interest policy is very specific. You seem to b defining it too broadly - so broadly that you would undermine or do away with NPOV. NPOV assumes that all editors have conflicting interests, and that their views of what is best for an article are in conflict. Accepting this, rather than forbidding it, is the starting point of Wikipedia. Anything that interferes with this would threaten the whole project. We do have a COI policy, it serves us well, and it is clear and specific. Slrubenstein | Talk 17:06, 19 January 2009 (UTC)

It's important for me to understand exactly what you are saying and that we therefore take things step by step. Let's focus for the moment only on Conflict Of Interest Held By A Wikipedia Editor. I think you agree that:

1a. An editor should follow such COI(HBAWE) principle (in addition to e.g. NPOV, V, and NOR).

1b. Such principle specifies "Do not edit Wikipedia to promote your own interests ... unless you are certain that the interests of Wikipedia remain paramount."

1c. That principle seeks to dissuade some editing attempts by editors around the world. Yet the whole project of Wikipedia is not threatened by such principle.

So far so good? Bo99 (talk) 17:31, 19 January 2009 (UTC)

COI lists exemplary COIs: Financial, Legal antagonists, Autobiography, Self-promotion, Promotional article production on behalf of clients, Campaigning, Close relationships. I think it is important to bear this in mind when discussing "interests." My liking and intending to vote for John McCain creates no COI in my editing the articles on McCain or Obama. My being a paid or volunteer employee of the McCain campaign may. COI policy encourages me to disclose this; if I do and am careful to comply with our core policies, I am not violating COI. Slrubenstein | Talk 18:07, 19 January 2009 (UTC)

You mention some things that i think will be useful, but not just yet. First, the following is not a huge deal, but it really would be nice if you might Answer The Direct Yes/No Question i posed in my prior posting. (And when you do, fear not: even if i ever do wind up making a good case for reliable sources excluding e.g. highly-financially-interested sources, i'm just going to leave it as Discussion, and will not insist on anyone editing the guideline page.) Bo99 (talk) 18:40, 19 January 2009 (UTC)

I have already made it clear I support our COI policy. But the policy addresses editors, not sources. I do not see the relevance. Policies guiding editors' actions should not necessarily apply to sources. Slrubenstein | Talk 18:55, 19 January 2009 (UTC)

So it's fairly clear that you agree with 1a and 1b above, but it's not clear that you agree with 1c above, and even if you think it's clear, it would be efficient and nice for you to Answer The Direct Yes/No Question, right? Bo99 (talk) 19:17, 19 January 2009 (UTC)

I I do not understand what 1c means. Look, this is not about niceness, or efficiency. This talk page is for discussing improvements to the Verifiability policy. So far you seem to be avoiding that. Unless you have a proposal to improve the verifiability policy, this discussion is not appropriate. You did make a proposal above, which I disagreed with. So far no one else seems to support your proposal. Slrubenstein | Talk 19:22, 19 January 2009 (UTC)

Hi Bo, it would be better to ask questions about COI on WT:COI. This page is only for discussing the sourcing policy. SlimVirgin ^{talk|contribs} 20:29, 19 January 2009 (UTC)

There are serious problems with Slrubenstein's response above. I will detail them in due course. I continue not to see any valid argument supporting the idea that conflict of interest principles should apply to editors, and to essentially everyone in the world on whom others rely, but should not apply to sources cited in Wikipedia. I will detail counterarguments why generally-accepted COI principles should apply uniformly to all, and why such principles thus should deem-unreliable some sources, e.g. those with a financial stake. (SlimVirgin, thanks for commenting.) Bo99 (talk) 22:23, 19 January 2009 (UTC)

The reason is the same reason for NOR, why Wikipedians do not put their own arguments into articles, but do put in arguments that come from verifiable sources. Slrubenstein | Talk 23:11, 19 January 2009 (UTC)

Bo99, I wouldn't bother with your planned statements, because it's based on a misunderstanding of the existing rules, and it's not going to result in a policy change.

For one thing, it's not just the source. It's also how you use the source. So imagine that you have the owner of a publicly traded company trying to pump up his stock. He's got an obvious financial conflict, and he's obviously a primary source, his statements are self-published in this instance: At a stockholders meeting, he says something like, "In the coming quarter, sales will exceed anything the industry has ever seen before, beyond your wildest imagination." And the stock soars, because fools and their money are soon parted.

You already can't use his statement to support,

"Sales in the second quarter of 2009 will break all previous records.<ref>It's True™ because the CEO said so!</ref>"

You can however use his statement to support,

"CEO John P Umpindump projected increased sales during the second quarter of 2009.<ref>in front of four video cameras, and the Securities Exchange Commission is going to be having a friendly little chat with him.</ref>"

The fact that the CEO is inherently "biased" should not prevent you from reporting the actual, verifiable facts. Existing policy, however, limits how you report those facts. WhatamIdoing (talk) 03:45, 20 January 2009 (UTC)

"WhatamIdoing", there are various problems with your posting, but at least for now i'll limit my comments to the following:

(Bo99 (talk) 15:23, 20 January 2009 (UTC))

2a. (This numbering starts at "2" to distinguish it from my prior posting's use of "1a" etc.)

I think you are unintentionally supporting what i believe is the view i have written (and the worldwide view):

The first thing i wrote above is that

"Articles should rely on reliable, unbiased, third-party published sources with a reputation for fact-checking and accuracy." The word "unbiased" incorporated the given link, and therefore was clear shorthand for the more cumbersome phrasing

"Articles should rely on reliable, third-party published sources that have a reputation for fact-checking and accuracy, and that do not have a conflict of interest on the proposition for which the source is being cited."

I think you agree with such view.

2b. I think you unintentionally oppose SIrubinstein and the current policy. (SIrubinstein believes that a source is a Wikipedia Reliable Source even if the source has a conflict of interest on the proposition for which the source is being cited. And the policy never simply says that a source, to be "reliable", must not have a conflict of interest on the proposition for which the source is being cited.) Bo99 (talk) 15:23, 20 January 2009 (UTC)

Bo99 are you being a disruptive editor? You seem to be deliberately confusing the issue. The word "source" has two (and more) meanings and you are mixing up the two. In the article on conflict of interest, you just added the word "source!" (and now you refer to it as if it supports your argument? How sleazy). And I forced you to clarify that you meant a human being, a person serving as a source of information for a reporter. That i indeed one meaning for the word "source." But here, a source is a book or article. The point is that books and articles are verifiable and reliable in ways that confidential informants are not. If you tell me your source for an edit was a friend of yours who works in the Pentagon, that is not a "source" in the way that this policy means, and none of us can verify it. But if you mean that your source is one of Jane's guides to armor, well, that is a verifiable source all of us can check and we know its reliability. You can't put lipstick on a pig and you have just spent a lot of talk here trying to do just that. Stop screwing around with this talk page, it is meant for improving the policy, not subverting it. Slrubenstein | Talk 16:16, 20 January 2009 (UTC)

SIrubenstein, you're wrong in all of your accusing and name-calling. I note that you continue to violate Wikipedia principles; if you care about that, please request a detailed list and i will provide specifics of your conduct and the Wikipedia texts you have violated.

You might be focusing on the detail that the author speaking in a document might have a COI but technically the document itself (which is the alter ego of the author) might not have a COI, very arguably. Either view is tolerable for me. To handle that alternative, the policy could simply be supplemented a bit further (see the following italics), to read as follows, "Articles should rely on reliable, third-party published sources that have a reputation for fact-checking and accuracy, and whose authors do not have a conflict of interest on the proposition for which the source is being cited." Bo99 (talk) 17:01, 20 January 2009 (UTC)

I already explained how that would violate NPOV. There is nothing wrong ith using sources that exspress strong views, if anything we need more of them. Slrubenstein | Talk 17:35, 20 January 2009 (UTC)

SIrubenstein: I believe your point about NPOV was already rebutted: even under the proposed improved Reliable Source definition, all relevant points of view would be presented in Wikipedia. It would just be that the cited support would be from actually-reliable sources i.e. sources that do not have COIs (on the specific view for which the source is cited) (or significant COIs if anyone desires that further word). Bo99 (talk) 17:53, 20 January 2009 (UTC)

Bo, read the WP:COI policy again... It focusses on editors, not sources. An editor might have a conflict of interest when he or she edits a particular Wikipedia article... sources do not have a conflict of interest. A particular source may have a bias, but that is handled by following the NPOV policy. Blueboar (talk) 15:07, 21 January 2009 (UTC)

Blueboar:

No, i don't need to read again or glance again at the guideline you cite, Wikipedia Conflict Of Interest guideline. Yes, such guideline focuses on editors, not sources. The guideline is a subset of conflict of interest situations. Another subset is source documents whose authors have COIs.

You seem to be claiming that the author of a cited source document can't have a COI (if that's not what you mean, please clarify). The following seems to be a counterexample. (If anyone wants to make the example more realistic and/or detailed, please do.)

The wiki writer is any disinterested individual.

The subject is a company that has issued securities.

The source person is a supposedly independent securities analyst.

The source document is a document produced by the analyst that says that the company has excellent prospects.

The writer writes in wiki that the company has excellent prospects, and in support of that cites the analyst's source document.

It turns out that the company is paying the analyst.

So such source person has a serious financial COI that makes such source person unreliable.

And so the source document is unreliable too, even though the Wikipedia Verifiability Policy seems to label such document a Reliable Source, strangely. Bo99 (talk) 17:43, 21 January 2009 (UTC)

The wiki writer's "error" was that he should have written "Analyst X reported that the company has excellent prospects. (cite for the analysts report)". Then a later editor can add "On 21 January 2009 the Securities and Exchange Commission arrested Analyst X on a charges related to the collapse of the company." without having to completely rewrite the entire section based on false information. Roger (talk) 18:59, 21 January 2009 (UTC)

I (Bo99) agree with these example variations or elaborations, regarding such things as what the writer or the USA SEC should or could do.

Regarding the source(s), and the example's conclusions: the conclusions seem to remain true:

The source person has a serious financial COI that makes such source person unreliable.

And so the source document is unreliable too, even though the Wikipedia Verifiability Policy seems to label such document a Reliable Source, strangely. Bo99 (talk) 19:28, 21 January 2009 (UTC)

To clarify my comments... Yes, an author can have a conflict of interest in writing a source. Wikipedia does not really care about this. It certainly has nothing to do with WP:COI, which is purely concerned with the situation that arrises when the author of the source (or someone directly tied to that author) edits a Wikipedia article.

In your example above... I agree that we need to take into account the fact that the Subject company is paying Author for his analysis, but the fact of payment does not make the analysis unreliable... For one thing, there we should not assume that payment influenced the analysis... the assignment could have been to give an independant expert critique. Even if the assignment was to give a glowing report, that simply shifts the analysis into the "Self-Published" category. It becomes no more reliable, but also no less reliable than an analysis issued directly by the company.

To put this another way, your concerns may affect what we say about the source, but they do not affect whether we can use the source.Blueboar (talk) 19:19, 21 January 2009 (UTC)

Blueboar: Yes, the fact that a source document's author can have a conflict of interest in writing a source document is not covered by WP:COI, which would be better titled as Conflict Of Interest Held By A Wikipedia Editor,

Your other statement, that Wiki does not care about the source document's author's COI, is problematic. Such COI is covered by the Wiki article Conflict Of Interest Held By Anyone (and, for SIrubenstein, such COI was covered by such article before you and i jointly edited such article).

It's good that you Blueboar agree (and probably everyone above agrees, other than maybe SIrubenstein) that Wiki needs to take into account that in my above example the subject company is paying the author for his favorable writeup. No need to reply on the following, but keep in mind that yours is probably a distinct minority view worldwide when you say that the fact of payment does not make the author's favorable analysis unreliable necessarily. Yes it's true that conceivably the favorable analysis might be borne out in time, but the author would very likely be subject to financial penalties and prison, according to the earlier poster Roger and Roger's description of USA law, and i agree that penalties and prison would be the worldwide general-accepted approach.

Yes, your wording is moving in the right direction about generally-accepted COI analysis, as you write/think-through "[COI] concerns may affect what we say about the source, but they do not affect whether we can use the source." But Conflict Of Interest Held By Anyone can shortcut your thinking directly to the worldwide generally-accepted answer: in shorthand, "Disallow, Or Disclose"; in longhand, the best thing to do is to disallow use of the editor, professional provider, source document, etc on which others might rely, or as a second choice, at least disclose, in a prominent way, that such person, on whom others might rely, has (or if it's a document, then its author would have) e.g. a significant financial interest (and spell out the publicly known details if any) on the specific view/proposition/etc that such person or document is to be used for, and therefore is unreliable on such view/proposition/etc.

By the way, the proposed improved Reliable Source definition will rarely affect a Wikipedia article, probably. It will be fairly rare probably that it is publicly known at a timely date that there is such a triggering COI, for purposes of Wikipedia. For example, in my example further above, at the time that the wiki writer is writing, he would not know that the securities analyst and the analyst's document are tainted by a financial COI, so, for a time, that actually-unreliable source would remain a Wikipedia Reliable Source (even under the proposed improved Reliable Source definition). And no one has come forward with a real-life example in a Wikipedia article in which the proposed improved Reliable Source definition would prevent the article from voicing any view. Bo99 (talk) 03:25, 22 January 2009 (UTC)

Tell me, do you have any idea how many fringe theory promoters claim that mainstreeam sources are biased against them?

All of them. This proposal would cause a severe increase in tendentious editing. Shoemaker's Holiday (talk) 15:39, 21 January 2009 (UTC)

There are problems with your posting. I will reply when appropriate. Bo99 (talk) 19:38, 21 January 2009 (UTC)

to the above poster "Shoemaker's Holiday": There are problems with your posting, but for now i'll limit my comments to the following. There would seem to be a simple and short Wikipedia response to your contemplated fringe theory promoter, as follows: Wikipedia requires reliable sources; if the promoter has no sources, either because mainstream sources are biased or for any other reason, then the promoter cannot write their proposition in Wikipedia. ("Shoemaker's Holiday", if you envision the dialogue with the promoter as being more complicated, please spell it out a bit more, for further analysis.) Bo99 (talk) 18:00, 27 January 2009 (UTC)

Bo99, I see that you have edited your comments today to assert that requiring unbiased sources will not prevent any articles from voicing any views. I submit to your attention Feminine essence theory of transsexuality, where even the best kinds of sources (peer-reviewed papers in scientific journals) have been denounced as "biased". As far as I can tell, exactly zero 'unbiased' sources exist on this subject. WhatamIdoing (talk) 02:30, 27 January 2009 (UTC)

to poster "WhatamIdoing": There are problems with your posting. I will reply when appropriate. Bo99 (talk) 18:27, 27 January 2009 (UTC)

to poster "WhatamIdoing":

There are problems with your above posting, but progress since your posting further above (2009-Jan-20).

Your first sentence overstates/misrepresents significantly what i wrote (a few inches above, at start of a paragraph): "By the way, the proposed improved Reliable Source definition will rarely affect a Wikipedia article, probably. It will be fairly rare probably that it is publicly known at a timely date that there is such a triggering COI ... . ... And no one has come forward with a real-life example in a Wikipedia article in which the proposed improved Reliable Source definition would prevent the article from voicing any view." (Such last sentence's adding, in this Diff page, seems to be what you are referring to in your first sentence above. Such adding was not a change in my view on the proposed improved Reliable Source definition.)

Your second sentence in your posting above seems to indicate that you are not yet talking about the subject of this entire Conflict-Carrying Sources discussion. The generic, broad word "bias" was never the subject. I used such word at the top of this discussion, not as the generic broad word, but rather as a link, as shorthand, with express emphasis to incorporate the COI discussion in Conflict Of Interest Held By Anyone. So one needs to look at that link, and/or Conflict Of Interest Held By A Wikipedia Editor. Of the many COI examples given in such two links, for most of the COIs there seems to be the rough commonality of a reasonably-specific, reasonably-direct, money-related involvement of the subject person (or a non-money-related involvement that is relatively equally serious) (in the specific proposition/view/project for which the person is being used). (I don't recall either of those articles expressing fear that their COI analysis is likely to, for example, prevent there being sufficient conflict-free editors to write Wikipedia, prevent any industry from functioning, prevent there being sufficient citable sources for an encyclopedia to exist; such prevention would indeed probably occur if the generic, broad word "bias" were the test.)

Regarding your third/last sentence in your posting above, please do say so if in the transsexuality article/subject-area you cited above (which i have not yet looked at) you still feel that 100% (or even 10%) of the sources in that area have the type of bias/COI we are talking about (i.e., again, a publicly-documented, reasonably-specific, reasonably-direct, money-related involvement, or a non-money-related involvement that is relatively equally serious, in the specific proposition or view for which the source would be used). Bo99 (talk) 19:43, 28 January 2009 (UTC)

Bo, please don't edit your comments or majorly reorganize threads after the fact. It is frowned upon. If you notice a mistake in something you wrote, or wish to indicate a change in your thoughts, may I suggest that you use the strike through function (<s>text</s> which looks like this: "use the strike through function.") Thanks. Blueboar (talk) 17:15, 27 January 2009 (UTC)

Blueboar, there are a number of problems with your posting immediately above. I will reply further when appropriate. For one thing, i don't recall having made (a) any major change to a prior posting by me, or (b) regarding any posting by anyone else, any change at all to the content, or even the meaning (which i recognized could be affected by some change to a prior posting by me), or (c) any change that made this Discussion-page topic less of a good record of the chronology of thoughts expressed. (And: (d): i already used a highlight-and-date-the-change technique (of which your suggested strike-through is a subset), near the top of this section.) Bo99 (talk) 18:27, 27 January 2009 (UTC)

I am referring to edits like this one. I have no problem with you changing your mind and deciding that you should have worded something differently... the problem is that others responded to your original wording, so makeing later changes impacts on the comments of others. By striking through what you want to change, you allow others to see your original wording, as well as your new wording. It isn't a big thing, but is a courtesy to others. Blueboar (talk) 19:00, 27 January 2009 (UTC)

Blueboar, maybe i previously expressed and covered your above point, in my above clauses a through c, especially b and c; the Diff page you cite shows a change to the topic title that was not intended to change the meaning significantly (especially within the context of title shorthand, in a title that was and is already too long for retaining struck-through words, a technique i was familiar with) (please ask if you would like me to write out the closeness in meaning of the former title words to the newer title words). I too have no problem with you or anyone changing the mind and deciding that one should have worded something differently, so long as one follows courtesy/clarity techniques/considerations like e.g. my above clauses a through d (including your suggested strike-through technique). Bo99 (talk) 19:32, 27 January 2009 (UTC)

To the Bo99 person: There is zero validity in the above 99 verbiage. To everybody else: If you liked this thread, you will also like Talk:Debate over the atomic bombings of Hiroshima and Nagasaki#allegation by the BillyTFried person of view-pushing. --Hans Adler (talk) 17:21, 27 January 2009 (UTC)

Hans Adler, you're deeply wrong, in terms of Wikipedia principles (and in terms of the substance of that other page); if you care about your violations of Wikipedia, please request a list and i can provide specifics, with links. Bo99 (talk) 18:27, 27 January 2009 (UTC)

I do in fact care very much about my violations of Wikipedia. I hereby request a list with specifics and links so that I can mend my ways. Thank you in advance.

(I take it that your kind offer refers to an expertise about my entire recent editing history rather than just the calculated WP:CIVILITY and WP:AHI violations that I committed with the above single post, which I am already well aware wasn't the best way of saying WP:DFTT.) --Hans Adler (talk) 00:15, 28 January 2009 (UTC)

Bo99, Feminine essence theory of transsexuality, which you have apparently not looked at, has recently been accepted into formal mediation over allegations that the people writing all the reliable sources on the subject have a conflict of interest (namely, that they're sexologists who support a different idea of sexuality, and therefore couldn't possibly know what hundreds of transwomen clients mean when they say that they're women trapped in men's bodies, have a female soul in a male body, and so forth).

So, yes, either your "unbiased" language (which you will find in the second sentence of this thread) or your "conflict of interest" concerns would prevent this entire article from being written. (To the delight of several POV-ish and COI-affected Wikipedia editors, by the way.) WhatamIdoing (talk) 01:24, 29 January 2009 (UTC)

to poster "WhatamIdoing": There are a number of strange problems in your above posting (but such posting still is progress over your first posting further above). For example, for the third time, you pretend that a relevant subject is the broad generic word "bias", written without a link built into the word, and now for the third time i explain that the subject is e.g. "bias", with a link built into the word, incorporating all the general COI principles of the link (e.g. COI Held By Anyone and/or COI Held By An Editor), e.g. speaking grossly: Disallow, or Disclose. So i gather that precise full analysis by me of your postings does not progress things. So i'll just paint with your type of broad brush. Regarding the transsexuality article you cite above, i'll make the huge assumptions that the universe of potential editors and sources is 100% composed of:

(a) sexologists who have a true COI on the subject (not just lack of knowledge as you say above, which lack is generally not a true COI) (which COI means that such persons are unreliable on the statements they write on the subject),

and (b) transwomen who have a true COI on the subject (which COI means that such persons are unreliable on the statements they write on the subject),

and (c) no other person in the universe.

Still, under the above-grossly-summarized COI approach of Disallow Or Disclose, the sexologists and transwomen could e.g. disclose their COIs and thus be editors and sources. So the article you cite could be written. Bo99 (talk) 15:46, 29 January 2009 (UTC)

I think this is an issue that needs to be addressed on an article by article basis, depending on whether individual editors or groups of editors are letting their "conflict of interest" affect the neutrality of their editing. It certainly does not belong in WP:Verifiability policy. (and Bo... please stop saying that there is "a problem with (someone's) posting" without further explanation... If you see a problem, explain what that problem is, rather than just stating that a problem exists... without explanation, other editors will simply assume you are saying WP:IDON'TLIKEIT and will ignore your concerns.) Blueboar (talk) 18:04, 29 January 2009 (UTC)

Editing policy

At the moment the Wikipedia:Editing policy (see WP:PRESERVE) contradicts WP:PROVEIT. See Wikipedia talk:Editing policy#Demote to a guideline --PBS (talk) 10:36, 28 January 2009 (UTC)

Reasonable efforts

AFAICT this edit on the 14 December 2008 was not discussed before it was inserted. "and it has always been good practice, and expected behaviour of Wikipedia editors (in line with our editing policy), to make reasonable efforts to find sources onesself that support such material, and cite them."

Although the intentions were fine, I think it causes more problems than it solves, because it sends out a confusing message:

• A: deletes a sentence that (s)he considers dubious and it has no citation.
• B:"I am reinserting the fact because you did not make an reasonable effort to find a source"
• A:"I'm deleting it again because there is no citation."
• ...etc.

It is better that the policy is clear "The burden of evidence lies with the editor who adds or restores material."(WP:PROVEIT) and if B wants to restore it (s)he does so with a citation.

I think that a comment like this would be better off in a guideline. So I am removing the addition. --PBS (talk) 17:47, 28 January 2009 (UTC)

How about this april 2008 version? Or so many versions in between or before? 212.200.243.116 (talk) 19:17, 28 January 2009 (UTC)

From all the random versions in the past that I looked at, "Uncle G's version" of paragraph is present in most of them. You need consensus for your change, as WP:V is such an important policy. 212.200.243.116 (talk) 19:19, 28 January 2009 (UTC)

My error. I mixed up the difs. it's "your" version actually. 212.200.243.116 (talk) 19:25, 28 January 2009 (UTC)

To clarify the diffs between the version you gave and the current one are here and it does not show that the 14 December 2008 addition was present before that date. --PBS (talk) 09:51, 29 January 2009 (UTC)

Redesign of the edit box?

Hi, I don't know for sure if this is the right place to ask this but I believe it is relevant to verifiability. Well lately I've been thinking about the edit box used for articles plus how annoying it is undoing vandalism on pages and, in doing so, I managed to think up a possible way in which the edit box could be changed. At the moment, there is the ability to add, edit and remove using the edit box (like I am at the moment). However, there is nothing to stop those without verifiable material adding to any article they wish. So to help combat this issue (directly targeting those who intentionally vandalise and those who simply do not understand the need to source their content) I thought that the edit box should be redesigned for article pages. I don't know if this has been discussed already but if it has please let me know.

Idea: Firstly, the edit summary should be moved to the top and possibly renamed with a better description of its purpose (possibly even have the wikipedia copyright line at the top as well). Now this is where it gets harder to describe but I will try my best (I don't know all the technical terms so bear with me). So you are on the edit page like you usually are when modifying an article. However, you do are not given the ability to edit anything this time. Instead, you see the text in the same format as the article itself but in a edit box (ie links still show as [1] etc). Now, to edit you need to highlight the paragraph/line/word you want to edit and click another edit button below the box, which opens a edit box this time under the one you are currently viewing (this helps to further define the edit being made). Okay so now we have two boxes: one acts as a preview displaying the change immediately and the second one is used to edit. Okay so the second one (the edit box) will have special functions like being able to identify verifiable material. For instance, say I'm in the middle of creating a new line of text, well the edit box will identify the full stop at the end and ask me to insert a reference before I can submit my edit (cause really everything must be verified). Well then I insert a reference like I usually do and click submit. So the edit box will then close and an updated preview box (the first box) will show my edit and the original text around it. Well before I can submit my change to the article I then must create a edit summary before the submit button will become available. Once that is done, I submit and the article is modified.

Obviously, I can't discuss everything involved with this ideo so, for everyones sake, I've just talked about one type of edit scenario. Like I said earlier, if this has been discussed already please let me know. Furthermore, please let me know if this is just not possible. If you are wondering why I came up with this idea it's cause I believe the current way of editing is simply not a viable solution for wikipedia if it intends to become "encyclopedic", especially since vandalism is all too common on wikipedia. Thank you for reading.- Jack ^(Talk) 10:21, 29 January 2009 (UTC)

Nice idea. But what if I just edit the spelling, or wikify the text, or make some minor edit? Another problem is that people could add unrelated reference with the edit. 212.200.243.116 (talk) 00:16, 30 January 2009 (UTC)