Wikipedia talk:Verifiability/Archive 29

page numbers and quoted material

Several weeks ago Crum posted his thoughts on quoted material. Since he is still edit warring over his opinion today I am pulling from the archives to hopefully enable more discussion and less edit warring.

What you can do if the referenced item is contentious and the source not readily available (i.e. online with no subscription required), is to ask the editor adding the material to provide more specific details. For example, you can ask for exact quotes, page numbers, specific URL and/or other locating information. The quoted material provided (which need not necessarily go into the article) should be sufficient to understand the context and validate the reference. You can also ask for evidence that the publisher is reliable, e.g. name of publisher, number of people involved in the fact checking process, examples of previous publications, citations in other independent publications, etc. If the provided context material is insufficient, and the inserted material is contentious, it should be removed pending proof of verifiability and reliability. Crum375 (talk) 03:52, 19 May 2008 (UTC)

Personally my thoughts on this is that while it is acceptable to expect someone adding new information to provide these details on request, it is not acceptable remove material from an article some months after it was added (and the library books long ago returned) because while enough information is given to locate a book the editor did not provide page numbers, etc. Editors will use such a statement about page numbers in this policy to remove uncontraversial information that came from a reliable offline sources. If anyone doubts that read Talk:Marcel Lefebvre where such uncontraversial information as when a notable churchman was ordained, the title he was given and the number of parishes under his authority was up for removal because the sourced information wasn't cited well enough for the person wanting of nominate the article as a Good Article despite there being no dispute about the actual content. My offers to format the citations properly were dismissed because simply formatting them as footnotes would not provide the page numbers he believed to be required. Here is an edit [1] making the argument for removal of information that only lacked page numbers in Sept 2006 based on the guideline at WP:CITE. Adding simlar language to this policy will certainly make the activities seen in this example become more widespread. Can you alter this language somehow to address my concerns about editors who do not contest the actual content and yet will insist on removing sourced material on the technicality of lacking page numbers?--BirgitteSB 21:14, 11 June 2008 (UTC)

If someone insists on inserting material which is properly sourced, that means that they have the source handy. If the source is not readily available online, and someone asks for a full quote and page number to evaluate this, I see that as very reasonable, and in fact, this is what we do on a daily basis on Wikipedia. I just can't think of a reason why anyone insisting on inserting (or keeping) the material would refuse this good-faith request. Crum375 (talk) 21:38, 11 June 2008 (UTC)

I gave an actual example showing just how such a situation is possible. The book was no longer in the editor's possesion months after he had used it as a resource for the article. I joined him in insisting the sourced material not be removed because no one believed the material to be incorrect and the article was a better article with detailed information from a high quality source than it was without it. Do you believe my actions in that example were not in good faith, or do you simply fail to understand how language meant to forestall disputes over content that has disagreement can be problamatic when dealing with content where there is no disagreement?--BirgitteSB 21:51, 11 June 2008 (UTC)

Without getting into specifics, I think if no-one has access to a source, even if someone claims to have seen it in the past, the claim which rests on that source should be removed, or embedded in edit comments, pending finding of the source. I am not casting any doubt on anyone, and assuming good faith, but we need to have at least one editor with access to the source. Our entire policy of WP:V rests on having such access. Crum375 (talk) 22:07, 11 June 2008 (UTC)

You need access to the source while writing the material. As I said above I support the expectation that when adding new information to an article you must provide any details about the source that are asked of you. However you are incorrect that the Verifiability policy is based on at least one editor having access to any sources used at all times. The Verifiability policy is based on the material in our articles being supported by published sources. An important facet of the policy is that when material is controversial or questioned it must be either have a source provided or else be removed. However that facet is not the whole policy. Most information in Wikipedia is not controversial and requiring that these details be provided any time someone chooses to remove the material will be detrimental to the encyclopedia. Certainly these details should be required where there is a dispute over the accuracy of the material. However there is no foundation for requiring this at any time information is removed for any reason.--BirgitteSB 22:46, 11 June 2008 (UTC)

This is all guideline-type stuff. Too detailed, and flexible, to be put on the policy page. If someone doesn't have a page number, etc., that's too bad. Maybe poor sourcing, maybe poor style. Could be better. But it's not a policy violation and it doesn't meant he information is unverifiable. Wikidemo (talk) 23:26, 11 June 2008 (UTC)

The real issue is that not all challenges are equal and Crum is insisting this policy address challenged material in a way that is only appropriate for extremely strong challenges. What he is advocating is a helpful approach with very strong challenges (i.e. I know the information I just removed in inaccurate or unreliable) but it is a harmful approach for medium or weak challenges ({{dubious}} or a request to upgrade to a newer citation style). Of course most challenges on Wikipedia fall in the middle so advocating an extreme approach does more harm than good overall. I feel such an extreme approach should definitely not be the exclusive one described on a policy page when discussing challenged material in general.--BirgitteSB 23:47, 11 June 2008 (UTC)

What Crum proposes would be reasonable only if it were time-limited. It would be reasonable to expect the editor who added material to be able to supply page numbers and quotes on request for a very short time after the edit (perhaps 24–48 hours). If the material is not challenged in that time, I would say that it's up to the challenger to obtain a copy of the source and verify the citation him/herself. Part of the problem with what Crum proposes is that it does not assume good faith. When an editor adds information to an article and provides a source, we must assume that the information is actually supported by that source unless there is strong reason to think otherwise. If the information seems questionable or controversial, it's reasonable to ask for confirmation from the editor who posted it. Otherwise, the person who wants confirmation can go to the library and look it up if it's not available online, or can ask other editors to do so.--Srleffler (talk) 00:04, 12 June 2008 (UTC)

What Crum proposes is reasonable for any material within a short time frame, and reasonable for controversial material at an time. (Controversial material = Material with discrepancies between different sources) But applying the extreme approach he advocates to challenges involving uncontroversial material in a long time frame will be at odd with assuming good faith and in some cases writing an article from a neutral point of view. But above all it would be detrimental to the process of collaboration, and I find that aspect most harmful to the encyclopedia overall. --BirgitteSB 00:39, 12 June 2008 (UTC)

Brimba just re-added this controversial paragraph, with the edit comment "RV, 'The burden of evidence lies with the editor who adds or restores material.' It is specifically NOT upon the editor who makes the challenge to do the leg work" Besides the plain fact that this policy change does not have consensus, Brimba has made a mistake in interpretation. The policy clearly states that the burden upon the editor is to attribute material that may be challenged to "a reliable, published source using an inline citation". Once the editor has done that, the burden is satisfied. A reliable source is all that is required. There is no further burden. If someone wants to verify that the cited source supports the material in the article that's fine—but it is up to that person to do the legwork unless the addition is very recent. It is not at all reasonable to expect an editor to be available and able to defend contributions made at any time in the past when he or she has already provided a reliable source for the information. The whole point of requiring sources is so that someone else can check the validity of the information contributed. We don't require reliable sources to be available online, and for good reason. Crum and friends seek to impose an entirely new burden on editors, one not supported by existing policy or consensus.--Srleffler (talk) 04:41, 12 June 2008 (UTC)

Someone coming along months after the material is entered into Wikipedia, and having enough knowledge of the subject to reasonably question a statement, has the right to challenge the material. In so doing, they are not required to disprove or otherwise research the statement. “When an editor adds information to an article and provides a source, we must assume that the information is actually supported by that source unless there is strong reason to think otherwise.” No, does not work that way. Anyone can ask for a source if they have a reasonable expectation that the material is incorrect. “Otherwise, the person who wants confirmation can go to the library and look it up if it's not available online, or can ask other editors to do so.” Bunk. “Controversial material = Material with discrepancies between different sources” Statements based upon a single source can most certainly be controversial; there’s no requirement to have two sources. Brimba (talk) 04:43, 12 June 2008 (UTC)

I disagree. Someone coming along months later certainly has the right to challenge the material, but if the material is sourced it is up to the challenger to provide evidence to dispute the statement. That might involve looking up the original source to confirm that it says what is asserted. It might instead involve finding other sources and seeing if they agree. Any other policy makes no sense. There is no reason to assume that the original editor will still be available months later, and even if so he or she may no longer have access to the cited source. Wikipedia would be harmed by encouraging deletion of sourced material merely due to the failure of the original editor to remain available and keep the source in hand indefinitely.--Srleffler (talk) 04:53, 12 June 2008 (UTC)

OK, I see what you are getting at, you are saying material that already is cited. I was under the impression that this was material that was attributable to a RS, but not cited. I will go back and read the whole thing again. Brimba (talk) 04:48, 12 June 2008 (UTC)

It would be a huge new policy requirement - basically requiring people to quote (which is a potential copyvio) or cite page numbers for, sources that are not freely available online. "Being able to" basically means you have to extract that information at the time you source the material. You may not be around later, or you may not have easy access to it. Whatever library book you had may be returned. You may in a new city, etc. That's not an unreasonable requirement but it would be a huge change and impose considerably more burden on sourcing. Plus we'd have to find a system for tracking it. In a set of new citation templates? Then you would have to grandfather in or phase out all of the material currently cited to sources that are not freely available - otherwise, anything currently in the encyclopedia on those sources could be discarded. Then there's the question of material that's currently available that becomes unavailable later. Overall, I would say it's too much burden for a volunteer project like ours, but it would improve sourcing quality. At any rate it's a major enough change affecting so many articles that a project-wide discussion of some length would be in order. And as I said before, I don't think this is the correct page for that requirement. It's a citation style, which is a guideline issue, not a question of verifiability. Wikidemo (talk) 06:04, 12 June 2008 (UTC)

What I see here isn't a need to change the policy... but a better way for people to challenge "incomplete" citations than just removing the information... Something that fills the gap between completely uncited and fully cited. Do we have a tag that can be added to cover a request for a more complete citation? The {{refimprove}} tag is really for entire articles or sections... is there something that can be applied for individual citations? If not, perhaps something like {{please improve citation}} sould be created. This would then tell editors that they need to re-obtain the book from the library, and add the missing information, but would also give them a reasonable amount of time in which to do so. Blueboar (talk) 11:57, 12 June 2008 (UTC)

I'm inclined to agree. I understand some of Crum's feeling about this, having just gone through a protracted struggle over a reference with severe WP:SAYWHEREYOUGOTIT issues-- to the point where it appeared that the supposed original source didn't appear to exist. It's reasonable to expect someone citing from a 700 page book to give some indication of where in the book the material lay. But it's disruptive to encourage people to stir up fights over otherwise uncontroversial material. And the page number issue is just one aspect of defects, and defects but one aspect of citation problems. We really don't have mechanisms for dealing with verification issues in general.

Also, I would point out that books often have indexes; if the work can be obtained by another, I would imagine that for uncontroversial material it's probably not too hard for someone to find the material and update the reference. Verification is better than redaction, as far as building the encyclopedia is concerned. If we are going to emphasize better referencing, we need to do so in a way that doesn't encourage legalistic destructiveness. We want to encourage people to take the effort to fix bad references, not just delete stuff out of a sense of self-righteousness. Mangoe (talk) 13:08, 12 June 2008 (UTC)

(ec)Yes there are several such templates that might work for that ({{verify source}}; {{verify credibility}}; {{request quote}}; and {{dubious}}). However please don't forget that not everyone removing information from an article is actually disputing the accuracy and reliability of the information. I gave an example of such a weak challenge where information was being removed because requirements for citations had changed since the material was originally added and the editor removing the info would not accept the original contributers assertion that the material he added within the time frame of a few weeks came from the book that he added to the reference section in that same time period. When I verified that the material was covered in the version of the book I was able to access he still would only accept citations with page numbers, which meant I had to cite all the material from a French language edition with page numbers rather than make citations for the English version the original editor used without page numbers. Also the book was a biography on the subject of the article so it is not like they So now that article contains citations with page numbers to a French language book when anyone reading the article is unlikely to read French and the person who added quality material from a quality source stopped participating at the article. In this case, after the migration to footnotes that article was more poorly sourced than before as many unreliable webpages, which did not even fully support the statements, were cited for material that originally came from books because WP:CITE was interpreted in a way the preferred easily internet accessible sources and page numbers for books. If we migrate this language from a guideline to a policy examples like that one will become all the more common to the detriment of the encyclopedia. So please don't imagine that all challenges to material will be based the fact the material is truly disputed as inaccurate or unreliable and please don't write policy as if that were true.

And just to in case you think my experience above means I don't have a good general idea of the sort of challenges that happen across the project I will explain a little more of my experiences. One reason I claim that most challenges in Wikipedia are not strong challenges is because I am aware of the real numbers of challenges throughout the encyclopedia. There are 110,000+ articles that contain either {{fact}} or a general request for citations, and probably about half the articles with {{fact}} have at least three instances of the template. Many of these direct challenges have been in place since before February 2007 when a bot went around and dated on the undated {{fact}} templates. In the last week I went through all the articles in the December 2007 category and moved any statements still tagged with {{fact}} to the talk page. I think it is very safe to say that when material is challenged and the challenges are abandoned for over a year it is not a strong challenge, nor are they weak challenges since someone took the time to flag some particular statement(s) within an article with one of the stronger templates. We definitely have 30,000 such challenges with {{fact}} alone probably more than 50,000 if I were to consider other templates as middling challenges. Since I read a good number of these abandoned challenges recently I can tell you with confidence that it was very rare to find a statement that I felt should have been a strong challenge (for NPOV reasons or simply because it was an exceptional claim) and removal pursued more aggressively. None of them struck me as being blatantly inaccurate and I would hate to see any of my removals refused back into the article because the re-inserting editor does not include page numbers. I know of at least one re-insertion without a proper citation but with the editor saying what the source was in a very general way that I will not dispute any further since I do not strongly doubt the accuracy of what I removed. As for strong challenges, I would imagine we have less than 1,000 active disputes over the inclusion of material, but that is a pure guess. Certainly I cannot believe there are 5,000 challenges that are strong enough that an editor is actively involved getting material removed. Certainly looking at those numbers, I am confident that the large majority of challenges on Wikipedia at given moment are the middling sort of challenges and it would be a mistake to write policy treating all challenges as if they were strong challenges.--BirgitteSB 14:34, 12 June 2008 (UTC)

I've not read through this whole discussion (I will later), so apologies if I'm repeating something already said, but if the issue is just whether we need page numbers for book citations, we do, as CITE made clear when I last looked at it. They're also required at FA level.

That doesn't mean something should be removed immediately if we're dealing with a good-faith editor who just forgot to add the page number, and now doesn't have the book anymore. He can be given some time to find it again, but it does need to be added at some point, unless it's the kind of very general point that sometimes it's okay to cite an entire book for. SlimVirgin ^talk|edits 17:38, 12 June 2008 (UTC)

I would urge you to review the discussion before changing a policy. For the reasons discussed I would not support adding the page number requirement to the policy page at this time, and appears to be no consensus for that here or in Wikipedia at large. The best practice may be to include page numbers, and we have a guideline for those best practices. But it has not been a policy requirement that information without page numbers is deemed unverified.Wikidemo (talk) 17:55, 12 June 2008 (UTC)

Is it a change? If so, we could add that best practice is to add page numbers, and leave it there. SlimVirgin ^talk|edits 18:07, 12 June 2008 (UTC)

I can agree with that when dealing with controversial material. But I do not understand why it would ever be appropriate to remove or oppose restoration sourced material that is uncontroversial simply because the page numbers are absent. The change that attempted here was requiring page numbers for any information that is challenged with no restriction to only controversial or disputed information. I give a some (long-winded) details above about in which sorts of cases I find the approach you are advocating helpful and where I find it harmful and the range of seriousness that exists for challenges on Wikipedia.--BirgitteSB 18:03, 12 June 2008 (UTC)

If it's entirely uncontroversial, I'd agree that it would be best not to remove it, but better still to find a page number. Sorry for not reading everything you wrote higher on the page, Birgitte. I'll do it shortly. SlimVirgin ^talk|edits 18:07, 12 June 2008 (UTC)

Is the page number addition a change? It says "including page numbers for book sources and direct quotes if challenged." The policy already said that inline citations are needed for anything challenged, likely to be challenged, and quotes. And inline citations include page numbers for books.

We are only saying "if challenged," so I'm not sure I see how this is a change. SlimVirgin ^talk|edits 18:11, 12 June 2008 (UTC)

SV, I only visit this page every couple weeks because otherwise my head would spin, but I think it was added about a month ago and reverted in and out since then. My condensed version is that enforcing this more rigorously may be a good balance of extra burden to extra reliability of content, but we have a possible grandfathering problem for existing page content added to date, and also in cases where a formerly accessible online link becomes inaccessible. We've never required strict compliance with the citation templates. Some day we may start enforcing accessdate...who knows? I'm fine with referring to it as a best practice. Wikidemo (talk) 18:18, 12 June 2008 (UTC)

Hi WD, my head is spinning too, as I've not looked in here for awhile, and I can't work out what's new and what's old. The thing about page numbers -- we've long asked for sources for anything challenged/likely to be etc. By "sources," we pretty much always meant inline citations, and that was added, as I recall, maybe a year ago (guessing wildly, could be wrong). As I see it, the page number requirement is just expanding on what we mean by inline citations. Remember, this is only if challenged/likely to be.

Maybe I'll try to formulate a sentence that makes things clearer and includes the "best practice" phrase, so that people don't go around wikilawyering material out of articles -- though bear in mind that we have a paragraph that makes it clear people shouldn't do that. SlimVirgin ^talk|edits 18:23, 12 June 2008 (UTC)

Random break

I went to add "best practice is to include page numbers for book sources and direct quotes, particularly if challenged," but the problem is that this makes it sound as though it's okay not to provide a page number. But if someone is using a direct quote, they must have that book in front of them, in which case why wouldn't they add a page number? Quotes really do always have to be directly attributed. All material does, but quotes in particular. If you're saying that a person X said certain words, you really do have to say where, exactly. SlimVirgin ^talk|edits 18:30, 12 June 2008 (UTC)

We are only saying *if challenged,* so I'm not sure I see how this is a change. *if challenged* adds nothing. *If challenged, you cannot restore material from books without page number* is equivalent to *you cannot restore material from books without page number*. Since removing information counts as a challenge it is basically two words that add no meaning. Any restoration is a challenged one. Please don't be confused into thinking only of new material being added. This will equally apply to all of Wikipedia. So question isn't why wouldn't someone add a page number as why hasn't everyone over the last four years added page numbers. If you would this restrict "provide details on demand or face removal" to new material or disputed material, I would not object. But this policy applies to everything in Wikipedia which is why changes need to be well-vetted and not forced through by edit-warring.--BirgitteSB 18:44, 12 June 2008 (UTC)

Okay, back to basics: the policy has been for a long time that quotations and any material that is challenged, or likely to be challenged, needs a source. "Challenged" means disputed or questioned by someone.

By "source," we mean an inline citation, and that was made explicit at some point (I recall about a year ago, but maybe it wasn't that long ago). It's also in CITE, and it's in the FA criteria (or was when I last looked). An inline citation for a book includes a page number, otherwise how will people find the material? So this really isn't a change; it's just an expansion of what was already there.

Having said that, no one should be running around removing uncontentious material just because a page number is missing; instead, the person who wants to keep the material should be given plenty of time to find one. I've often added a citation with "will add page number soon" in an invisible comment, if it's a situation where I've read a book and recall that this was in it, but can't put my hand on it right now. I almost always remember to add the page number, but sometimes I've forgotten and people have had to prod me, and on a couple of occasions, the material has been removed because I couldn't find the book or quote or whatever it was, which is fair enough.

We do have a paragraph that says material should never be removed in order to be difficult without giving people a fair chance to find the sources (or words to that effect).

Birgitte, is there an actual example of this requirement having caused a problem e.g. edits being removed inappropriately? SlimVirgin ^talk|edits 18:52, 12 June 2008 (UTC)

I gave some examples of the range of challenges I have seen above, but I don't think my position is as radical as you are thinking. Just to clarify I don't object all mentions of page numbers and would agree to Markskell's example version which is more general. I only object when we begin adding very specific language mentioning restoring information. (I would nit-pick Marskell's that version and prefer "when requested" to replace "if challenged" and put a ";" after page numbers but I wouldn't object to his version now realized that this not the only interprtation possible and therefore more than a nit-pick)--BirgitteSB 18:59, 12 June 2008 (UTC)

OK, that's not quite my version. The last five words of the second were added by someone else and I agree that "challenged" is problematic as it can be seen to imply that sometimes it's OK not to cite quotes. Marskell (talk) 19:07, 12 June 2008 (UTC)

As noted below, I would simply leave it as "... including page numbers for book sources." The opening of the Burden section already makes clear quotes must always have a cite. Marskell (talk) 19:10, 12 June 2008 (UTC)

It is not my dream version and is ambiguous without the fix you suggest (are the quotes in the article or within the disscussion about the source?), but it is not anywhere near as objectional to me as the edit I started this heading with. --BirgitteSB 19:12, 12 June 2008 (UTC)

I'd be fine with Marskell's suggestion too, minus "if challenged." Having glanced at the discussion (not yet read carefully, so apologies if I've missed something important), Crum's concern was that editors should provide quotations on talk with page numbers if asked to do so, so that other editors who've not seen the book can judge whether the edit corresponds to the source. What I suggest is that we add in a footnote that "as a matter of courtesy," editors should provide quotations on talk on request for any challenged material. That way, we're not expanding the policy. SlimVirgin ^talk|edits 20:02, 12 June 2008 (UTC)

I have no issue with a footnote about quotations being provided on request. Although the more I look into it the more I think that the whole issue about providing quotations from the source is a misunderstanding of the orignal language about extra careful citations for direct quotes in the article. (telephone game via edit-war rather than whispers)--BirgitteSB 20:16, 12 June 2008 (UTC)

An "as a courtesy" footnote could work. At once, we might include the countervailing point that requesting editors should allow sufficient time for page numbers to be brought. What I oppose is explicitly demanding extra talk page hoops for editors to jump through wrt to quotations; I agree with Brigitte that this seems very problematic. Marskell (talk) 20:22, 12 June 2008 (UTC)

Proposal

Please check out User:BirgitteSB/Burden of evidence. Feel free to edit at will, but please put links to your changes here for easier discussion.--BirgitteSB 20:36, 12 June 2008 (UTC) I hope this doesn't really need to be said, but . . . If anyone would be willing to make this revert to the policy page, please object in this section first.--BirgitteSB 20:47, 12 June 2008 (UTC)

Late for me. Will comment on waking. Marskell (talk) 21:16, 12 June 2008 (UTC)

• I changed the original footnote to now read "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." diff from live policy--BirgitteSB 21:50, 12 June 2008 (UTC)

Hi Birgitte. My first opinion is: I dislike footnotes in policy strongly, they are generally better served as inline text; "any other details requested should be provided" would leave too much latitude for the challenger; "should" and "courtesy" appear to be in opposition; and in general, the whole question of party duties in insertion disputes is far more complex than this sentence. The effort represented by this sentence is valid but I wouldn't be able to comment better without thinking through the whole statement of duties as one set. JJB 15:35, 13 June 2008 (UTC)

Changes to Policy

How many forums on the project have been informed of these changes? Policy is a project wide concern and needs the consensus of the community. Any and all changes should be discussed on the talk page first, and proposals should be flagged on all the project forums. --Domer48 (talk) 15:47, 12 June 2008 (UTC)

As I explained, you are one of the people changing the policy, why don't you begin by explaining where you have discussed your changes?--BirgitteSB 16:06, 12 June 2008 (UTC)

Hi Domer48, it looks like the language you reverted back to was only recently added in less than 2 weeks ago diff. I don't know how much discussion was had about it, but it the way it reads now, it looks game-able -so, personally, I'm against it. Thanks. R. Baley (talk) 16:22, 12 June 2008 (UTC)

Probably nothing whatsoever to with the fact that Domer48 has been quoting the addition on a page he's involved in an edit war on (Talk:The Great Hunger)... Bastun^{BaStun not BaTsun} 17:29, 12 June 2008 (UTC)

It's a long standing thing where if a source is disputed the editor provides [the exact quote(s) from it to prove that it sources the text in question. From an earlier discussion some might not be particularly happy about this being policy, but it needs to be, otherwise we are just opening the door to editors adding anything they want and citing all sorts of things. The encyclopaedia will suffer if we allow that to happen. Hypothetical situation:

Actual text added to article: The British Conservative Prime Minister Robert Peel, immediately recognizing that the circumstances in Ireland meant that this crop failure could cause famine, ordered Indian corn and meal to be sent from the United States and a Relief Commission set up. David Ross, Ireland: History of a Nation, Geddes & Grosset (Scotland 2006) ISBN 13: 978 1 84205 164 1 , p. 311

Now the editor who added the text is adamant that the text is supported by the source. (the source is provided below) How dose one challenge this information if you are not sure it’s properly sourced? Requesting direct quotes from the source provided is the most logical way. It’s logical that the information we add places obligations on us as editors. The obligations are WP:V , WP:OR and WP:NPOV? Based on these policies, which should be the most difficult, adding or removing information?

How is the embellished information removed, if the editor insists that it is correctly sourced?

Actual Source verbatim: Famine begins. Prime Minister Robert Peel orders corn and meal to be sent from the United States. A Relief Commission is set up under Edward Lucas. David Ross, Ireland: History of a Nation, Geddes & Grosset (Scotland 2006) ISBN 13: 978 1 84205 164 1 , p. 311

"Citing sources and avoiding original research are inextricably linked: to demonstrate that you are not presenting original research, you must cite reliable sources that provide information directly related to the topic of the article, and that directly support the information as it is presented." From our policy on WP:OR. How are we to raise the highlighted section of our policy if we challange sources.

It has everything to do with the fact that a policy I have been using is changing every other day. Nice to see you trolling my edits Bastun. --Domer48 (talk) 17:39, 12 June 2008 (UTC)

Trolling? I repeat, you've been quoting the recent addition on a page where you're involved in an edit war, (Talk:The Great Hunger), which you've been warned about. Bastun^{BaStun not BaTsun} 18:11, 12 June 2008 (UTC)

Yes Bastun, Trolling. Now you've made your point, would you like to contribute to the discussion?--Domer48 (talk) 18:42, 12 June 2008 (UTC)

I've not had time to look at who has changed what, but it's really inappropriate that the policy can't stay stable for more than a few days at the moment. Editors have to be able to trust that the sentence they quote on Monday is still there on Tuesday.

Would people please leave it alone, and only change things if they're either completely uncontroversial, expanding succinctly on something already in the policy, or simple copy editing improvements (so long as you are certain that it's an improvement)? Otherwise, please wait on talk to make sure you have consensus and make sure the change is consistent with best practice; if it isn't, no consensus on this page should override that. SlimVirgin ^talk|edits 18:19, 12 June 2008 (UTC)

I quite agree. --Domer48 (talk) 18:42, 12 June 2008 (UTC)

Me also. I'd add that where an editor is quoting a policy in an ongoing dispute, it is wholly inappropriate for that editor to simultaneously be involved in editing the policy. Bastun^{BaStun not BaTsun} 07:58, 13 June 2008 (UTC)

and direct quotes if challenged.

For the reasons already stated in the discussions above, the wording “and direct quotes if challenged.” causes problems as longstanding policy states that all quotes are cited -no challenge is required.

We currently have three paragraphs within the Burden of Evidence section that prior to July of last year where configured as a single paragraph reading as:

The burden of evidence lies with the editor who adds or restores material. Material that is challenged or likely to be challenged needs a reliable source, which should be cited in the article. Quotations should also be attributed. If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it.

In July of last year the wording was changed to require the use of inline citations, which resulted in the paragraph being split in two:

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged should be attributed to a reliable, published source using an inline citation. The source should be cited clearly and precisely to enable readers to find the text that supports the article content in question.

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

That wording held for most of a year, being changed early last month and in so doing the two paragraphs became three. The current status is:

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged should be attributed to a reliable, published source using an inline citation.

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 for book sources and direct quotes if challenged.

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

I think that is probably better English per say, and yet that’s the source of the problem; the last “if challenged” would not be necessary if the first two paragraphs where recombined to form a single paragraph. In so doing we drop the last two words “if challenged” and which eliminates the ambiguity, and allow the earlier wording “All quotations and any material challenged” to be the last word on the subject. I will update User:BirgitteSB/Burden of evidence to reflect this. Brimba (talk) 02:51, 13 June 2008 (UTC)

Agreed. The final "if challenged" is misleading. SlimVirgin ^talk|edits 17:32, 13 June 2008 (UTC)

Thanks for researching what exactly happened. I am impressed, because everytime I try to do something like that on this policy I give up in frustration. Restoring old format is a nice fix and I don't object to removal of the hidden comments instructions nor any of the copyedits.--BirgitteSB 19:17, 13 June 2008 (UTC)

Very well done. --Domer48 (talk) 21:01, 13 June 2008 (UTC)

Exceptionaly high-quality reliable sources?

Where is the community consensus to make such a drastic and far reaching change to a policy page? The net effect of all those qualifiers would raise the bar so very high, that very few sources would qualify. To me that would include only the very most prestigious peer reviewed journals, with very well known and unimpeachable authors. The criteria has been: Exceptional claims require exceptional sources. Sorry, but this requires wide and deep community consensus before being changed. — Becksguy (talk) 23:34, 20 May 2008 (UTC)

I changed "exceptional sources" to high quality reliable sources, exactly as it is mentioned further on in the text. The reason is that "exceptional sources" can be misinterpreted as being exceptional in some other respect. E.g. there are "exceptional" journals that exclusively deal report on anomalous phenomena. Needless to say, such journals often have very bad reputation in the scientific community. This clearly is not the intended interpretation of "exceptional".

So, I thought, why not simply say "high-quality reliable sources" as is done further on in the text. But then someone else demanded that the qualification exeptional (in the correct interprepretation, of course) is still needed. If this is changed back I would prefer "high-quality reliable sources" over just "exceptional sources", because pseudoscience POV pushers are guaranteed to misinterpret the meaning of "exceptional" :) Count Iblis (talk) 03:14, 21 May 2008 (UTC)

Fine, but we still need exceptional community consensus before a policy page can be changed. This change will have far reaching effects anywhere this policy is cited, especially in XfDs. It needs to be reverted. — Becksguy (talk) 21:03, 22 May 2008 (UTC)

What is an "exceptionaly high-quality reliable source"? For example, would investigative reports in major newspapers count? Wire service reports that have been widely published? Mainstream scholars writing for large publishers? ·:· Will Beback ·:· 08:37, 1 June 2008 (UTC)

If we want to make Wikipedia "exceptionaly high-quality", "exceptionaly high-quality reliable sources" are necessary. Masterpiece2000 (talk) 03:34, 10 June 2008 (UTC)

Sure, and if we want Wikipedia to be "super-duper" then we'd only use "super-duper" sources. But that doesn't answer my question. What is an "exceptionaly high-quality reliable source"? ·:· Will Beback ·:· 19:24, 10 June 2008 (UTC)

since there's not obvious consensus for this change, and since ""exceptionally high-quality reliable sources" is undefined, I'm going to change it back to "high quality reliable sources." ·:· Will Beback ·:· 03:52, 15 June 2008 (UTC)

Looks like somebody else already got it. ·:· Will Beback ·:· 03:57, 15 June 2008 (UTC)

Self-published and questionable sources about themselves

The section Self-published and questionable sources about themselves says among other things that such sources can only be used when "the material used is relevant to their notability". Shouldn't that be when the material used is not relevant to their notability?? We should not rely on a company's publications, for example, to establish that company's notability, but a company's website and publications can provide all kinds of information about the company and its products. As long as the statements used in the article are uncontroversial and comply with the other restrictions on self-published sources, such references should be acceptable. For example, one might cite a notable company's website as the source for the company's current products. That is often done, but appears not to comply with the policy as written. Am I missing something?--Srleffler (talk) 03:20, 9 June 2008 (UTC)

I notice that the text in this section initially required that the information be notable, and was changed to require that the material used "is relevant to the person's notability". I think I understand now what was intended here: the intent was not to imply that the material must be used to establish the source's notability (which would not comply with Wikipedia:Notability), but rather that the material must be relevant to the thing for which the person (or organization) is notable. I think this should be changed to require that the material be relevant to the article in which it is to be used. Notability is the wrong standard here. A source should not even have to be notable to be used for information about itself: if it is appropriate to mention a person or organization in a Wikipedia article, it is surely appropriate to use things they have written about themselves as references for information about that person or organization (assuming it complies with the other requirements).

Originally this section restricted the use of self-published sources to articles about those sources, which would automatically require the source to be notable. The current text, however, does not have this restriction: self-published sources can be used as sources about themselves in other articles as well. Because of this change, though, it is no longer automatic that the self-published source be notable.

The criterion "the material used is relevant to their notability" should be rephrased to something like "the material used is relevant to the article". Comments?--Srleffler (talk) 21:33, 9 June 2008 (UTC)

The material, if self-published by the topic of an article, is inherently relevant to the topic of the article, so you're really arguing about deleting the clause, which should not be done. And the reason we want this clause there is that we don't want articles to be filled with endless piles of crap published only by the topic of the article. At present, a notable physicist's article (who is actually notable for being a physicist) can include sections on his personal theories/analyses of physics sourced to his own writings. The criterion you are objecting to currrently prevents editors from writing in his article diatribes on every issue in politics or sports or medicine that this physicist has ever ranted about on his blog. Someguy1221 (talk) 22:08, 9 June 2008 (UTC)

Take a different example... Say a person is notable for writing a book, and makes comments on that book on his personal web page (a self pulbished source) ... surely his comments are relevant to his notability, and should be citable. Blueboar (talk) 22:23, 9 June 2008 (UTC)

I see your point and agree, but we seem to be thinking about very different cases. The case I have in mind is a corporate website. Such a site is a self-published source, is it not? It's certainly not a "reliable source". My concern is that we should have clear rules on when and how information from a company's website can be used on Wikipedia. Clearly if a company is notable, it is appropriate to source material such as company history or a summary of products from material published by that company, as long as constraints 2–7 of WP:SELFPUB are met and the material is relevant. The current wording does not allow that, unless such material is relevant to the company's notability. I think we need to adjust the wording to cover this case, and also to cover what you describe above.--Srleffler (talk) 03:32, 10 June 2008 (UTC)

Um... I would think that material such as the company's history or a summary of its products would be relevant to the company's notability... and thus allowed. Thus, the current wording is fine. Blueboar (talk) 12:25, 10 June 2008 (UTC)

I've wondered about that particular phrases meaning myself. Now i understand it better after reading this, thanx. How about rewording it to read "the material used is germane to the notability of the article topic." ? 208.43.120.114 (talk) 13:12, 10 June 2008 (UTC)

Since no one objected to the rewording i proposed above i think this means silent consensus so i went ahead and edited the policy page. I think this helps make stuff more clearer. 208.43.120.114 (talk) 14:37, 12 June 2008 (UTC)

It looks like my change under wp:self was wiped out in a massive revert of the whole page because of rapid changes under "burden of proof" (wp:proveit) causing problems. I dont think anyone objected before the wipeout and so i am going to put it back again. Once again i am assuming in good faith that the silence represents no objection thus silent consensus. 208.43.120.114 (talk) 16:09, 14 June 2008 (UTC)

Proposal of changes

I am posting this from User:BirgitteSB/Burden of evidence so that if anyone has missed it there, they will have a chance to review it here.

---

Burden of evidence

For how to write citations, see Wikipedia:Citing sources

The burden of evidence lies with the editor who adds or restores material. All quotations and any material challenged or likely to be challenged should be attributed to a reliable, published source using an inline citation. 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.

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

Any material lacking a reliable source may be removed, but editors may object if you remove material without giving them sufficient time to provide references. If you want to request a source for an unsourced statement, consider tagging a sentence by adding the {{fact}} template, a section with {{unreferencedsection}}, or the article with {{refimprove}} or {{unreferenced}}. Alternatively, you may leave a note on the talk page requesting a source, or you may move the material to the talk page.

Do not leave unsourced or poorly sourced information that may damage the reputation of living persons or organizations in articles and do not move it to the talk page (See Wikipedia:Biographies of living persons for details of this policy). As Jimmy Wales has put it:

I can NOT emphasize this enough. There seems to be a terrible bias among some editors that some sort of random speculative 'I heard it somewhere' pseudo information is to be tagged with a 'needs a cite' tag. Wrong. It should be removed, aggressively, unless it can be sourced. This is true of all information, but it is particularly true of negative information about living persons.

–Jimmy Wales

---

Changes

1) Added one footnote:

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.

Wording was added to WP:V on May 8th ,and has remained in one form or another from that time, to address sources that are previously cited, but are deemed “incomplete” (if this word is not to your liking, or seems vague, please read where this is discussed in detail under other headings on this same page).

This footnotes does three things; it does not give anyone a new to remove information that they simply disagree with, it does not move the wording to any guideline where it can be copy-edited out of existence the first time some editor finds it problematic, and lastly it does not add words to the main section, it's a footnote, not a main clause.

We don’t want the wording used as a means of censoring information. In other words, we don’t want to give editors that are pushing a particular POV a new tool/technicality with which to remove information that runs counter to their own views. It imposes no new requirements upon anyone; it prods but does no demand. This is particularly important when either the editor who gave the citation is no longer actively involved in WP, or the editor no longer has direct access to the cited source.

It does not shift the wording to WP:CITE or any other guideline. Guidelines should follow, back-up, and support policy. Too often they do not. Better to have the basic wording included on this page, and if needed an expansion included elsewhere.

2) Invisable HTML comments: it drops the line “an invisible HTML comment on the article page.” Editors can still do that, but we are not openly encouraging it. I think that is self explanatory.

3) Copyedits: Contains various copyedits to simplify the wording. As far as I can tell no changes in policy occurred.

Please feel free to make use of User:BirgitteSB/Burden of evidence.

Brimba (talk) 14:42, 14 June 2008 (UTC)

Forgot to say that it includes the changes proposed in "and direct quotes if challenged." Brimba (talk) 14:56, 14 June 2008 (UTC)

Exceptional claims: Is a change of heart exceptional?

Let's say someone follows a religious movement and while a follower they make statements praising the religious leader, which are widely reported. Later, they leave the religion and make negative comments about the leader, which are also widely reported. Do the later remarks trigger WP:REDFLAG because they are contrary to a position previously held? A similar example would be Scott McClellan. Say a member of a politician's entourage serves as spokesman and defends the politician to the media, but then leaves the job and later writes a book (published by a mainstream publisher) depicting the politician in a poor light. Do his later remarks trigger REDFLAG? And in both cases, even if the claims are considered exceptional, don't mainstream media reports of the comments count as exceptional sources? Do all claims by "apostates" and "disgruntled employees" automatically trigger REDFLAG? If so, how exceptional do the sources need to be for us to report them? ·:· Will Beback ·:· 04:18, 15 June 2008 (UTC)

I could probably write a long essay on this, but I am too tired atm. A) Do the later remarks trigger WP:REDFLAG because they are contrary to a position previously held? No; B) Do his later remarks trigger REDFLAG? No; C) don't mainstream media reports of the comments count as exceptional sources? Yes, notably if the person gave a first-person interview; D) Do all claims by "apostates" and "disgruntled employees" automatically trigger REDFLAG? NO.

None of these answers will satisfy anyone one because I have not explained the answers (not up to it tonight). But in short, if he person sat down with a NYTimes reporter, or gave an interview on 60 MINS, or otherwise made it possible to directly connect the person with the statement, then that is all we require. When “reports of a statement by someone that seems out of character, embarrassing, controversial, or against an interest they had previously defended” and there is no video tape (meaning news reports) or something irrefutably published with that persons blessing, such as publishing a book through a major publishing house in there own name (meaning listed as the author), then we have problems. With living people you have problems when its “Mcain/Obama said xyz” and there’s no major media reports to back the statement, as would be expected. If the NYTimes came out with a story saying so and so said such and such and attributed it to an anonymous source, I could see where depending on what was supposedly said, that might be off-limits, but not if there is no dispute as to the person actually saying it. Even if it was reported in an academic text book later on, what would that text book use as a source? The NYTimes interview; and so should we. Brimba (talk) 05:02, 15 June 2008 (UTC)

Thanks for the short essay, which covers the points well. Let me expand on the McClellan example make sure I've got it. If someone back in 2004 had been asserting that McClellan thought his boss was being untruthful that would have been an exceptional claim, and it would have required very good sources indeed (not "unnamed sources close to X say..."). The same assertion today would not be exceptional, because McClellan's change of heart (or whatever) has been expressed first-hand in highly-reliable sources. Now that his change of heart is known, his statements against his old boss no longer trigger the REDFLAG clause and only require the normal, high-quality sources needed for any BLP. So we could note in Bush's biogaphy that his previous press secretary has since written that he thinks the case for the war was intentionally overstated (or whatever).

This came up in regard to a leader of a new religous movement. The person was made head of his late father's movement at age eight, and his mother and brothers acknowledged his spiritual leadership while they took over administrative control. Eight years later he became an adult and upset the applecart. His mother disowned him by saying he was no longer the chosen one due to his behavior. Some editors here claim that the minimal details of the mother's complaint, as reported by the Associated Press and in numeous books, represents a REDFLAG violation. In the same case a second claimed violation comes in reporting what the founding president of the movement's U.S. branch said in an interview with UPI, also repeated in numerous books, about the leader several years after leaving the movement. In both cases their statements were widely reported by wire services and picked up by countless newspapers, and in both cases their comments have been summarized in at least a half dozen mainstream books, not to mention scholarly papers. Also, both the mother and the past president make some of the same basic assertions, as have other current and past followers.

Even if we were to assume that a single person making remarkable claims about another living person raises a fresh red flag (on the person being discussed), having multiple people independently saying the same thing makes the claims less exceptional to begin with. We wouldn't say that McClellan's assertion that Bush overstated the case for war is itself a REDFLAG in the Bush article because it's a an assertion that has been made by enough relevant people that it isn't exceptional anymore. ·:· Will Beback ·:· 08:07, 15 June 2008 (UTC)

Example 1

• 1978: "50,000 adherents in the U.S., of which 10,000 to 12,000 were considered very active" (Religious Requirements and Practices of Certain Selected Groups: A Handbook for Chaplains, a publication for which Melton would produce later versions)
• 1982: "a U.S. following of no more than 3,000 committed followers in 1982 out of some 50,000 who had been initiated into the Knowledge meditation" (Melton & Moore)
• 1986: "[From February 1979] ... progress of the Mission" (Melton, in a tertiary source)

→ WP:REDFLAG on the 1986 source: Melton didn't change sides, but writes something that is out of sync with his prior analyses, and with all other known sources covering the same period. All other sources indicate in the same period (i.e. from February 1979) a decline of the movement or a status quo at best (Downton, Rudin&Rudin, LATimes in January 1979, Schnabel,...). As usual, Melton collaborates with others when writing his 1986 tertiary source (BTW accidently also misspelling one of the names appearing in the DLM entry), which seems not very well researched, at least not strong enough to verifiably support the "exceptional claim" of progress in the period from February 1979. --Francis Schonken (talk) 11:32, 15 June 2008 (UTC)

Example 2

• "However, as the group withdrew from the public eye, little controversy followed it except for the accusations of Robert Mishner [sic] the former president of the Mission, who left in 1977. Mishner complained that the ideals of the group had become impossible to fulfill and that money was increasingly diverted to Maharaj Ji's personal use. Mishner's charges [were] made just after the deaths at Jonestown, Guyana [...]" (Melton 1986)
• "Suddenly there were new reports from people who'd actually managed the Divine Light Mission—Robert Mishler, the man who organized the business side of the mission and served for 5 1/2 years as its president, and Robert Hand Jr., who served as a vice president for two years. In the aftermath of Jonestown, Mishler and Hand felt compelled to warn of similarities between Guru Maharaj Ji and Jim Jones. They claimed the potential for another Jonestown existed in the Divine Light Mission because the most fanatic followers of Maharaj Ji would not question even the craziest commands. As Jim Jones convincingly demonstrated, the health of a cult group can depend on the stability of the leader.
  Mishler and Hand revealed aspects of life inside the mission that frightened the Deitzes. In addition to his ulcer, the Perfect Master who held the secret to peace and spiritual happiness 'had tremendous problems of anxiety which he combatted with alcohol,' Mishler said in a Denver radio interview in February 1979." (Washington Post, February 15, 1982)
• Transcript of the Feb. 12, 1979 radio interview with Bob Mishler

→ Do we call redflag on the source mentioned in the third bullet above? I'm too mixed up in the imbroglio to ascertain whether redflag applies or not, but here's some additional info: (1) the given link is, as far as I know, the only place where the *full* transcript of the interview has been published; (2) the website publishing this full transcript of the radio interview does currently not pass WP:V#Questionable sources (that is, without unanimity, the current consensus appears to be that that is how the website should be qualified - its webmaster is a Wikipedia editor, writing this morning: "I also believe that [the website] should be recognised as a reliable source but I do not have the time to attempt to argue my case.").

Uninvolved advise would be greatly appreciated! --Francis Schonken (talk) 11:32, 15 June 2008 (UTC)

While these examples are from the same case, they are different from the examples I gave above. As for Francis' second example, I was referring to an interview given to the UPI, which was published in mainstream newspapers. The transcript of a radio interview that exists only on a self-published site is entirely different. I personally don't think that the source is reliable enough to use, regardless of REDFLAG. Regarding your first example, I don't think that changing estimates sets off a REDFLAG. The author isn't doing anything surprising as all of the estimates are vague and other writers make estimates in the same ballpark. For a variety of reasons one estimate may be more accurate than another, but REDFLAG doesn't apply, IMO. ·:· Will Beback ·:· 16:36, 15 June 2008 (UTC)

<<< These specific content disputes should be dealt in article's talk page and not in a policy talk page, as with any other content disputes sources and attribution sources have to be assessed in the context of the entire article to have a chance to comply with all core content policies of WP:NPOV, WP:OR, and WP:BLP in addition to WP:V, and discussed with editors in order to achieve consensus.

As for the generic "change of heart" argument presented here, again, it has to be evaluated in the context of how these were made, by whom, where these were expressed, how these were picked up by reputable sources, as well as how widely these were covered.

Allow me here to give a couple of of fictitious examples that may help illustrate my point.

Take living person X, the CEO of a company which was founded in 1960. In that organization, person Y was president and spokesperson for a few years. After strong disputes about the direction the company, person Y gets fired. Five years later, person Y makes a remark to a journalist that person X abused his employees, poisoning some of them, and beating others, as well as other derogatory comments about person X' character. A few months later, person Y dies in a car accident, leaving person X with no recourse to defend himself. The many historians and scholars that studied and reported on this company and person X, do not mention person Y, besides only a short comment about him in one source, that he made some critical comments after his departure. Now, in the WP article about this living person, should we repeat the allegations of person X, and by virtue of that give voice to these exceptional claims per WP:REDFLAG without taking into consideration the context? or should be use good editorial judgment and simply mention that person Y made some critical comments after being fired and leave it at that?

Another example: Precocious 5-year old is described by his mother to be a better opera singer than Enrico Caruso, claiming amongst other things that he is no other that the re-incarnation of the famous opera singer. As the child is a minor, the mother runs this child's media empire, carefully staging his performance and being the de facto owner of the business based on this child's talent. She does this for 12 years. When the child reaches maturity, he takes control of the business, and pushes the mother out of it. At that point, the mother starts a campaign to discredit him, by making derogatory comments about him such as his is a fake singer, that he does not really sing well at all, using dubbed tapes and lip-syncing while pursuing legal action to try and keep a portion of the business. Now, in the WP article about this singer, who is a living person and continued with his career as a singer, can these exceptional claims by the mother be repeated? If so, can these be made without carefully attributing these derogatory comments to the mother and providing the context in which she made them? For example, should the lead of the article say "he was criticized for being a fake and lip-syching to a dubbed tape" without mentioning the mother? Furthermore, should the mother exceptional claims be made in the lead, or would it be more appropriate to present iit in the article's body where it can be properly contextualized? ≈ jossi ≈ (talk) 00:06, 16 June 2008 (UTC)

Your examples would be more fitting if they mentioned that both the ex-president and the mother were making some of the same criticisms, criticisms (or observations) which have also been made by other parties. ·:· Will Beback ·:· 00:19, 16 June 2008 (UTC)

These are fictitious examples. For discussion of specific articles, use article's talk page. Now, lets assume that person Y was the president of the media empire controlled by the mother, OK? So for the purpose of this fictitious/generic example lets say that the president was fired in the same period as the mother, and that both the mother and the president of the company were in collusion and had a massive ax to grind as a result. Let's also assume that no other reputable sources exist besides what was reported about these two. ≈ jossi ≈ (talk) 00:44, 16 June 2008 (UTC)

Those assumptions would not make the examples closer to the case. (And we don't have a 3rd-party source saying that the president was fired, IIRC.) It'd also be a big mistake to say that no other reliable sources exist that cover the same ground. ·:· Will Beback ·:· 00:52, 16 June 2008 (UTC)

If that is your stance, this should be discussed in article's talk page, and not in this page which is a discussion page about the policy of Verifiability in general, and not a specific case that is a complex one under dispute and under mediation. ≈ jossi ≈ (talk) 00:56, 16 June 2008 (UTC)

• I already laid out the basics of the case. An uninvolved editor replied that so long as the change of heart is well-known then statements which are opposed to the speaker's previous interests are not exceptional claims. It doesn't matter if the speaker was fired or not. It doesn't matter if the subject if 5 or 50. to go back to the McClellan comparison, once the change of heart is well-known it's no longer surprising for him to make negative remarks about his old boss. ·:· Will Beback ·:· 01:02, 16 June 2008 (UTC)

Of course it matters. The context in which these statements were made are essential for NPOV. In particular when the statements are libelous and outrageous. The McClellan book does not libel Bush. ≈ jossi ≈ (talk) 01:07, 16 June 2008 (UTC)

We're not talking NPOV, we're not talking libel, we're talking about "Exceptional claims require exceptional sources". A) These are not exceptinal claims. B) Even if they were, they are supported by exceptional sources. ·:· Will Beback ·:· 01:22, 16 June 2008 (UTC)

Surey exceptional claims: reports of a statement by someone that seems out of character, embarrassing, controversial, or against an interest they had previously defended;, and the sources are not exceptional: no historian, sociologist, biographer, or scholar, repeat these claims, for the same very reasons we have this wording in policy. ≈ jossi ≈ (talk) 04:13, 16 June 2008 (UTC)

Wikipedia:Verifiability is one of Wikipedia's core content policies. The others are Wikipedia:No original research and Wikipedia:Neutral point of view. Jointly, these policies determine the type and quality of material that is acceptable in Wikipedia articles. They should not be interpreted in isolation from one another, and editors should familiarize themselves with all three. ≈ jossi ≈ (talk) 03:51, 16 June 2008 (UTC)

<<< We cannot piecemeal policies and guidelines, as these are not "laws"... these have to be applied with good editorial judgment and in the context of specific material for a specific article. ≈ jossi ≈ (talk) 04:10, 16 June 2008 (UTC)

It is incorrect that no one has repeated the assertions, but that's not relevant. This query is about how to interpret the "exceptional claims" clause. Other policies may apply to particular cases, but those other policies aren't relevant to this discussion over the meaning of a clause of a policy. To get back to this question, if someone held one position at one point, and then make a very public change of heart, their new assertions don't trigger the "exceptional claims" clause anymore. So Scott McClellan's negative comments about his old boss are not exceptional claims today, while they would have been four years ago. Yes it would have been out of character and against an interest he defended then. But now negative comments about his boss wouldn't be out of character and he's made it clear he no longer defends that interest. People have changes of heart, and so long as the change is reported in highly reliable sources, their new views are no longer exceptional. ·:· Will Beback ·:· 06:31, 16 June 2008 (UTC)

McClellan he does not make negative comments about his boss. He made negative comments about his policies. Big difference. As argued before, the application of this specific aspect of V is better discussed in article's talk page and in the context of the specific subject, sources, etc. 03:11, 17 June 2008 (UTC)

It doesn't make a difference. He is contradicting a position he formerly held, but it doesn't violtate REDFLAG because his change of heart is well known. We can only discuss this sensibly on the article talk page if we agree on the interpretation of the policy, which you seem to be interpreting incorrectly. ·:· Will Beback ·:· 03:35, 17 June 2008 (UTC)

WP:V's built-in notability criterion

Re "If no reliable, third-party sources can be found for an article topic, Wikipedia should not have an article on it." [2] - this is a built-in notability criterion in WP:V, aka: an article for a topic on which only primary sources exist should, well, at least be checked per WP:AFD (I don't think there's a "speedy" criterion in this sense nor would I advise there to be one). WP:NOR discusses the applicability of primary sources to support article content. There's no implication by WP:NOR that articles can be built on topics for which only primary sources exist. --Francis Schonken (talk) 12:00, 15 June 2008 (UTC)

Third party does not mean secondary sources it means non Wikipedia sources. It make much more sense to put this sentence into WP:NOR#Sources. Perhaps it should be moved into the subsection of WP:NOR#Sources called WP:NOR#Using sources but that detail can be decided on the NOR's talk page. --Philip Baird Shearer (talk) 12:31, 15 June 2008 (UTC)

One of the concerns is that people should be able to find it: so I oppose moving without leaving a pointer for at least some months. And BTW, although an inane luiguistics determination would be possible (didn't we have that already more than once?) in 99,9% of the cases "third party" is either secondary or tertiary, not primary. --Francis Schonken (talk) 12:40, 15 June 2008 (UTC)

Third-party is ambiguous. When one of the policies had a detailed discussion about not using Wikipeda as a source, it called non-Wikipedia sources "third-party". However at NOR the old mention of third-party sources was seen as non-primary sources and was lost when the primary/secondary/tertiary section was re-written. Personally I think we should not use the term third party and simply describe what we mean. If third part is truly meant to mean "non-Wikipedia" it can be dropped altogether in this sentence as "reliable" already excludes Wikipedia. If the intention is broader it should be probably be rewritten as If no reliable sources can be found which are independent of the article's subject; Wikipedia should not have an article on the topic.--BirgitteSB 14:10, 15 June 2008 (UTC)

"Third-party" has always meant "independent of the article's subject." So, for a company, a reprinted press release (even if printed in a newspaper) would be first-party, as would the company's website and the like, whereas an independent piece written by Consumer Reports or a journalist is third-party. Third-party does not necessarily mean reliable. A blog run by someone who is not an employee or stakeholder in the company, and talks about it, is third-party, but it is not reliable. (Indeed, Wikipedia articles, provided that they are not created by someone with a conflict of interest, are third-party, but they too are not reliable for our purposes.) One would think it's common sense that such third-party sources need to be reliable to count, but I can see where it would be valuable to make such a distinction. Seraphimblade ^{Talk to me} 14:39, 15 June 2008 (UTC)

Seraphimblade is correct; while it is often misunderstood, first-party/third-party and primary/secondary/tertiary are orthogonal classifications. First-party sources are sources that are under the control of, or directly related to, the subject of the source or the claims being supported, and third-party sources are those that, well, aren't. In this sort of instance, there's no such thing as second-party, at least not in any well defined way. Primary/secondary/tertiary are as described in WP:PSTS. At one point I was pushing for a separate page on classification of sources that would clarify these orthogonal classifications, but things went a different way. SamBC(talk) 15:00, 15 June 2008 (UTC)

To help clarify, consider the following cases:

1. An actor's own statements, say on a blog, about their experiences making a film.
2. A studio-produced synopsis or summary of the film.
3. Official statements from the Academy indicating that the actor won an Oscar for the film.
4. A magazine review of the film, where the publisher and author have no ties to the film, studio, Academy, or anything like that.

There are two of these that are primary, two secondary, two first-party, and two third-party, forming 4 distinct combinations (when considered from the point of view of an article about the film or the actor). I hope it is clear which is which, but if people seem to be having trouble understanding, I'm happy to give the answers. SamBC(talk) 15:05, 15 June 2008 (UTC)

However one wants to classify third party I think the sentence should be in WP:NOR#Sources and not in WP:PROVEIT. --Philip Baird Shearer (talk) 16:08, 15 June 2008 (UTC)

I actually agree with the logic of that, but one cannot ignore the fact that people often link here for that very phrase; adding it to NOR makes perfect sense, IMO, but removing it from here has more complex factors to weigh up. SamBC(talk) 16:13, 15 June 2008 (UTC)

• "Third-party" has always meant "independent of the article's subject. followed by Seraphimblade is correct; while it is often misunderstood. It is often misunderstood and therefore it can't always mean anything. I really don't care to dig through the archives and prove how the meaning had changed over at NOR. Personally I think we should clear up the ambiguity that leads to misunderstandings. But above all I do not have as strong feelings on how exactly you handle this sentence as much I do about the edit warring over it. Stop edit warring.--BirgitteSB 16:27, 15 June 2008 (UTC)
  • Consensus seems to be forming that the sentence in question should remain here, and probably be in NOR as well; however, I think that it is worth trying to clear up misunderstandings. Given that the terms "third party" and "independent" are used in many places, how do people feel about trying to put together some definitions and explanations that can be put on a new WP-space page to be referred to from the various policies and guidelines? SamBC(talk) 16:33, 15 June 2008 (UTC)
  • NB: the previous attempt to do this is at Wikipedia:Classification of sources. SamBC(talk) 16:34, 15 June 2008 (UTC)

I agree with the idea of including it in both policies. Blueboar (talk) 19:42, 15 June 2008 (UTC)

Useless if we don't apply it

Even admins want to keep articles there are no sources for... and when I try to remove unsourced and dubious claims I am told it's vandalism to not just add a {{fact}} tag and wait years for sources that don't exist. The unenforcability of this policy is ridiculous... it's as if people think sourcing is still optional, and all too often admins claim that the person asking for sources is no more correct than the person wanting to restore unsourced content, it's merely a difference of opinion. --Rividian (talk) 03:01, 16 June 2008 (UTC)

It is enforced. But it is verifiability and not verification that is being enforced. Citations are not required by this policy -- that's a misinterpretation that is very easy to make, and a great many people do. Only the potential of sourcing is required, that is, sources must be out there. If something doesn't have sources out there somewhere, then it isn't verifiable and that item is subject to being deleted. It's a subtle distinction, yes, but very carefully maintained by those who watch over this policy. The wording of the policy is misleading, perhaps intentionally so, and because of this it provides the image that Wikipedia seeks reliability by requiring that everything in it be referenced. But it is just an appearance - because verifiability is much different than actual verification. So people are allowed to jump to this conclusion, as no clarification is provided in the policy. It's a political balancing act, between being largely unreferenced and wanting to be fully referenced, and between verification and the wholesale deletion of a huge proportion of Wikipedia. The reality is that most of Wikipedia is unreferenced. We can't allow that material to be deleted, because that would be too damaging and costly. So, if someone goes on an enforcement rampage to delete every article they come across that lacks sources, they will likely be the focus of an RfC, and shut down. Therefore, the policy is not useless, and serves an extremely important purpose. The Verifiability policy is the fulcrum point of a check and balances system. It is a compromise between deletionism and inclusionism, and is at the very center of Wikipedia's highly political structure. Once you understand this and embrace it, you will have achieved Wiki-Zen.  :) Become one with the wiki. Good luck. Have fun. The Transhumanist 06:51, 18 June 2008 (UTC)

Comments please - SPS rewording

I was under the impression that the accepted/prefered procedure for changing an article (mainspace or wp) is the bold, revert, discuss cycle. I know that is is always good wikiquette to explain all edits and reverts via summarys and talkpages. I made a good faith effort to implement clarifying language for a single point under SPS (this was not an attempt to change the rule but only to give it clearer language), i discussed my proposed changes on the talk page (here), received ZERO disagreement which i interpretted as strong silent consensus, made my edits and used the edit summary to clearly point to the discussion that supported it. I believe i did everything right and did it in the right way only to see my contribution reverted without discussion or explanation (see this diff) so i am respectfully asking now for an explanation and/or if there is any problem with the proposed language change? PS: I am aware there is an ongoing effort to rewrite this entire policy and that may (or may not) result in changes to the existing SPS sections but (a) if the efforts to rewrite fail then this point still needs clarification and (b) the new rewrite if successful might benefit from the proposed language change if not significantly different on this point. 208.43.120.114 (talk) 20:28, 16 June 2008 (UTC)

If the above description of correct editing practices is correct, then you could say policy editing is broken and has been for years. Or you could also say the those practices lead to poor results on policy pages, so they have not followed here for years. There is no silent consensus on policy pages. The talk pages are simply too high volume to have every idea responded to in a meaningful way. The ideas that do not provoke strong agreement or disagreement get little response, but an idea usually needs substantial agreement to be incorperaated into the policy. I would first suggest avoid policy completely. If you wish to disregard that (I cannot manage to follow my own advice there), then try to engage whoever reverts you on their talk page. You also might want to look at who has posted on that section of the policy in past and engage them on their talk pages. But most of all don't edit war. Realize that everyone has a rather low chance of seeing the policy changed to their prefered wording and you are no different.--BirgitteSB 21:13, 16 June 2008 (UTC)

I guess that sounds like pretty good advice BrigitteSB, i will talk with Crum375 on his talkpage. Sadly what you say about policy editing being broken seems true ... if non-controversial minor changes cannot be made without strong support (i think that is what you said, yes?). Such a process prevents careful and incremental improvements and only allows for dramatic and possibly dangerous changes. If silent consensus does not work, what does? If WP:BRD does not work, what does? 208.43.120.114 (talk) 21:51, 16 June 2008 (UTC)

There just aren't that many people with time to keep track of these things. Honestly, you should probably do a RFCpol (scroll down). I don't particularly like adding a large word (germane) which is essentially a synonym to relevant. I don't think it is necessary. Could you explain, succinctly, how your change improves things? ImpIn | (t - c) 22:04, 16 June 2008 (UTC)

Sure, three improvements:

1. The biggest problem with the current wording is the use of the imprecise pronoun "their". Because the case sentence preceding the conditions refers to sources as the 3rd person pronoun "themselves" the subsequent use of another 3rd person pronoun "their" confusingly can cause the sentence to read "Self-published and questionable sources may only be used ... if ... the material used is relevant to [the sources'] notability. Giberish of course but newbies are trying to grasp a ton of other policy text written in wikispeak and so here is an opportunity to eliminate some giberish. By explicitly stating the material must be relevant/germane to the article's notability this pronoun ambiguity is removed.
2. The use of germane is preferable for two reasons. (1) according to websters dictionary "relevant implies a traceable, significant, logical connection ... germane may additionally imply a fitness for or appropriateness to the situation or occasion", and in fact the definition of germane is "being at once relevant and appropriate". The reason for this point in SPS is because without it we would get unrelated and inappropriate material being included in articles.
3. Finaly, i think the word germane, because it is a less common (but not unknown) word, will cause editors to think about its meaning a tiny bit harder. That extra effort may help to reduce inappropriate uses of SPS.

Comments? 208.43.120.114 (talk) 00:36, 17 June 2008 (UTC)

To comment only on #2 and #3: Considering the number of regular editor who do not use English as their first language, I think is a good goal to use terminology which requires no effort for native speakers. In any case definitely debatable whether to use the most easily understandable or most exacting wording in a policy. And therefore is debatable over whether that change makes the policy slightly better or slightly worse.--BirgitteSB 02:32, 17 June 2008 (UTC)

(outdenting) Well, arguably you may be right. :) On the idea of understandable versus exacting where do you put notability? :) OK, so germane is a hard sell, i accept that reluctantly for the moment, what about just changing the sentence ending to eliminate the pronoun problem? 208.43.120.114 (talk) 09:36, 18 June 2008 (UTC)

Sure, fix the pronoun problem. ImpIn | (t - c) 09:46, 18 June 2008 (UTC)

WP:SELFQUEST

A living person who has an article in Wikipedia, is the founder of a 501(c)(3) charitable organization that carries his name. In the official website of this organization there is a page with biographical information, in which they list a series of Freedom of the City (a.k.a "Keys to the city") received by this individual. Would mentioning these fall under the restrictions of WP:SELFQUEST if properly attributed to the official website? ≈ jossi ≈ (talk) 16:03, 18 June 2008 (UTC)

Note further that the peson, Prem Rawat, does not hold any position in the foundation, and that the material on the foundation website is unsigned with no known author. WP:SELFQUEST requires that the author be known. Also, the assertions in the foundation page are self-serving. There are no 3rd-party sources to confirm the awards and honors claimed there. ·:· Will Beback ·:· 16:39, 18 June 2008 (UTC)

(I was hoping to have a generic discussion, but oh well. The foundation webpage discussed is Prem-Rawat.org). The argument by Will above is pretty strange, to say the least. Applying that argument would mean we cannot use material from www.apple.com about the number of iPhone units sold, because there we do not know who the author of that text is. I do not see how a Freedom of the City award can be construed as "self-serving". Self-serving is another thing altogether: for example, saying that "Jeff Beck is the most accomplished guitar player in the world" in the article about Beck sourced from his website. ≈ jossi ≈ (talk) 16:51, 18 June 2008 (UTC)

The number of iPods sold is reported in the business media. These assertions are reported solely by this website. The subject has a website of his own, this website belongs to an independent foundation where he has no official role. Claiming to have received dozens of awards and honors from around the world is self-serving, in my opinion, unlike simple biographical data like place of birth or education. Let's see what uninvolved editors think. ·:· Will Beback ·:· 17:02, 18 June 2008 (UTC)

Yes, let's hear from others, just note that he is the founder of this organization as per the page linked above. ≈ jossi ≈ (talk) 17:38, 18 June 2008 (UTC)

My reaction in real life is that I would prefer an independent source. This is exactly the sort of thing that is prone to puffery. Mangoe (talk) 18:03, 18 June 2008 (UTC)

How can this be "puffery"? We are just reporting a fact [3] ≈ jossi ≈ (talk) 19:59, 18 June 2008 (UTC)

I think that is the issue here... is it fact? That the website claims that Prem Rawat was given all of these "Freedom of the City" awards is a fact... but do we know if this is actually the case? I think that if there is any doubt, we should get a source that is completely independant of the subject to verify it. As to the Puffery issue... what does this award actually mean? Is it significant? ... I know that New York City hands out "Keys to the City" like candy. It really does not mean all that much. The same may not be true for other cities. Context folks.... we need more context. Blueboar (talk) 21:47, 18 June 2008 (UTC)

I do not think that there is any evidence that these keys to cities have not been awarded, or any reason to doubt it. This is information that this foundation has made available via official press releases, such as [4], [5], for example. ≈ jossi ≈ (talk) 00:55, 19 June 2008 (UTC)

The standard for inclusion in Wikipedia is a reliable source. Not a poor source that no one has disputed. Press releases aren't known for their objectivity. ·:· Will Beback ·:· 02:10, 19 June 2008 (UTC)

Wikipedia:Usage of diacritics

See the new proposal Wikipedia:Usage of diacritics.

"For a placename or person that is well known in the English-speaking world, i.e. is widely mentioned in English-language sources: ... " and then goes on to lay down rules of usage rather than relying on usage from reliable English language sources. This seems to contradict this policy. -- Philip Baird Shearer (talk) 19:05, 19 June 2008 (UTC)

I am not sure I understand the problem. Blueboar (talk) 19:11, 19 June 2008 (UTC)

Direct Quotes

With regard to this recently added note 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."

If the cited text is some five or ten pages is one obliged to reproduce all of it? Its just that some editors are now using this as an edit war tactic: requesting lengthy quotations with each citation given. Is one in contempt of this rule if one refuses to reproduce pages and pages of copyright material on demand? Colin4C (talk) 20:25, 19 June 2008 (UTC)

I’m sorry, but it dose not take 5 pages to quote the passage that sources “infamous,” and “Emigration reached new heights.” You were not asked to produce 5 pages. --Domer48 (talk) 20:35, 19 June 2008 (UTC)

I'll ask again here why it took you many edits from 13:36 to 19:46 on Talk:The Great Hunger to clarify that you weren't actually looking for 5 pages of a quote and were instead only looking for 1 word. Or, after the edit, one word and one phrase. You did, however, manage to quote the policy you yourself had edited in here to justify your request. Bastun^{BaStun not BaTsun} 20:38, 19 June 2008 (UTC)

Actually it was there before I edited this policy, but don't let the truth get in the way of your trolling. Domer48 (talk) 20:44, 19 June 2008 (UTC)

The previous policy was different. Viz:

"If a reliable source is not readily available (e.g., freely accessible online), the editor inserting or restoring the challenged material should be able to provide page numbers and/or direct quotes from the original text supporting the material, upon request."

The current policy (if that's what it is) has absurd consequences if it requires you to reproduce many pages of copyright text. There seems to be some fallacy involved as well which presumes that citations only refer to a single page rather than 5, 10, 15, 20 or whatever. In the latter case an assumed obligation to reproduce reams of text on demand is a boon gift for any editor wishing to game the system. If a single sentence in an article summarises 20 pages of text in the source are we obliged to type out those twenty pages on demand? Colin4C (talk) 21:48, 19 June 2008 (UTC)

5 pages to quote the passage that sources “infamous,” and “Emigration reached new heights.” ?--Domer48 (talk) 22:01, 19 June 2008 (UTC)

• I glanced at this and I strongly suggest that you all work with the Arbcom appointed mentors to come to a solution. Forum shopping issues from an article that has previously been through Arbitration is never a good idea.--BirgitteSB 22:40, 19 June 2008 (UTC)

Maybe we should add "subject to copyright laws" to

"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."

It can't be right that the wikipedia obliges us to break the law. Colin4C (talk) 09:31, 20 June 2008 (UTC)

Fair use is fair use and legal under US law. Wikilawyering may also be legal under US law, but it is not fair or appropriate; especially when done on a talk page unrelated to the page where the dispute originated, and should be stopped. Brimba (talk) 13:06, 20 June 2008 (UTC)

It is also worth noting that a very quick check of the talk page of the article where the dispute actually resides shows that more then once it has been pointed out that there is no direct obligation to provide the requested material, is simply an expected courtesy to other editors (says so right there in the text). Unless you feel the editor is acting in bad faith, something should be presented to substantiate the material, although it may not encompass all that has been requested. Brimba (talk) 15:04, 20 June 2008 (UTC)

If an editor who keeps on requesting you to provide lengthy quotations from books calls you a 'liar' and a 'dick' on the Talk page in question is that evidence of bad faith? Colin4C (talk) 16:06, 20 June 2008 (UTC)

When an editor has recently been shown to have misrepresented sources not once but twice it is more than reasonable that they provide direct quotes for other sentences and/or passages of disputed sourcing. Domer48 (talk) 16:35, 20 June 2008 (UTC)

one think at least is clear--people who violate WP:NPA are in danger of being blocked for that, regardless of the merit of their case. The precondition for working this out is an agreement to use civilized discourse, & anyone who wants to work here at Wikipedia must accept the obligation to limit themselves to that. DGG (talk) 18:07, 20 June 2008 (UTC)

Burlington Resources is America's largest Natural Gas company rather than having been replaced by Google in the S&P 500 ( "an oil company" ) and is owned by Conoco. Pc32 (talk) 20:02, 22 June 2008 (UTC)pc32

Rules, rules, rules ....

I see that Wikipedia already has more than 200 pages of rules and instructions explaining what is wrong or right. Also there are the rules 'supplemnets' - "Single purpose account for example" - written in the essay form. Could you people stop inventing and writing rules? The whole discussion above and proposal of the new rule is simply meaningless. English over non-English sources? Wikipedia is all about knowledge and resources shall be regarded only as good or bad, existing or non existing.--72.75.24.245 (talk) 03:36, 24 June 2008 (UTC)

Welcome to life. Life has rules. Rule #1: play nicely.--Srleffler (talk) 05:25, 24 June 2008 (UTC)

And, of course, remember this rule: Wikipedia:Ignore all rules --Lukobe (talk) 06:35, 24 June 2008 (UTC)

Um, you applying that to sockpuppets and banned users also (the IP editor seems to be Velebit aka Stagalj aka Standshown aka Pederkovic Ante aka Guivon aka J.A.Comment aka..whoever that is). Doug Weller (talk) 21:15, 24 June 2008 (UTC)

When are citations not necessary?

I posted a comment here, asking for inline citations to offences that soccer players (who are still living persons) may have committed. in response, users have said that citations are not necessary because the information "is freely available from numerous sources". If this is true, then it comes as a surprise to me, as I assumed that citations are required for all content in article space. Thoughts?Bless sins (talk) 14:20, 26 June 2008 (UTC)

Not all. See WP:WHEN; we don't need to source common knowledge. It doesn't seem like this is the sort of thing at issue here, but that is a question of fact; saying that "some soccer players have broken the rules of soccer" (at some time in the last century) would be stating the obvious. Septentrionalis PMAnderson 20:12, 26 June 2008 (UTC)

Non-English sources

As more and more articles are relying on non-English sources, and the policy seems widely misunderstood, I suggest that the text be amended for clarity and brevity. This arises out of a discussion at WT:MHCOORD#Non-English sources.

Unless anyone objects, I propose replacing the existing text with the revised text in a few days time. Thoughts? --ROGER DAVIES ^talk 04:14, 23 June 2008 (UTC)

Disclosure: I've left messages on the talk pages of all editors participating in previous discussions here during the last six months. --ROGER DAVIES ^talk 04:46, 23 June 2008 (UTC)

• I don't see a compelling improvement, and the line "Only reliable sources should be used" implies that non-reliable sources can be used if they are English-language. This is not something we want to encourage. If you do change it, please reword this line to something like "*Extreme care should be taken to only use reliable sources," which does not imply that unreliable sources are acceptable in English, only that some level of care less than "extreme" is used to keep out English-language non-reliable sources. davidwr/(talk)/(contribs)/(e-mail) 04:36, 23 June 2008 (UTC)

I've tried to make the text more direct and provide a checklist of requirements. Good point about "extreme care", --ROGER DAVIES ^talk 04:46, 23 June 2008 (UTC)

• The reliable sources line is instruction creep, even more so on this page which is about reliable sources; it should be struck entirely. Also, I don't agree that the line, "Translations published by reliable sources are preferred over translations made by Wikipedia editors" should be eliminated; that introduces too much chance of original research by Wiki editors, of the type I've had to deal with many times. Translations from reliable sources are preferred. SandyGeorgia (Talk) 05:18, 23 June 2008 (UTC)

I support the change that says explicit translated quotes come into play only when an issue is actually "subsequently challenged", rather than being obligatory from the start. I've always found that requirement unrealistic and an unnecessary burden, and not really matching consensus about actual good practice in the areas I work in (where non-English sources are extremely common). Other than that, I don't see a substantial improvement in the new text but nothing I'd seriously oppose either. Fut.Perf. ☼ 05:26, 23 June 2008 (UTC)

I support as flexible approach to foreign language sources as possible. If everything in the encyclopedia had to be verifiable by every editor we would be dumbing down. As a non-scientist I cannot see the difference between complex science articles which need a basic knowledge of the subject to understand the sources, and sources in a foreign language, I certainly think we should change things top make it easier to use foreign language sources without being challenged by someone merely because they do not understand that language or think, as I have seen time and again, that only English language sources should be used. I think foreign language sources should be dealt with in exactly the same way as English language sources. Thanks, SqueakBox 05:36, 23 June 2008 (UTC)

Agreed in principle. There are many cases where a direct translated quotation doesn't even help substantially. Many statements are based on a summarising reading of large passages of text that can't be pinned down to a single quotable sentence or paragraph. If you don't trust me summarising a non-English source correctly, why would you trust me translating (and possibly ripping out of context) some passage to back it up? And on the whole, it is true that verifiability must be made possible, but that doesn't mean it must be effortless or possible for anyone. If I work from a rare book, it's only verifiable for someone who can afford traveling to a library. If I work from a highly specialised academic text, it's only verifiable for someone who has the relevant education. If I work from a non-English text, it's only verifiable for someone who has a friend who speaks that language. All the same thing. Fut.Perf. ☼ 05:53, 23 June 2008 (UTC)

You have done more than merely amend the text for "clarity and brevity". You should have been clearer about what you are trying to achieve, and why. I don't mind the reduction of the burden of providing quotes, but I do have some other concerns:

1. The first sentence is a run-on sentence and should be broken up. (Yes the old policy also starts with a run-on sentence. Two wrongs don't make a right.)
2. You have (inadvertently?) tightened the requirement for when a non-English source can be used, from when no English source is available to when no English source exists. This is a significant tightening of the rule, and I oppose it entirely.
3. I disagree with the removal of the prescription that "Translations published by reliable sources are preferred over translations made by Wikipedia editors." Translation by WP editors is dangerously close to original research, especially when material is being "directly" quoted, or when it is controversial.
4. You have removed the ability to directly quote a foreign source in the original language without providing a translation. This is rarely useful, but I don't think policy should forbid it.

--Srleffler (talk) 05:40, 23 June 2008 (UTC)

• No need to rush into it. Give the discussion some time. Mostly, I agree with davidwr's statement. I do like the "of equal quality" concept, but I'm unhappy with the inadequate way we handle citations and the reliability of sources when articles are translated from other wikis. Other languages on WP often take different approaches, such as a simple list of refs for the whole article. We need a better way...LeadSongDog (talk) 05:48, 23 June 2008 (UTC)

• The revised text is much improved. I still think "Care" should be "Extreme care." It would also help to either say or link to a statement that says that if an unfamiliar non-English source is used, the author should either link to an English-language Wikipedia page or reliable English-language web site that makes it clear that it is a reliable source, or should make the claim himself on the article talk page. davidwr/(talk)/(contribs)/(e-mail) 05:52, 23 June 2008 (UTC)

I would say only the last item is needed. Foreign language sources (for instance - in English on :fr ... ;o) correspond to two different problems with respect to verifiability: (1) find the book, and (2) check the translation. There is nothing specific in "finding the book", the general comments on verifiability apply to that case as well. In a way, if the source can be easily found in a broad choice of places, it is close to be common knowledge for which a specific source is not required in the first place, so why bother to use a foreign one (unless the article is translated, of course)? When there is no choice, use whatever source you can get, when there is a choice, use the most reliable and easily found, and in all cases, use common sense... The only specific problem is the translation, which can be biased and must be checked from this point of view. Indeed, if this verification is needed, the original quote must be somewhere in the article, so that any reader fluent in the corresponding language can verify it, without having to refer to the original source. —Preceding unsigned comment added by Michelet (talk • contribs) 06:38, 23 June 2008 (UTC)

• I largely agree with Roger's proposed rewarding, noting however that unless one knows the language, the source will always be unreliable. On the other hand, if the article can not be constructed without even one major English source for the subject, is it notable to the English user? It seems to me that non-English sources can only be used as supporting sources, and not as primary sources for the Wikipedia entry. English Wikipedia is after all meant to be used largely by the English speakers. It would be far better if editors used one solid English source to base their article on, and used other non-English sources as supporting sources, than to try and piece it together from a multitude of relatively unavailable works.--mrg3105 (comms) ♠♥♦♣ 06:48, 23 June 2008 (UTC)

• Some of the statements in this discussion are alarming because they misinterpret the very policy they are about: a Wiki article can't be used as an indication of reliability because Wiki articles aren't reliable, and nothing about an article being in another language makes it non-reliable. (Without mentioning the issue of misunderstanding of the notability guideline.) Re-iterating, I don't support repeating wording that non-English sources should be reliable on a policy page that deals with reliability of sources (all sources should be reliable, and the way we measure that is the same in any language, and has to do with fact-checking and editorial oversight), and I don't support deletion of the wording that would encourage editor translations over reliable source translations. SandyGeorgia (Talk) 06:59, 23 June 2008 (UTC)

I suggest a slight change because often there is more than one translation possible. This is especially true of old sources. We should make it policy that all translations in reliable sources available to the editors are included.

Regarding material not sufficiently covered in English, one has to keep in mind that this encyclopedia is not limited to the Anglosphere. Wandalstouring (talk) 07:43, 23 June 2008 (UTC)

I am sure you don't mean what you said about including all available reliable translations. If it's a quotation from an ancient Greek philosopher there may easily be dozens of slightly different translations, including some obviously incorrect ones in otherwise reliable sources. Your proposal would force us to include all the major variants, including any non-notable wrong ones, even in situations where the differences don't even matter and no sane person would challenge anything. --Hans Adler (talk) 09:51, 23 June 2008 (UTC)

• Thanks for heading up but sorry, I disagree, so oppose. I, living in a non-English country, may know an English translated or original version exist of the resource I may rely. But it doesn't mean it is easily available for me. It sounds just for me "if you are non English and not living in English speaking country, go away". It may reduce the usability and reliability of the Enlgish Wikipedia in entire, despite of the intention of the proposer, I'm afraid. --Aphaia (talk) 07:44, 23 June 2008 (UTC)
• As the proposal to cite "reliable English-language web site that makes it clear that it is a reliable source", it is not practical, since not in every field English has the reliable source. They just could have not been introduced properly into the anglophone world, particularly the matter in concern is not popular in English speaking countries. It is not a way of encyclopedia, but rather a "reference for dummies". in my opinion.--Aphaia (talk) 07:50, 23 June 2008 (UTC)

• Comment: I agree with the comments above. I currently live in Spain and it's almost impossible for me to gain English source material in a Spanish-speaking country. I am limited to what I brought over from the United States and the little that is sold in Spain in English. That said, sometimes Spanish sources are much more reliable than their English counterparts, especially when in regards to specific topics. For example - the Spanish Civil War. But, I would say that this isn't only true for Spanish-language books, but for every language. By suggesting that an English source should be used over a foreign-language source is to effectively shut out a large portion of Wikipedia's editors. In any case, I'd like to know what exactly makes an English source more verifiable than a Spanish source? Who, really, will buy either book to verify that the facts are true? Why should we assume that there are no Spanish-speakers on Wikipedia? My point is that there is as much of a chance that another editor will be able to verify the Spanish source, than one would be able to verify the English source. I guess I'm talking more about a specific range of articles, where book sources are the only plausible sources (such as military history), otherwise I'd agree that an English media source (if available) would be probably be better. JonCatalán (talk) 09:55, 23 June 2008 (UTC)
  • What makes an English source more verifiable than a Spanish one? That more of us can verify it. You don't need to buy a book to check it; many books are available, in whole or in part, on the Web, and almost all books are available through Interlibrary Loan. Septentrionalis PMAnderson 13:52, 23 June 2008 (UTC)
    • Okay, so are you suggesting that to pass FAC we should have people look at the books being used? Because, that's the only way an English source is truly more veriefiable. 99.9% of the time the sources will not be checked by hand, and so in the end a Spanish source is just as reliable as an English source. JonCatalán (talk) 14:58, 26 June 2008 (UTC)
      • That would be an improvement over many present FA comments (Looks good to me/ Oppose, it has normal spaces in the footnotes). But the point is not our procedures; it is that reader should be able to verify, if he wishes to. Septentrionalis PMAnderson 20:23, 26 June 2008 (UTC)

• Comment If a foreign laguage source is being used, I think it preferable to reference it untranslated unless a quote is eplcitly needed. In working on Troilus I found extracts from both the Iliad and the Aeneid whose meanings were disputed. There was also an issue with verse translation tending to be freer than some (but not all) prose translations. Quoting translations, as suggested by some editors, is therefore often a means of introducing unreliabilty and a very literal and unidiomatic transation by a Wikipedian could well be more reliable than a published translation intnded o be easu n theeye or ear. There is also an issue with copyright So early twentieth century works culd involve the foreign language source being out of copyright but the best English translations might be in copyright. This might be particularly problemtic to poetry where a quotation might constitute a significant proprotion of the text.--Peter cohen (talk) 10:53, 23 June 2008 (UTC)
• Comment Agree largely with srleffler's suggested new wording and with everything said above except for Wandalstouring and mrg3105. Are we going to say "Care should be taken", or "Extreme care", or leave it out because it's redundant with other things on this page? (I'd suggest "Care", and leave it to guideline pages to work out over time what that means.) How do we handle the editor who translates from a language no one else on Wikipedia knows, who points out that our new policy doesn't allow us to reject his reliable source, because there's no alternative? I've heard people in the German and French irc channels say that they have to deal with the issues around borrowing from English all the time; perhaps we should ask them, because apparently, we don't struggle with this issue that often (Sandy could only remember one case at WP:FAC, which she mentions at WT:MHCOORD#Non-English sources). I like Sandy's point that some things matter more than others (direct quotes, BLP issues), and I think that's probably important enough to make it into this text, even though of course we talk about the same issues as pertaining to English here and at WP:BLP. - Dan Dank55 (talk)(mistakes) 13:05, 23 June 2008 (UTC)

A few points. Available to whom? Surely each editor can only put in citations available to them. Other editors may find better ones. Sometimes it makes sense to give more than one.

I'm not sure what mrg means by primary & supporting sources. It simply isn't always practical to follow the approach suggested. Let me illustrate with my work on Pali Canon. There can be no question of notability here: this is the scripture of 100,000,000 people. Yet I've had to draw citations from all sorts of sources. One was a German book, untranslated as far as I know. Another illustrates a point not mentioned in the page or discussion I think: a case where the foreign original is likely to be more readily available to many people than the English translation. This is because the original paper was published in a leading German journal, which would be available in most major university libraries, while the translation appeared in an obscure Indian publication, which is likely to be less readily available in most of the English-speaking world (except maybe India). I put both in, English 1st. Peter jackson (talk) 15:35, 23 June 2008 (UTC)

Available means available to the editor adding the material, of course. If you have a book in German, use it. You might want to do a Google Books search though to see if the relevant section is available online in an English translation. If your source is a German website, definitely do a web search to see if comparable information is available elsewhere in English. If you use the German reference, another editor may come along and replace it with a reference to an English translation, or perhaps to some other work in English that provides the same information.--Srleffler (talk) 05:15, 24 June 2008 (UTC)

I agree with User:Michelet:

• The first two caveats are redundant as they apply equally to English-language sources and are expressed in the main body of our policies on sourcing already.
• The last sentence of the third caveat can be shortened to "Translations published by reliable sources are preferred".

Furthermore in the main body, "This allows other readers and editors to more easily verify that the source material has been used correctly." is imprecise - most readers may be able to more easily verify the English language version, but not necessarily all (due to geographical factors, language knowledge, etc.), so who are "other" editors in the sense of the original text? What is actually meant is that all English-speaking people who read the article will not have any additional barriers in checking the source due to language; the most precise wording would say something to that effect. Knepflerle (talk) 18:14, 23 June 2008 (UTC)

Surely to avoid "additional barriers in checking the source" we would simply choose to both quote and cite the original language material and the translation(s) used. If no published translation is available, I see no great problem in offering one, so long as it is clearly identified as "loosely translated". Side by side with the original, translation errors will be corrected soon enough if the article is truly notable. WPBabel can assist in identifying suitably-equipped editors to make, vet and repair translations. This is one way we can make significant inroads in reducing systemic bias against non-anglosphere content.LeadSongDog (talk) 19:13, 23 June 2008 (UTC)

Knepflerle, this is the English Wikipedia. I think it's fair to assume that the vast majority of readers of this Wikipedia can more easily deal with an English-language source than one in another language. I don't see any need for the more convoluted text you propose.--Srleffler (talk) 05:15, 24 June 2008 (UTC)

I did not propose a wording. You may feel any tightening of the wording is unnecessary, but if other people do they are free to suggest a wording which is hopefully convolution-free. Knepflerle (talk) 08:45, 24 June 2008 (UTC)

I am not sure how many respondents here are or have been professional translators, but I am one. There is a basic issue here that no one seems to have addressed, which is that each word in a given language can cover a multitude of meanings, and more importantly, feelings. It is rare for any word or phrase in the target language to correspond exactly, in every way, with an original word or phrase. Even if the basic meaning is approximately correct, the associated feelings can be 180 degrees away, and the correct translation might not just be hard to find, it may, probably will, be non-existent. I have seen on several articles recently that editors have attempted to support their biases by introducing terms that hit the reader in the face with their emotional impact, claiming (rightly) that they are giving the "correct" dictionary translation of the source's words. In my opinion, translated sources should be avoided almost entirely. I would support a change to this section, but not starting, "As this is an English encyclopedia..." but rather something like, "Bearing in mind the great difficulty in accurately rendering controversial ideas from one language to another..." Rumiton (talk) 15:05, 24 June 2008 (UTC)

Thanks for a thoughtful entry, Rumiton. I've dabbled in translating enough to know I don't want to try to do it for a living :-)

First, a question. Why exclusively controversial? That is the most delicate, but certainly not the only situation in which one might want/need to use a non-English source. One might want to reference a news article or a scientific paper, for example.

Second, people are kidding themselves if they believe that an English language text prepared by groups of people, scientific researchers for example, who don't use written English much and perhaps don't speak it at all, is going to impart exactly what they mean to say. I've many times smiled to myself reading things that one really has to know both languages to even understand. Probably everyone has done so, re the directions which come with some imported products. Sometimes so-called English texts need "translation", too.

So, controversial subjects are one thing, but citing sources in foreign languages doesn't always need to be problematic. IMO. --Hordaland (talk) 19:53, 24 June 2008 (UTC)

I am not sure of your point. You seem to be accurately illustrating the difficulties in all translational work. If technical people and other non-linguists knew how bad their own earnestly made translations often are, they would be horrified. And I agree, we could leave out the word controversial. Rumiton (talk) 16:26, 25 June 2008 (UTC)

Yes, let's discourage extremist readings in non-controversial situations as much as possible. I have just witnessed the deletion of an article on a notable subject as unsourced, when nobody actually challenged the content. I see a certain danger that, e.g., an article about a fifteenth century French count will be deleted by an admin not interested in the subject, based on the rationale that the article is sourced only to a French book, while a Google search shows that twenty years earlier an English book about the same subject had appeared. In general expert editors should have the option to discard all available English sources as unreliable. It makes no sense to restrict this before someone actually challenges the neutrality of an article. After all, the vast majority of our articles are not controversial, and we have no interest to make them so. --Hans Adler (talk) 20:28, 24 June 2008 (UTC)

But in controversial article -- eg the New Chronology (Fomenko-Nosovsky) article alleging no human history before 800 AD, a number of references are entirely in Russian - despite the existence of English translations of his books. I have no way of knowing whether his claims are being reported correctly, what the context is where the claim is something written in such a way as it is to vague to check without further information, etc. I am seriously tempted to remove all the Russian language references (and whatever is referenced) and ask other editors if they want to replace them to replace them with references I can verify somehow (and in this case the books are searchable on Amazon, which is great). That would be drastic I guess, but tempting. Doug Weller (talk) 21:20, 24 June 2008 (UTC)

I agree that that's a good example for a controversial topic. Did you try asking for help from uninvolved Russian-speaking editors? I imagine we have plenty of them. --Hans Adler (talk) 21:31, 24 June 2008 (UTC)

I appreciate what Rumiton is saying, which is why I suggested we talk with people who have to deal with this issue all the time, such as people who regularly import content from the English Wikipedia into their own languages. I don't think we need to discard all non-English sources. As Sandy pointed out, some things are more important than others; I would prefer not to rely on non-English sources for BLP material or direct quotes. As Rumiton points out, we also don't want to rely on non-English sources for contentious or controversial material, since it's very difficult to verify that we're being precise and nuanced. (Hell, that's difficult enough to verify in English!) - Dan Dank55 (talk)(mistakes) 16:30, 25 June 2008 (UTC)

There is one amendment we really need. As stated, it says that any quotation from a source in a foreign language (even the Bible, or Tacitus) must also quote the original, even when the quote is from a published, scholarly, translation. Surely that's not what we mean? Is citing the translation quoted enough? (Is it necessary?) Septentrionalis PMAnderson 20:17, 26 June 2008 (UTC)

I think your specific examples, where the original version doesn't even exist any more, wouldn't be a problem in practice. But the two proposed revisions above would probably lead to a lot of activity related to chasing down the original text in other cases, because read out of context they can (and will, by some people) be interpreted as saying: Whenever you quote a translation anywhere, even a published one, you have to quote the original as well. And even in the case for which it is intended, i.e. quotations that we first found in a non-English source, it will be completely unnecessary in many cases.

The word "original language" is problematic here. Suppose I source an article on a Sumeric king to a French book. Suppose this book quotes the king in (cuneiform) Sumeric and gives a French translation. Is Sumeric or French the "original language"? We shouldn't have a policy that forces me to write cuneiform in order to quote the translation. Then I source an article on Giacomo Casanova, who wrote in French, to an Italian book. Is French or Italian the "original language"? We shouldn't have a policy that forces me to quote the Italian translation; obviously it would be much better to quote exactly what Casanova wrote.--Hans Adler (talk) 20:58, 26 June 2008 (UTC)

Huh? Of course the Latin text of Tacitus and the Greek and Hebrew text of the Bible exist. But this is a matter of detail. Septentrionalis PMAnderson 01:09, 27 June 2008 (UTC)

Sorry for the blunder. I thought Tacitus was a back-translation from Arabic, but I was wrong. And I don't know what I was thinking about the bible (perhaps something wrong about Aramaic.) --Hans Adler (talk) 08:19, 27 June 2008 (UTC)

Don't we really want something like the following? 1) If there is an adequate published English translation, use it. 2) If there is no published English translation, make your own, but quote the basis of your translation as well. 3) If published English translations exist but are not adequate for the article (e.g. wrong), make your own, quote the basis of your translation, and explain the problem. --Hans Adler (talk) 20:58, 26 June 2008 (UTC)

That's a reasonable approach, although again we may want to say available. Some texts were last translated in the sixteenth century, and other translations only exist in University libraries. Septentrionalis PMAnderson 01:09, 27 June 2008 (UTC)

Regarding your point above, Sept, I think we're talking about sources whose translated versions are not reliable sources in the same way that the original versions are. We wouldn't have to quote the Aramaic from the Gospels, since the English-language versions are reliable sources in their own right. (I'm sure some might claim that the translation can't be as good as the original since the original was the Word of God, but Wikipedia's standards are a little lower.) - Dan Dank55 (talk)(mistakes) 15:17, 28 June 2008 (UTC)

Blogs

I just wanted to get some opinion on blogs. Surely we can take some blogs as RS? There are many which have build up a reputation for reliable reporting. Secondly, could it not be said that a blog reproducing information from RS which is not otherwise available on-line is acceptable?Traditional unionist (talk) 15:09, 27 June 2008 (UTC)

In my opinion (and that's what you asked for :-)), some blogs are absolutely respectable, reputable, reliable and trustworthy. I'd include most or all of the bloggers at ScienceBlogs, for example. That said, we'll never have a list of which blogs are RS, and the question will have to be hashed out over and over again, topic by topic. I don't see any other way, do you? --Hordaland (talk) 20:20, 27 June 2008 (UTC)

No, but I don't see anything wrong with that. Clearly some blogs are not intented to be impartial, but then again I've seen books that describe themselves as biased on the topic at hand being cited on WP in support of controversial additions. Slugger O'Toole in Northern Ireland strikes me as another example of a mostly RS source on many things. But shouldn't that be listed on the relevant wikiproject as acceptable RS?Traditional unionist (talk) 23:14, 27 June 2008 (UTC)

Impartial and (factually) reliable are two different things. Regards, HaeB (talk) 03:47, 29 June 2008 (UTC)

Not the first time this is being brought up... The fundamental problem with blogs is that they are basically designed to make publication as quick and easy as possible, which in most cases means without (mandatory) review by an editor. In other words, self-published. A blog posting by an author will always be considered less reliable than a peer-reviewed article by the same author in a reputed scientific journal. That being said, there are certainly many blogs where the reputation of the author or the blog itself (as evidenced by quotes from the blog in reliable sources, for example) makes them appropriate as sources in the right context (other examples: Freedom to tinker, Google Blogoscoped).

I think Jimbo Wales described it rather well in this interview:

... a too ruthless consistency really tries to chop through the messiness of the world in a way that may not make a lot of sense and leads to you know, sort of strange conclusions. You know, an example of overly, an overly broad principle would be you know, a very simple statement that blogs are not valid sources.

Well, at first glance a lot of people would say that sounds about right. I mean anybody can write a blog and post anything they want -- is that really a valid source? But then again once you start getting into the nuisances of it, you say well, John Edwards is blogging. He's a famous politician and if he writes something about his own views on his own blog, it sure sounds like a pretty good source for what his own views are.

And, you know, on the other hand a random blog on MySpace offering some opinion doesn't seem like much of a source for anything. So when you have these attempts to write down the principles in an overly broad way it doesn't really work.

(It goes on a bit about the problem of context in determining what original research and reliable sources are, worth a read.)

Regards, HaeB (talk) 03:47, 29 June 2008 (UTC)

Removing challenged content

Removing challenged content is currently a very frustrating process, as fans of that content will often re-add it without sources because this policy says:

Any material lacking a reliable source may be removed, but editors may object if you remove material without giving them sufficient time to provide references.

"sufficient time" basically means forever, in my experience. This is a sad state of affairs... material an editor deems incorrect or unsourcable should be removed... the burden should be on people wanting to restore the content to find sources, there's no right to include challenged, unreferenced content. This language needs to be corrected to point out that immediate removal might rub people the wrong way, and you can opt to be nice and add {{fact}} tags, but that you should always be able to remove challenged and unsourced content, rather than just have to leave it in the article forever but with a tag.

Allowing material that doesn't seem sourcable to stay in perpetually is really a terrible idea... it lets people create utterly unencyclopedic content so long as they don't mind a [citation needed] tag. --Rividian (talk) 02:15, 29 June 2008 (UTC)

You already got a response to this ≈ jossi ≈ (talk) 02:22, 29 June 2008 (UTC)

Which basically told me to leave the unsourced, challenged content in so I can "achieve Wiki-Zen". I considered that response utterly useless and incoherent. --Rividian (talk) 02:26, 29 June 2008 (UTC)

See also Wikipedia:Eventualism ≈ jossi ≈ (talk) 02:23, 29 June 2008 (UTC)

And BTW, you can always try and find sources for material, rather than deleting it or even adding a {{tag}}. And if the material remains unsourced, you can always delete it. ≈ jossi ≈ (talk) 02:25, 29 June 2008 (UTC)

I always do try to find sources... when I can't, I remove that content. I find that if the article has owners, they just add the content right back when they think I've gone away... and often as not an admin sides with them since it's only a content dispute, nevermind that the burden is supposed to be clearly on them. If I can "always delete it" then why do you oppose clarifying the wording? It apparently isn't clear right now. --Rividian (talk) 02:27, 29 June 2008 (UTC)

It is bad practice to try to "fix policy" because you had some encounters with people that keep adding unsourced materials. DO you think that having that wording will make them desist? Wishful thinking. Better, would be that you try and educate these users on why do we need sources, and if they are unable/unwilling to understand this basic principle, well ... time to post a report at WP:AN/I about these users' behavior. ≈ jossi ≈ (talk) 02:31, 29 June 2008 (UTC)

The language is still unclear... people do get the idea that it's rude to remove incorrect unsourced information, and correct policy is to just leave it in the article for several years with a {{fact}} tag. So I do believe the policy needs clarification. And it's admins that get this policy wrong too. At any rate, you're right, I should jump through the hoops and go to ANI when I encounter article owners who add unsourced content... no amount of policy wording will probably stop some of them. That doesn't mean the policy doesn't need tweaking though. --Rividian (talk) 02:33, 29 June 2008 (UTC)

people do get the idea that it's rude to remove incorrect unsourced information, and correct policy is to just leave it in the article for several years with a tag. Who? Where? What? There is nothing in this policy that support what these unnamed people say. ≈ jossi ≈ (talk) 02:41, 29 June 2008 (UTC)

"editors may object if you remove material without giving them sufficient time to provide references." They certainly do object, and prefer you just add a fact tag and wait and wait, even if you seriously doubt those sources exist. --Rividian (talk) 02:44, 29 June 2008 (UTC)

Independent third-party sources

Why is the word "independent" completely missing from this page, in particular, from WP:SPS? SandyGeorgia (Talk) 16:58, 24 June 2008 (UTC)

I take "third-party" to mean a disinterested party, and past discussions have suggested that it's easier to define what that means than "independent", but "independent" is certainly most of the sense of what we're going for here, and I personally wouldn't mind adding the word in front of "third-party". - Dan Dank55 (talk)(mistakes) 15:59, 25 June 2008 (UTC)

Ah, I see; third-party is supposed to cover independent? The only place I could find a discussion of independent is at Wikipedia:Independent sources. SandyGeorgia (Talk) 16:02, 25 June 2008 (UTC)

• I do not understand it. If I am writing a book about, say, the enzymes - am I a third party or the second or the first one? Also, if a book of the first-hand testimonies about the holocaust era is written - shall we disqualify this book for the sake of coming from an interested-in party???--72.75.24.245 (talk) 19:40, 28 June 2008 (UTC)

If you are using your own data about enzymes, you not third-party. You're also in a wierd region between secondary and primary sources. If you're analyzing other people's data, you're a third party, and a secondary source. If you're compiling such analyses, you're third party and tertiary. Someguy1221 (talk) 03:28, 1 July 2008 (UTC)

It seems to me that every use of "third-party" in this and other policies and guidelines needs to be critically reviewed. In every case the question should be asked: "Who are the first and second parties?"

• From WP:PROVEIT

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

  Who or what would be the first and second parties here? What would be the distinction? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)
• From WP:SOURCES

  Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy.

  The first party presumably would be the author, but who is the second party? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)

I am thinking that perhaps "third-party" should be changed to "secondary" in the above two instances. The phrase "reliable secondary sources" is already found in WP:RS. PSWG1920 (talk) 02:56, 1 July 2008 (UTC)

• From WP:QS

  Articles about such sources should not repeat any contentious claims the source has made about third parties, unless those claims have also been published by reliable sources.

  Again, who would be the second party? Maybe "third parties" could be changed to "living persons"? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)
• From WP:SPS

  Self-published material may, in some circumstances, be acceptable when produced by an established expert on the topic of the article whose work in the relevant field has previously been published by reliable third-party publications.

  So the first party would be the author, but again, who or what would be the second? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)
• Further from WP:SPS

  Self-published sources should never be used as third-party sources about living persons, even if the author is a well-known professional researcher or writer;

  In this instance it is clear who the first and second parties are. But how is this different from simply saying "Self-published sources should never be used as sources about living persons..."? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)
• From WP:SELFANDQUEST

  Self-published and questionable sources may only be used as sources about themselves, and only if ... the material used ... does not involve claims about third parties;

  And what about claims regarding second parties? PSWG1920 (talk) 19:30, 30 June 2008 (UTC)

I'll answer these in order:

• If it's a person, then that person, and his friends, family, coworkers, and lawyers are not third-parties by Wiki standards. If it's a company, then the same for its employees. If it's a scientific theory, then the people who came up with it are also not third parties. Also, there is generally no first or second, there is just third and not third.
• But if you'd like to think of second parties, then I suppose his publisher, his lawyer, his family, his friends...
  • Secondary definately should not be substituted in these two cases. A primary source is some manner of raw data, or an original record of an event. A secondary source reports or analyzes the former. This is entirely seperate from third-party, a person can produce a secondary source on his own company, theories, life, etc...But such a source would certainly not be third-party, and we certainly don't want to give the illusion that they are generally reliable.
• Should not be changed to living persons. We already have BLP, as I see you are aware of. But in general, we do not use self-published sources for any negative information about third parties. This applies more broadly than BLPs; we also don't take self-published content to reference criticism of companies, for example.
• Third party actually means something here distinct from in the rest of the guideline. It specifically implies that the publisher is third-party to the author, whereas previously this would not be the case. That is, if an author happens to own the publishing company, then it is considered a self-published work when the company puts out one of his own books.
• Self-published sources can be used for non-contentious information about living persons. This is all just a train of exceptions: Self-published sources are generally not OK outside articles about the author/publisher. Ignore the previous sentence if the author is an expert in the field. Ignore the previous sentence if the content of the source refers to a living person, negatively.

I think that makes it clear. Maybe it could still be better worded, though. Someguy1221 (talk) 03:28, 1 July 2008 (UTC)

What about the use of "independently published" instead of "third-party published"? That would seem to make it more clear what it means. PSWG1920 (talk) 03:43, 1 July 2008 (UTC)

I thought the one unclear thing about "independently published" is that people who don't already know what it means are going to ask, "from what?" Or maybe that would be clear to them, I don't know... Someguy1221 (talk) 04:00, 1 July 2008 (UTC)

What about "published independently of the author"? PSWG1920 (talk) 04:04, 1 July 2008 (UTC)

There's also the need for the author to be independent of the topic. Don't get me wrong, "third-party" seems to confuse quite a few people as well...Someguy1221 (talk) 04:28, 1 July 2008 (UTC)

I don't see what's wrong with how it is phrased currently. Also, there have been some discussions where people have concluded that an expert talking about his subject -- and refuting the work of another expert or source -- does not violate BLP. That should probably be discussed here first and put into policy, but it's been said anyway. There was a discussion about this dealing with history over at WP:RS/N a while ago -- here is link. But it seems reasonable to me that a self-published source by an expert can contradict and refute other sources. Otherwise we might as well say an expert is uncitable. II | (t - c) 04:08, 1 July 2008 (UTC)

Listing just what we mean by "third party" on various pages is a good idea, and thanks for that work, but it doesn't mean the party after the first two parties; in Wikipedia policy and guidelines, it seems to me to mean "uninvolved", "independent", "not having any obvious reason to be biased". - Dan Dank55 (talk)(mistakes) 20:24, 2 July 2008 (UTC)

Challenged

A short while ago i asked about defining the often referenced but undefined phrase "challenged or likely to be challenged" (its in the archives of this page somewhere) and got some mixed response ranging from "yeah its vague" to "its supposed to be vague" and also "its plain common sense". I went away for a little while and thought about this then came up with what i think is a pretty good idea cuz i dont think common sense alway prevails here, if it did we would probably not need so many gosh darned rulez.

Anyways i want your comments on the following idea to existing policy (my words are just first draft so i welcome any suggestions):

Any editor may ask for reliable source citations for any material in any article per this policy however such a request (either by talk page or by tag) is not by itself a "challenge" to the material, but rather it is only a request for sources and nothing more. To actually challenge material an editor must state on the article's talk page that in good faith they believe there is an material error in it, and whenever possible give guidance to the other editors as to what they believe that error to be. This will enable other editors to understand what issues may need to be corrected with the material in addition to simply providing required citations.

I think this strikes a nice compromise between those who want to improve wp by getting citations and those who want to actually argue about what is verifiable. It changes no procedures but it does add a level of responsibility to editors who wish to challenge material to actually go to the talk page and talk.

208.43.120.114 (talk) 22:18, 30 June 2008 (UTC)

As to "only a request for sources": The policy states Any material lacking a reliable source may be removed (after allowing for sufficient time to respond to the request for sources).

Your proposal would shift the burden of evidence - or at least, the burden of research - completely from those wishing to add material to those wishing to remove material. It also amounts to a bullet-proof protection of well-done hoaxes and fakes, as it is often very hard or impossible to prove (and therefore also to state in good faith) that something does not exist. "I hereby state that I honestly believe that this is false, and am therefore entitled to request a reference" is a statement which is much harder to make than "I am not sure that this is true, please provide a reference".

Regards, HaeB (talk) 00:04, 1 July 2008 (UTC)

There is already "a level of responsibility to editors who wish to challenge material". I'll repeat what I said recently at WP:VPP: the only tool you've got here is the balance between WP:V and WP:POINT: if someone asks you to source something (with an acceptable source, not a wiki page), you have to do it (or else they can remove the material); but if that editor has a habit of asking for things to be sourced that are not "likely to be challenged", whatever that means, and if they're obnoxious enough about it, that's actionable at WP:ANI. I'm not bringing this up to suggest that we start arresting people; I'm saying that that's the only available recourse. WP:V says that you have to provide a cite when challenged. - Dan Dank55 (talk)(mistakes) 13:45, 1 July 2008 (UTC)

I agree with HaeB, "challenging" material and waiting before removing it is basically a courtesy, not a requirement. Using similar logic, if people ignore the {{fact}} tag I can interpret that as ignoring a request and they don't care if the material is removed. I really don't see what the substantive difference between a "request" and a "challenge" is. Mr.Z-man 20:50, 1 July 2008 (UTC)

Sorry my internet connection has been very bad the past few day so i am sorry for being late responding. There seems to be a serious misunderstanding that what i am proposing changes the existing process of tagging and removal. That is absolutely false!!!!!

• OMG! I just realized that i never made it clear where this paragraph was supposed to be used. It is NOT going to replace any existing text, i just wanna ADD it under the section on "likely to be challenged" i dont wanna substitute it for anything.

There are many editors out there who rely on the catchphrase "any material challenged or likely to be challenged" to justify their actions without meaningful dialog by saying "the material has been removed therefore it is challenged and wp:v says challenged material cant be restored without sources". Search through the talkpages and you will find that argument over and over again (as I said at the beginning this phrase is used over 2000 times here, often along this line of argumentation). This has become in practice an unhelpful "canned response" to good faith editors trying to understand how wikipedia works and why the tag was added in the first place. It often leaves them with a (wrong) impression that the material was tagged just because the editor "felt like it".

This canned argument leaves the definition as "challenged==removed" and "removed==challenged" which means that when the policy says

Editors should provide a reliable source for quotations and for any material that is challenged or is likely to be challenged, or the material may be removed.

it really means

Editors should provide a reliable source for quotations and for any material that is removed or is likely to be removed, or the material may be removed.

Tell me that aint stupid!

Good faith editors deserve some minimal guidance on what material is "likely" to require a source. Right now that guidance is "source it or lose it" which is not informative or helpful. If i am working on calendars on a subsection called "workweeks" and i write "The traditional workweek in the USA is Monday through Friday" ... do i really need to spend hours looking for a an obscure source to support that obvious and non contentious statement? Hours that could be spent on more significant ways to improve articles. (See my examples later below)

Adding this simple definition of challenged does not prevent an editor from adding a fact or unsourced tag, it simply prevents the editor from saying "because i added the tag that makes this material challenged". Instead what happens is the tag still goes on, the editor who wants to keep the material still needs to find a source, but the editor who added the tag loses a "canned response" and when asked "why did you tag/remove it?" might (note i dont say must) actually have to explain what their issue with the material is. The burden of proof is still on the editor who wants the material included.

Let me give some examples (all of these examples assume good faith that the editors are not being needlessly argumentative or deliberately editwarring). Imagine i am a new user at wikipedia:

---

Example A (as things are now, the hard way):

1. I am working on calendars on a subsection called "workweeks" and i write "The traditional workweek is Monday through Friday". Finishing this and some other work i log off.
2. An editor comes along the next day and adds a fact tag to this statement.
3. A couple weeks later (i am not on wikipedia all the time) i come back and find my statement removed.
4. I look at the history and see an editsummary saying "no refs - removed". I look at the talk pages and see nothing.
5. I think to myself (wrongly) "that statement didn't need refs" so i restore it with an editsummary of "obvious fact, no refs needed".
6. Now the other editor comes to my talkpage and accuses me of violating wp:v for restoring "challenged" material and undoes my restore so i find myself in a potential edit-war.
7. I go to an admin looking for help and he says "The other editor was just following policy, he challenged your material so now you have to source it. I know it's a really dumb thing to have to source but thats the rules."
8. I look for hours until i find an old college textbook saying exactly what i said. I add the citation and restore the material, wp:v satisfied.
9. The other editor comes along and writes in the talkpages "your view of the workweek is too usa-centric, in some other countries the workweek is very different" and then adds a POV-section tag to the material.

• 9 steps

---

Example B (as things could be, the easier way):

1. I am working on calendars on a subsection called "workweeks" and i write "The traditional workweek is Monday through Friday". Finishing this and some other work i log off.
2. An editor comes along the next day and adds a fact tag to this statement.
3. A couple weeks later (i am not on wikipedia all the time) i come back and find my statement removed.
4. I look at the history and see an editsummary saying "no refs - removed". I look at the talk pages and see nothing.
5. I think to myself (wrongly) "that statement didn't need refs" so i restore it with an editsummary of "obvious fact, no refs needed".
6. The other editor comes along and reverts my edit with an editsummary that says "CHALLENGE: see talk page".
7. I look at the talk pages and see "your view of the workweek is too usa-centric, in some other countries the workweek is very different".
8. Now i understand why the statement is a problem and not only that but also i realize that the section can be greatly expanded to include discussions of the many different kinds of workweek traditions around the world. Now i can begin to really add value to the article at last.

• 8 steps, a little better

---

Example C (as things could be, the best way):

1. I am working on calendars on a subsection called "workweeks" and i write "The traditional workweek is Monday through Friday". Finishing this and some other work i log off.
2. An editor comes along the next day and adds a fact tag to this statement. He also adds an editsummary that says "CHALLENGE: see talk page".
3. A couple weeks later (i am not on wikipedia all the time) i come back and find my statement removed. I see the editsummary so i look at the talk pages and see both my removed text and "your view of the workweek is too usa-centric, in some other countries the workweek is very different".
4. Now i understand why the statement is a problem and not only that but also i realize that the section can be greatly expanded to include discussions of the many different kinds of workweek traditions around the world. With this focus i begin to quickly add value to the article.

• 4 steps total, much better

---

As Z-man said "challenge & wait" is a courtesy, not a requirement, and that does not change. Obviously what I am proposing editors could (and some do) already do now, but there is no motivation to do so versus just slapping a tag with a pathetic battle cry of "per wp:v". By clarifying that "a tag is just a tag" and "a removal is just a removal" and that "a challenge is something more than just a tag or removal" we encourage editors to continue requesting citations and removing unsourced material but to also explain the issues that make them feel a source is needed. This leads to better sources being found the first time and lowers the frustration level and edit wars.

208.43.120.114 (talk) 18:47, 2 July 2008 (UTC)

"The threshold for inclusion..."

Hi.

It keeps bugging me: I see that phrase "The threshold for inclusion" is verifiability, not truth. But I don't like this because it suggests that anything over this threshold may be includable, which is obviously not the case. In order for something to be included on Wikipedia it must meet all the relevant content policies and guidelines: not just WP:V, but WP:NOT, WP:NOTE, WP:NOR, and others. I'd suggest this be changed to something like "One of Wikipedia's primary criteria for inclusion...", "A necessary condition for inclusion in Wikipedia..." etc. but not the current phrasing. If one wants to argue that the current phrasing is good because if something does not pass WP:V it cannot be included, one could claim this argument also justifies similar wordings to be included in WP:NOR, WP:NOT, etc. as the same applies to them. But that would be confusiong: which one is "the" threshold for inclusion? Well, the answer is: they all are! They all define and refine the bar of inclusion on Wikipedia. Together, they establish a bar for inclusion, not any one of them on its own. mike4ty4 (talk) 04:04, 6 July 2008 (UTC)

Appropriate sources

We can read here that:

Wikipedia:Verifiability and Wikipedia:Reliable sources require that information included in an article have been published in a reliable source which is identified and potentially available to the reader. What constitutes a reliable source varies with the topic of the article, but in the case of a scientific theory, there is a clear expectation that the sources for the theory itself are reputable textbooks or peer-reviewed journals. Scientific theories promulgated outside these media are not properly verifiable as scientific theories and should not be represented as such.

The fact that as a rule only reputable textbooks or peer-reviewed journals should be used as sources for statements about scientific theories should be spelled out explicitely. More easily accessible sources in addition to peer reviewed journals or textbooks can also be given, of course. But what we should avoid at all costs are statements in wiki articles about some novel scientific claims made in some newspaper when there are no peer reviewed journals to back up such claims. Count Iblis (talk) 18:09, 6 July 2008 (UTC)

Verifiability per article

I have had several arguments with individuals about whethere or not a source in another wikipedia counts as verification within another article. Can the policy be clarified to flat out state sourcing is required for each article independently of whether the material is sourced elsewhere in Wikipedia? -- The Red Pen of Doom 08:01, 7 July 2008 (UTC)

WP:V#Self-published sources says: "Articles and posts on Wikipedia may not be used as sources." As I recall,it used to say something like "Articles and posts on Wikis may not be used as sources." -- Boracay Bill (talk) 11:49, 7 July 2008 (UTC)

The original question here is not too clear to me. What does "a source in another wikipedia" mean? If I'm translating an article from another Wikipedia (language), I'll do my best to check whether the sourcing is correct and reliable and if it is, I'll use the same references here.

But the question whether "sourcing is required for each article independently" seems to me to be another question entirely. If I'm writing an article (or adding to one) and use a concept which is well-explained and documented in another article in English, I certainly don't need to define and document the article title I'm linking. (I may not have explained that any better then the original. Perhaps we need a concrete example.) --Hordaland (talk) 18:24, 7 July 2008 (UTC)

I never thought this was a very significant consideration. I can think of two regular (and one random) form of this: Listing subjects in a list article that are themselves sourced as belonging in that list on their own articles, and a summary of a subject in the article on its parent topic. In both cases, the facts being claimed are easily verified by visiting the subject's own article, but by the same token, it's quite trivial to port the sources over, so I never honestly understood why some people (not necessarily the editor who opened this thread, though) seem to find that requirement draconian, if it does exist, which I don't believe there is any agreement on. The only other form of this I can think of is making a point about a subject in an article on another subject. But in that case it's quite clear that a relevant source should be cited, to demonstrate that you are not exercising original synthesis. So as my actual answer, I would say do what seems like it would help the 'pedia and prevent unnecessary disputes. Just provide a citation wherever someone might challenge a statement. It's easy enough to copy-paste a reference if need be. 151.152.101.44 (talk) 20:57, 7 July 2008 (UTC)

Plot points: primary source?

Hello, In an article I've been working on, we seem to be stuck. In general, assuming WP:N is met for the article, is it generally acceptable to use primary sources (book, movie, video etc.) as a source for a fact? So could a statement like "In episode #1 Bob uncovers a conspiracy to kill the President" be sourced to the episode? Hobit (talk) 19:02, 7 July 2008 (UTC)

Works of fiction are undoubtedly reliable sources for their own contents, and if you are quoting the work, or giving an uncontroversial and unoriginal explanation such as the above, citing the work itself is usually no problem. If what you are saying about the work is controversial or likely to be challenged, however (e.g., something like "Chapter 1 is a veiled political statement about the Iraq War"), then you need a source other than the work itself. COGDEN 19:57, 7 July 2008 (UTC)

problem with the basic sourcing policy

This point is not new -- most likely -- but it's been long enough for the current policy to demonstrate that it doesn't work in significant ways. The current policy requires sourcing of factual statements. Obviously not a problem, as any statement that the Moon is made of green cheese is obviously the sort of thing that should be sourced. Though I'm sure somewhere there is a printed source which says just that. Perhaps best removed if not sourced, but on the other hand, if sourced, it's still wrong. But the present policy also requires the following to be sourced: Iron is a solid at room temperatures. It's a factual statement and so should be sourced. But this is silly as it's an obvious fact to anyone who has ever seen, touched, experienced, an iron object. But, for those on patrol for policy violations, this is an eminently taggable item. Some good sense on the part of those patrolling would obviate the problem, but a change in the policy would work better. The present policy results in much wrangling, irritation, and waste of time when misused with abandon. Wikipedians, though numerous, are not an infinite resource and should be 'used' to best advantage in the interest of a better Wikipedia.

Proposed change: Factual statements, except those obviously true or common knowledge, require source citation.

Comment from others on a serious point causing trouble throughout the Wikipedia? ww (talk) 20:42, 4 July 2008 (UTC)

Check out wp:Common knowledge. Cheers. L0b0t (talk) 20:48, 4 July 2008 (UTC)

Also, wp:FACTS. L0b0t (talk) 20:52, 4 July 2008 (UTC)

I've looked into both, thank you, and neither addresses the issue, quite. Common Knowledge is a gloss on the citation policy, and no need to cite the sky is blue is a (perhaps carefully chosen) bad example of something very obvious, with assorted gotchas if treated as obvious.

The problem I've been seeing is with those who wield cite sources without common sense; it seems for purposes other than merely increasing Wikipedias's reliability and reducing the guff level. . I've not used actual examples to preclude offense, and the "iron is a solid" example above was as simple as I can contrive to more or less illustrate the issue.

But I'll try again with a somewhat disguised actual example. I made an edit to an article in which I wished to help the reader see that there were several conceptual cases. The article in question mixed technical and ordinary user content to some extent and so I was at pains to disentangle the feet of Readers inclined to trip over technical content. One case, artificially simple, was x, and certain unfortunate things followed from using x. Another case, tempting from some perspectives, was y, but certain things also unfortunate followed from y, making y an unsatisfactory choice as well. Thus we can see that neither option x nor y are satisfactory for the purpose intended. And we blah blah, rest of article content... Another editor objected that I had not sourced the example I'd used (in this case y). I sense 3RR in the road ahead.

It may be said that this editor is so exceptional that one must merely cope, general rules not being capable of covering outliers. But I've come across this in other contexts as well of late. Another real though disguised example is the following. An editor looked into a medical article I monitor, left two fact tags (for well established clinical realities), neglected to tag other equally unsourced well-known clinical realities, and left. I queried that editor about this, and received a comment that being unfamiliar with medicine, I thought these needed tagging for fact. I strongly believe in the citation rule and live by it, ... Drive by tagging as it were!

Citations are a Good Thing, but as we all know, a Good Thing can be carried too far. I am suggesting that we revise our definition of citation need to preclude at least some kinds of excess in the use of this Good Thing. I'm sadly wise enough to expect that no such change will eliminate all excess and misuse. Maybe a note that things can Go Too Far? Something, please, folks....

More comment, please! ww (talk) 21:39, 6 July 2008 (UTC)

In the interest of full disclosure I must confess that I firmly believe every single declarative statement or statement of fact should have an inline citation. I do, however, feel your pain. It brings to mind the time I was tagging some WP:OR and WP:SYN in a trivia section on a television episode article and was chastised that I "...would demand a source for the ocean being blue." Well of course I would, the ocean isn't blue. What seems patently obvious to some might be mysterious and strange to someone else. I would tend to think that if the claim is being made about the subject of the article then a cite would be appropriate, if the claim is being made about something tangential to the subject of the article then a wikilink to an article about the tangential item would fine, using your example, iron is a solid at room temperatures would need a cited source in the article about iron (I don't think I've ever seen a general undergrad chemistry, or even a junior-high science, textbook without a mention of iron as a solid and mercury a liquid at room temperatures.) but an article about something made from iron would just wikilink iron (I hope that makes sense). On the other hand, there are articles that were tagged for sources back in 2006 that have yet to be fixed, so I don't see the tags accomplishing much. I agree this could use some discussion. Anyone? L0b0t (talk) 22:48, 6 July 2008 (UTC)

A couple points...first, since this is verifiability, not every statement need be sourced. And so no one should be removing a statement merely because it is unsourced; if there is legitimate grounds for doubting the accuracy of a statement, then by all means. And by all means tag the statement or ask on the talk page for a source. But there is a policy that acts as a universal reminder that the blunt force of policy should never be used merely because it can be. And given that, it's not really necessary to riddle the rest of the policies with disclaimers about using common sense; if you see someone who seems to be a bit too zealous, maybe he just needs a gentle reminder of all that... Someguy1221 (talk) 01:58, 7 July 2008 (UTC)

WP:IAR is quite relevant, but the sort of person who weilds sourcing demands as a censorship enforcement (for whatever reason) is not likely to recognize a more 'liberal' rule as taking precedence (or perhaps as even legitimate) in constrast to a more restrictive and so more useful for their purposes rule. Still buffaloed on this aspect.

It occurs to me though, rereading the various comments above, that part of the problem is caused by the ease with which fact tags may be strewn about. Very little effort required. Perhaps, if the fact tag must be placed on the talk page, with a snippet of the statement which is missing the fact tag, the kind of drive by tagging I've seen of late might be reduced to something more like the set of statements which actually require sources as opposed to the much larger set of statements which can be hit with a fact tag during a drive by.

Perhaps this would address at least part of the issue I've been sufficiently exercised by to post here about? Ideas, reactions from others? ww (talk) 14:44, 8 July 2008 (UTC)

Next go-round on poor "reliable sources":Gabriel’s Revelation

I have tagged the article on Gabriel's Revelation because all the sources given are newspaper or general news magazines, which have a poor track record on handling these sorts of textual issues. (See for instance the hype around the Gospel of Judas, which when the dust settled turned out to be a typical sort of gnostic text.) There has been some discussion of this elsewhere (e.g. at GetReligion) but one hardly needs to go that far. Of the four sources, one is worthlessly short, and one is a copy of one of the others; of the remaining two, only one bothers to produce a doubter of the claims made. As far as I can tell, there is no scholarly article behind this, at least one that has actually be printed and reviewed by experts. There is no mention of the body of intertestamental apocalyptic and messianic material that has already been discovered, and of which this is all too likely to prove to be just another example. There is only slight acknowledgement in one of the articles that the reading of these documents is not certain anyway, and none at all in the other.

But enough of that: we're back to the problem that the New York Times is being declared a Reliable Source in spite of their poor record and in spite of the fact that the article bears all the signs that tell anyone with much familiarity with the field that the claims are probably empty hype founded in a one-sided article. Can we find some way to end this madness? Sources should be declared reliable by actually being reliable, and where they have a poor track record, they shouldn't be considered reliable. That is all there should be to it. Mangoe (talk) 04:11, 8 July 2008 (UTC)

As I am continually shocked that I have to point out, there is a considerable difference between claiming X is true and claiming John says X is true. In the latter, the NYT is undeniably reliable, as that is what it was actually doing. The same is true of the remaining news sources cited by GetReligion. The worst a newspaper can do in this situation is give undeserved attention to a minority opinion, and this is something that has to be handled at the editorial level with an eye towards due weight. Someguy1221 (talk) 06:52, 8 July 2008 (UTC)

This is old news, I'm adding some better sources from last year and earlier this year and a translation. Doug Weller (talk) 14:54, 8 July 2008 (UTC)

Thanks for the additions; the article is much improved.

The dispute started because I placed a POV tag on the page. I did so because examination of the articles, particularly of the NYT article that appeared to be the predominant source, showed a lack of balance characteristic of mainstream "revelations" of these sorts of documents. Perhaps it's because I take a more natural, non-policy view of "reliability" which factors in not only accuracy of reporting, but balance; in any case, the problems which I could identify showed that there was a definite balance issue, which is why I put the tag on. The answer I was given was that since the NYT is officially reliable, I could not apply the tag. Given that we have a more extensive version of the same news filing from the IHT, there is room for specific critique of the NYT version; but even if one doesn't dispute that the NYT reported accurately what the reporter was told, that reliability doesn't make the balance issues go away.

Let me put it another way: when the NYT is being used as a source for the attempt to make a splash with this announcement, then yes, it is a reliable source in all senses of the word. However, when it is the only source used about the state of the field concerning the issue being reported, it is only as reliable as the range of figures it surveys. In this case Brannon seems to have only spoken to the people pushing this particular interpretation, which is why a POV-tagging was warranted. Mangoe (talk) 19:34, 8 July 2008 (UTC)

Let me put it more succinctly: the position is being pushed that, because the NYT is considered "reliable", that is also therefore neutral. Mangoe (talk) 21:30, 9 July 2008 (UTC)

It may be worth a sentence in NPOV, specifically WP:UNDUE, that while news coverage establishes notability, it does not establish significance, but then we're on the wrong page. In any event, I have no opinion on the specific dispute at hand. Someguy1221 (talk) 23:20, 9 July 2008 (UTC)

Something's missing: reliability of a source sometimes depends on what you are citing it for.

It has been a while since I've read this page, and think it is coming along well, but something is still missing: You get the impression reading this page that you cannot ever cite unreliable sources. Actually, you can--you just can't use them for the purpose of supporting the claims and propositions set forth in the article. You can, however, use totally unreliable sources for the purpose of showing what the sources say, which may be totally uncontroversial. For example, it is perfectly acceptable to cite the text of a notable hoax, such as The Protocols of the Elders of Zion, to discuss details about the hoax, such as what the hoax said, examples of racism found within it, etc. I think we need to make that clear, because I think some editors might go astray, thinking that some sources simply cannot be cited for any reason in Wikipedia.

On a related note, that fact that something is a reliable source does not mean that citing it complies with Verifiability. Sometimes, perfectly reliable sources simply parrot or quote totally unreliable sources. Sometimes, an otherwise-reliable is totally off-the-wall on the point you are citing. Sometimes, seemingly-reliable research is later shown to be uncontroversially-false (e.g., Pons & Fleishman, whose cold fusion article was published in an otherwise-reliable source). Sometimes, scientific journals accept hoaxes. Just because it appears in an extremely reliable source does not mean you have complied with Verifiability. In these cases, where you are citing a source for a point on which that source is unreliable, you should treat it as unreliable. (If, on the other hand, the subject of your article is how some otherwise-reliable professor was bamboozled by a hoax, then you can cite her article, because for that purpose, the source is reliable.) COGDEN 01:07, 4 July 2008 (UTC)

If someone has concerns about a source, the concerns should be resolved on the article talk page. No policy is ever perfect or without defect. Wikipedia relies upon reliable sources, with the unwritten caveat that no sources is ever unquestionable reliable; this is normally resolved by explaining both sides of any POV issue within the article itself. To do otherwise is to open the flood gates to mob rule. If ten editors state that X is true, and 4 editors state that x is a fraud, then x becomes true by simple majority rule (instead we cite reliable published sources). “The threshold for inclusion in Wikipedia is verifiability, not truth” is there to avoid this outcome. We as editors do not decide truth, we simply report what reliable sources have stated. Brimba (talk) 04:41, 9 July 2008 (UTC)

I hear you, but why not make that "unwritten" caveat written? Editors often turn to this page as a reason to admit or deny content, and if that caveat is truly non-controversial, we out to make that explicit, to provide editors with guidance. We ought to just say what you just said, because the current article gives the impression that what you are saying is false. COGDEN 02:11, 15 July 2008 (UTC)

Self-published sources and quotes from them on GNU/Linux naming controversy

Can I get some opinions on this article? The subject is an acrimonious debate (want evidence? Talk:Linux/Name) and many of the sources on this page are either self-published (e.g. by Richard Stallman on the FSF's GNU website) or not even properly published but drawn from mailing list postings. The current argument seems to be that they are "experts" in their field. 76.10.148.211 (talk) 17:38, 11 July 2008 (UTC)

In the case of Stallman's publications (by an organization of which his is not the only member, merely an important one) I would suggest that unless the editor can be identified as Stallman himself, there is not a case of self-publishing in the sense WP cares about it.

I can write something, post it on any Web site (including one I run), and if you cite my work, everything will be fine. So, if I cite my work openly, there's a possible problem. If the Web site is a public one, it may be permissible here. If it's purely mine, then there's a conflict with WP policy. A middle case is me citing this last web site, but doing so as a sock puppet. Not always easy to detect, but if discovered, there'd be the same problem re WP policy.

In your example case, is the editor suspected, or proved, to be a sock puppet run by Stallman? I'd think it unlikely, given my impression of his perspective on various things, but perhaps it's possible. If so, the question re self-publishing citation is not so easy. The FSF is clearly a large and influential organization, and so it's not clearly obvious that Stallman's writing on its web site is self-publishing in this WP sense. Likely anything he feels strongly enough about to write up will find sufficient support to put on the FSF web site, but that's not a very clear case of self-publishing.

From a self-published citation perspective, being an expert in a subject isn't sufficient. Evaluating expertise is very difficult in a WP context. ww (talk) 12:42, 16 July 2008 (UTC)

I'm pretty sure Stallman did write much of the material in question. However, I suppose I see it more like GNU and the FSF are the self-publisher, not Stallman. After all, the publisher in this case is by no means neutral: GNU and the FSF have a vested interest in the OS being called GNU/Linux. The same arguments were used elsewhere to remove citations to Microsoft's Total Cost of Operation studies, even though it wasn't Bill Gates himself who authored them.

I think another question with a more obvious answer is this: are Linus' and friends' posts to a mailing list unacceptable? I tried to delete references to them twice and got reverted. Check the recent page history. 206.248.134.130 (talk) 07:12, 17 July 2008 (UTC)

Suggest moving this discussion to Wikipedia:Reliable sources/Noticeboard which discusses sourcing in individual articles. This page is for discussion of the policy. Best, --Shirahadasha (talk) 12:22, 17 July 2008 (UTC)

Done, thanks. 206.248.134.130 (talk) 07:16, 18 July 2008 (UTC)

Multiple uses of the same number

If an article uses a challengeable piece of data multiple times, does it have to be cited each and every time it's used? I'm doing a GA review, and trying to wrestle with a timeline that provides context. Citing the elements of the timeline in the timeline causes a mess. Is it sufficient to just leave the timeline as uncited if all of the data used to create the timeline is in the main body of the text? SDY (talk) 17:31, 18 July 2008 (UTC)

Yes. If anyone challenges something in the timeline, they can just be directed to the reference in the text. Since that would prove the material is verifiable, there is no WP:V concern. — Carl (CBM · talk) 16:28, 19 July 2008 (UTC)

What happened to the "no gratuitous challenges" language

I swear that last time I was talking about this page it had a passage that told editors not to challenge for the sake of challenging, and noting that gratuitous challenges were disruptive. And yet when I went to tell an editor that they needed to actually explain why they were challenging the information they were challenging, I was quite surprised to find out that, no, that language has vanished.

Would someone please direct me towards the explanation for why we've lifted that prohibition and allowed such a potent weapon of disruption? Phil Sandifer (talk) 15:10, 19 July 2008 (UTC)

Challenging for the sake of challenging is not very productive, but challenging to produce a better referenced article is a good idea. I have not seen the wording you speak of, what was the gist of it? Chillum 16:38, 19 July 2008 (UTC)

The gist of it is that you should only add cn tags and remove information that you are skeptical of the accuracy of, and not demand citations for the sake of citations. Phil Sandifer (talk) 16:48, 19 July 2008 (UTC)

Right now, WP:V does say "Any material lacking a reliable source may be removed, but editors may object if you remove material without giving them sufficient time to provide references." I don't know if there would be agreement to say anything stronger than that, but the consensus has always been that outside of BLPs, editors should not hurry to remove unsourced material unless they genuinely believe it is incorrect. — Carl (CBM · talk) 17:07, 19 July 2008 (UTC)

Yeah. I'd really like something that shored up the issue of good faith and challenges. Any ideas for language? Phil Sandifer (talk) 15:03, 20 July 2008 (UTC)

Album Leaks

would having info on an album leaking be non-verifable if the only way to verify it is to show a download link to it? Reply on my talk page - -[The Spooky One] | ^{[t c r]} 22:49, 21 July 2008 (UTC)

Similar shortcuts SPS and SELFPUB

Hi. I changed the shortcut to SELFPUB to SELFQUEST because I noticed a couple people linking to it and taking it as the authority on self-published sources, missing SPS. Brimba recently changed it back.[6] I still think these shortcuts are unsatisfactory and confusing, and thus hard to remember. I don't find SELFQUEST hard to understand once you read the title, and I'm wondering what other people think. Perhaps we should use SELFEXPERT for the first rather than SPS, and SELFPUB or SPS for the second? Or maybe we could have maybe just one shortcut, so people are forced to read the entire section and understand what is going on. Or maybe I'm making a fuss about nothing, since eventually editors' will (hopefully) figure out, or be told, that there is a diff. between an expert and a non-expert. Still, I think it is important to have logical shortcuts, and these two are too similar currently to be considered logical. II | (t - c) 02:09, 22 July 2008 (UTC)

Note: I investigated this because I noticed someone switching a link from SPS to SELFPUB. See a current discussion over at RS/N for an examplediff of confused IP of people confused by these two shortcuts. II | (t - c) 02:13, 22 July 2008 (UTC)

An idea occurred to me. Why not make the two self-published questions subsections of an overarching "Self-published sources" section. Then SPS or SELFPUB goes to the overarching section, SELFEXPERT goes to expert, and SELFQUEST goes to the other one. II | (t - c) 02:16, 22 July 2008 (UTC)

Proposal for reliable sources

When we think about what it means to be a reliable source, can't everything be summed up in the principle that all information must be citable to the type of sources generally relied on in things like encyclopedias and scholarly publications? Also, we should be citing the most authoritative sources for any given point, given a choice. So, for example, what if we edited the following paragraph as follows:

Articles should only present rely on reliable, information that is cited and credited by at least one third-party published sources publication with a reputation for fact-checking and accuracy. When several publications provide such information, a Wikipedia article should generally cite the most authoritative among them.

COGDEN 20:44, 7 July 2008 (UTC)

Well it's more complicated than just that. Sometimes, a legitimate dispute exists amongst sources of similar reliability and the only neutral way around it is to cite all of the sources and frame the dispute. Even further, there is the "fact" being claimed by a source, and the fact that the source made that claim (X is true versus John says X is true). It may take a scholarly journal to "verify" the first, but any old newspaper could verify the second. But we don't discount the latter if it happens to be notable, although we won't mention in the main article as it may be insignificant. So what I'm saying is, I'm concerned your wording gives the impression that insignificant or merely less significant viewpoints may be disregarded entirely, which is certainly not always the case. 151.152.101.44 (talk) 21:08, 7 July 2008 (UTC)

That's a very good point. What about something like this:

Articles should only present rely on reliable, information that is cited and credited by at least one third-party published sources publication with a reputation for fact-checking and accuracy. When multiple publications provide such reliable information, a Wikipedia article may cite any such publication. However, when two sources provide exactly the same information, and editors disagree over which source should be cited, preference should be given to the source that is cited with greater frequency by academic writers in the relevant field.

I guess the last sentence is optional, but it might provide some guidance on an issue that crops up now and again. COGDEN 01:24, 8 July 2008 (UTC)

I am all for WP:V and WP:RS, but this is too radical a change in the policy and guideline. As a work in progress, Wikipedia has 1) thousands of articles that cite no sources, 2) thousands more that only cite sources that fall short of the 3PFC&A standard, and 3) still thousands more that have part of their information that is uncited. This changes the policy from verifiability (someone must have the ability to verify the information, even if the article does contain every citation needed to show them exactly how to do that) to verified (everything must be verified in the article) - not a desirable change, IMHO. What would you propose we do with all of the articles that fall into 1), 2) and 3) above: delete them? UnitedStatesian (talk) 02:02, 8 July 2008 (UTC)

I think you misread the language above, but it can perhaps be made clearer. The above language says that the information has to be cited in a relilable source, but does not require that the source actually be cited. This is no different from the current policy. If nobody challenges the information, all that is required is that somewhere, somebody makes that point in a reliable source, and the editor could cite that source if called-for. Below is a revision to make that clearer. COGDEN 02:29, 15 July 2008 (UTC)

By what method could we determine what "the most authoratative source" is? It may be possible in a very limited number of situations, but doubtless many people will find it to be the one one agrees with. A second problem with the proposal is that authorativeness doesn't necessarily have anything to do with reputation for fact-checking. What gives a source a reputation for authorativeness depends on the subject, and a definition based on only a limited number of fields would be problematic. In a great many subjects, opinions, theories, philosophies, concepts, etc., not facts, are what is at issue. The proposal seems to assume a world where questions have clear and definite answers. Not all of us live in such a world. Best, --Shirahadasha (talk) 03:12, 8 July 2008 (UTC)

The second version of the draft removes the "authoritative" language, and I have further edited it to incorporate UnitedStatesian's comments above:

Articles should only present rely on reliable, information for which a researcher would be able to find at least one third-party published sources publication with a reputation for fact-checking and accuracy. When multiple publications provide such reliable information, a Wikipedia article may cite any such publication, or no publication at all, if the point is not likely to be challenged. However, when two sources provide exactly the same information, and editors disagree over which of the two sources should be cited most prominently, preference should be given to the source that is cited with greater frequency by academic writers in the relevant field.

COGDEN 02:29, 15 July 2008 (UTC)

Wikipedia has made a conscious decision to be different from other encyclopedias in a number of respects, and our sourcing policies reflect that. Whereas most encyclopedias explicitly take a scientific point of view, Wikipedia has decided that its neutral point of view policy means neutrality between scientific and non-scientific points of view even in disputes between them. The sourcing policies have always reflected this. The community has consistently resisted attempts to change the policy to to special authoratativeness to scientific points of view, and they have resisted numerous attempts to use the sourcing policies to achieve such a goal or to make a de facto change to the WP:NPOV policy through the back door. Although scientific publications are authorative on science matters and points of view, we have articles on religion, politics, popular culture, and many other matters that simply wouldn't appear in most encyclopedias or scholarly works, and we include points of view that simply wouldn't appear in most encyclopedias or sholarly works. We've explicitly rejected the goal of becoming an exclusively "scholarly" encyclopedia in the academic sense, in favor of the goal of a neutral, reliable encyclopedia which nonetheless reflects a wider range of subjects and points of view. For example, in Archaelogy#Decendent peoples and related disputes (examples of situations where scientists may be involved in conflicts with non-scientists), we wouldn't necessarily give special weight to the scientists' opinions by handicapping non-academic sources such as journalism dissenting indiginous peoples might rely on, nor should the article or its sourcing imply the indigenous peoples' views, so long as they are recorded in sources meeting the general reliable sources criteria, are somehow invalid or less valid unless first vetted by the scientists. A different kind of encyclopedia would doubtless handle things differently, but we're not that publication. It could be that scientists' claims to always act objectively and without self-interest are always correct and people who oppose them are always in the wrong. but Wikipedia policies don't require assuming this and leave it to the reader to decide for themselves on a case-by-case basis. The sourcing policies have historically reflected this neutrality, including not assuming that sources reflecting the scientists' point of view in a notable dispute with a non-scientists will always be the most reliable globally (as distinct from reliable for reflecting scientific points of view.) Best, --Shirahadasha (talk) 19:30, 20 July 2008 (UTC)

I agree. We ought to re-focus on the verifiability of the information presented in the articles. If the information is simply verifiable (i.e., it can be found in a source that does fact-checking), we don't make any further judgment about the information (at least not regarding verifiability). One of the problems, though, that I have with the current language of the policy is that some people mistakenly interpret it to say that we cannot cite "unreliable" authors. Actually, we can cite notable unreliable authors, but only if what they say can be found (or is also found) in a reliable source like a journal article or a newspaper, etc. I couldn't quite tell whether you agree with my present draft, but I'm revising further, now that I've thought about the issue more. I don't think the last sentence was necessary, and I've streamlined it:

Articles should only present rely on reliable, information that could be found in at least one third-party published sources publication with a reputation for fact-checking and accuracy. Such publications are considered to be "reliable sources". Information is verifiable even if the reliable source citing it or presenting it does not agree with the information. While citations are strongly encouraged, Wikipedia editors may present uncited information if the information could be found to a reliable source, or is not controversial. A publication that is not initially reliable may become reliable if subject to fact-checking or analysis by later reliable sources.

COGDEN 23:24, 23 July 2008 (UTC)

Citations for only "any material challenged or likely to be challenged?"

Has anyone raised the concern that if Wikipedia aims to approach a truly academic encyclopedia, than every idea that is not common knowledge (which is almost nothing), needs to be cited? I feel that the policy of citing material that is only likely to be challenged limiting, because I believe citations are important to know where all information comes from, not only what may be controversial. At least, that is how I was raised, academically.

I suppose my main problem is that the current verifiability policy, allows far too much material to go uncited in the average article, and seems to encourage editors to neglect doing proper citation and references.--Ducio1234 (talk) 01:44, 15 July 2008 (UTC)

I don't think it is realistic to require that everything be cited right off the bat. I think it is valuable to allow editors to treat articles as works-in-progress, adding material first, then finding citations later. If, when an editor adds uncited material, some other editor is always free to force an immediate challenge to the verifiability of the information. If nobody challenges the material, the editor should be allowed time to go back and add citations later. Imposing a citation requirement would, I fear, force editors to do much of their work offline. COGDEN 02:39, 15 July 2008 (UTC)

Ducio1234- When would you require citations? Hyacinth (talk) 03:03, 15 July 2008 (UTC)

I don't currently see why it would be a large problem to ask editors to cite every idea they put on wikipedia as they add new material, I mean, he or she is getting the information from somewhere, so why not cite where it comes from? It only takes an additional second to correctly cite what you are writing. I'm not proposing that every sentence be required to have a citation, since several sentences could share a general idea with one citation.--Ducio1234 (talk) 16:58, 15 July 2008 (UTC)

There are many things that are not be cited in academic work - for example, that the Invasion of Normandy was in France in 1944, or the definition of a group in a mathematics text. For things like this it's better to just include some general references (for example, to a history textbook and a math textbook, respectively) rather than add individual footnotes. This is why we only require specific citations for things that are controversial, challenged, or quotes. There is some additional discussion about this at Wikipedia:SCG#Uncontroversial_knowledge. — Carl (CBM · talk) 17:13, 15 July 2008 (UTC)

The Uncontroversial Knowledge link helped clear up some of my concerns. However, I still feel the "any material challenged or likely to be challenged" is far too vague and that a policy of at least asking for citations be placed with material as it is posted could ideally help Wikipedia gain credibility and lead to easier improvements in articles in the future.--Ducio1234 (talk) 00:06, 16 July 2008 (UTC)

An excessive amount of citation is in fact not appropriate to the style of a general encyclopedia. Since we write using principally accepted tertiary sources, not sources requiring detailed analysis, it should not be necessary to cite things which can reasonably be found in standard works--though of course such works should in fact be given as general references, just as Ducio says. We must of course give evidence for anything controversial or negative BLP, or to any quotations or paraphrases. But we are not writing a scholarly treatise, and are not responsible for finding exhaustive citations to every fact in every sentence. There is a place in the world for such work, and it is not in general at Wikipedia .DGG (talk) 04:42, 16 July 2008 (UTC)

I'm not advocating any change in the amount of citations in Featured Articles, my main concern is the citations in articles like Jeff Wall, where obviously a lot of work has been done, but no one bothered to cite their sources, presumably, because there is no blatantly controversial material. No, the article is not a scholary treatise, but if the editors had taken a bit of extra time to add proper citations, improving the article would be enormously easier and Wikipedia would be better and more credible. The policy of citing only material viewed controversial, in my opinion, encourages the thousands of articles like Jeff Wall.--Ducio1234 (talk) 17:06, 16 July 2008 (UTC)

Ummm, don't know if I dare butt in here, but it seems to me that the above can't be a genuine conversation without better defining of terms, and probably some (more) examples. What material is likely to be challenged? I learned the hard way that newer scientific stuff may be challenged. On a topic I've read a lot about the last couple of years, I added new (like 5-10 years old) knowledge to a couple of articles. Seemed to me that everyone knew that stuff, since I'd read so much about it. Well, I got challenged. Very unpleasantly. Wikipedia isn't the place for one's pet ideas, I was told. Bah, I added 3-4 references and my challenger must have been satisfied by that - but did not apologize! So, I (now) lean toward agreeing with Ducio. If in doubt, reference it!

However, I can't agree with Ducio that it takes only a second. --Hordaland (talk) 10:08, 16 July 2008 (UTC)

I don't know... It's pretty easy to add citations from books... --Ducio1234 (talk) 17:06, 16 July 2008 (UTC)

Unless you are an expert, and have the books at hand, it is very hard to add good references from books. I'd wager that 80% of the references from printed sources in wikipedia were made from either the mere listing of a book in a catalog, or the abstracts of an article in an index, or by copying from a bibliography in another source--or, at best, as a very general reference--all with the hope that nobody inquires for the exact details. DGG (talk) 01:36, 18 July 2008 (UTC)

I still don't feel its that hard to add proper citations, but perhaps I'm biased from my own background and experiences. I still feel that Wikipedia could be better and more credible if the policy asked people to cite when they add new material. But I seem to be in the minority and it's probably unnecessary to continue this discussion any further, since I'm not willing to organize some sort of Wikipedia mutiny or political upheavel as I'm still new and learning about how Wikipedia works. Thanks to those who addressed my concerns and clarified my understanding of Wikipedia.--Ducio1234 (talk) 16:43, 22 July 2008 (UTC)

Verifiability Over Truth

I think that statement is a bit misleading in many ways. Some people may get the idea that you can post erroneous information if it can be verified. I think everything that is added to an article should be truth and verified as well and not just verifiable. Mr. C.C. (talk) 18:50, 18 July 2008 (UTC)

The place to discuss this is at Wikipedia talk:Verifiability --Philip Baird Shearer (talk) 19:29, 18 July 2008 (UTC)

(The above is moved from Wikipedia talk:Attribution)

The issue is just that some people are uncomfortable with the word "true". Adding material which you know to be erroneous, simply because it can be verified, is inappropriate. The most common solution is to change from a direct statement of fact ("Jones was an Irishman") to an attributed statement ("Smith says Jones was a Irishman"). Everyone can agree about the correctness of the attributed statement, even if they don't believe the claim Smith makes. — Carl (CBM · talk) 14:36, 23 July 2008 (UTC)

One problem here is that, particularly on matters that are in dispute, an editor's position, even one the editor might think obvious, could also be wrong. If you're adding material, you have some flexibility in what to add and what sources to use. But taking sourced material out because one thinks it wrong can be problematic. Best, --Shirahadasha (talk) 15:34, 23 July 2008 (UTC)

You have to ask whether or not there is an active controversy about whether the information is erroneous or correct. If somebody in an earlier work made an error, like an erroneous date or a math error, that was proven incorrect by later researchers, and there is no active, genuine controversy that the original information was in error, then the cited erroneous information is unverifiable, even if the error originally occurred in the journal Nature or some other impeccably reliable publication. COGDEN 20:00, 28 July 2008 (UTC)

Concerning archives

Do we have a policy with regards to the use of websites like internetarchive.org to "recover" certain versions of a webpage previously used as a source in an article, or does this violate an established policy or guideline? I'm asking becuase the article Iowa class battleship has got info sourced to websites going back 3+ years, and the use of interentarchive.org or other similar websites may enable the recovery of the original sourced material. TomStar81 (Talk) 03:27, 24 July 2008 (UTC)

So long as an internet archive can be trusted to not fabricate old information (which I assume at the moment is all of them, since I've never heard of that), they are as fine a source of information as the website you are using it to view. Someguy1221 (talk) 07:05, 24 July 2008 (UTC)

Exemptions from Verifiability and BLP?

Why are the days of the year articles, e.g. July 28 exempt from sourcing? Is it okay to delete entries where the destination links also are unsourced? What other pages are exempt? -- Jeandré, 2008-07-28t11:57z

Good question, why should they be exempt? And of course, they are used as links in other articles... I'd definitely say delete entries where destination links are unsourced. Doug Weller (talk) 13:17, 28 July 2008 (UTC)

Agree... the information should be sourced, either at the linked article or on the day of the week article. I would advise not rushing off to delete, however... before you do that, raise the issue on both talk pages and give people a reasonable time to correct the problem. (with, obviously, the caveat that if there is a serious BLP issue involved you should delete immediately) Blueboar (talk) 14:00, 28 July 2008 (UTC)

The standard is not whether they are unsourced, it is whether the information is unverifiable. If anything is not controversial or not likely to be challenged, nobody has any business deleting it, even if there is no citation to a specific source. I wouldn't make any wholesale deletions here, unless you are sure that the linked information is unverifiable, on a case-by-case basis. COGDEN 19:45, 28 July 2008 (UTC)

And if it is controversial or not obviously accurate? No one is suggesting wholesale deletions so far as I can see. Doug Weller (talk) 21:04, 28 July 2008 (UTC)

I'm just saying that you shouldn't just go through a page like July 28 and delete all the entries where the underlying article does not cite a reference for the date in question. Before deleting a date that you think is not verifiable, the protocol is: (1) that you genuinely have reason to doubt that the date is accurate, and are not just being anal about citations, and (2) that you tag the disputed date, and allow somebody else sufficient time to establish verifiability by producing a citation, if one exists. COGDEN 22:00, 28 July 2008 (UTC)

Note: BLP is an exception to this general rule. If material could put someone in a bad light (defamatory) etc.) or could misrepresent someone, it needs to be deleted on sight unless well-sourced. It's unlikely that potentially getting a date wrong would have this effect, but not impossible. Also, links to other articles (such as date links) simply provide more information about the linked topic. They don't normally imply that the linked articles contain sources for the way the subject is used in the article containing the link. (For example, if an article on John Smith says Smith was born June 28, 1921, and contains a link to this date, the date article isn't expected to discuss Smith's birthday). It's possible that the date could be sourced within the biographical article, for example, there might be a reference or reading list at the bottom which might, if checked, confirm the date. Best, --Shirahadasha (talk) 23:18, 28 July 2008 (UTC)

Actually, when it comes to BLPs I don't think the problem would be the date... A BLP violation is more likely to occur with the statement about what happened on the date. Let us assume that five nuns and a baby seal were murdered in LA yesterday, that Joe Blow has been accused of the murder, but has not gone to trial yet... it would clearly be a BLP violation to enter: "Joe Blow murders five nuns and a baby seal in LA" in the list of events that occured on July 27. It would be perfectly fine to enter: "Five nuns and a baby seal are murdered in LA".

Translate non-English

“Where editors use a non-English source to support material that others are likely to challenge, or translate any direct quote, they need to quote the relevant portion of the original text in a footnote or in the article, so readers can check that it agrees with the article content.”

Shouldn't this read “...they need to quote a translation of the relevant portion...?” —Michael Z. 2008-07-23 16:36 z

No... you should provide a quote from the original non-english source, so that editors can determine if it is being translated/interpreted correctly in the context of the article text. Blueboar (talk) 17:03, 23 July 2008 (UTC)

I don't think any editors should be engaging in their own direct translation of text. Translating things depending on the language it is going to and from is not always an exact science.--Crossmr (talk) 06:51, 31 July 2008 (UTC)

WP:SELFANDQUEST revisited

I think I understand WP:SELFANDQUEST aka WP:SELFPUB a bit better now than I did before. The material which it restricts (that which is contentious, unduly self-serving, involves third-parties, etc.) can be referenced where relevant, it just needs an independent source. But what if we simply stated that anything cited from questionable sources should relate directly to what is cited from independent sources? That would be a simpler way of preventing an article from becoming a forum for the promoters of a particular viewpoint, and at the same time would not idiosyncratically exclude genuinely relevant information not explicitly referred to by an independent source, because its source is "contentious", "unduly self-serving", etc. PSWG1920 (talk) 23:36, 26 July 2008 (UTC)

In your case, the source is not the self-published source, but the independent source, so I do not see what is the problem. If a reputable source describes a self-published source, then use the former. ≈ jossi ≈ (talk) 00:17, 27 July 2008 (UTC)

I meant to add "not explicitly referred to by an independent source" to the last sentence. My concern is that genuinely relevant bits of information can currently be excluded for idiosyncratic reasons. PSWG1920 (talk) 00:57, 27 July 2008 (UTC)

That will change substantially the current formulation, so yo will need to seek wide consensus for that change. ≈ jossi ≈ (talk) 01:47, 27 July 2008 (UTC)

Well, I'm not going to push too hard for this; I'll just state my point that something along the lines of "the material used is directly relevant to what is cited from independent sources" seems like a far more reasonable restriction than "it is not contentious", "it is not unduly self-serving", "it does not involve claims about third-parties", etc., which could idiosyncratically exclude genuinely relevant information which may not be explicitly referred to by an independent source. On a side note, someone once told me that sources promoting a fringe theory were not allowed even in an article about the fringe theory, because, he said, such material is by its very nature unduly self-serving. I'm fairly certain that interpretation was in error, but that makes me think that this policy needs to be simpler. PSWG1920 (talk) 03:47, 30 July 2008 (UTC)

Fringe theory: What you quote makes sense to me. If you read through a fringe theory promotion and pick out bits to quote, that is OR. If a respected newspaper or journal (etc.) has done that research and reported it, that would be fine to quote. (Maybe that's just the same as jossi above said...) Sounds like you want to be able to add something like "The source that the NYT is describing also says XYZ," sourced to the fringe theory promotion site (or article, etc). Can't do that. --Hordaland (talk) 17:38, 30 July 2008 (UTC)

See WP:RS on extremist and fringe sources. It's clear there that such sources can be cited directly; how they're used is the issue. PSWG1920 (talk) 00:40, 31 July 2008 (UTC)

Not sure what "case" is being referred to. If this is coming out of a specific article and there is a specific source that is at issue, perhaps an explanation of the context might help. Best, --Shirahadasha (talk) 21:38, 30 July 2008 (UTC)

---

Legal terms and common use

An issue has been raised in Talk:Illegal immigration to the United States. Namely, several editors have decided to use the common use meaning for a legal term (spec. immigrant) instead of the legal term throughout the article. I said it might be a good idea *if* we could verify what the common use meaning is. I also said that I commonly see the term used to mean the same thing that the [United States Citizenship and Immigration Services] means when it uses the term.

These editors have replied that verification isn't needed in determining what the common use meaning is. Apparently, they are content to stick to whatever they (these editors) say the common use meaning is. I find this particularly troubling because the term is highly contested in society due to the politics surrounding the issue. So, here's the question. Does the Wikipedia policy regarding verifiability apply in this case?-198.97.67.58 (talk) 13:24, 4 August 2008 (UTC)

Given that this is not a Legal reference but a general encyclopedia, I agree with the concept that we should be using common meanings of words in our articles (although, we can and in this case should discuss the legal definition as well.) The common meaning is certainly verifiable... We just have to look the word up in a dictionary. According to the OED (considered one of the more definitive dictionaries of the English Language) the word in question is defined as follows: "Immigrant - One who or that which immigrates; a person who migrates into a country as a settler." Sounds good to me. Blueboar (talk) 15:06, 4 August 2008 (UTC)

Reputation

How do we know if a source has a "reputation for fact-checking and accuracy"? Richard Blatant (talk) 16:10, 4 August 2008 (UTC)

It's a matter that has to be determined by discussion on article talk pages. There is no one rule that could be used. — Carl (CBM · talk) 17:07, 4 August 2008 (UTC)

But how would the discussion determine what the reputation is? Richard Blatant (talk) 17:43, 4 August 2008 (UTC)

Evidence & consensus are helpful. Antelan 17:48, 4 August 2008 (UTC)

So is it the reputation among us editors or we would be determining the reputation among others besides us? Does it matter what reputation we personally have for a source? Or should we be looking outward for signs of reputation among the general population or scholars or what? Richard Blatant (talk) 18:25, 4 August 2008 (UTC)

The opinion that individual editors have about the source will of course be relevant, but convincing others is often easier if you present external evidence of reliability (for example, is the source cited in academic work?). There's no simple formula, because there are many shades of reliability, and a particular source may be more reliable in some areas than in others. — Carl (CBM · talk) 19:09, 4 August 2008 (UTC)

Continued: proposal for first para of "2.1 Reliable sources"

This was archived without any opinions expressing disagreement with the current proposal, but I wanted to continue the discussion before making any edits, because I know that people sometimes check this page infrequently. The proposal as it currently stands is as follows:

Articles should only present rely on reliable, information that could be found in at least one third-party published sources publication with a reputation for fact-checking and accuracy. Such publications are considered to be "reliable sources". Information is verifiable even if the reliable source citing it or presenting it does not agree with the information. While citations are strongly encouraged, Wikipedia editors may present uncited information if the information could be found in a reliable source, or is not controversial. A publication that is not initially considered reliable may become reliable if verified by subsequent fact-checking or analysis by a reliable source.

And once again, the point of the proposal is just to make clear certain things that are pretty obvious to most everybody who is familiar with the policy and how it works in practice, and which ought to be spelled out more clearly. There's no intent here to make any changes in the current practice, so if something would be a departure from present practice, please let me know, so we can conform. COGDEN 02:21, 5 August 2008 (UTC)

In general I fail to see anything here that is a constructive addition to the current policy. Your taking a single sentence “Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy.” and expanding it out to a paragraph. Why? “the proposal is just to make clear certain things” I don’t see how this accomplishes its stated goal.

Here is a short list of problems that I see just on first reading:

1) We rely upon a source, because we are summarizing the information. 2) “could be” is unnecessarily vague. 3) What’s the values of stating “in at least one”? 4) We relay upon “published sources” and not simply publications. Publication to a lot of people equates to “printed” even if you and I see it as being more encompassing. Instead we use more encompassing language at the outset for clarity. 5) “Such publications are considered to be "reliable sources".” No, “fact-checking and accuracy” are the threshold and serve us well, but it’s left to the judgment of the editors working on and particular article. No sources is ironclad in its reliability, which is what the statement is saying. 6) “Information is verifiable even if the reliable source citing it or presenting it does not agree with the information.” No. I have no idea as to how you came to include this. If the source does not agree with the information, then we have no reliable source for the information. Please feel free to elaborate on this point, as I must be missing something. 7) “While citations are strongly encouraged, Wikipedia editors may present uncited information if the information could be found in a reliable source, or is not controversial.” This needs a rewrite. As written this would allow the introduction of OR. 8) “A publication that is not initially considered reliable may become reliable if verified by subsequent fact-checking or analysis by a reliable source.” No, the information may be verified; that a single source used the publication as a source does not make the publication itself any more reliable. Again, please feel free to elaborate. Brimba (talk) 05:19, 5 August 2008 (UTC)

Agree with Brimba's assessment. I do not see how this proposed wording helps, on the contrary: it obfuscates. ≈ jossi ≈ (talk) 05:21, 5 August 2008 (UTC)

Okay, obviously there are too many concerns here for the current language to pass the inertia bar, but let me address some of these issues, so that maybe there can be some dialogue on this:

1. "Rely" is the wrong word. That we are to "rely" on sources does not tell us whether we can, or must, cite them. If we "rely" on a source but fail to cite it, is that okay? Yes it is, but the present language might confuse me on that point. I might think that in addition to "reliance" on the source, I actually have to cite it, but Wikipedia has never required the citation of sources as a blanket proposition.
2. "Could be" simply reflects the fact that you do not have to have already found a reliable source in order to present information in a Wikipedia article. It is good enough that one could be found.
3. agreed
4. I don't oppose using "published sources" rather than "publication"
5. I don't understand your comment, but agree the sentence in question is not important.
6. A reliable source can discuss something without agreeing with it, and still be a reliable source for the topic. For example, if you are making a point about creationism, there is no reason you cannot cite a work by a rationalist author who is describing creationist views in an NPOV way. I've seen lots of these kinds of citations, and they are generally non-controversial as long as they are NPOV.
7. "While citations are strongly encouraged, Wikipedia editors may present uncited information if the information could be found in a reliable source, or is not controversial" (let's change "controversial" here to "likely to be challenged") would not allow OR. The standard is whether the information can be found in a reliable published source. If it can, it is by definition: #1 verifiable, and #2 not original research.
8. The point of this sentence is that reliability may change over time. If something is not verifiable now, further verification might verify it. If something is not reliable now because of a lack of independent fact checking, when further publications that do the independent fact checking can prove the original source was reliable. It's as simple as that, and really, it all depends on how the source is treated by modern literature by experts in the field. If they consider it reliable now, it doesn't matter that 20 years ago, people considered it to be questionable or even unreliable.

COGDEN 23:16, 6 August 2008 (UTC)

1) "Rely" is the correct word, because we are summarizing, sometimes from more then a single source. Present implies a more direct transfer.

2) Material that is challenge must have a source. In a dispute concerning sourcing, the wording “could be” implies that there is a reasonable assumption that someone could find a source if they where sufficiently diligent; but that it is not required. Your statement “"Could be" simply reflects the fact that you do not have to have already found a reliable source in order to present information in a Wikipedia article. It is good enough that one could be found.” is correct for marital that has not been challenged and falls under common knowledge; however, it is entirely incorrect for material that is or is likely to be challenge. That distinction is not clear or more accurately, not there.

6) Your point may be valid, but is not reflected in the wording you are proposing.

7) I am going to reply to this later. I wrote out a response and now I look at it, I see that I responded to your response, instead of to the actual wording as you proposed it.

8) I fully disagree. “when further publications that do the independent fact checking” we cite those as the source. If the reliability of the original source is in dispute, we don’t cite that as a source. “If they consider it reliable now” there would have to be a paper trail, and we would cite to that. Brimba (talk) 01:59, 7 August 2008 (UTC)

Transclusion

I've transcluded two intro paragraphs so that they may be transcluded elsewhere as examples (the lead being probably one of the most quoted paragraphs on Wikipedia). This will allow changes to be made in one place, so that every page where the section is shown as an example would be updated automatically. (An example is at WP:AADD#I like it.)

As for this page, whether the text is directly on the page, or transcluded to the page should make no difference to anyone reading it.

So all of this said, (and since, per WP:BRD, someone reverted the transclusion - so I'm posting here) are there any thoughts or concerns about this? - jc37 06:28, 6 August 2008 (UTC)

Most people cut and paste. Yes they could use your system, but really how many would? I would venture to say few. As it fits seamlessly into the main page, most would be entirely unaware that it even existed.

This creates duel talking pages for a single policy page; and people wishing to edit that section of the policy may not be able to if they did not understand what was going on. I am familiar with the process and it took me a few minutes to figure out what was going on.

“This will allow changes to be made in one place, so that every page where the section is shown as an example would be updated automatically.” This would eliminate the paper trail that normally exits on talk pages. If a straight cut and paste is used, then someone coming along after could still follow the conversation even if the policy had changed or material that was addressed had been moved further down the page. Brimba (talk) 02:21, 7 August 2008 (UTC)

You can talk page a sub-page on the main page's talk page. WikiProjects do that all the time. (As a matter of fact, a redirect can be placed on a sub-page's talk page. Another "common practice".)

And eliminate a paper trail? Every edit is saved. So I'm not sure what you're talking about.

And to give you an example of a place where it seems to have worked fairly well: Wikipedia:Userboxes. It's a rather large page that's split into several subpages transcluded to the main page. (And the subpages have an edit history. Editors seem to be able to find them just fine.)

Or how about the various XfD pages? They've been operating on transclusion (sometimes 5 or more pages deep) for quite some time.

I also set up the noticeboard at WP:CMC/NB. And it seems to be working quite well.

So no, I don't think this is "out-of-the-ordinary". And since the section is just that important, it would be useful for quoting, and quoting consistant to the current version.

That said, after you look over WP:UBX, WP:CMC/NB, WP:AFD, WP:MFD, WP:DRV, etc., I welcome further discussion on this. - jc37 08:06, 7 August 2008 (UTC)

And eliminate a paper trail? Every edit is saved. So I'm not sure what you're talking about.

No, under your system only the link is saved, what that link actually showed at the time it was used in a discussion would not be. The link would only show the current wording at the time the link was followed, which may or may not be the same wording as when the discussion took place.

That it works other places is great. In relation to a core policy page, I do not see how the positives out weigh the negatives. Brimba (talk) 12:16, 7 August 2008 (UTC)

Well, as yet I haven't heard any negatives. (Perhaps I'm missing them?)

For example, the sub-page also has a contribution history. So yes, I would presume that it would be actually very easy to determine how any page looked at any particular time.

But all this aside, what's your actual concern? Is this merely a question of preference? - jc37 04:30, 8 August 2008 (UTC)

My concerns are sufficiently laid out above. Your response that: “the sub-page also has a contribution history. So yes, I would presume that it would be actually very easy to determine how any page looked at any particular time.” fails to even acknowledge that a legitimate question (or negative if you prefer that word) has been raised. In reality looking at two separate page histories to determine how one page looked like on a particular date is certainly not “very easy”, and is an undue burden that would be imposed upon editors if we adopted your system. Having a single uninterrupted history that someone could look at without having to mentally superimpose the history from one page over the history a second page, that’s what’s easy and that’s why we use the current system. As with any overly engineered solution, your systems complexity and the unintended consequences of that complexity are the root negatives. See the KISS principle (acronym for "Keep It Simple, Stupid"). Brimba (talk) 14:26, 8 August 2008 (UTC)

So you oppose the use of transclusion on any policy page? WHat about headers? What about navboxes? I can think of quite a few pages which transclude introdutory sections. (WP:RFA is the first which comes to mind.) So far this sounds like personal preference.

Undue complexity? This makes it easier to use the section for examples, and makes those examples consistant everywhere used. So I would presume that "accuracy" would be of more value that presumed simplicity.

And again, WP:UBX has had no problems having editors find that page, and edit the various transcluded sub-pages. And I would venture to suggest that many who edit there are brand new Wikipedians. (Having only just made their userpage, and added one or more userboxes. Then discovering the guideline page, and editing from there.)

So no, I think that quite easily disproves your "complexity" theory.

Are there any other reasons? - jc37 22:08, 8 August 2008 (UTC)

I agree with Brimba here... for proposed changes to Policy it is best to keep everything on one page. Blueboar (talk) 14:34, 8 August 2008 (UTC)

Everything will still show on a single page. And it will be simple enough to edit a transcluded page. Note that all transclusion links are displayed at the bottom of a page being edited. So this isn't any more difficult than clicking a link, and editing another page. Wikipedians do that all the time. - jc37 22:08, 8 August 2008 (UTC)

Web sources that block some web browsers or operating systems?

A debate has arisen at Talk:Silkroad Online over whether a notice should be included in an External Links or References section, if the linked web sites have been demonstrated to be blocking certain web browsers or operating systems. I wasn't able to find anything in the policy that specifically addresses this, but perhaps the policy on avoiding, or in the alternative, directly quoting, non-English sources -- similarly non-readable to some Wikipedia visitors -- can serve as a guide? Matt Fitzpatrick (talk) 14:32, 6 August 2008 (UTC)

In the same vein, many sources use add-in formats, like Flash, that are inaccessible to many readers. It comes down to the question of whether every reader should be able to do the verification or if it's good enough that some editors can. For paper sources, we clearly settle for the latter even though the former is a nice to have.LeadSongDog (talk) 15:10, 6 August 2008 (UTC)

The user-agent thing is a minor inconvenience, since it's so easy to send a fake user agent. Firefox has extensions to make it a one-click process. The more serious concern is that the article is sourced only to the game's website. If the game is notable, it has been covered in reliable sources independent of the game itself, such as major media outlets. Those reliable sources should be used, rather than forums on the game's website. — Carl (CBM · talk) 15:31, 6 August 2008 (UTC)

Verifiability has never meant that "you" can verify something. Nor does it require that things be verifiable right this second. As long as a source is able be verified by the imaginary "anyone" after a reasonable amount of dilligence, it meets our criteria. So... the fact that a website can not be accessed by a specific browser or operating system does not really change its verifiability. It is not that difficult for "anyone" to borrow another computer (one with a browser or OS that is compatable). Alternatively, "anyone" can post a message at the Village Pump, asking some other Wikipedian to verify it for him/her.

That said, I do think it is common courtesy to include a note in the citation or link if there are known browser/OS limitations... that way those who can not access the information will not waste their time trying to doing so. Blueboar (talk) 20:51, 6 August 2008 (UTC)

For external links that aren't references you might look at What not to link #7 of the external links guidelines. If it's not accessible to readers because of their browser etc. then it probably shouldn't be linked. -- SiobhanHansa 14:56, 8 August 2008 (UTC)

WP:V and WP:ELNO address different issues. If one has to choose between a source from a well-regarded expert using outdated web technology and a source from a marginally notable private web site that's readily accessible, WP:V doesn't really care much about the delivery technology, the source's reputation and expertise is much more important. WP:ELNO takes technology and accessiblity into account. Best, --Shirahadasha (talk) 16:15, 8 August 2008 (UTC)

Questionable sources

This is not clear: "Questionable sources are those with a poor reputation for fact-checking. Such sources include websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions." Is it saying that ALL "websites and publications that express views that are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions" are questionable sources? Do these all have a poor reputation for fact checking? Or is it saying that things that have a poor reputation for fact-checking MAY include ..? That is, can something extremist or promotional, be a reliable source? Richard Blatant (talk) 21:14, 2 August 2008 (UTC)

Speaking in general terms, I think there are very few websites that "are widely acknowledged as extremist, are promotional in nature, or rely heavily on rumors and personal opinions" but are nevertheless reliable sources. Websites like that should be treated with great caution. — Carl (CBM · talk) 00:57, 3 August 2008 (UTC)

Thanks for you opinion, but I'm not sure if you're telling me the policy or if you're disagreeing with the policy. Does the official policy allow "widely acknowledged as extremist, are promotional in nature" to be used as reliable sources? Richard Blatant (talk) 02:56, 3 August 2008 (UTC)

No, sources that are promotional in nature or widely acknowledged as extremist are not the sort of sources that the policy calls "reliable". — Carl (CBM · talk) 03:10, 3 August 2008 (UTC)

Thanks. It looks like it could use more precise wording then. Richard Blatant (talk) 03:43, 3 August 2008 (UTC)

Actually, such sources can be considered reliable in very limited situations... for statements as to the beliefs or opinions of the group that holds them. It really depends on why and how you are using the source. If the article on a particular extreamist neo-Nazi group (for example) states that they believe that "Arians" are smarter than other people, it is appropriate to cite their website, where the group actually states this belief. But it would not be reliable for the statement that "Arians" actually are smarter than others. In other words, such sources can be reliable as a citation for statements of opinion, but are not reliable for statements of fact. Hence the caution. Blueboar (talk) 14:15, 3 August 2008 (UTC)

I might add that Paul Krugman's New York Times opinion pieces and, on the other side, the Wall Street Journal's editorials, are obviously "promotional" and "rely heavily on personal opinion" but that doesn't mean they may NEVER be used for simple facts like income for Americans in the 99th percentile grew 1.2% last quarter. It's one thing to advise caution and another thing to have a blanket prohibition that doesn't permit editors to examine source reliability on a case-by-case basis.Bdell555 (talk) 14:33, 9 August 2008 (UTC)