Wikipedia talk:Mediation Cabal/Cases/27 February 2012/Wikipedia:Verifiability/Group 2

This is the discussion page for work group two in the verifiability mediation. Note: The previous guidelines for this page are no longer in effect. If you would like to add new threads, feel free. —  Mr. Stradivarius ♫ 22:21, 8 April 2012 (UTC)

Improvements to draft 5
From the results of step 5, there seems to be a rough consensus to use draft 5 from this group in the RfC. This is not to say that the wording of the draft is now set in stone, far from it. What this means is that now we can forget drafts that are substantially different from draft 5, and concentrate on making incremental improvements to the draft 5 wording so that we can form a strong consensus. This process was already underway during step five, so all you need to do is continue it here. I know that I've archived those threads, but you shouldn't read anything into that. The archiving was just to make the page look neater, and you should continue where you left off. The page protection on the drafting page has now expired, so please start submitting more drafts using the improvements that you discuss here (and have already discussed in the archives). Best —  Mr. Stradivarius ♫ 22:21, 8 April 2012 (UTC)


 * I've been closely watching the comments about Draft 5 since the group draft page was protected. I'm going to try to summarize the things that I'm aware of here, and also ask some questions that I hope other editors will try to answer.
 * Paragraph 1, sentence 3: There seems to be a strong agreement amongst Blueboar, North, and me, that "nothing, such as perceived truth or personal experience" should be revised to "nothing, such as your personal experience or what you know to be true". In my opinion, we should do that. Are there any objections to that?
 * Paragraph 1, sentence 2: On the statements page, Kalidasa objected to the phrase "verifiable but inaccurate". I'm not exactly sure how to fix this, or whether it's a serious enough issue to require change. One possibility would be to shorten it to "influence when verifiable material should not be included"; another would be to shorten it to "influence when inaccurate material should not be included". The first keeps the focus on V but does less to explain the "transcription monkey" concern. The second explains the issue, but implies wrongly that inaccurate material sometimes should be included, so I don't like it. Thoughts?
 * Paragraph 1, sentence 2: Kalidasa also would prefer not to refer in a general way to policies, etc., but instead to name NPOV, DUE, NOR, and copyright policy specifically. On the other hand, Blueboar has argued for including the word "considerations" precisely because not everything comes directly from policies. We can't have it both ways. Paragraph 3 does spell out the policies. Is that good enough? Are there alternative solutions?
 * Paragraph 1, sentences 3 and 4: On the statements page, Mangoe objected (in general, about multiple drafts from multiple groups) to language about editors not substituting personal beliefs for verifiability. My take is that it would be better to address that issue in the Group 3 or Group 4 drafts than here. Thoughts?
 * --Tryptofish (talk) 21:23, 9 April 2012 (UTC)

Para 1 Sent 3 I think that the first change should be made, but still be open for tweaking. Not sure that the wording is yet optimal but it resolves something that both Blueboar and I feel is a problem.

Also in the footnote I think that "we are verifying the statements in Wikipedia" is mystifying/unclear and needs a tweak to clarify. North8000 (talk) 00:55, 10 April 2012 (UTC)
 * For the first point, it seems to me that we have consensus for that, but subject to improving it further. (I don't have any ideas about how to do that, nor do I really see a need, so someone else will need to make specific suggestions about the wording.) I've crossed out the existing wording, below, and shown the tentative new wording in green.
 * For the second point (about the footnote), now that you point it out, I agree. How about changing: "In so doing, we are verifying the statements in Wikipedia, not the facts on which they are based, nor whether the sources are correct that they are true" to: "In so doing, we are verifying that the statements in Wikipedia exist in the cited sources, as opposed to verifying the facts on which those statements are based or verifying that the sources are correct that those statements are true"? --Tryptofish (talk) 21:09, 10 April 2012 (UTC)
 * Yes, that helps a lot.   Let's try it. North8000 (talk) 21:33, 11 April 2012 (UTC)

Should we preserve 5 and make a copy to Version #5.1 for these edits? North8000 (talk) 01:03, 10 April 2012 (UTC)
 * You should preserve draft 5, yes, but there's no need to make up new numbering schemes for step six. Just submit the new text in the next available draft slot. I trust that everyone here can recognise if a new draft is based on draft 5 or not without calling it draft 5.1, 5.2, etc. —  Mr. Stradivarius on tour ♫ 08:48, 10 April 2012 (UTC)
 * I copied draft 5 to draft 9 to allow tweaks while preserving the original. North8000 (talk) 18:54, 10 April 2012 (UTC)


 * Paragraph 1, sentence 2: This is about what the policy should say about treatment of verifiable material whose accuracy is challenged. Under the heading "Accuracy" on the general talk page of this mediation, there is a large and growing thread about this very question, though it begins from another draft wording issue. People who have written in that thread include BeCritical, Mangoe, North, Unscintillating, Pesky, Blueboar, and myself. Probably not necessary to repeat here everything that's been said there. Just to say that yes, the treatment of "verifiable but inaccurate" content is indeed a serious issue, a hugely serious one... Kalidasa 777 (talk) 02:41, 10 April 2012 (UTC)
 * I am thinking that we need two separate statements to deal with the issue of accuracy and verifiablitiy... one to deal with adding information (a variation on the VNT language), and another to deal with when it is ok/not ok to remove information.
 * We all seem to agree on a statement that says (in effect) "don't add unverifiable information, even if you are 100% positive that it is true/accurate/etc." ... what we probably need is another statement that says (in effect) "If you think verifiable information that is already included in an article is inaccurate, you may remove or change it as needed... but only after discussion and only with consensus".  (Note... I am not suggesting this as specific language... I am talking about concept and structure at this point). Blueboar (talk) 13:26, 10 April 2012 (UTC)
 * I have to confess that the "accuracy" discussion is one of those things that I've tuned out as tl;dr. I'm not really sure what we should be doing – specifically – about the wording of this draft. What I took from Kalidasa's statement on the main page seems to be something different from what Blueboar is talking about (objection to the phrase "verifiable but inaccurate", as sounding too much like a slogan, versus adding advice about when to remove material).
 * As for the additional material that Blueboar outlines, I think the best place to put it would actually be in the second paragraph, right after the sentence that says: "Any material that requires a citation but does not have one may be removed." One option would be to make a footnote at the end of the sentence, which would allow us to make the footnote more lengthy without making the lead overly wordy. In that case, the new footnote could actually be pretty close to what Blueboar wrote just above. --Tryptofish (talk) 21:25, 10 April 2012 (UTC)

I've revised Draft 9 to reflect the changes that are shown below. Before we decide that we are done here, let me ask whether we should also add that sentence that Blueboar proposes above (or something similar to it). I'm fine with it either way, so, Blueboar, if you feel strongly, now would be a good time to say so. --Tryptofish (talk) 20:48, 14 April 2012 (UTC)
 * I don't know how Blueboar feels, but I do feel strongly that more needs to be said about the "verifiable but inaccurate" question, if the page is going to mention it at all. Draft 9 does mention this question but does it do so in a helpful way? Please put yourself into the shoes of an editor who has found something he/she considers "verifiable but inaccurate"; or put yourself into the shoes of an editor who has been told by another editor, "hey, that point may be verifiable, but it is not accurate". So the editor goes to the policy page and is told that "other policies, guidelines, and considerations... influence when verifiable but inaccurate material should not be included". How helpful would you find that? Wouldn't you want to know what those "policies, guidelines and considerations" actually said about "when..." etc??
 * The point has been made several times that the lede should not talk about topics that are not in the body. Well, right now, Draft 9 does just that -- the lede mentions the accuracy topic, though the body does not. So why not include the sentence proposed by Blueboar as part of the body ? That way the lede is kept short, and lede and body cover the same topics. Kalidasa 777 (talk) 02:26, 15 April 2012 (UTC)
 * Three thoughts. First, could you clarify which sentence you feel should be added?  Second, I agree with you that that is an immensely important topic that must be dealt with in Wikipedia. Third I'm just not so sure that the lead of wp:ver is the place to try to tackle that, but maybe one well written new or tweaked sentence on it here would be good. Sincerely, North8000 (talk) 10:50, 15 April 2012 (UTC)
 * As I said, it's fine with me either way. I just would like to see someone suggest exactly how to say it. --Tryptofish (talk) 16:31, 15 April 2012 (UTC)
 * North, I agree that the lede of WP is not the main place where this should be tackled. I think there should be a new short section in the body of WP:V.


 * What specifically wants to be said:


 * Quoting Blueboar, from the general talk page: "If you feel that verifiable information is inaccurate, do not remove it until the issue has been fully discussed on the talk page and a consensus for removal has been established. Keep in mind that rewriting how the material is presented is often a better choice than removing it entirely."


 * Even though Blueboar said he was proposing a concept, rather than specific language, I actually think the language he's used is rather good. Still, if anyone can improve the language while keeping the concept, so much the better.


 * My own preferred option would actually be to say a bit more, as has been done in Group 3 Draft 15. However, if you want to keep it down to a sentence or two, then I would suggest Blueboar's words above. Kalidasa 777 (talk) 00:23, 16 April 2012 (UTC)
 * I'm really torn on this. I'd love to see that very important material get added to a major policy.   But I think that anything that tries to go very far in 1-2 sentences in the wp:ver lead is going to do it problematically and I fear could be a poison pill for acceptance of this version. Even Blueboar's version has technical flaws (does not cover major exceptions), due only to it's brevity. North8000 (talk) 00:33, 16 April 2012 (UTC)
 * How about making that passage a footnote, instead of putting it in the main text? That way, we aren't adding it to the main text of the lead, but we are still providing the information. I think we could put the footnote in the second paragraph, right after the sentence that says: "Any material that requires a citation but does not have one may be removed." --Tryptofish (talk) 22:41, 16 April 2012 (UTC)
 * Looking at it in situ (below), something doesn't look right to me. That "do not remove it" is probably too absolute, too prescriptive, and that could be where North's concern about a poison pill could materialize. I'd rather make it more like a suggestion of good practices, instead of a seemingly mandatory policy. At the risk of complicating the color scheme, I'll suggest a fix below. --Tryptofish (talk) 22:48, 16 April 2012 (UTC)
 * Looks good. A footnote is safer, as is the less prescriptive wording. North8000 (talk) 01:00, 17 April 2012 (UTC)
 * Good, thanks, it looks good to me too. One additional, and very minor, tweak that occurs to me now is that only the word "inaccurate" really needs to be in italics, so I'd be inclined to remove those other italics. I'm going to put it into Draft 9 that way now. We then should all scrutinize that draft, and see whether we think there is anything remaining to be fixed. --Tryptofish (talk) 15:20, 17 April 2012 (UTC)

Footnote #2
I found this footnote to be very confusing. After I read it about 10 times I finally figured it out and why it is confusing. It covers / blends about three different topics, all good points:
 * 1) The distinction between sourcable and sourced
 * 2) The distinction between verifying that the text is sourcable vs verifying the facts stated in the text
 * 3) The logical distinction between saying "XXXX" vs. saying "Some people say XXXX" statements

Only #1 is relevant to the sentence in the lead which cited it. Possibly we should disentangle this footnote? North8000 (talk) 10:42, 11 April 2012 (UTC)
 * Sure, I'm very receptive to that. I cobbled it together from several comments by other editors, basically just trying to accommodate what they seemed to want. But I'm lost as to what (if anything!) to change it to, so someone else is going to have to suggest specific wording. And please see also what I said about it in my reply to you above, where I did suggest one aspect of a possible rewrite. --Tryptofish (talk) 19:49, 11 April 2012 (UTC)
 * I am all for untangling policy. The rub is always in how we try to untangle it, but I am open to ideas. Blueboar (talk) 22:55, 11 April 2012 (UTC)
 * I think that the edit in the draft fixed the primary problem. But I think that we should also drop the last sentence in that footnote.  It is trying to get into the topic of  the logical distinction between saying "XXXX" vs. saying "Some people say XXXX" statements, but conflates an incorrect statement about verifiability into it.  A false statement (e.g. the moon is made of green cheese)may indeed be wp:verifiable.  North8000 (talk) 12:49, 12 April 2012 (UTC)
 * I think the policy should discuss the distinction between the verifiability of a statement of fact vs the verifiability of a statement of opinion. In many cases, information is unverifiable (and thus unacceptable) if expressed as a statement of fact ("Jesus is the Son of God")... but can be made very verifiable (and very acceptable) if turned into an attributed statement of opinion ("Christians believe that Jesus is the Son of God").  The question for me isn't whether to discuss the distinction... but where to discuss it... I could definitely see the argument that it does not belong in the lede... perhaps we should propose moving it to a point later in the policy? Blueboar (talk) 14:14, 12 April 2012 (UTC)
 * Agree that such is an immensely important topic that should be covered in Wikipedia. Maybe in wp:ver would be fine, maybe even in the lead.   But sentence that I BRD-struck conflated (and even your post did so) the topics of wp:verifiability wording to properly handle false or questionable statement.   For example, if a WP:RS said simply "The moon is made of green cheese"  then that statement as-written is wp:verifiable, and, if cited, it is wp:verified. So wp:verifiability allows it to be put in.   If there is an ongoing dispute about the material of the moon and one side says "green cheese" then wp:npov requires that the statement be put in.  Otherwise the editors can decide to simply leave the material out; no policy should prevent them from doing so.  IMHO they should be free to do so.   North8000 (talk) 14:46, 12 April 2012 (UTC)
 * Here is where I strongly disagree with how North interprets several policies and guidelines... First, is unlikely that any source that asserts "the moon is made of green cheese" as a fact would ever be considered an RS in the fist place... but, assuming for the sake of argument that for some reason it was considered reliable, the fact that the source asserts something as fact does not automatically make the assertion itself wp:verifiable... Instead, it makes the statement "According to   , the moon is made of green cheese" wp:verifiable.
 * Second: WP:NPOV does not necessarily require that we include the statement (we can exclude it if the statement gives WP:UNDUE to a fringe view). (Or rather, to make this more complex... that second "statement of opinion" might be required in one article and not in another - for example it might be required in Alternative views on lunar composition, but not in Geology of the moon).
 * The problem, of course, is that this is all quite complicated... to unravel the complexity editors have to understand how multiple policies and guidelines interact and relate to each other. I do agree that it is too complex to deal with in a one sentence foot note... it may even be too complex to discuss in the lede.  However, it does need to be discussed somewhere in the WP:V policy. Blueboar (talk) 15:43, 12 April 2012 (UTC)
 * Well, we disagree less than you think .   But on your first paragraph we do.  IMHO, if the statement is sourcable/sourced to a wp:RS, then it is wp:verifiable/wp:verified. On your second paragraph, I agree that you are certainly right. My statement was an oversimplification (= inaccurate) just to make the point that there are special  cases where wp:npov mandates inclusion of material even if contrary to editor consensus.  And I 100% agree on your last paragraph. North8000 (talk) 16:24, 12 April 2012 (UTC)
 * I'm closer to agreeing with Blueboar than with North on this one, but my instinct is to look for a middle ground. I'm also thinking about comments from another editor who, earlier in this mediation process, said that they strongly wanted something like this sentence to be here. Therefore, I'd rather try to fix the sentence to be (I hope!) responsive to North's concern, instead of deleting it entirely. I've given that a try, below, so please take a look and see whether or not it helps. --Tryptofish (talk) 17:49, 12 April 2012 (UTC)
 * A final thought... I think the issue of how to deal with an RS that makes a potentially inaccurate statement relates more to how we cover the information than it does whether we cover it. Blueboar (talk) 18:35, 12 April 2012 (UTC)
 * Trptofish, Either way, please don't hesitate to undo my strike if you disagree. North8000 (talk) 20:32, 12 April 2012 (UTC)

Working area for changes to the text
OK... I am finding it very annoying to have to keep switching pages whenever someone refers us to "Paragraph x, sentence y"... To avoid this, I am posting the text of Draft 5 onto this talk page. My suggestion is that this be a working space... which means we can make changes here as we go along (but please don't be overly bold and make changes unless they have been fully discussed). Blueboar (talk) 12:58, 10 April 2012 (UTC)
 * Hmm, you're not the first person here to say that they are annoyed with the draft page structure... So let's give this a try and see if it helps us be more productive. As long as everyone can use the process to get some results, that's the main thing. —  Mr. Stradivarius ♫ 13:20, 10 April 2012 (UTC)
 * This is altogether a minor comment, verging on one you all can safely disregard, but I have found it useful to open a second tab in my browser, and display the talk page in one tab and the drafts page in the other. --Tryptofish (talk) 20:46, 10 April 2012 (UTC)
 * Responding to Blueboar's comment.   I think we have to find the middle ground on boldness.    If folks are too bold we'll make a mess out of it, and this version (as already drafted) it think received the strongest support of any version in any group.   But, if everyone is too cautions, we'll never get this done, and never get the little bugs worked out of this version.    My suggestion:  if you have support for your exact proposed change from at least one other person, BRD it in here on the working copy.  And the ones that stick we'll move to draft 9 which is the new copy of the now-preserved draft 5. North8000 (talk) 09:13, 11 April 2012 (UTC)
 * My point was to say that things should be discussed a bit before we impliment... I think we can be bold in discussion, but cautious in editing. Give people a chance to reply to an idea (and potentially raise objections or concerns) before you rush off and implement it. Blueboar (talk) 23:00, 11 April 2012 (UTC)
 * Let me barge in here and say that I think that you should feel free to create new drafts on the draft page even when the difference between your draft and another one is really tiny. This way you can be bold in editing and bold in discussion, as it were. I also want to warn against too much use of small incremental changes, as this process tends to increase the length of drafts. One common objection at the Oct-Dec 2011 RfC was that the proposed version increased the length of the policy... —  Mr. Stradivarius ♫ 10:22, 12 April 2012 (UTC)
 * About your warning concerning incremental edits lengthening the draft too much, are you saying that you think that we are doing that here? If you would be comfortable saying so, can you point to where in the draft you feel that it is becoming a problem? --Tryptofish (talk) 17:53, 12 April 2012 (UTC)
 * Ah... creating new drafts (to show others what you desire and how it would fit into the existing text) is indeed fine (and helpful)... what I was talking about is slowing down on implementing desired changes into the text of the "final draft" - to give others a chance to hold a reasonable discussion, and determine if there is a reasonable indication that the desired change has consensus. Blueboar (talk) 14:42, 12 April 2012 (UTC)
 * I want to emphasize that everything I'm bringing up at this point on this draft is just minor tweaks in areas that I consider secondary. Just hoping to get this near-perfect (for what it is directed to be) version to perfection. I heard that there was a a huge $$$$$$$ prize for the selected version and figured that I would need to suggest at least some little changes in order to get a part of the prize money.   My only concern is that everyone here is being so polite and so cautious that we might never get this done.  :-) North8000 (talk) 21:23, 12 April 2012 (UTC)
 * Well, let me (politely and cautiously!) suggest that, with the revisions shown below, we in this group are done. I think it's good enough to go forward with, and maybe there's nothing more to be gained by staring at it further. --Tryptofish (talk) 20:20, 13 April 2012 (UTC)
 * I think so. North8000 (talk) 20:23, 13 April 2012 (UTC)

{| style="background:#DCFFA0"
 * Verifiability on Wikipedia is the reader's ability to check reliable sources that directly support the information in an article. All information in Wikipedia must be verifiable, but because other policies, guidelines, and considerations also influence content, and particularly influence when verifiable but inaccurate material should not be included, verifiability by itself does not guarantee inclusion. Verifiability, not truth, is one of the key requirements for inclusion in Wikipedia— nothing, such as perceived truth or personal experience, can be a substitute for meeting the verifiability requirement. No matter how convinced you are that something is true, do not add it to an article unless it is also verifiable.
 * Verifiability on Wikipedia is the reader's ability to check reliable sources that directly support the information in an article. All information in Wikipedia must be verifiable, but because other policies, guidelines, and considerations also influence content, and particularly influence when verifiable but inaccurate material should not be included, verifiability by itself does not guarantee inclusion. Verifiability, not truth, is one of the key requirements for inclusion in Wikipedia— nothing, such as perceived truth or personal experience, can be a substitute for meeting the verifiability requirement. No matter how convinced you are that something is true, do not add it to an article unless it is also verifiable.

It must be possible to attribute all information in Wikipedia to reliable, published sources that are appropriate for the content in question. However, in practice it is only necessary to provide inline citations for quotations and for any information that has been challenged or that is likely to be challenged. Appropriate citations guarantee that the information is not original research, and allow readers and editors to check the source material for themselves. Any material that requires a citation but does not have one may be removed. Unsourced contentious material about living people must be removed immediately. For help on adding citations, see Citing sources. This policy applies to all material in the mainspace.

Verifiability, No original research and Neutral point of view are Wikipedia's core content policies. They work together to determine content, so editors should understand the key points of all three. Articles must also comply with the copyright policy.

Comments on the current draft
As of 4/16/12 - foot note 1 seems to simply repeat the same words that are said in the main text. Personally, I prefer it in the footnote... if others agree, we should reformulate what is stated in the main text. Blueboar (talk) 15:45, 17 April 2012 (UTC)
 * The origin of that footnote is North's concern about specifically defining "not truth". North can speak to whether or not he would support deleting the footnote, but I'm guessing not. I would strongly oppose moving the wording of the footnote into the main text, for the reasons that I've already stated gazillions of times before (basically, that it sounds like a circular definition when put there). --Tryptofish (talk) 15:50, 17 April 2012 (UTC)
 * Yes, I consider the "means" statement to be VERY VERY important, and Tryptofish doesn't want it in the main text. And I'm trying to be nice to Tryptofish so that I can get some of the large $$$$$$$ prize if this version is used, so I settled for it in the footnote. :-)  North8000 (talk) 15:56, 17 April 2012 (UTC)
 * I'd like to add another footnote, making "being nice to Tryptofish" a new core policy. --Tryptofish (talk) 15:59, 17 April 2012 (UTC)
 * Nah... its more of a guideline
 * As I said, I prefer it as a footnote... but whether we say it in the main text or in a foot note, we should not repeat ourselves. The entire point of a footnote is to either expand on what is stated in the main text, or to restate what is said in the main text using different language... so that it clarifies the point being made. Merely repeating the same language is not helpful. Blueboar (talk) 16:04, 17 April 2012 (UTC)
 * Policy I say, policy!!
 * Well, it doesn't exactly repeat the same language, as far as I can see. In fact, I do see it as restating using different language, in a manner that expands upon the main text. At this point, I think we need you to state specifically what language changes you would propose to accomplish that. --Tryptofish (talk) 16:09, 17 April 2012 (UTC)
 * Main text:    nothing ... can be a substitute for meeting the verifiability requirement.
 * Footnote text: nothing ...is a substitute for meeting the verifiability requirement.
 * Seems repetitive to me. Blueboar (talk) 16:12, 17 April 2012 (UTC)
 * OK, now I see what you mean. How about leaving the main text as is, and changing the footnote to a slightly different wording, with the same essential meaning? Perhaps changing the footnote from "is a substitute for meeting" to "can take the place of". Would that work for everyone here? --Tryptofish (talk) 16:25, 17 April 2012 (UTC)
 * Why not simply omit it from the main text entirely and state it as a foot note?
 * Main text: Verifiability, not truth, is one of the key requirements for inclusion in Wikipedia[1]. No matter how convinced you are that something is true, do not add it to an article unless it is also verifiable.
 * Footnote: [1]In this long-standing description, "not truth" means that nothing (such as truth) is a substitute for meeting the verifiability requirement. See also the essay Verifiability, not truth.
 * (Note: If the real issue is the language of the parenthetical, I could accept either formulation) Blueboar (talk) 16:34, 17 April 2012 (UTC)
 * (edit conflict, responding only to previous post) Blueboar, when you are saying that they are the same, you you are overlooking the main operative word and verb in the footnote which is "means". Because of this (only) the text in the footnote is a positive statement that can be used to deflate other creative mis-interpretations of VNT, which is the most common problem with VNT. Sincerely, North8000 (talk) 16:40, 17 April 2012 (UTC)
 * , responding to Blueboar's latest post: That, then, is a different issue than the repetition of "can be a substitute for meeting", and it would put me strongly in the oppose position. Here's why. We do need, in the main text, to explain what VnT does and does not mean. As I see it, that's the entire point of everything that has happened from the previous RfC through the present date. If you want to pursue an approach like that by way of Group 4, go right ahead. But what you would leave out of the main text here is what I see as right at the core of what has to be in this draft, or we would be wasting our time. --Tryptofish (talk) 16:43, 17 April 2012 (UTC)
 * I totally agree that we need to explain what VnT means... but I think the sentence: "No matter how convinced you are that something is true, do not add it to an article unless it is also verifiable" does this quite adequately and clearly. I find the insertion of the "nothing is a substitute" sentence as distracting from that explanation. I guess I just don't see why the sentence you wish to add is so vital. Blueboar (talk) 17:00, 17 April 2012 (UTC)
 * OK, so (just trying to break this down to the specific issue) there is the question of repetition of the same idea between "nothing, such as your personal experience or what you know to be true, can be a substitute for meeting the verifiability requirement" and "No matter how convinced you are that something is true, do not add it to an article unless it is also verifiable." I guess I agree with you that the two sentences repeat the same concept, in somewhat different words. Yes, I agree with that. My take, however, is that this isn't a problem, but rather, a good thing. It's such an important point that the repetition makes it clearer, by, first, by explaining what the user needs to understand, and, second, explaining what the user needs to do. I see what we are doing in writing this as explaining V to the least sophisticated of editors, not to those of us who have been spending too much time in these discussions. It's kind of like good teaching. I accept what you say as a matter of logic, but as a matter of explanation I don't see it as a problem. --Tryptofish (talk) 17:26, 17 April 2012 (UTC)
 * (edit conflict, responding only to Blueboar's) What we're talking about isn't just about getting the general intent across, it is about clearly avoiding common accidental mis-interpreations of it, and to provide a tool to positively deflate deliberate misinterpretaiotns of it. The statement in the body does the former, the more explicit "means" statement in the footnote does the latter. Sincerely, North8000 (talk) 17:32, 17 April 2012 (UTC)
 * I suppose my problem is that I don't see the "nothing is a substitute" language as being such an important point to begin with ... It isn't wrong, which means I don't object to it being in a footnote ... but I do dislike it as part of the main text. It certainly isn't (to my mind) the key to understanding what VNT means. And I definitely don't think it is worth saying twice (with very subtle changes that will go over the heads of most readers).  Also, I don't really understand where the misinterpretation or misrepresentation could occur if we omit it from the main text and state it in the footnote instead. Blueboar (talk) 21:34, 17 April 2012 (UTC)
 * Although most people don't realize it (because it takes a rather abstract structural view to see it) that particular sentence (specifically the stronger version in the footnote) addresses and solves THE main problem with VNT, which is that it is prone to creative misinterpretations and misuses.  After spending about 500,000 words on the topic, and since we're dancing around the two most infamous logically superfluous words of all time ("not truth") IMHO I think we we could spare a few extra words in the foot note to double resolve it, :-)  especially since it is a fundamentally different statement with "means" as the verb.  Sincerely, North8000 (talk) 21:49, 17 April 2012 (UTC)
 * I understand (and respect) that this is how you feel... but I don't understand why you feel so strongly about it (and I am trying to do so). How does it address and solve the main problem with VNT?  What creative misinterpretations and misuses does it prevent?  I just don't see it.  Blueboar (talk) 21:54, 17 April 2012 (UTC)

Well, the starting point is to agree on the premise (as I think 95% do) that the ONLY thing that VNT means is to create and reinforce the rule that verifiability is a criteria for inclusion. Nearly all of the other crap that 95% want to get rid of (e.g. "accuracy NEVER matters", "our job is just to be transcriptions monkeys", that inaccuracy may not be discussed during content discussions, that sourcing is a force or mandate for inclusion) arise from somebody asserting that VNT means other additional things besides defining a requirement for inclusion. The footnote statement (with the operative word "means") says that VNT does not create or support those additional creations. Sincerely. North8000 (talk) 22:12, 17 April 2012 (UTC)
 * I kept that brief for clarity, but I would be VERY happy to discuss this in depth with you. I respect you very very very highly, and realize that you and I are talking two different languages when we analyze these things. I think that you view everything in terms of communicating the main intent, and figure that doing that is enough.  I view policies and their effects structurally, including an analysis of how policies get mis-used, with a focus on preventing that misuse. North8000 (talk) 22:19, 17 April 2012 (UTC)
 * And I, in turn, view all this stuff from the rather pedestrian perspective of just wanting to get consensus and be done with it! So, I'm very much in favor of achieving in the footnote what North is talking about. From what Blueboar is saying, part of the issue is not really seeing why North sees it as so important, and another part is disliking the use, twice, of the phrase "nothing is a substitute". I can't help with the first part, and For the first part, Blueboar, can you explain exactly what you mean about "but I do dislike it as part of the main text", where "it", specifically refers to the words "nothing is a substitute"? For the second part, please let me suggest, again, what I said earlier about using a slightly different wording ("can take the place of") the second time. --Tryptofish (talk) 23:24, 17 April 2012 (UTC)
 * I missed that the first time. Sounds good. North8000 (talk) 00:05, 18 April 2012 (UTC)
 * I don't understand how the sentence does what you think it does (even with the changed words)... HOW does adding....
 * "nothing (such as truth) can take the place of meeting the verifiability requirement"
 * ...actually tell someone not to interpret VNT as: "accuracy NEVER matters", "our job is just to be transcriptions monkeys", "inaccuracy may not be discussed during content discussions", "sourcing is a force or mandate for inclusion", or any of the other things you are talking about? Blueboar (talk) 01:13, 18 April 2012 (UTC)
 * Well, first, we're also talking about the more explicit "means" statement in the footnote. And my " "accuracy NEVER matters", "our job is just to be transcriptions monkeys", "inaccuracy may not be discussed during content discussions", "sourcing is a force or mandate for inclusion"" is the essence of their arguments, not their actual arguments.  Answering your question, they would help at two levels.  First the softer statement in the body which gives the core meaning of "not truth"   will tend to minimize the creation of other meanings.  Second, once one of the zillions of discussions starts  where someone has created a new unintended (for 95%) meaning for "not truth", someone can say "wikipedia doesn't say that, where did you get that?" and then they quote VNT, and then the other person can point to the clear "means" statement in the footnote and say "No, this shows that VNT does NOT say that" — Preceding unsigned comment added by North8000 (talk • contribs) 12:13, April 18, 2012‎
 * I disagree that it would minimize the creation of other meanings... In fact, I think it could actually encourage those other meanings. I could easily see someone who was (in essence) saying that "accuracy does not matter" pointing to this sentence and saying: "The text I want to add meets the verifiability requirement, and the policy explicitly says nothing is a substitute for that... this includes your argument that the text is inaccurate/untrue.  So, according to WP:V we must include it - because it is verifiable."
 * In short, I don't think the sentence actually achieves the goal you want it to achieve. And because I also find the wording a bit clunky, I question (not oppose) its inclusion. Blueboar (talk) 13:11, 18 April 2012 (UTC)
 * I suspect that it will always be possible for a willfully difficult editor to seize on a particular sentence and be, well, difficult. But I also think that the scenario to which Blueboar points is solvable by directing that difficult editor to the sentence just before the sentences we are discussing here, the one that points to the other policies etc., and says, explicitly, that verifiability does not guarantee inclusion. For me, what the sentence under discussion does accomplish is telling that difficult editor that, even if you have personal experience, even if you have what you are convinced is certain knowledge, you still need to show the rest of us that it is verifiable. That tells the difficult editor more than just what the final sentence of the paragraph does. And I think I have a pretty good ear for clunky versus lucid writing, but I really don't find it all that clunky. I'm glad that Blueboar is questioning rather than opposing it, and I hope that maybe we can get to a comfortable consensus. --Tryptofish (talk) 23:46, 18 April 2012 (UTC)
 * No... but perhaps an uncomfortable acceptance. (I really don't like the sentence, but if others insist on it I will not oppose.) Blueboar (talk) 23:50, 18 April 2012 (UTC)
 * Oh dear, I certainly don't want you to be uncomfortable! But I'm at a loss for what to do, since we are not going to simply delete it. If you can think of alternative wording (not deletion, but alternative wording!), I'm still happy to listen. --Tryptofish (talk) 23:54, 18 April 2012 (UTC)
 * Well, I never advocated for deletion... I have no problem with it in a footnote. Blueboar (talk) 00:40, 19 April 2012 (UTC)

I think it's time to turn our work on the markup into a draft. (?) North8000 (talk) 12:08, 23 April 2012 (UTC)
 * I couldn't agree more! In fact, I've been doing that as we go along, once I've been reasonably sure that we have agreed about the proposed changes here. Please see Draft 9. And please look at it carefully, to make sure I haven't missed anything (which I probably have... ). --Tryptofish (talk) 23:36, 23 April 2012 (UTC)
 * Good. So I think that draft 9 is it. (?) North8000 (talk) 20:48, 27 April 2012 (UTC)
 * I agree with that, subject to views from other editors, and the need for eyes other than mine (which have stared at it too long) to double-check it. --Tryptofish (talk) 19:59, 28 April 2012 (UTC)